Home

Understanding Society: Waves 1-4, User Manual

1. Non X X X employment Second jobs X X X Discrimination X EMB X EMB GPC LDA GPC LDA Childcare X X X Parents and X X responsible Children responsible mother and mother and responsible responsible father of father of children children Family X X Networks Remittances X Unearned X X X Income and State Benefits Household X X X Finances Politics X X X Harassment X EMB X EMB GPC GPC LDA LDA Environmental X Behaviour Consents for X linkage to health administrative records Consents for X linkage to education administrative records Consents for linkage to benefit administrative records Interviewer X X X observations Proxy X X X Childhood X language Ethnic identity X EMB GPC LDA General Health X X see Health and Disability module 33 Nutrition Physical Activity Smoking History Disability lt gt lt gt lt gt lt X X Annual Event History X ae interviewed before X interviewed before X a interviewed before Commuting Behaviour X X Physical Work Work Conditions Voluntary Work Charitable Giving Personal Pensions Savings lt X X XxX X lt lt X X X lt X lt Retirement Planning X age 45 50 55 60 65 and not retired X age 45 50 55 60 65 an
2. Wave 1 household design weight Wave 1 household design weight for GPS only Design weight Design weight GPS only Extra 5 minutes design weight BHPS inclusion weight for OSMs issued into UKHLS Analysis Weight a_hhdenus_xd a_hhdengp_xd a_psnenus_xd a_psnengp_xd a_ind5mus_ xd b_psnenbh_li BHPS 2010 longitudinal enumerated person weight b_psnenbh_lw BHPS GPS and EMB combination inclusion weight Typically for advanced users see technical details 56 b_psnenub_li 3 7 1 1 Not Using Weights Note that an unweighted analysis does not correctly reflect the population structure unless the assumptions below are true It is suggested that researchers publishing or presenting unweighted estimates make these assumptions explicit If no weighting is used an analysis of the UKHLS assumes that all estimates of interest are the same in Northern Ireland as in the rest of the UK that people who live at an address with more than three dwellings or more than three households are the same as those who don t that people who responded at Wave 1 are the same with respect to your estimates as those who did not that people who continued to respond at later waves are the same as those who did not and that people who responded to each particular instrument used in the analysis individual interview self completion questionnaire etc are the same
3. cccceeeeeeeeeeeeeeeeeeeeeeeeeeees 10 2 3 Data Collection and Response Outcomes cceeeeeeeeeeeeeeeeeeeeeeeeeeeenaaees 11 Bese le NOVGIVIGW Sct pisces el E T 11 2 3 2 Data Collection 2 00sec erent tee eo ete a 12 2 3 3 Panel Membership and Panel Maintenance eseeeeeeeeeeeeeees 15 2 3 4 Response Outcomes 2 c ccc hisecedSecehiehadendiindeelidadeanestadetdeledetideneeeiiee 16 2 4 Data Processing and Cleaning cceeeeceeecceeeeeeeeeeeeeeeenneeeeeeeeeeeeeeeseaaees 29 2A GOING eregi A nee S 29 2 5 Documentation of the Survey Instruments eeeeeeeeeeneeeeeeeeeeeeeeeeeeneees 30 2 5 1 Reading the Questionnaires sn kanccisiernlee einen eeiineaed 30 2 5 2 Summary of Questionnaire MOdules cccccceeeeeeeeeeeteeeeeeeeeeeeeees 32 2 5 3 Content Highlights by Waves 2 ccc sn a eg tee ee ee tee as 37 2 5 4 Changes to the Questionnaire 0 cceeeeeeccccee eee eeeeeeeeneeeeeeeeeeeeeeeees 39 2 5 5 Other Fieldwork Materials ccccceeeeeeseeccceeeeeeeeeeeeeeneeeeeeeeeeeeeeeees 39 3 Understanding Society Data 0 cccccceeccccecceeeeeeeeeeeneeeeeeeeeeeeeeetenaeeeeeeeeeeeeneees 40 3 1 Information About Data Files cece cccscccceeeeeeeeeeeeeeaeeeeeeeeeeeeeeeeenaeeees 40 Bll PARA OA cninn aree sass EEEE PACE ae TEE EE ECEE EACAN NEARER 41 3 2 Information About Variables nosesneeeeeeeeeeneennnreeseeerrnnnnnrnneserrrnnnnnneeeet 42 3 2
4. 3 2 3 VARIABLE VALUES AND LABELS The detailed variable view provides information about valid and invalid responses Additional codes denote different types of reasons for the lack of a valid response These values have not been specified as missing in Stata or SPSS However these statistical packages have commands to assign values to missing for many variables simultaneously Table 19 describes the missing value codes Table 19 Missing value codes Value Description 9 Missing by error or implausible 8 Not applicable to the person or because of routing 7 Proxy respondent The question was not asked of proxy respondents or derived variable cannot be computed for proxy respondents 2 Refused 1 Don t know The meaning of other values is explained with the variable s value labels There may also be notes in the detailed variable view of the online documentation system on the website https www understandingsociety ac uk documentation mainstage dataset documentation Note that the default missing value code for derived variables tends to be 9 missing or wild Missing value codes on the youth self completion questionnaire also tend to be less accurate because the instrument was administered as a paper and pencil questionnaire and processing was therefore not as closely monitored e g respondents may not have followed the question routing We recommend that users carefully read the questionnaires and compare
5. refers to child age 10 child age 10 Work conditions X Young adults X age 16 21 The content of the youth self completion instruments is summarized in the long term content plan https www understandingsociety ac uk system uploads assets 000 000 018 original Long Term Content Plan Nov2011 3 pdf 1355920157 A number of articles in the Understanding Society Findings publications focus on measures available see https www understandingsociety ac uk research publications findings earlyand https www understandingsociety ac uk research publications findings 2012 2 5 3 CONTENT HIGHLIGHTS BY WAVE 2 5 3 1 Wave 1 Wave 1 collected important baseline data Some Wave 1 measures are stable that is not time variant In subsequent waves we try to collect this type of information from individuals who are new entrants to the study see modules marked NE in Table 14 Notice that some modules are covered annually beginning in Wave 1 These represent the strongest areas for examining annual change See for example Disability Caring employment related information childcare politics and income and benefits These have been the focus of major longitudinal research in the BHPS and should also be a prominent focus with Understanding Society Wave 1 also saw the first rotating modules These included Parents and Children Family Networks and Environmental Behaviour Within the ethnicity strand that is within the Extra 5 Minutes qu
6. 91 BHPS original sample starting in 1991 England Scotland and Wales 01 BHPS sample starting in 2001 original sample Scotland and Wales boost NI ip Innovation Panel Options for weight type aa lw longitudinal analysis weight xw cross sectional analysis weight Id longitudinal design weight xd cross sectional design weight li longitudinal inclusion weight 3 7 2 1 Examples a_indinus_xw is the cross sectional analysis weight for individual interview data from Wave 1 representing the population of persons aged 16 or older b_indscus_Iw is the longitudinal analysis weight for individual self completion interviews from Wave 1 and Wave 2 representing the adult population who continuously lived in UK at the times of Wave 1 and 2 3 7 3 TECHNICAL DETAILS In this section we describe in turn how the weights were derived for e GPS and EMB Wave 1 weight e GPS and EMB longitudinal weights e GPS and EMB cross sectional weights after Wave 1 e BHPS longitudinal weights e BHPS cross sectional weights e Combined sample BHPS GPS and EMB longitudinal weights e Combined sample BHPS GPS and EMB cross sectional weights 3 7 3 1 Wave 1 GPS and EMB Weights The Wave 1 household level weights consist of two components a design weight and nonresponse adjustment for household level nonresponse Wave 1 individual 58 level weights consist of four components the design weight nonresponse adjustment for househol
7. XP ui Y A209 XB 914 uz Y3 A309 XP3 317 32Y gt U3 Yk ago XPy akiYi yaYo Akk 1Yk 1 Uk Here Y1 Y2 Yk are the income and auxiliary variables to be imputed ordered from the one with the smallest percentage of missing values Y1 to the one with the largest percentage of missing values Yk X is a set of auxiliary variables observed for all individuals a s and B s are parameters and u1 u2 uk are random errors Such a recursive system allows us to carry out the imputation separately for each variable and sequentially The sequential procedure is given by the following steps e estimation of the first equation and imputation of the missing values for Y1 e estimation of the second equation using the imputed values to replace the missing values of Y1 and imputation of Y2 e repetition of estimation and imputation steps sequentially for each of the following equations until when all k variables Y1 Y2 Yk have been imputed We use stochastic imputation that is we draw the imputed values from the posterior predictive distribution of the variable to be imputed conditional to the observed data For more details about stochastic imputation we refer to Rubin 1987 Schafer 1997 and Kenward and Carpenter 2007 This sequential estimation is consistent only if the recursive system is valid Since this is not necessarily a valid assumption ICE uses the imputed values produced using the abov
8. 83 8 80 6 82 3 85 3 78 8 Proxy interview 706 16 201 99 22 37 34 1 115 2 1 0 9 4 0 1 5 1 4 2 1 1 7 2 2 Other non interview 295 48 109 56 14 22 27 571 0 9 2 7 2 2 0 9 0 9 1 3 1 3 1 1 Refusal 287 30 77 47 10 17 15 483 0 9 1 7 1 5 0 7 0 7 1 0 0 7 1 0 Household non contact 832 44 212 158 57 51 34 1 388 2 5 2 5 4 2 2 4 3 7 2 9 1 7 2 7 Household refusal 2 919 108 418 312 99 82 59 3 997 8 9 6 0 8 2 4 8 6 4 4 7 2 9 7 8 Household other non 258 21 40 42 11 15 22 409 interview 0 8 1 2 0 8 0 7 0 7 0 9 1 1 0 8 Household untraced 1 285 52 352 230 52 54 78 2 103 3 9 2 9 6 9 3 6 3 4 3 1 3 8 4 1 Household ineligible 483 29 142 108 36 31 33 862 1 5 1 6 2 8 1 7 2 3 1 8 1 6 1 7 Total 32 918 1 788 5 070 6 476 1 550 1 743 2 049 51 594 25 2 3 4 4 Wave 4 The Wave 4 main survey fieldwork started on 8 January 2012 and ended on the 19 June 2013 including the re issue period Table 10 shows the household response rates for Wave 4 of the UKHLS The table separates the different samples as in Table 7 above As before ineligible households have been removed from the table these would include households where all sample members had died consist of only TSM individuals emigrated from the UK or which have merged with a previous wave household for example an adult moving back to live with his or her parents who are also part of the sample The fully responding household response
9. During the breaks in day there were stalls and displays of media coverage research findings and information about the study a Twitter stand and an area where interviewers could write questions on post it notes for discussion later in the day The content of the briefing consisted of plenary sessions where an overview of progress on the study so far was presented along with researchers from ISER or the LSE talking about how they used Understanding Society in their research videos of the Chief Executive of NatCen Penny Young and the ISER Director Professor Heather Laurie MBE were shown and medals awarded to interviewers who had achieved 100 response rate in any of their allocations at Wave 3 Once during the morning and once in the afternoon there were a number of break out sessions with small groups of interviewers to share best practice and experience of i contact and co operation and ii how to deal with household splits and allocating outcome codes The discussions of these break out sessions were then discussed at the plenary sessions Interviewers were assigned to specific areas For Wave 1 911 interviewers were employed to cover 3 517 areas in the sample The number of interviewers briefed in Wave 2 was 819 and 746 at Wave 3 At Wave 4 692 interviewers worked on the study 2 3 2 4 Fieldwork When beginning fieldwork for Wave 1 we did not know who was in the sample Interviewers mailed an introductory card f
10. Michaela Benzeval Nick Buck Jon Burton Paul Fisher Laura Fumagalli Olena Kaminska Gundi Knies Peter Lynn Stephanie McFall Alita Nandi and Jakob Petersen Many people participated in preparing and processing the questionnaires and data From the information technology side we recognize the contributions of Geoffrey Angel Tom Butler Jeannette Chin Paul Groves Elaine Prentice Lane Paul Siddall and Catherine Yuen From the survey research team we recognize Noah Uhrig Sarah Budd Emily Dix and Deborah Wiltshire A small group was active in contributing code for derived variables and flagging issues in using the data They include Gundi Knies Jakob Petersen Alita Nandi Cara Booker Mark Bryan Alexandra Skew and Mark Taylor 87 6 REFERENCES Berthoud R L Fumagalli et al 2009 Design of the Understanding Society Ethnic Minority Boost Sample Understanding Society Working Paper 2009 02 Burton J and Ed 2012 Understanding Society Innovation Panel Wave 4 Results from Methodological Experiments Understanding Society Working Paper 2012 06 Hayes C and H Watson 2009 HILDA Imputation Methods HILDA Project Technical Paper Series University of Melbourne No 02 9 Kenward M and J Carpenter 2007 Multiple imputation current perspectives Statistical Methods in Medical Research 16 3 199 218 Knies G and S Menon 2014 Understanding Society Waves 1 3 2009 2012 Special Licence Access Geo
11. Q1 Q2 Wave 2 Wave 3 Q3 year 2 year 1 Q4 2011 Q1 Q2 Wave 3 Q3 year 2 Q4 Q1 Q2 Q3 Q4 BHPS becomes sample component in Wave 2 year 1 2012 2013 2 3 2 DATA COLLECTION 2 3 2 1 The players who does what ISER together with NatCen and the Central Survey Unit of NISRA work closely together on all aspects of data collection implementing an agreed set of survey procedures designed to ensure adequate response and effective data quality ISER has the primary responsibility for design work NatCen manages fieldwork editing and coding and data entry It also advises on the design of all research instruments NISRA collaborates with NatCen and is responsible for fieldwork in Northern Ireland ISER plays a major role in quality control through specification of fieldwork practices survey materials editing and coding requirements and monitoring and analysing weekly fieldwork progress reports This working relationship is reinforced by an agreed set of survey specific procedures to ensure adequate response and effective data quality Full details of these and other technical aspects of the data collection and fieldwork coding and data processing are found in the Technical Reports published each wave on the Understanding 12 Society website see https www understandingsociety ac uk documentation mainstage technical reports 2 3 2 2 Getting Ready for Fieldwor
12. the country and year of the respondents selection the respondent s report on the most recent change of address for those not born in Britain the year of their arrival to Britain the country and year of school or university studies For adults whose residence remains unknown after using the above information it is inferred from other household members if it is consistent Note that for those who were born after the sample was selected we make use of residency information from their mother In this way we obtain residency information with 5 categories England Wales Scotland Northern Ireland or abroad for 1991 1999 2001 and 2009 2010 time points for each single respondent Total probability The important point is that an estimate of each of the following probabilities is available for each person regardless of when and where the person was selected Pix if resident in England in 1991 Pix if resident in Scotland in 1991 Pix if resident in Wales in 1991 Pox if resident in Scotland in 1999 Pox if resident in Wales in 1999 P3x if resident in Northern Ireland in 2001 Pax if resident in England in 2009 10 Pax If resident in Scotland in 2009 10 Pax If resident in Wales in 2009 10 Pax if resident in Northern Ireland in 2009 Psr If resident in England in 2009 10 Psg if resident in Scotland in 2009 10 Psg if resident in Wales in 2009 10 For each time point 1991 1999 2001 and 2009 2010 the probability is no
13. 1 4 8 4 1 Household non contact 3 338 125 1 156 555 210 155 111 5 650 6 3 4 4 10 8 5 6 8 3 5 6 4 0 6 7 Household refusal 7 229 350 1 743 1 493 400 427 203 11 845 13 6 12 3 16 2 15 0 15 9 15 4 7 2 14 0 Household other non 118 5 60 29 9 6 4 231 interview 0 2 0 2 0 6 0 3 0 4 0 2 0 1 0 3 Household untraced 4 178 159 1 207 754 172 208 178 6 856 7 9 5 6 11 2 7 6 6 8 7 5 6 4 8 1 Total 53 254 2 840 10 742 9 967 2 517 2 769 2 802 84 891 20 Table 6 Longitudinal individual re interview rates adults by sample origin Full interview at Wave 1 UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain Scotland Wales BURRS Joa Full interview 29 646 1 640 4 200 5 633 1 335 1 507 1 875 45 836 74 3 81 0 62 2 69 4 67 8 67 6 83 3 72 4 Proxy interview 775 16 188 97 17 38 15 1 146 1 9 0 8 2 8 1 2 0 9 1 7 0 7 1 8 Telephone interview 184 59 57 300 7 lt 2 3 3 0 2 6 0 5 Other non interview 334 22 157 73 16 28 32 662 0 8 1 1 2 3 0 9 0 8 1 3 1 4 1 1 Refusal 316 25 94 53 13 15 11 527 0 8 1 2 1 4 0 7 0 7 0 7 0 5 0 8 Household non contact 1 890 68 500 376 126 96 34 3 092 4 7 3 4 7 4 4 6 6 4 4 3 1 5 4 9 Household refusal 4 144 167 734 965 245 260 109 6 633 10 4 8 3 10 9 11 9 12 4 11 7 4 8 10 5 Household other non 65 2 31 18 3 2 2 123 interview 0 2 0 1 0 5 0
14. 2 0 2 0 1 0 1 0 2 Household untraced 2 252 63 639 439 105 140 105 3 744 5 6 3 1 9 5 5 4 5 3 6 3 4 7 5 9 Household ineligible 507 22 208 280 51 86 68 1 222 1 3 1 1 3 1 3 5 2 6 3 9 3 0 1 9 Total 39 929 2 025 6 751 8 118 1 970 2 229 2 251 63 285 21 2 3 4 3 Wave 3 The Wave 3 main survey fieldwork started on 7 January 2011 and ended on the 12 July 2013 including the re issue period Table 7 shows the household response rates for Wave 3 of the UKHLS The table separates the different samples as in Table 4 above As before ineligible households have been removed from the table these would include households where all sample members had died consist of only TSM individuals emigrated from the UK or which have merged with a previous wave household for example an adult moving back to live with his or her parents who are also part of the sample Household response rates including partial household response were higher in Northern Ireland than in the rest of the UK as at Wave 2 The household response rate for the continuing Understanding Society GPS was 75 3 in Great Britain and 79 1 in Northern Ireland The household response rates for the former BHPS samples were higher than the Understanding Society GPS both in terms of overall household rates and fully responding households although there is only a slight difference between the overall household rate for Living in Scotland and the British GP
15. 2001 are used for creating weights for longitudinal analysis starting in 2001 Analysis using these weights will include all the BHPS samples For more information on the BHPS weight calculation please refer to BHPS documentation Taylor 2010 For each of the Wave 18 weights an additional adjustment is applied to correct for attrition between Wave 18 of the BHPS and Wave 2 of Understanding Society when the BHPS joined Understanding Society The adjustment is the reverse of the estimated probabilities of participation enumeration or response to main questionnaire based on logistic regressions predicting participation at Wave 2 of UKHLS conditional on participation at Wave 18 of BHPS The covariates used in the model predicting enumeration are from the BHPS Wave 18 household grid and household questionnaire The same covariates plus covariates from the Wave 18 main questionnaire are used for predicting response to the UKHLS Wave 2 main questionnaire Enumeration weights for newborn babies biological step or natural born to an OSM mother since the time of the BHPS Wave 18 interview are equal to their mother s enumeration weight For rising 16 year olds OSMs who turned 16 between the time of the BHPS Wave 18 interview and the UKHLS Wave 2 interview and who could therefore be aged 16 17 or even 18 at the time of UKHLS Wave 2 main response weights consist of the relevant longitudinal enumerated person weight with an adjustment for the p
16. For example at Wave 1 we have 59 466 individuals for whom at least the household questionnaire is available and among these individuals 80 3 provided a personal interview 5 5 have a proxy interview whereas 14 2 had neither a proxy nor a personal interview The item nonresponse rate for individuals who provided an individual questionnaire varies across income variables It goes from a maximum of about 50 for self employment earnings to zero for some of the benefit variables and it is generally below 20 for the remaining income variables 3 8 1 WHAT DO WE IMPUTE In Understanding Society we do not impute income variables for non responding households Responding households are households for which the household questionnaire and information on the household composition structure household grid module are available We suggest that the user take account of household nonresponse via weighted estimates described in Section 2 Weighting adjustments For individuals who respond to the individual questionnaire but do not provide answers to all income questions item nonresponse we impute the following personal income variables wages self employment earnings second job earnings interests and dividends pensions benefits and other income sources For individuals for whom a proxy questionnaire is available we impute total earnings and total income whenever missing The proxy questionnaire is a short version of the individual questionnai
17. The first included common predictors for England and Wales and eligibility was predicted for these two countries The second was based on England Wales and Scotland using a more limited number of predictors Eligibility for Scotland was predicted only from this model Following this the probability of responding was estimated using backward stepwise logistic regression weighted by eligibility status where the ineligible were excluded those known to be eligible had an eligibility of one and those with unknown eligibility had a weight proportional to the predicted probability of being eligible obtained from the above model The predictors used in this model were the same as for the eligibility model and are described in detail below Given that administrative neighbourhood data differs between England Wales Scotland and Northern Ireland a separate model was implemented for each country GPS and EMB response propensity was modelled together which allowed us to model nonresponse within each country separately but the indicator of EMB was retained in the model even if it was not statistically significant Predictors used for eligibility model and household level nonresponse correction come from the following sources e Sampling frame information including such variables as sample month and geographical region e Predicted ethnic density of the postcode sector for five main ethnic groups in England Scotland and Wales as described in Berth
18. UKDS we encourage users to consult the Understanding Society webpage The documentation will develop over time We have developed specific guides about major content areas such as the biomeasures or cognitive measures and guides for issues that are frequently problematic for users such as selection of appropriate weights We will continue to develop specific user guides over time 4 1 RELEASE VERSIONS 4 1 1 END USER LICENCE EUL DATA Most of the Wave 1 Wave 4 data has been released according to the conditions of the regular UK Data Services End User License EUL http ukdataservice ac uk get data how to access conditions aspx tab end user licence The data are listed as SN6614 Understanding Society Waves 1 4 2009 2013 In 2010 2012 Understanding Society augmented survey questions with direct health assessments and the collection of blood samples Data from this collection are released under the standard EUL through the UKDS SN7251 Understanding Society Waves 2 3 Nurse Health Assessment 2010 2012 4 1 2 SPECIAL LICENCE SL DATA A number of sensitive data are released under Special Licence SL Researchers can apply for access to SL data through a UKDS application procedure where they are required to justify their research objectives and explain why EUL data alone would be inadequate to reach those objectives Researchers are asked to report publications resulting from using the data The conditions for using SL dat
19. and EMB for Wave 2 The cross sectional enumerated individual weights are based on the longitudinal enumerated individual weights which are allocated through a weight share method to temporary sample members TSMs and permanent sample members PSMs who entered the sample at Wave 2 Note that only new TSMs and PSMs entering the study after Wave 1 receive a shared weight TSMs who were present in Wave 1 in the EMB sample are given a cross sectional weight of 0 This is done as the GPS part of the sample does not have an equivalent TSM group OSM non ethnic minority members living with TSM ethnic minority members Giving a cross sectional weight of 0 to Wave 1 TSMs maintains the balance of the whole sample These cross sectional enumerated individual weights then serve as the base for the other cross sectional individual level weights each of which main main or proxy self completion youth involves an additional adjustment for nonresponse to the relevant instrument conditional on enumeration The nonresponse models are therefore based on all eligible persons enumerated at Wave 2 including TSMs and those OSMs who did not respond to the respective instrument at Wave 1 with covariates taken from responses to the UKHLS Wave 2 household grid and household questionnaire The cross sectional weights for households b_hhdenus_xw are set equal to the minimum nonzero longitudinal enumerated person weight b_psnenus_Iw amongst adults in the house
20. as those who did not see Lynn Burton et al 2012 An unweighted analysis of the former BHPS sample assumes that all estimates of interest are the same in each of England Scotland Wales and Northern Ireland that people who live at an address with more than three dwellings or more than three households are the same as those who don t that people who responded at Wave 2 of UKHLS in 2010 are the same with respect to your estimates as those who may have become non respondents at any time since Wave 1 of BHPS in 1991 that people who keep responding in later waves of UKHLS are the same as those who stopped responding at any point of time between 1991 and the last year in your analysis We therefore strongly suggest conducting weighted analyses of the UKHLS data 3 7 2 NAMING CONVENTIONS FOR WEIGHTING VARIABLES The naming conventions will help users to select the weight they need or to interpret the purpose of some weight variables The structure is as follows W_XXxyyzz_aa where w wave xxx target population yy instrument zz sample aa weight type options for xxx hhd household psn persons 0 ind persons 16 yth persons 10 15 57 options for yy en enumeration grid in interview px interview or proxy 5m extra 5 minutes items sc self completion ns nurse visit bd blood options for zz us the GPS and EMB samples bh BHPS sample ub BHPS GPS and EMB samples gp GPS sample
21. by _dv or _ if suffixes However most questionnaire variables which are carried in both surveys will have the same main variable name though with a different wave prefix Since the last wave of BHPS was Wave 18 the wave prefix is R Thus if we wished to match Wave 2 work status b_jbstat on the file b_indresp to previous wave values for the GPS sample we would match using pidp to a_indresp and use the variable a_jbstat while for the BHPS sample we would match using pid to rindresp and use the variable rjbstat 3 4 NOTES ON USING THE EXTRA 5 MINUTES QUESTIONS An Extra 5 minutes of question time was set aside for questions of particular interest to ethnicity related research e g ethnic identity and remittances To provide sufficient sample sizes to allow analysis of these questions separately by ethnic minority groups the questions were asked of the EMB sample see Section 2 2 3 the GPC sample see Section 2 2 2 and ethnic minority individuals living at Wave 1 in an LDA i e members of ethnic minority groups who had a selection probability of 0 to be in the EMB sample Those eligible for the Extra 5 minutes can be identified using w_xtra5min_dv They also received the standard questionnaire that the rest of the sample received The flag w_xtrajminosm_dv identifies OSMs who are eligible for these questions In other words w_xtra5min_dv is 1 and w_xtra5minosm_dv is 0 for TSMs who have joined the househol
22. component and they should be included in analyses of the GPS component 2 2 3 ETHNIC MINORITY BOOST SAMPLE The EMB sample was designed to provide at least 1 000 adults from each of five groups Indian Pakistani Bangladeshi Caribbean and African The initial step was identifying postal sectors with relatively high proportions of relevant ethnic minority groups based upon 2001 Census data and more recent Annual Population Survey data The set of 3 145 sectors constituted approximately 35 of the sectors in Great Britain and covered between 82 and 93 of the population of the five ethnic minority groups The 3 145 sectors were sorted into four strata based on the expected number of ethnic minority households that would be identified by the sampling and screening procedures see Berthoud et al 2009 for details All sectors were included for the stratum where a yield of three or more households was expected In the other three strata sectors were sub sampled at rates of 1 in 4 1 in 8 or 1 in 16 respectively This was done to constrain the number of sectors that might have just one or two eligible sample households or even none The total number of postal sectors selected for inclusion in the EMB sample was 771 Of these 6 were in Scotland 7 were in Wales and the remaining 758 were in England with a concentration in London 412 sectors The number of addresses selected per postal sector ranged from 15 to 103 Sampling fractions varie
23. do the following for more waves add wave specific prefix in the foreach statement Figure 8 Stata Code Merging individual files across waves into wide format use pidp a_johas using a_indresp_ip clear sort pidp ave temp replace oreach win b use pidp w_jbhas using w _indresp_ip clear sort pidp merge 1 1 pidp using temp drop _merge sort pidp save temp replace Save final6 replace erase temp dta 3 10 PRESERVING CONFIDENTIALITY In preparing the data for the general release we have taken steps to maintain the confidentiality of responses These include not releasing the full date of birth and not releasing the most detailed job related SOC and SIC codes Household income has been top coded Open or narrative text e g names of schools or employers has not 83 been released since it may indirectly identify individuals Geographical identifiers below the level of GORs are also not included in the general release Analysts may apply to gain access to restricted resources see Section 4 below The study has a Data Access Committee to take decisions on applications requesting access to restricted electronic data and biological samples from Understanding Society Its aim is to allow important research to proceed while minimising risks particularly to study participants 4 DATA ACCESS The data are released through the UK Data Service UKDS in SPSS and Stata formats While documentation is released through the
24. grid from wave n 1 Newborns are given the enumeration weight of their mother The enumeration weights are then scaled to a mean of one For longitudinal main interview weights n_indin91_lw and n_indinO1_Iw response to the main interview is predicted using logistic regression with predictors obtained from the wave n 1 household questionnaire household grid and main questionnaire The probability is then inverted The response of rising 16 year olds is predicted in a separate logistic regression in which response to the main questionnaire for all adults 16 and over is estimated using predictors from the wave n household questionnaire and household grid and conditional on enumeration in current wave The response probabilities are then inverted and multiplied by the base weights n 7_indin91_Iw or n 7_indin0O1_Iw for adults and n_psnen91_Iw or n_psnen01_Iw for rising 16 The weighted ratio of rising 16 to others is scaled to reflect the ratio of these age groups as estimated using the longitudinal enumeration weight in wave n 3 7 3 5 BHPS Cross sectional Weights for Wave 2 The BHPS cross sectional weights are created as follows we first model the chance of each BHPS OSM being issued into the UKHLS reflected in b_psnenbh_li then the chance of being in a responding household complete the household grid and the household questionnaire at Wave 2 of the UKHLS conditional on being issued reflected in b_psnenbh_Iw The weight b_psnenbh_Iw is t
25. missing value distributions across waves before using the substantive information contained in them 3 2 4 IDENTIFIERS AND OTHER USEFUL VARIABLES Households are identified by w_hidp a wave specific variable with a different prefix for each wave It can be used to link information about a household from different records within a wave but cannot be used to link information across waves Since the composition of households can change between waves the data do not include a longitudinal household identifier Individuals are identified by the personal identifier pidp which is consistent in all waves and can be used to link information about a person from different records belonging to one wave or to link information from different waves Individuals are 43 also identified by w_pno the person number within the household The combination of w_hidp and w_pno is unique for each individual Table 20 lists some variables commonly used in analysis and may help the analyst to begin planning Recall that the variables with the prefix w_ have the values for that wave There is also the file xwavedat which has variables with stable values Variables in that file do not have the wave prefix Analysts should also remember to consult the section on specifying the complex sampling variables from Section 3 6 and on weighting from Section 3 7 Table 20 Some useful variables Variable name Description
26. record contact information and how to deal with the complexities of multiple dwelling units and multiple households The afternoon was spent discussing the survey content and reviewing and working with the Blaise computer aided personal interview CAPI instrument At Wave 3 there were two types of briefing for interviewers experienced with the study or for interviewers with experience who were new to the study The latter briefing went into more detail about the background of Understanding Society early findings the more technical details of the sample and the task of enumerating the household At Wave 4 the style of briefing changed in Great Britain Interviewers had worked on the study for three waves and were familiar with the mechanics of how to conduct the survey For those interviewers returning for Wave 4 the focus of the briefing switched from the survey procedures to motivating the interviewers and giving them information to enable them to motivate the sample members when making contact Interviewers who were new to the survey still attended a standard briefing as did interviewers in Northern Ireland These standard briefings were held in Belfast 3 and London 1 Experienced interviewers attended conference style briefings These briefings were much larger than standard briefings with 150 250 interviewers attending each event There were three such events held prior to the start of Wave 4 in Birmingham Liverpool and London
27. relative to the average person selected at the same time in the same country This is the real unique component available for the time points when the person was actually selected To obtain tv for the time points when a person could have been selected but wasn t we used the following procedure For each country and time point we ran a 69 multiple stepwise regression with v as the dependent variable and covariates from the household questionnaire and household grid from Wave 2 of the UKHLS as explanatory variables Note that identical predictors are available for all samples of interest We can therefore use the regression model to infer vj for sample members selected through another sample Residency In order to assign a correct probability for each person we need to know where they lived in 1991 England Scotland Wales or other 1999 Scotland Wales or other 2001 Northern Ireland or other and 2009 2010 postcode sector if England Scotland and Wales Northern Ireland other Because no interviewing was conducted among BHPS members in 2009 we treat their address at the time of Wave 2 of the UKHLS 2010 as if it were their address at the time of selection of GPS and EMB We do not have perfect information for each enumerated respondent on their residency in the four years of interest but we use all the available information to us to infer residency as closely as possible Specifically the following information is used
28. the different samples The GPS consists of respondents in Great Britain and Northern Ireland The EMB households are only located in Great Britain The former BHPS sample consists of the Living in Britain sample started in 1991 the Living in Scotland and Living in Wales boost samples started in 1999 and the Northern Ireland Household Panel Survey NIHPS started in 2001 also a boost sample Ineligible households have been removed from the table these would include households where all sample members had died consist of only TSM individuals or emigrated from the UK For the former BHPS component ineligible households would also include households which have merged with a previous wave household for example an adult moving back to live with his or her parents who are also part of the sample Fully responding households are those in which the household is successfully enumerated the household questionnaire is completed and all eligible adults give an individual interview Partially responding households are those where the household is enumerated and a household questionnaire is done and at least one eligible adult but not all eligible adults complete an individual interview Household response rates were higher in Northern Ireland than in the rest of the UK The household response rate for the continuing Understanding Society GPS was 76 8 in Great Britain and 81 9 in Northern Ireland The household response rates for the former BHPS
29. the registration form provided at the top right of the following page https www understandingsociety ac uk support projects support The Understanding Society website has a Frequently Asked Questions FAQ https Awww understandingsociety ac uk documentation fag The User Support forum also has a FAQ about data related questions https Awww understandingsociety ac uk support projects support wiki 85 4 3 LINKS TO OTHER STUDIES IN THE UKHLS FAMILY 4 3 1 THE BRITISH HOUSEHOLD PANEL SURVEY Data from the BHPS prior to joining the UKHLS can be obtained from the UK Data Service SN5151 British Household Panel Survey Waves 1 18 1991 2009 http www esds ac uk findingData snDescription asp sn 5151 The study documentation is available at http www iser essex ac uk bhps 4 3 2 THE WAVES 2 3 NURSE HEALTH ASSESSMENT In 2010 2012 Understanding Society augmented survey questions with direct health assessments and the collection of blood samples see McFall Petersen et al 2014 Data from the Wave 2 and Wave 3 health assessment are released through the UK Data Service SN7251 Understanding Society Waves 2 3 Nurse Health Assessment 2010 2012 http discover ukdataservice ac uk catalogue sn 7251 4 3 3 THE UKHLS INNOVATION PANEL The Understanding Society project incorporates the UKHLS Innovation Panel IP a separate survey intended to support methodological research https www understandingsociety ac uk about innovatio
30. the time of selection of sample j This is in effect to the inverse of the average individual attrition rate jx S the individual specific variation from the mean continuous enumeration probability This is explained in detail below For the sample in which the individual was selected we know the selection probability and modelled response probabilities these can be obtained from the available weights and the method is described below These probabilities will be called real For the samples that the individual could have been selected through but wasn t we can infer each of the components Such probabilities are called inferred The inference method is described below Selection probability s j For j 1 2 3 sj for respondents known or inferred to have been resident in relevant country at the time of selection is equal to the number of eligible households divided by the number of residential households in the population at that time Such probabilities are calculated separately for each country England Wales Scotland and Northern Ireland The population information was obtained from Census 1991 for the 1991 sample Office for National Statistics 1991 and from Census 2001 for samples selected in 1999 and 2001 Office for National Statistics 2003 For respondents known or inferred not to have been resident in relevant country at the time of selection sj 0 For respondents actually selected into the GPS or EMB sample the rea
31. 0 is not available for the current and last job Information about how the derived variable is produced is shown in the notes for derived variables in the detailed variable view of the online documentation The view provides descriptive statistics and in the Origin field lists of the variables used in the computation of the derived variable For variables that were computed during the interview additional information is available in the questionnaires Analysts should 47 consult the description of any derived variables that they plan to use in their analysis 3 6 SAMPLE DESIGN VARIABLES AND ANALYSIS As the sample design involves stratification clustering and weighting these design features affect standard errors and should therefore be taken into account in analysis Appropriate variables are provided to allow the analyst to do this The weighting variables are described in Section 3 6 Here we describe the stratification and clustering variables 3 6 1 W_psu PRIMARY SAMPLING UNITS PSUs This is an indicator of the primary sampling unit PSU to which the sample member belongs The prefix w_ denotes waves in general The value of w_psu does not change between waves but for new sample entrants it is only defined from the wave at which they enter the sample Values on the variable w_psu are further described in Table 21 Table 21 Description of the UKHLS Primary Sampling Unit variable w_psu Value Sample Not
32. 1 3320 UKHLS GPS in Corresponds to groups of two or more England Scotland and PSUs in selection order as they were Wales selected systematically from an implicitly ordered list see Lynn 2009 3321 UKHLS GPS in Northern Ireland treated as a single stratum Northern Ireland 3322 5117 UKHLS EMB Corresponds to the postal sectors in the high minority density domain as selections were made independently from each see Berthoud Fumagalli et al 2009 3 6 3 ANALYSIS EXAMPLE USING STATA In Stata to obtain estimates that correctly take into account the sample design the user need only specify the design variables using the svyset command for example svyset a_psu pweight a_indpxus_xw strata a_strata Then any compatible command simply needs to be prefixed with svy for example svy logistic depvar variable1 variable2 variable3 3 7 WEIGHTING ADJUSTMENTS FOR THE WAVE 4 RELEASE A number of weights are provided for data users They adjust for unequal selection probabilities differential nonresponse and potential sampling error A weighted analysis will adjust for the higher sampling fraction in Northern Ireland and for different probabilities of selection in the EMB sample as well as for response rate differences between subgroups of the sample Separate sets of weights are provided for a the combined GPS and EMB sample components b the former BHPS sample and c the combined GPS EMBS and BHPS components The available
33. 1 Learning About the Study Variables c ccccceeeeeeeeeeeeeeeeeeeeeeeeeeeees 42 3 2 2 Variable Naming and Labelling Conventions eeeeeeteeeeeees 42 3 2 3 Variable Values and Labels cccceeeseceeeeceeeeeeeeeeeeeeneeeeeeeeeeeeeeeeas 43 3 2 4 Identifiers and Other Useful Variables c ccccccccecceeececcecceeeeeeeaes 43 3 3 Notes on Using the BHPS Sampple cccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 45 3 4 Notes on Using the Extra 5 minutes questions ccceeeeeeeeeeeeeeeteees 46 3 9 Derved VanlaDles e er aee e naa ea Ee Eke a E AE AE EEEE EE AAE EAA EEEE ENEAN A CERES 46 3 6 Sample Design Variables and Analysis ccccceeeeeeeeeeeeeeeeeeeeeeeeeeeeaeees 48 3 6 1 Primary Sampling Units PSUS ociecccereteiiersicctieitcors eitenieledicertcesieeeaace 48 3 62 pole seco ht tee tA et eee Ree 49 3 6 3 Analysis Example Using Stata cceeececeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 49 3 7 Weighting Adjustments for the Wave 4 Release ccccceeeeeeeeeeeeeeeeeees 49 3 7 1 Selecting the Correct Weight for Your Analysis cceeeeeeeeeee 50 3 7 2 Naming Conventions for Weighting Variables cceceeeeeeteee 57 S732 Technical DetallS ix gecce cco ccever aaea aae aa EEA Ea aia 58 3 8 Imputation of Income Variables ccceeeccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeaeeees 72 3 8 1 What Do W
34. 164 15 57 32 30 7 21 326 interview 0 6 1 0 1 3 0 5 2 2 0 4 1 1 0 7 Household untraced 857 40 167 221 50 88 41 1 464 2 9 2 6 3 7 3 7 3 6 5 4 2 2 3 1 Household ineligible 315 23 58 69 19 17 32 533 1 1 1 5 1 3 1 2 1 4 1 0 1 7 1 1 Total 29 532 1 568 6 016 1 391 1 644 1 876 4 527 46 554 28 2 4 DATA PROCESSING AND CLEANING The data for a sample month is delivered to ISER in batches scheduled for 4 months following the beginning of the fieldwork process This interval allows time for interview re issue coding and data entry from paper documents e g the self completion instruments Data is delivered as SPSS system files which are then exported to triple S data exchange format and imported into a SIR database Quality control processes include extensive data checking to ensure that the data conform to the expected structure and to the routing and range constraints defined by the questionnaire specifications Data anomalies are investigated to determine whether they are related to e the invalid specification of the questionnaire e the incorrect scripting of the questionnaire e a failure to specify that a particular constraint should be included in the questionnaire e anincorrect implementation of the check or e aproblem in exporting and or delivering the data After investigation steps may include correcting the specification data editing reporting the error to NatCen to be fixed in
35. 43 50 56 56 1 799 2 6 3 3 4 3 1 9 2 7 2 6 2 4 2 8 Total 41 252 2 208 8 143 7 559 1 828 2 153 2 362 65 505 2 7 Table 12 shows the individual re interview rates by sample The re interview rate at Wave 4 is higher than it was at Wave 3 82 7 of those who gave a full interview at Wave 3 also gave one at Wave 4 compared to a 78 8 re interview rate between Waves 2 and 3 In the UKHLS GPS the re interview rate is higher for the British sample than the Northern Ireland sample The opposite is the case for the former BHPS samples the Northern Ireland sample has the highest re interview rate Table 12 Longitudinal individual re interview rates adults by sample origin Full interview at Wave 3 UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain d Wales MIRES Total Full interview 24 705 1 246 3 375 5 070 1 095 1 351 1 641 38 483 83 7 79 5 74 6 84 3 78 7 82 2 87 5 82 7 Proxy interview 609 10 137 101 26 30 7 920 2 1 0 6 3 0 1 7 1 9 1 8 0 4 2 0 Other non interview 198 25 86 30 18 11 24 392 0 7 1 6 1 9 0 5 1 3 0 7 1 3 0 8 Refusal 183 18 62 29 12 8 12 324 0 6 1 2 1 4 0 5 0 9 0 5 0 6 0 7 Household non contact 663 60 201 79 25 26 41 1 095 2 3 3 8 4 4 1 3 1 8 1 6 2 2 2 4 Household refusal 1 838 131 384 385 116 106 57 3 017 6 2 8 4 8 5 6 4 8 3 6 5 3 0 6 5 Household other non
36. 8 Wave 3 cross sectional individual adult response rates by sample origin UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain Seatiand Wales es Total Full interview 29 135 1 550 4 432 5 959 1 377 1 604 1 846 45 903 61 3 59 5 47 8 70 4 66 0 67 6 72 5 61 3 Proxy interview 2 516 47 667 333 65 107 70 3 805 5 3 1 8 7 2 3 9 3 1 4 5 2 8 5 1 Other non interview 1 039 150 426 144 54 60 107 1 980 2 2 5 8 4 6 1 7 2 6 2 5 4 2 2 6 Refusal 2 051 207 490 322 68 111 142 3 391 4 3 7 9 5 3 3 8 3 3 4 7 5 6 4 5 Household non contact 2 274 253 834 418 176 135 134 4 224 4 8 9 7 9 0 4 9 8 4 5 7 5 3 5 6 Household refusal 7 826 295 1 611 1 005 260 255 149 11 401 16 5 11 3 17 4 11 9 12 4 10 7 5 9 15 2 Household other non 414 34 107 67 15 22 25 684 interview 0 9 1 3 1 2 0 8 0 7 0 9 1 0 0 9 Household untraced 2 262 71 698 219 71 80 73 3 474 4 8 2 7 7 5 2 6 3 4 3 4 2 9 4 6 Total 47 517 2 607 9 265 8 467 2 086 2 374 2 546 74 862 24 Table 9 Longitudinal individual re interview rates adults by sample origin Full interview at Wave 2 UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain Scotland Wales MAPS TOA Full interview 25 853 1 440 3 519 5 424 1 249 1 434 1 747 40 666 78 5 80 5 69 4
37. C 2000 1 digit number employed at the current job workplace for employees number of employees if self employed whether is self employed and hires employees whether the employment organization is private or not only for employees type of ownership if self employed sole ownership or partnership an indicator for whether annual business accounts are prepared for the Inland Revenue for tax purposes if self employed e household variables reflecting economic situation log amount spent on food from food shops in four weeks prior to interview log amount spent on 77 food eaten outside the home in four weeks prior to interview log last year expenditure on domestic fuel e g electricity and gas number of bedrooms in the house number of other rooms in the house Council Tax band e Government Office Regions GOR The imputation of the income sources in the income file pensions and benefits is performed in a second model where each income source is imputed using as predictors the other income sources of the income file a set of demographics age age squared number of children number of children squared sex ethnicity marital status GOR the income sources imputed in the previous stage earnings for first and second job and investment income and information on benefits and pensions in the previous year total value and total number of benefits All variables are imputed as reported except for wages and self employment i
38. Given the complexity and multi purpose nature of the UKHLS design we provide multiple weights to meet the different needs of users The weight for your analysis reflects the survey instrument which is the source of the data being used in the analysis the analysis level household or individual and the combination of waves involved Each weight has been scaled to have a mean of one amongst cases eligible to receive the weight The naming conventions for weights are intended to help users to pick the correct weight The name of each weight reflects the wave for which the weight is calculated level of analysis data source and its nature design weight cross sectional analysis weight or longitudinal analysis weight The rules are described in the Naming Conventions for Weighting Variables section below If your analysis uses only data from Wave 4 select the xw cross sectional version of the weight This weight is defined for all sample members who responded to the relevant survey instrument at Wave 4 If your analysis uses data from multiple waves select an appropriate Iw longitudinal version of the weight For individual level analysis you may want to combine information from different questionnaire sources In this situation please select the weight suitable for the lowest level according to the hierarchy below Table 23 Selecting the correct weight Hierarchy of analysis levels Level of Analysis Questions availab
39. Interviewers returned to the same sample that was interviewed at the Wave 1 and 2 pilots The Wave 4 dress rehearsal pilot took place September November 2011 Interviewers returned to the same sample that was interviewed at the previous waves pilots 2 3 2 3 Interviewers We have tried to use interviewers of above average levels of experience and ability because of the demanding nature of Understanding Society The majority of interviewers in Northern Ireland had worked on the BHPS Northern Ireland component the Northern Ireland Household Panel Survey and were familiar with the design and operation of Understanding Society In addition to general interviewer training interviewers working on the study attended a one day survey specific briefing Generally around 12 20 interviewers attended each briefing along with two or three briefing managers or area managers The briefings were led by at least one researcher from NatCen with the majority also attended by ISER staff The briefings in Wave 1 took place across the UK Belfast Birmingham Brentwood Bristol Derby Edinburgh Glasgow Leeds London and Manchester Similar topics and locations were used for the Wave 2 briefings At Wave 3 the Edinburgh briefing was dropped and two briefings were held in Glasgow Additional briefings were added in Bury St Edmonds Liverpool and Gateshead The morning sessions were devoted to fieldwork procedures for example the administrative forms to
40. KHLS household The division by two reflects the idea that these newborns had double the chance of becoming BHPS OSMs relative to people born to both OSM parents as they would have been included had either their mother s or father s 1991 household been sampled For newborns observed with a single parent ina household in the first wave after their birth the weight given was the parent s weight This reflects a close to zero likelihood for the baby to be sampled via the other parent The adjustment for household nonresponse at UKHLS Wave 2 was derived from a model of enumeration at Wave 2 conditional on entering the UKHLS sample i e being issued to the field for UKHLS Wave 2 in which covariates came from the Wave 9 household instruments for England Scotland and Wales and the Wave 11 household instruments for Northern Ireland The weight which reflects the chance of a BHPS OSM of being selected into the BHPS to be issued into UKHLS and to be enumerated at Wave 2 of UKHLS is the BHPS 2010 longitudinal enumerated person weight b_psnenbh_lw Finally the BHPS cross sectional enumeration weight b_psnenbh_xw was created through a weight share method by sharing the BHPS 2010 longitudinal enumerated person weight to TSMs and PSMs The BHPS cross sectional weights for main proxy or telephone interview respondents b_indpxbh_xw main interview respondents b_indinbh_xw and self completion respondents adults b_indscbh_xw and youth b_yth
41. S rate Among the samples in Great Britain the Living in Britain households who have been part of the sample for the longest time had the highest response rate at 81 9 The Living in Wales households had a similar response rate to the Living in Britain sample 80 9 whilst Living in Scotland had a lower response rate at 76 5 The NIHPS had a higher household response rate than the Great Britain samples with 86 Table 7 Household response rates Wave 3 UKHLS GP EMB Former BHPS sample Living Living was ums ORAU ine aies Tot Britain Wales Fully 13 629 680 1 566 2 762 664 726 833 20 860 responding 57 1 55 3 42 9 66 2 62 2 62 2 66 4 57 3 Partially 4 341 293 951 653 153 218 246 6 855 responding 18 2 23 8 26 0 15 7 14 8 18 7 19 6 18 8 All 17 970 973 2 517 3 415 817 944 1 079 27 715 responding 75 3 79 1 68 9 81 8 76 5 80 9 86 0 76 1 Non 1 056 91 277 186 77 55 54 1 796 contact 4 4 7 4 7 6 4 5 7 2 4 7 4 3 4 9 Untraced 1 458 44 351 133 48 50 51 2 135 mover 6 1 3 6 9 6 3 2 4 5 4 3 4 1 5 9 Refusal 3 088 102 461 395 115 105 55 4 321 12 9 8 3 12 6 9 5 10 8 9 0 4 4 11 9 Other non 295 19 47 44 11 13 15 444 interview 1 2 1 6 1 3 1 1 1 0 1 1 1 2 1 2 Total 23 867 1 229 3 653 4 173 1 068 1 167 1 254 36 411 Base is all households
42. THOUT PROXY QUESTIONNAIRE For individual non respondents in responding households with no proxy questionnaire we impute total personal income using information from the household questionnaire only The procedure used is again hot deck More precisely whenever available we use e individual socio economic variables age sex marital status ethnicity work e household socio economic variables household size number of children in the household whether there is nobody in the household who speaks English whether the interview had to be translated house type an indicator for whether the person is owner of the house the external condition of the address relative to the others number of bedrooms in the house number of other rooms in the house value of the property for home owners number of cars number of durables log last year s expenditure on domestic fuel e g 78 electricity and gas amount spent on food eaten outside the home in four weeks prior to interview amount spent on food from food shops in four weeks prior to interview weekly rent paid whether the household can keep the accommodation warm enough e GOR and LDA 3 8 6 NET INCOME ESTIMATES The data also include estimates for income net of tax and national insurance As far as possible we have tried to replicate the approach used by DWP in developing their Households Below Average Income HBAI estimates There are however some deductions and some incomes source
43. UK Data Archive Study Number 6676 Understanding Society Secure Access Understanding Society The UK Household Longitudinal Study Waves 1 4 User Manual Edited by Gundi Knies Institute for Social and Economic Research University of Essex Colchester Essex Version 1 1 October 2014 Understanding Society THE UK HOUSEHOLD LONGITUDINAL STUDY CONTENTS Table of elo ere eer meere a ee ery erm ma rT mre EERE er ERRES 3 MaIG Ot FIQUIES a bari ea Beeb eA eA be Le hee Abe eek eae 4 List of Abbreviations cece cae tt ett he te tt late th 5 dic gt WIMFOGUGCUOIN cosececesnus let ecncoetaiacsitteccusleteiceehestiece lath ceeiassaaue bette ctediemiceeheecacuenet eer a 6 1 1 Whatis Understanding Society cccccccccceeceeececeeeeeeeeeteeeeeseeeeeeeeeeeeeeeees 6 1 2 Howto Navigate this User Manual ceeeececeecceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 6 2 Understanding Society Study DeSIQN cccccceeeeeeeeeeeeeeeeeeeeeeeeeeeneeeeeeeeeeeeeees 7 i EE VTAT EIN AAAA E AE A EAE A EA A A A A A AT 7 2 2 Sample Bsc 6 9 eer eng ee ee ee 8 2 2 1 General Population Sample c 3 cecccnnceeceeeinuaineedennes 8 2 2 2 General Population Comparison Sample cccccccceceeeeeeeeeeeeeeeeeeees 9 2 2 3 Ethnic Minority Boost Sample cceeeeeeeeeeeeeeeeeeeeeeeeeeeeneeeeeeeeeeeeeeees 9 2 2 4 Former BHPS Salmpleict aust ta Ata ete econ 10 2 2 5 Sample Status and Following Rules
44. a are provided at http ukdataservice ac uk get data how to access conditions special licence aspx A SL version of the UKHLS Main Survey data that contains the month of birth full occupational coding full uroban rural classification rare country of birth nationality occurrences and uncapped income variables is available through the UKDS SN6931 Understanding Society Wave 1 4 2009 2013 Special Licence Access Look up tables containing the official school reference code for children in the UKHLS are released under SL see SN7182 84 Look up tables linking Understanding Society survey data via the wave specific household identifier with official geographical units are also released under SL They include Local Authority Districts SGN6666 Census 2001 Area Classification for Output Areas SN6674 Travel to Work Areas SN6675 Westminister Parliamentary Constituencies SN6668 Rural urban Indicators SN7454 Local Education Authorities SGN6671 and Primary Care organisations SN6673 For further information about these geographical units see Office for National Statistics ONS Geography website http Awww ons gov uk ons quide method geography beginner s quide index html The geographical datasets in the UKHLS are described in Rabe 2011 and in annual release notes Note that we provide official geographical codes for Census 2001 but also for Census 2011 as they become available A SL data set linking Department for Tran
45. a subsequent delivery and or a quality feedback report suggesting changes to the questionnaire or field practice in subsequent waves Batch specific databases are merged into a single database from which anonymised data is exported for the creation of public use files Data distributions are also checked for theoretical and statistical plausibility This checking is done through direct scrutiny and by analyses which road test the data Last but not least data are being routinely checked in the process of creating added value content such as the respondent s age and sex based on the information collected across all waves or pointers to specific other members in the household such as a biological parent See Section 3 2 4 2 4 1 CODING Understanding Society collects freetext information on respondents job titles and the industry of the job held Industry descriptions are coded to ONS Standard Industry Code 2007 or SIC 2007 Job titles are coded to the ONS Standard Occupational Classification 2000 or SOC 2000 Coding is undertaken using the Computer Assisted Structured Coding Tool CASCOT system Several questions e g country of birth religion political party national identity and citizenship had an other please specify option These responses were coded using an automated process Coding was also done for an open ended question which read We ve asked you a lot of questions but we also want to know what has happe
46. act 4 3 1 7 7 3 4 6 5 9 4 8 2 5 4 6 Untraced 1 450 50 411 181 49 50 43 2 235 mover 5 6 3 7 10 0 3 9 4 0 3 9 3 2 5 6 Refusal 3 359 162 600 648 199 185 117 5 281 13 0 12 1 14 6 13 8 16 1 14 2 8 7 13 2 Other non 94 8 28 20 6 5 12 173 interview 0 4 0 6 0 7 0 4 0 5 0 4 0 9 0 4 Total 25 910 1 336 4 117 4 682 1 234 1 300 1 348 39 942 Base is all households issued to the field for wave 2 minus any found to have become ineligible Non contact rates were lower in Northern Ireland than in Great Britain The level of untraced movers was higher for the UKHLS GPS in Great Britain than in the former BHPS The levels of non contact and untraced movers were highest in the EMB samples possibly reflecting their younger average age concentration in large urban areas and higher level of mobility Within the former BHPS the level of untraced movers was higher than in the past This is likely to be due to the longer gap between waves of interview The interviews for the former BHPS sample for Wave 2 of the UKHLS took place throughout 2010 and into the early months of 2011 The previous interview for most of these households was between September and December 2008 As the gap between the Wave 18 BHPS interview and the Wave 2 Understanding Society interview increased so did the level of untraced movers Refusals as well were generally higher in Great Britain than in Northern Irela
47. ajor module on work conditions that encompasses such topics as payment mechanisms unions pensions work times autonomy and security and work stress 2 5 3 3 Wave 3 There are multiple new modules for Wave 3 including those on local neighbourhoods content on social networks in the main survey and the self completion questionnaires groups and organizations use of news and media and political self efficacy There is a major module on cognitive ability see McFall 2013 for more detail about the concepts and measures of cognitive ability Important data related to family ties can be found in the parents and children and family networks modules both of which are repeated from Wave 1 There is also data about child maintenance payments and relationships with children who do not live in the household The self completion modules have parents reports on children including a version of the Strengths and Difficulties questionnaire and parenting styles The self completion modules also include a Big 5 personality measure sexual orientation and several modules of questions for young adults which bring questions from the youth questionnaire into the 16 21 age group The EMB and related samples have repeat modules on discrimination and harassment and a new module on Britishness 2 5 3 4 Wave 4 Most modules for Wave 4 have appeared in earlier waves They include major modules on work conditions covering e g transport behaviour and job sat
48. al and household level variables obtained from the household questionnaire such as age and gender marital and employment status household size and presence of children in the household as well as household expenditure on food and food outside consideration of use of environmental energy among others The individual level nonresponse adjustment was obtained as the inverse of the predicted probability and was then multiplied by the relevant either individual or Extra five minutes design weight and by the household nonresponse correction No truncation was deemed necessary as there were no extreme values substantially impacting the design effects The post stratification was implemented as described above in the individual level enumeration weight section except that a greatly reduced matrix was used in the case of the Extra five minutes weight due to the much smaller sample size for which this weight applies After multiplying by the post stratification adjustment each of the obtained weights was then scaled to a mean of one 62 3 7 3 2 GPS and EMB Longitudinal Weights Each of the five types of longitudinal weights enumerated persons proxy or main interview main interview self completion and Extra five minutes interview is based on the corresponding previous longitudinal weight except in Wave 2 where it is based on Wave 1 cross sectional weight An additional adjustment for nonresponse since the last wave is
49. and telephone interviews from individual questionnaires including self completion w_youth Substantive data from youth questionnaire age 10 15 40 Table 16 lists some additional data files analysts may need to access In particular we would like to point users to the data file xwavedat in Table 17 which contains stable characteristics of individuals such as ethnicity which is typically collected only once in the lifetime of the study The complete list of files and their descriptors can be seen in the online documentation system https www understandingsociety ac uk documentation mainstage dataset documentation Table 16 List of select data files Data from enumerated sample members Filename Description w_indall Household grid data for all persons in household including children and non respondents w_child Childcare consents and school information of all children in the household w_egoalt Kin and other relationships between pairs of individuals in the household Table 17 List of select data files Cross wave files Filename Description xwavedat Stable characteristics of individuals xivdata Interviewer characteristics xwaveid Individual and household identifiers across all waves 3 1 1 PARADATA Some paradata i e additional data collected about the interview process is available These consist of call records timings data and other information collected by the interviewers during the interview The w_callre
50. and has been run by ISER since it began in 1991 To learn more about the BHPS and other components of Understanding Society see Section 4 3 below The User Manual is structured as follows We first present the general aspects of the study design Section 2 which covers sample design Section 2 2 data collection Section 2 3 data processing Section 2 4 and questionnaire content Section 2 5 Section 3 provides a description of UKHLS data files and variables covering derived variables Section 3 5 weighting adjustments Section 3 7 imputation of income Section 3 8 and example code for matching information contained in different files Section 3 8 Information on how to access the data is provided in Section 4 Section 5 provides additional information about further studies in the Understanding Society family currently the BHPS the UKHLS Nurse Health Assessment which took place in Waves 2 and Wave 3 and the UKHLS Innovation Panel As an introduction to the UKHLS Main Survey data and documentation we particularly recommend the following reading e The summary of the general questionnaire content Section 2 5 and notes on naming conventions Section 3 2 2 e The sections on sample design Section 2 2 weighting adjustments Section 3 7 and data collection and response outcomes Section 2 3 e Variable level descriptions of the data can be found on the study website https www understandingsociety ac uk documentation main
51. and untraced movers in this sample The refusal rates for the EMB at Wave 4 are slightly higher than those of the GPS in Britain and only a couple of percentage points higher than the Living in Scotland sample Table 11 shows the cross sectional response rates for adults in Wave 4 Where a household responded we have an individual level outcome for all adults Where a household did not respond we have assigned the household nonresponse outcome to the adults who were issued to that household 26 Table 11 Wave 4 Cross sectional individual adult response rates by sample origin UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain Scotland Wales Ne Total Full interview 27 643 1 341 4 236 5 547 1 222 1 479 1 749 43 217 67 0 60 7 52 0 73 4 66 9 68 7 74 1 66 0 Proxy interview 2 654 36 666 330 74 115 40 3 915 6 4 1 6 8 2 4 4 4 1 5 3 1 7 6 0 Other non interview 740 113 290 104 53 52 88 1 439 1 8 5 1 3 6 1 4 2 9 2 4 3 7 2 2 Refusal 1 676 180 394 298 62 115 153 2 878 4 1 8 2 4 8 3 9 3 4 5 3 6 5 4 4 Household non contact 1 958 150 756 243 84 95 110 3 396 4 8 6 8 9 3 3 2 4 6 4 4 4 7 5 2 Household refusal 5 209 281 1 328 842 243 224 139 8 266 12 6 12 7 16 3 11 1 13 3 10 4 5 9 12 6 Household other non 296 35 127 52 40 17 27 594 interview 0 7 1 6 1 6 0 7 2 2 0 8 1 1 0 9 Household untraced 1 076 72 346 1
52. aper 2011 01 Ragunathan E T J M Lepkowski et al 2001 A Multivariate technique for multiply imputing missing values using a sequence of regression models Survey Methodology 27 1 85 95 Rubin D B 1987 Multiple imputation for nonresponse in surveys New York Wiley Schafer J 1997 Analysis of Incomplete Multivariate Data London Chapman amp Hall Spanier G B 1976 Measuring dyadic adjustment new scales for assessing the quality of marriage and similar dyads Journal of Marriage and the Family 38 15 27 88 Taylor M F 2010 Weighting imputation and sampling errors British Household Panel Survey User Manual Volume A Introduction technical report and appendices J Brice N Buck and E Prentice Lane Colchester University of Essex A5 1 13 van Buuren S H C Boshuizen et al 1999 Multiple imputation of missing blood pressure covariates in survival analysis Statistics in Medicine 18 681 694 89
53. applied Each adjustment is based on a model of Wave n response conditional on Wave n 1 non zero longitudinal weight for the instrument in question For the enumerated person model covariates are taken from the Wave n 1 household grid and household questionnaire In the model for proxy and main interviews covariates were taken from the Wave n 1 proxy interview or the equivalent items from the main interview household grid and household questionnaire In both the model for main interviews and the model for adult self completion questionnaires covariates are taken from the Wave n 1 main interview household grid and household questionnaire The adjustment weight is calculated as the reciprocal of the model predicted response propensity The Wave n 1 weight is then multiplied by the Wave n adjustment to create the Wave n longitudinal weight Newborns born to an OSM mother since the Wave n 1 interview receive the longitudinal enumerated person weight of their mother reflecting the idea that the probability of observing the newborn is equal to the probability of observing the mother The principle behind the longitudinal weights is that they are defined for each person who is observed at all of the relevant waves for which they were eligible For this reason newborns observed at Wave n receive a Wave n longitudinal weight as they were enumerated at Wave n the only wave for which they were eligible 3 7 3 3 Cross sectional Weights GPS
54. ata associated with the missing information are used to define imputation classes In Understanding Society the hot deck method is used to impute missing information for proxies and non respondents For proxies where available imputation classes are defined by reported bands on income and earnings and limited covariates For non respondents and proxies where income bands are missing a richer set of covariates are used to define imputation classes including employment and benefit Carryovers age education sex ethnicity housing tenure marital status durable good ownership whether a parent and number of bedrooms Once a suitable donor is identified information on all income sources is carried over from the donor 3 8 2 2 Longitudinal imputation methods Methods for longitudinal imputation are used to take into account longitudinal patterns in the data The longitudinal methods we use are the Population Carryover and the Little and Su methods Population Carryover PC PC uses data from adjoining waves to replace missing wave information With only one adjoining wave of non missing data the information is carried over with probability one When two waves of adjoining information are available the information carried over is chosen based on proportions reported in the non missing population In Understanding Society PC is used to impute employment status and benefit eligibility for non respondents and proxies These varia
55. bles are then used as inputs into a hot deck procedure see above Little and Su LS The LS method imputes missing values using a multiplicative model see Little and Su 1989 The final imputation is the product of 3 terms a trend effect across waves the recipient s departure from the trend and a residual effect donated from 76 another respondent with complete information for the corresponding income component In Understanding Society a modified version of the LS method is implemented When identifying a donor for the residual effect rather than using the non missing information only we additionally make use of information imputed from the cross sectional methods described above In this way the LS procedure forms the final step in the Understanding Society imputation process where imputes from the other methods form inputs into the LS method Missing income sources imputed as inapplicable using the cross sectional methods do not receive a LS impute neither do respondents applicable for an income source in only one wave 3 8 3 ITEM NONRESPONSE ON INCOME VARIABLES IN THE INDIVIDUAL QUESTIONNAIRE The imputation of earnings wages and self employment earnings from the first job and earnings from the second job and investment income in the individual questionnaire is performed considering a separate equation for each of the income components With variables for which we have point information a single value we use e
56. buadeareuesavcsans 82 Stata Code Merging individual files across waves into long format 83 Stata Code Merging individual files across waves into wide format 83 LIST OF ABBREVIATIONS BHPS HBAI CAPI CASCOT CASI DWP ECHP EMB ESRC GOR GPC GPS ICE IP ISER LDA LS LSOA MMU MSOA NatCen NE NIHPS NISRA ONS OSM PC PMM PSM SIC SOA SOC TSM UKHLS British Household Panel Survey Households Below Average Income Statistics Computer Assisted Personal Interview Computer Assisted Structured Coding Tool Computer Assisted Self Interview Department for Work and Pensions European Community Household Panel Ethnic Minority Boost Economic and Social Research Council General Office Region General Population Comparison General Population Sample Imputation by chained equations UKHLS Innovation Panel Institute for Social and Economic Research Low density ethnic minority area Little and Su method Lower Layer Super Output Area Multi Mode Unit Middle Layer Super Output Area National Centre for Social Research New entrant to the study Northern Ireland Household Panel Survey Northern Ireland Statistics and Research Agency Office for National Statistics Original Sample Member Population Carryover method Predictive mean matching Permanent Sample Member ONS Standard Industry Code Super Output Area ONS Standard Occupational Classification Temporary Sample Member UK Household Longitudinal Study o
57. c data file has information on the number of calls made as well as the issue number time and date and the outcome of each call Information on the date of receipt of the case and the interviewer associated with each issue as well as the outcome at the end of each issue period is available in the file w_issue In addition to this information collected in the address response form ARF by interviewers while contacting each household and asking household members to participate in the survey is available in w_hhsamp This includes data on the area surrounding the address the type of accommodation and other information that the interviewer can observe about sampled addresses Reasons for refusal are also available Interviewers also collect some information about the quality of the interview and persons present during the interview process This is available along with substantive data collected during adult individual interviews including proxy interviews in w_indresp 41 Table 18 List of select data files Paradata Filename Description w_hhsamp Data from Address Record File for issued households w_callrec Information about interview outcome at each call w_issue Information about interview outcomes at each issue including interviewer number 3 2 INFORMATION ABOUT VARIABLES 3 2 1 LEARNING ABOUT THE STUDY VARIABLES There are multiple resources for learning about the study variables in order to plan analyses These include the questio
58. component were similar to the Understanding Society GPS Among the samples in Great Britain the Living in Britain households had the highest response rate at 77 2 The Living in Wales households had a similar response rate to the Living in Britain sample 76 8 whilst Living in Scotland had a lower response rate at 73 5 The NIHPS had a higher household response rate than in Great Britain with 84 8 The response rates for the BHPS samples in Great Britain were disappointing given that this was in effect Wave 19 for many households 17 However the lower response rate may have been due to the change in the fieldwork agency interviewers survey name and logo Interestingly in Northern Ireland where the survey name and logo changed but the fieldwork agency and so the interviewer stayed the same as in NIHPS the response rate was much higher Table 4 Household response rates Wave 2 UKHLS GP EMB Former BHPS sample Living Living ee tas in ee in NIHPS Total Britain Wales Fully 16 003 873 2 030 3 112 793 833 990 24 634 responding 61 8 65 3 49 3 66 5 64 38 64 1 73 4 61 7 Partially 3 888 221 749 504 114 165 153 5 794 responding 15 0 16 5 18 2 10 8 9 2 12 7 11 4 14 5 All 19 891 1 094 2 779 3 616 907 998 1 143 30 428 responding 76 8 81 9 67 5 77 2 73 5 76 8 84 8 76 2 Non 1 116 22 299 217 73 62 33 1 825 cont
59. cotland the information was linked at the data zone level from http www scrol gov uk scrol common home jsp and from http www scotland gov uk Topics Statistics SIMD From the Census 2001 information was obtained on population density mean age average household size and number of rooms per household in the data zone as well as the proportions in the data zone born in Scotland and outside the EU of different religious denominations employed unemployed and retired disabled those with different levels of qualification and types of occupation and different types of accommodation among others For Northern Ireland the information was linked at the Super Output Area SOA level and was obtained from http www ninis nisra gov uk Examples of predictors obtained from Census 2001 at the SOA level include the average hours worked by residents the average age of residents percentages of residents with different level of qualifications with different employment statuses and with different types of marital status among others The predictors also include 2007 2009 information on multiple deprivation indexes Note that using Understanding Society analysis weights all but design weights adjusts for household nonresponse bias in any estimate to the extent it is related to the above mentioned variables Enumerated Individual Weight The weight for analysis of enumerated individuals a_psnenus_xw is not equivalent to the household wei
60. d across the sectors in a way designed to deliver target numbers of respondents in each target ethnic minority group with adequate statistical efficiency see Berthoud et al 2009 for more details In sectors selected for both the GPS component and the EMB sample a single systematic sample of the required total number of addresses was selected and allocated in a systematic way to the two sample components thus ensuring that both sample components are spread throughout the whole sector The final stage of sampling was done by the interviewers The steps are described in the Project Instructions for Interviewers At addresses containing more than three dwellings or households the procedures to sub select dwellings or households were as described above for the GPS component Within each household rather than all resident persons becoming sample members there were three additional steps e A screen was carried out to identify whether there were any persons from target ethnic groups in the household e A random mechanism was applied to certain target groups identified by the screen in order to select only a desired proportion into the sample non mixed Indian Pakistani non mixed Caribbean African Far Eastern Middle Eastern For other target groups all resident persons were included in the sample mixed Indian Bangladeshi mixed Caribbean Sri Lankan Chinese Turkish e In households included in the sample in the previous two ste
61. d in modules Modules can be searched for in the online documentation system https www understandingsociety ac uk documentation mainstage dataset documentation Instruments and survey materials were translated into multiple languages Bengali and Punjabi in Urdu and Gurmukhi scripts Welsh Arabic Somali Cantonese Urdu and Gujarati Translated documents can be requested by email from info understandingsociety ac uk 2 5 1 READING THE QUESTIONNAIRES Figure 1 Mark up of household questionnaire shows a marked up sample page providing information for how to interpret the questionnaire text Note that the variable names in the questionnaire do not have the wave prefix a_ b_ etc or in the general form w_ Figure 2 Mark up of question with looping from individual questionnaire shows a marked up sample page from the individual interview The question is more complex The question is asked about each natural or biological child so multiple variables are associated with the question for each natural child The variables are located in the data file a_natchild which has one record for each natural child 30 Figure 1 Mark up of household questionnaire Variable name and Variable label Hsownd House owned or rented Note that there is no wave prefix Must add prefix to the variable name onres This variable has also been in the BHPS Text Does your household own this accommodation outright is it being bought with a mortgage
62. d level nonresponse individual level within household nonresponse and post stratification to population characteristics Each of the components is explained below Design Weight The design weight corrects for unequal probability of selection at a number of levels The household level design weight corrects for e Unequal selection probability due to the boost in Northern Ireland The GPS selection probabilities in Northern Ireland are approximately twice those in other parts of the UK e Unequal selection probability related to selection into the EMB Selection probabilities in the EMB part of the sample vary considerably between areas depending on the estimated ethnic mix of the area and ethnic composition of the household Additionally households in high density areas with at least one ethnic minority member were weighted to account for combined probability of being selected as part of the GPS or as part of the EMB samples e The selection probability of households in a dwelling with more than three households or at an address with more than three dwellings is adjusted for the fact that only three such households were selected from the same address Individual level design weights correct for all the above with one specific difference non EM persons who live with EM persons in the same household have a chance to be selected only via the GPS part of the sample and not via the EMB This means that non EM persons in the EMB who are TSMs a
63. d not retired X age 45 50 55 60 65 and not retired Domestic Division of Labour One item on hours of housework asked in first 6 months at wi X X Political Engagement X EMB GPC LDA General Election X May Dec 2010 Leisure Culture and Sports X Leisure Access X Positive and Negative Events X Britishness X EMB GPC LDA Own First Job X Educational Aspirations X Young Adults X age 16 21 X age 16 21 Local Neighbourhoods Urban rural Social Networks Groups and Organisations Health X XXIX X lt 34 Conditions Cognitive Ability X Family Access X Child X biological Maintenance parent of child other biological parent not in HH Migration X age Intention 45 50 60 65 and not retired Political Self X Efficacy News and X X Media Use Service Use X EMB LDA GPC recent immigrant Sleep X Twin X Biological mother of siblings with same birthday Mother s return to work X Transport behaviour Wealth assets and debt Olympics X X X NE new entrant EMB Ethnic Minority Boost GPC General Population Comparison LDA Low density ethnic minority area 35 The paper self completion questionnaires carried at waves 1 and 2 were n
64. d way the inclusion weight b_psnenub_ li is unaltered if it is non zero for all household members but the cross sectional weight is equal to the average inclusion weight for households where at least one member has a value of zero for the inclusion weight The obtained cross sectional enumeration individual weight is then scaled to a mean of one b_psnenub_xw For each household the lowest b_psnenub_li from OSM adults 16 years of age or older is selected This reflects the highest probability of enumeration among OSM household adult members The weight is then scaled to the mean of one b_hhdenub_xw 3 7 3 9 Longitudinal Weights for Total Sample BHPS GPS and EMB Components for Waves 3 and 4 A longitudinal enumeration weight for the BHPS GPS and EMB samples C_psnenub_Iw was created based on the inclusion enumeration weight into Wave 2 b_psnenub_li Conditional on a nonzero value for b_psnenub_li the enumeration is modelled using logistic regression and predictors from the wave 2 household questionnaire and household grid The estimated probabilities of conditional enumeration are inversed and multiplied by b_psnenub_li Newborns are given their mother s enumeration weight The weight is then scaled to a mean of one Longitudinal response weights for proxy and main interview main interview and self completion interview were also created at Wave 3 All of these are based on longitudinal enumeration in Wave 3 positive value of c_
65. ding Society General Population Sample households enumerated at Wave 1 including absent household members and those living in institutions who would otherwise be resident are Original Sample Members OSMs All ethnic minority members of an enumerated household eligible for inclusion in the EMB sample are OSMs In all of these samples any child born to an OSM mother after Wave 1 and observed to be co resident with the mother at the survey wave following the child s birth is an OSM In the former BHPS sample OSMs are those who were enumerated at the first wave of the sample from which they come Wave 1 for the original sample Wave 9 for the Scotland and Wales boost samples Wave 11 for Northern Ireland or who were subsequently born to an OSM mother or father or both From Wave 2 onwards of Understanding Society in the former BHPS sample as for the rest of the Understanding Society sample only children born to an OSM mother will themselves become an OSM OSMs of all ages are followed for interview and remain eligible as long as they are resident within the UK They remain potentially eligible sample members for the life of survey 10 The case may arise where the only OSM in the household is a child Other household members are then TSMs so long as they are co resident with the child and therefore eligible for interview even if the child is not yet old enough to be eligible for interview If the OSM child moves house they are followed
66. ds of these OSMs who are eligible for the Extra 5 minutes questions To analyse this sample use the appropriate Extra 5 minutes cross sectional and longitudinal weights see Section 3 7 for further details on the construction of these and other weights Note that there is no cross sectional weight for the Extra 5 minutes questions as at Wave 2 these were only asked of sample members who had completed the main interview at Wave 1 Thus the Wave 2 longitudinal weight should be used for Wave 2 cross sectional analysis For further details on ethnicity analysis using UKHLS data see Understanding Society Guide to Ethnicity Research forthcoming 3 5 DERIVED VARIABLES Derived variables are variables that are computed from one or more variables Some are computed by the Blaise CAPI program during the interview to control the routing within the questionnaire whilst others are computed post field for the purpose of analysts The suite of derived variables included in the UKHLS includes flags for whether or not a certain characteristic is true for a study member e g w_jbft_dv is a flag for whether or not a respondent has a full time job counts of the number of people in the household for whom a certain characteristic is true e g w_nemp_dv 46 is the number of employed people in the household and pointers to significant others in the household e g mnpid records the cross wave person identifier of the respondent s b
67. e fieldwork documents One example is Show cards which are used to help respondents with their answers Show cards are referenced in the questionnaire Project Instructions were prepared for interviewer training and to serve as a resource in data collection Documents for communicating with participants are also included on this portion of the website In Wave 1 we asked for consent to link to administrative health and education records The information leaflets and consent forms are in this section of the study website The Address Record Form ARF is an important document for recording information about responding and non responding households There are many different versions in Wave 1 Interviewers record the call record observations on characteristics of accommodation and households and household outcomes on the 39 ARF In Wave 1 there were several different versions of the ARF The first distinction is between the GPS and the EMB sample The versions labelled ARF EB for the EMB sample are longer because they include questions for screening household members for eligibility ARF s labelled 2 or 3 are for addresses with multiple households and or dwelling units Finally there are versions for ARF EB1 Year 1 or Year 2 This change in form was required by the change in selection criteria implemented in Year 2 of Wave 1 see Berthoud Fumagalli et al 2009 for more detail The ARF screening card was a show card used during the scr
68. e issue letter but the interviewers had discretion to offer the additional incentive on the door step if they felt that this would convert a non responding household to a participating household In addition during the latter quarter of Wave 3 fieldwork more effort was made to increase interviewer continuity for households across waves rather than prioritising interviewer efficiency It is estimated that these two procedures which were launched almost simultaneously increased household response rates by around 4 percentage points for the EMB sample and by around 2 5 percentage points in the GPS in Quarter 8 The procedures adopted in Wave 3 to maintain household response were continued in the Wave 4 fieldwork 2 3 3 PANEL MEMBERSHIP AND PANEL MAINTENANCE The rules for following individual respondents over time are based upon the composition of the household Individuals found at selected households in the first 15 wave were designated as Original Sample Members OSM We attempt to maintain OSM respondents as part of the sample as long as they live in the UK In addition births to an OSM mother are classified as OSM Individuals joining the household of an OSM after enumeration of the household at Wave 1 are Temporary Sample Members TSM One deviation from this is for individuals who were not an ethnic minority within the households selected as the EMB sample At Wave 1 these individuals were classified as TSMs We attempt to intervie
69. e lMMPUlG t ice tia ca ave Gee ar aes 73 3 8 2 Imputation Procedures ccccccceeeeeeeeeeeeeeeceeeeeeeeeeeeeeeneeeeeeeeeeeeeeeees 74 3 8 3 Item nonresponse on income variables in the individual questionnaire 77 3 8 4 Item Nonresponse for Income Variables in the Proxy Questionnaire 78 3 8 5 Individual Non respondents Without Proxy Questionnaire 78 3 60 60 Net income Estimates ai calive el aac ea eee ue AEE n A aera ene 79 3 9 Example Code for Matching Files cccceeeeeeeeeeeeeeeeeeneeeeeeeeeeeeeeeeenaaees 80 3 10 Preserving Confidentiality 32 2 0 ccic tscceieectcceceetceetseeneasbeeteceeheenteeeesecoctriet 83 4 Data ACCES eae ane ve ee ee 84 4 1 Release SV CESS 2 es ee Ss ls le a hat la hae eats ts tas ti 84 4 1 1 End User Licence EUL Data vrceiccscmsencscsoicsssuuedcvenstiodevearenleucnvoaevess 84 4 1 2 Special Licence SL Data eeccccceeeeeeeeeeeeeeceeeeeeeeeeseessseeeeeeeeeees 84 4 1 3 SOCUlC ACCESS ck ccece ie neta a aE 85 4 2 Revisions to Previous Releases ccccceecceeceeeeeeeeeeeneeeeeeeeeeeeeeeeenaeees 85 4 3 Links to other studies in the UKHLS family 00 eeeeeeeeeeeeeeeeeeeeeeeeeeeeees 86 4 3 1 The British Household Panel Survey cceeeeeeeeeeeeeeeteeeeeeeeees 86 4 3 2 The Waves 2 3 Nurse Health Assessment ccceceeeeeeeeeeetteeeeeeees 86 4 3 3 The UKHLS Innovation Panel vciccicccectccciceseerasicecccr
70. e recursive system as starting values in an iterative imputation process In other words the starting values are used to begin a new cycle of imputations where each equation is estimated sequentially but this time using as explanatory variables both X and all the imputed variables Y1 Y2 Yk excluding the 75 one used as dependent variable At the end of this new cycle a set of new imputed variables is produced and used to begin a further new cycle of imputations These cycles of imputations are repeated until convergence Notice that in practice some of the variables will exclude certain of the Xs and Ys variables in the imputations because it does not always make sense to use all variables as predictors Predictive mean matching PMM It is used to impute benefits pensions and other incomes For a given variable PMM replaces missing values with observed values from a donor i e a respondent with non missing information on the variable of interest This is done in four steps i regression models for the variable to be imputed are estimated ii fitted values are produced iii records with missing information recipients are matched to donors based on the fitted values computed in ii iv missing values are replaced with observed values from donors See also Little 1988 Hot deck HD For individuals with missing information the hot deck method identifies suitable donors within imputation classes Characteristics reported in the d
71. ed using a combination of cross sectional and longitudinal imputation methods There are two steps to the Understanding Society imputation procedure The first replaces missing values using cross sectional imputation methods with some exceptions where we make use of longitudinal information see carryover below A second step then replaces the first stage imputes using the longitudinal imputation method of Little and Su 1989 The overall approach is based on that used for the Australian household panel HILDA described in Hayes and Watson 2009 The various imputation methods used are described in more detail below 3 8 2 1 Cross sectional imputation methods Cross sectional imputation is carried out year by year through a range of parametric semi parametric and non parametric methods Parametric methods are linear regression for continuous variables interval regression for continuous censored variables logistic regression for binary variables ordered logistic regression for ordered variables multinomial logistic regression for non ordered categorical variables The semi parametric and non parametric methods are respectively predictive mean matching PMM and hot deck imputation Parametric methods and PMM are used to impute wages self employment earnings second job earnings interests and dividends plus their predictors for responding individuals For responding individuals wages self employment earnings second job ea
72. eeeeeeeeeeeneeeeeeeeenees 26 Table 11 Wave 4 Cross sectional individual adult response rates by sample origin27 Table 12 Longitudinal individual re interview rates adults by sample origin Full interview at Wave 3 Bere rpreperes Pere terrrrer errr erterenecererrrrtrrrre Eras ere erererrrererer rere errr rer ree 28 Table 13 Summary of questionnaire modules ee eee eeeeeeeee cette eee eeeeeeneeeeeeeeeeee 32 Table 14 Summary of adult self completion Content ccceeeeeeeeeeeeeteeeeeeeeeees 36 Table 15 List of select data files Data from responding sample members 40 Table 16 List of select data files Data from enumerated sample memberts 41 Table 17 List of select data files Cross wave fil S c cececeeeeeeeeeeeeeeeneeeeeeeeeees 41 Table 18 List of select data files Paradata ccccceeceeeeeeeeeeeeeeeeeeeeeeeeeeenneeeeeeeeeees 42 Table 19 Missing value codes iiss savrcavncn ea caesuh iy kinks cused ehusuk ie eausoe benenebn ten dant 43 Table 20 Some useful variables a crs es ana vus cxinst viccvins tires vay cxtgha unk Ceol acuwnsvontvenivertventeeenained ons 44 Table 21 Description of the UKHLS Primary Sampling Unit variable w_psu 48 Table 22 Description of the UKHLS stratification variable W_strata cee 49 Table 23 Selecting the correct weight Hierarchy of analysis levels 50 Table 24 Weight variables for analyses using hous
73. eening interviews Additional information about completion of the ARF can be found in the Project Instructions for Interviewers 3 UNDERSTANDING SOCIETY DATA 3 1 INFORMATION ABOUT DATA FILES The data release consists of multiple files in SPSS or Stata formats distributed by the UK Data Service Data for different waves are released in separate files File names begin with a prefix designating the wave of data collection a_ for the first wave b_ for the second wave in this user guide we have used w_ to denote waves in general Data collected from different sources i e e g the household interview the adult interview the youth interview are stored in separate files The root filename is fixed over time For example individual level data collected from interviews with responding adults in Wave 1 both years 2009 and 2010 is stored in the file a_indresp and individual level data collected from interviews with responding adults in Wave 2 both years 2010 and 2011 is stored in the file b_indresp Table 15 lists the data files that contain substantive information collected in interviews with responding households and individuals They are the most likely data files analysts will want to access Table 15 List of select data files Data from responding sample members Filename Description w_hhresp Substantive data from responding households w_indresp Substantive data for responding adults 16 including proxies
74. ehold grid or household interview as hase ac eu alae oto Satin eh eet alc et ok eaters Bante peo geen E E 52 Table 25 Weights for analysis using adult main and proxy interviews 05 53 Table 26 Weights for analysis using adult main interviews seeeeeeeeee 54 Table 27 Weights for analysis using adult Extra 5 minutes interview 5 54 Table 28 Weights for analysis using adult self completion eseeeeeeeeeeeees 54 Table 29 Weights for analysis using youth self completion e seseeeeeeeeeeeees 55 Table 30 Weights for analysis using nurse health assessment data 005 56 Table 31 Design and inclusion weights cccccceeeeseeeeeeceeeeeeeeeeeeeeneeeeeeeeeeees 56 TABLE OF FIGURES Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Mark up of household questionnaire eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 31 Mark up of question with looping from individual questionnaire 31 Stata Code Distributing household level information to individuals 81 Stata Code Summarizing individual level information to household level 81 Stata Code Matching individuals within a household cccceeeees 82 Stata Code Using the egoalt file to create household composition variables sige ugha A E EE E E AER ans Ga heaniuxaevepesnauanndusouuepiuvarae nibs
75. ehold level information keep a_hidp n1824 now merging this household information with the household level file sort a_hidp merge data files and save merge 1 1 a_hidp using hhinfo nogen drop _merge save final2 replace housekeeping erase hhinfo dta 81 Understanding Society UK Household Longitudinal Study Wave 1 4 2009 2013 User Manual 21 October 2014 In the fourth example we will match the information of wives onto that of their partners spouses Figure 5 Stata Code Matching individuals within a household Open the dataset with information on all persons in responding households and keep only those persons who have a spouse partner in the household se a_hidp a_pno a_hgpart a_sex a_dvage using 2_indall_ip if a_hgpart 0 clear rename the prefix a_to something that would indicate that this information relates Again open the data with information on all persons in responding households se a_hidp a_pno a_hgpart a_sex a_dvage using a_indall_ip f a_hgpart 0 dear rename the prefix a_to something that would indicate that this information relates as we want to match on a_hidp and a_pno rename r_hidp and r_pno back to these ename r_hidp a_hidp rename r_pno a_pno Now sort and merge with the spouse partner file sort a_hidp a_pno merge 1 1 a_hidp a_pno using spousepartner drop _merge save final3 replace erase spousepartner dta In the fifth example we will create a variable that measures the n
76. es with 110 sectors in each monthly sample Within each postal sector 18 addresses were selected using systematic random sampling The England Scotland and Wales sample in this data release is based upon an initial sample of 47 520 addresses In Northern Ireland 2 395 addresses were selected in a single stage from the list of domestic addresses In combination this data release is therefore based upon a total of 49 915 addresses At each address the final stage of sampling was carried out by field interviewers This consisted of identifying persons to be defined as sample members All persons resident at each sample address at the time the interviewer made contact were deemed to be a sample member with the exception of the small proportion of 8 addresses that contained more than three dwellings or households In those cases three dwellings or households were sub sampled at random 2 2 2 GENERAL POPULATION COMPARISON SAMPLE The General Population Comparison sample GPC has one sampled address for 40 of the selected postal sectors in General Population Sample GPS component for Great Britain In other words of the 2 640 general population sectors 60 of them 1 584 contain 18 GPS addresses and the other 40 contain 17 GPS addresses and one GPC address The persons in these households will be designated as members of the GPC sample regardless of ethnic group membership Members of the GPC sample are a random subsample of the GPS
77. es 1 575 former BHPS sample Identical to the BHPS variable wpsu in England Scotland and Wales 701 1999 former BHPS Northern Corresponds to initial BHPS wave 11 Ireland sample sampled households as these were selected in a one stage design 2001 4640 UKHLS GPS in Corresponds to the postal sectors used as England Scotland and PSUs see Lynn 2009 Wales 46424 UKHLS GPS in Corresponds to Wave 1 sampled 7035 Northern Ireland households as these were selected ina one stage design 7048 UKHLS EMB Corresponds to Wave 1 sampled 51789 households as these were selected ina one stage design within the high minority density domain see Berthoud Fumagalli et al 2009 Note there was an error in b_psu and c_psu for Northern Ireland BHPS households in the Wave 2 and Wave 3 data releases This has been corrected from the Wave 4 release 48 3 6 2 STRATA w_ strata This indicates the sampling stratum from which the sample member was selected The value of w_strata does not change between waves but for new sample entrants it is only defined from the wave at which they enter the sample The range of values on w_strata is listed in Table 22 Table 22 Description of the UKHLS stratification variable w strata Values Sample Notes 1 151 former BHPS sample Identical to the BHPS variable wstrata in England Scotland and Wales 701 former BHPS Northern Northern Ireland treated as a single stratum Ireland sample 200
78. es about these changes have been documented in the variable view of the online documentation system Also see Section 3 2 below 2 5 4 2 Across Waves There sometimes are changes to the questionnaire across waves This is the case for instance when routing errors only became known after data collection had been completed e g in Wave 1 only the proxy interview included the question for whether or not a respondent had access to a car w_drive and from Wave 2 onward this information is available for adult and proxy respondents Another example is the SF 12 which was asked in the main interview with adults in Wave 1 but was shifted to the adult self completion in Wave 2 Notes about these changes are documented in the variable view of the online documentation system Also see Section 3 2 below The switch in mode from paper self completion to the CASI on the lap top in Wave 4 meant that for some questions the response options were presented differently between waves For example where response options were arrayed horizontally in the paper self completion e g satisfaction questions they were presented vertically in CASI There is some evidence that this change in the way in which the response options were presented may affect how some people respond to the question Burton and Ed 2012 2 5 5 OTHER FIELDWORK MATERIALS Other fieldwork materials are also on the website https www understandingsociety ac uk documentation mainstag
79. estions there are modules about Remittances Harassment and Discrimination The Parents and Children module has content about attitudes and behaviours related to education activities and interaction with children and parenting practices Some interesting content in the Wave 1 self completion questionnaire includes measures of sleep behaviour and sleep quality and a subset of items characterizing relationships with partners from the Dyadic Adjustment scale Spanier 1976 There are also measures of generalised trust and attitudes to risk 37 2 5 3 2 Wave 2 In Wave 2 and subsequent waves there is the Annual Event History module which is asked of persons previously interviewed It asks about changing circumstances related to moves marital status or cohabitation new children including childbirth and pregnancy new health conditions educational experiences and employment changes Wave 2 saw the introduction of a set of rotating modules related to health behaviours nutrition smoking physical activity There are modules about voluntary work and charitable giving and important modules about savings and personal pensions Retirement planning is an age triggered module that is taken up again in Wave 3 Within the self completion questionnaire there is content on alcohol consumption dimensions of identity and gender role attitudes The EMB and related samples have modules on political engagement and ethnic identity Wave 2 also has a m
80. fficial acronym for Understanding Society 1 INTRODUCTION 1 1 WHAT IS UNDERSTANDING SOCIETY Understanding Society the UK Household Longitudinal Study UKHLS is a longitudinal survey of the members of approximately 40 000 households in the United Kingdom i e the geographical area of the countries England Scotland Wales and Northern Ireland Households recruited at the first round of data collection are visited each year to collect information on changes to their household and individual circumstances Interviews are typically carried out face to face in respondents homes by trained interviewers Data collection for each wave takes place over a 24 month period as shown in Figure 1 Note that the periods of waves overlap and that individual respondents are interviewed around the same time each year Understanding Society is funded by the Economic and Social Research Council ESRC and with funding from multiple government departments the Department for Work and Pensions DWP the Department for Education the Department for Transport the Department for Culture Media and Sport the Department for Communities and Local Government the Department of Health the Scottish Government the Welsh Assembly Government the Northern Ireland Executive the Department for Environment Food and Rural Affairs and the Food Standards Agency The scientific leadership team is from the Institute for Social and Economic Research ISER of the Uni
81. fit income w_inc3prben Receipts reported in income record where w_ficode equals 25 trade union friendly society payment 26 maintenance or alimony 35 sickness and accident insurance Component 5 investment income w_inc5inv w_fiyrinvinc_dv annual income from savings and investments divided by 12 Receipts reported in income record where w_ficode equals 4 a private pension annuity 28 rent from boarders or lodgers not family members living here 29 rent from any other property less estimated tax 79 Component 6 pension income w_inc6pen Receipts reported in income record where w_ficode equals 2 a pension from a previous employer 3 a pension from a spouse s previous employer Component 7 social benefit income w_inc7sben Receipts reported in income record where w_ficode equals 1 state retirement old age pension 5 a widow s or war widow s pension 6 a widowed mother s allowance widowed parent s allowance 7 pension credit includes guarantee credit amp saving credit 8 severe disablement allowance 9 industrial injury disablement allowance 10 disability living allowance 11 attendance allowance 12 carer s allowance formerly invalid care allowance 13 war disablement pension 14 incapacity benefit 15 income support 16 job seeker s allowance 18 child benefit including lone parent child benefit payments 19 child tax credit 20 working tax credit includes disabled person s tax credit 21 maternity allowance 22 housi
82. ght for all household members as often happens in other household studies This is because we have TSMs in Wave 1 who are not ethnic members selected into EMB part of the sample Thus the individual level design weight is not equal to the household level design weight for individuals in households containing a mix of EM and non EM persons The weight for the analysis of enumerated individuals is calculated as the product of the individual level design weight a_psnenus_xd and the household level nonresponse correction described above The design effect was tested showing that no truncation was necessary Weighted sample distributions were then compared to ONS mid year estimates with a correction for institutionalized population and post stratification was implemented for the fully crossed matrix of gender by geographical region by 5 10 year age groups Thus the individual level enumerated weight consists of 61 The individual level design weight the household nonresponse correction the post stratification adjustment The obtained weight is scaled to have a mean of one Individual level Nonresponse Adjustment Five different individual level weights were prepared for users reflecting nonresponse occurring at different levels and different questionnaire instruments Each individual level weight consists of this product The individual level design weight the household nonresponse correction the individual level nonresponse correctio
83. graphical Accessibility The UKHLS Accessibility Data File User Guide http doc ukdataservice ac uk doc 7533 mrdoc pdf 7533 ukhls accessibility userquide pdf Little R J A 1988 Missing Data Adjustments in Large Surveys Journal of Business and Economic Statistics 6 287 296 Little R J A and H L Su 1989 Item Non response in Panel Surveys Panel Surveys D Kasprzyk G J Duncan G Kalton and M P Singh New York Wiley Lynn P 2009 Sample design for Understanding Society Understanding Society Working Paper 2009 01 Lynn P J Burton et al 2012 An initial look at non response and attrition Understanding Society Working Paper 2012 02 McFall S 2013 Understanding Society UK Household Longitudinal Study Cognitive Ability Measures Understanding Society User Manual McFall S J Petersen et al 2014 Understanding Society Waves 2 and 3 Nurse Health Assessment 2010 2012 Guide to Nurse Health Assessment https www understandingsociety ac uk documentation health assessment Office for National Statistics 1991 Census Key Statistics for Local Authorities London HMSO Office for National Statistics 2003 Census 2001 Key Statistics for Local Authorities in England and Wales www ons gov uk ons rel census census 2001 key statistics local authorities in england and wales Rabe B 2011 Geographic Identifiers in Understanding Society Understanding Society Working P
84. h 2011 including the re issue period In total interviews were achieved in 30 169 households 26 089 in the GPS 4 080 in the EMB sample with full or proxy interviews with 50 994 individuals 43 674 in the GPS and 7 320 in the EMB sample Table 2 and Table 3 present the household and individual response rates for Wave 1 The individual response rates are for co operating households only Table 2 Household response rates among eligible households Wave 1 GPS EMB oe Northern Great Britain ireland Total Responding 57 1 60 9 57 3 39 9 Non contact 8 1 11 0 8 3 28 0 Refusal 33 9 27 4 33 6 29 0 Other 0 8 0 7 0 8 3 1 N 43 267 2 107 45 374 10 077 16 The response rates for the EMB sample component do not make any correction for the probability of non interviewed cases being ineligible The estimated response rate taking this factor into account is substantially higher Table 3 Individual response rates Wave 1 GPS EMB Great Northern Britain Ireland Total Full interview 82 0 77 3 81 8 72 4 Proxy interview 5 3 3 5 5 2 6 9 Refusal 6 5 9 2 6 7 8 7 Other non interview 6 1 9 9 6 3 12 1 N 47 615 2 584 50 199 9 237 2 3 4 2 Wave 2 The Wave 2 main survey fieldwork started on 12 January 2010 and ended on the 27 March 2012 including the re issue period Household response rates for Wave 2 of the UKHLS are shown in Table 4 The table separates
85. he BHPS sample a different identifier will need to be used the variable pid which is the BHPS cross wave person identifier The pid identifier is available in all person level files in the Understanding Society beginning in the Wave 2 release and in the 18 wave BHPS longitudinal data set While the great majority of BHPS sample cases who were interviewed in Understanding Society Wave 2 were previously interviewed at Wave 18 in 2008 9 there are a number who were last interviewed at an earlier wave Information about the response status of BHPS sample members at each of the 18 waves is contained in the BHPS file xwaveid The BHPS data set also contains a file called xwavedat which contains the values for stable variables e g ethnic group parent social class etc Because of some differences in variable definition this information has not been copied across to the new Understanding Society file also called xwavedat However in most cases values of these variables can be obtained by matching to the BHPS file We hope to produce a harmonized version at subsequent release 45 In matching to earlier waves of BHPS data it is important to be aware that variable names in the BHPS data set have slightly different formats e they are limited to eight characters e there is no underscore separating the wave prefix from the main part of the name e derived variables imputation flags weights and other special variables are not distinguished
86. hen extrapolated to TSMs and PSMs through a weight share method to create b_psnenbh_xw The detailed procedure for creating these weights as well as cross sectional individual response weights is described below The inclusion weight b_ psnenbh_li was calculated separately for a Northern Ireland and b England Scotland and Wales For each it has two components For Northern Ireland the first component consists of the BHPS Wave 11 cross sectional weight as this is the wave at which Northern Ireland first entered the BHPS This component encompasses a design weight post stratification and an adjustment for Wave 11 nonresponse The second component is derived from a model of the propensity to be issued at UKHLS Wave 2 conditional on being enumerated in BHPS Wave 11 This therefore adjusts for all the stages of dropout between BHPS Wave 11 in 2001 and UKHLS Wave 2 in 2010 Model covariates were taken from the Wave 11 household grid and household questionnaire This propensity was modelled as a single step from 2001 to 2010 because across wave response patterns varied greatly between the sample members There is no single BHPS wave since Wave 11 at which all the Northern Ireland sample members of those issued to UKHLS 65 responded and therefore no other survey instrument that can provide model covariates for all relevant sample members Similarly for England Scotland and Wales the first component consists of the BHPS Wave 9 longitudinal
87. hold reflecting the idea that the probability of observing the household is equal to or greater than the probability of observing the person in the household who has the greatest probability of being observed 63 3 7 3 4 BHPS Longitudinal Weights Four weights will be continued from BHPS with changed variable names The corresponding weight variables are wewght now called w_psnen91_Iw wiewtuk1 now called w_psnen01_Iw wirght now called w_indin91_Iw and wirwtuk1 now called w_indin01_lIw where w represents the most recent UKHLS wave These weights are based on Wave 18 BHPS longitudinal weights which account for the first wave household nonresponse the first wave within household individual nonresponse to enumeration or to an individual main questionnaire respectively and for individual nonresponse between the first wave and Wave 18 of BHPS The base weights which reflect continuous enumeration rlewght a BHPS variable name and continuous response to the main questionnaire rlrght a BHPS variable name since 1991 are used for creating weights for longitudinal analysis starting 1991 Note that such an analysis excludes Northern Ireland as it was added to BHPS in 2001 and will also exclude the Scotland and Wales boost samples that were added in 1999 Similarly the base weights which reflect continuous enumeration rlewtuk1 a BHPS variable name and continuous response to main questionnaire rirwtuk1 a BHPS variable name since
88. idowed mother s allowance or widowed pension e benefits severe disablement allowance disability living allowance war disablement pension attendance allowance carer s allowance incapacity 72 benefit income support job seeker s allowance national insurance credits child benefit child tax credit working tax credit maternity allowance housing benefit council tax benefit foster allowance guardian allowance rent rebate rate rebate employment and support allowance respond to work credit sickness and accident insurance in work credit for lone parents and pension credit and e other income sources educational grant trade union and friendly society payment maintenance or alimony payments from a family member not living together amount for rent from boarders or lodgers rent from any other property These personal income variables can be summed to obtain the total personal income Total household income can be computed from the personal total incomes of all household members Some of the income components can be missing More precisely there can be three types of missing cases e item nonresponse when individuals respond to the individual questionnaire but do not answer to some or all the questions on income components e individual nonresponse when individuals fail to respond to the individual questionnaire e household nonresponse when there is neither a household nor the individual questionnaire response
89. inal weight for BHPS GPS EMB but will have to use the weight for GPS EMB by looking at wave representing they should be able to find quickly which weight is best for their analysis 51 Table 24 Weight variables for analyses using household grid or household interview Wave s Wave ger Years starts Data source Analysis Weight representing from Household Household grid and or household ERW AENA Household grid and or n 2 household interview n_hhdenub_xw BHPS GPS and EMB Individual Household grid and or household interview Household grid and or n 2 household interview n_psnenub_xw BHPS GPS and EMB Household grid and or 1 2 household interview n_psnenus_lw GPS and EMB Household grid and or 1991 2 household interview n_psnen91_lw BHPS GB 1991 Household grid and or 2001 2 household interview n_psnen01_lw BHPS UK 2001 Household grid and or household interview BHPS GPS and EMB since 2010 2011 a_psnenus_xw 2 3 n_psnenub_lw 52 Table 25 Weights for analysis using adult main and proxy interviews Wave s Years representing 1 2 Wave starts from Data source Adult main and proxy interview Adult main and proxy interview Adult main and proxy interview BHPS Adult main and proxy interview BHPS GPS and EMB since 2010 2011 Adult main and proxy interview Adult main and pr
90. incapable of participating were contacted by the NatCen Multi Mode Unit MMU based in Brentwood The trained and briefed telephone interviewers at the MMU introduce themselves remind the sample members of the survey and ask whether they would be able to do the interview by telephone The purpose of the mop up was to increase participation among those who were hard to contact in person Analysis by NatCen indicates that the telephone mop up increased the overall household response rate for that period by about 3 percentage points for the EMB and by just less than 2 percentage points for the GPS This mop up was not conducted with the BHPS sample in Wave 3 since they are interview in the first year of each wave Towards the end of Wave 3 September 2012 a trial was conducted in two field areas in which an additional incentive was used at the re issue stage This was then rolled out across the sample from October and so covers the last quarter of Wave 3 In the implementation non responding households were reviewed by the NatCen Operations Department in Brentwood for re issue and possible re allocation to a different interviewer Households which had refused to participate in the initial fieldwork period but where the assessment was that this was a soft refusal were sent a re issue letter which mentioned an additional incentive if they participated during the re issue fieldwork period Other non responding households were sent a normal r
91. iological mother As a rule of thumb variables that are derived post field end on the suffix _dv and pointers to others in the household end on pno or pid they can therefore easily be identified on the data Note that a data file may offer alternative versions of a derived variable This is particularly true for derived variables that point to others in the household One set of variables e g variable names starting on hg has been computed based on information collected in the household grid module of the questionnaire during the interview The alternative version is computed post field after the information collected in the household grid has undergone extensive data cleaning See and compare for example w_hgbiom and w_mnpno for the person number of the respondent s biological mother in the household From Wave 2 onward proactive dependent interviewing was used to increase efficiency of data collection and lessen respondent burden Specifically information reported at an earlier time is fed forward to the respondent to personalize the question So rather than ask a question about current occupation with its complex probing by interviewers the question might say the last time you were interviewed you said you were specific occupation are you still specific occupation Feed forward variables are used at both the household and individual levels For example b_ff_hhsize feeds forward the household si
92. is it rented or does it come rent free Interviewer Instruction The text is what the interviewer reads F9 FOR HELP Options 1 Owned outright 2 Owned being bought on mortgage 3 Shared ownership part owned partrented Value labels 4 Rented S Rent free 97 Other Use Ask Hsownd Modules This question comes from Wave 1 ModuleHousehold_w1 Household Questionnaire Household Questionnaire module Figure 2 Mark up of question with looping from individual questionnaire Brfed Breastfeed Variable name amp Variable label Source UKHLS Leas Question may be asked multiple times Did you breastfeed vame even if only for a short time f About each resident child Options 1 Yes 3 No Values labels 3 Currently breastfeeding applies for children lt 5 in household only Use Ask BrFed Modules ModuleFertilityhistory w1 Fertility history module Question is from Wave 1 Sections Forti y Section1 individual interview Universe If LNPrnt gt 4 LPrnt 4 Parent of biological child Who is eligible to be asked this question And lf LChiv 7 Child resident And If resp is biological mother of resident child Resp is biological mother of resident ehiia And If resp is biological mother of resident child amp child lt 16 4 Resp is biological mother of resident child under 76 31 2 5 2 SUMMARY OF QUESTIONNAIRE MODULES About half of the questionnaire content is collected annually with additional modules collected at different in
93. isectensaaehienecipctenebatee 86 5 Citations and Acknowledgements cccceeeeeeeeeeeneeeeeeeeeeeeeeeennaeeeeeeeeeenes 86 5 1 Citation of the Data es Se Sats oe a Seat sta eM hE eh oe 86 5 2 Citation of the User Manual ctisicn basin cereal eet ened 86 5 3 ACknOwledgments cucuatinuateualaueiaitlaannialaanewalitne 87 6 References aidea ianen ice cnet tees a etapa tee eaea aaaea eanais 88 TABLE OF TABLES Table 1 Timing of data collection Start 600cceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeneeea 12 Table 2 Household response rates among eligible households Wave 1 16 Table 3 Individual response rates Wave 1 cc cece cece eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 17 Table 4 Household response rates Wave 2 eieccceceeeeeeeeeeeeeeeeeeeeeeeeeeeeeneeeeeeeeeeees 18 Table 5 Wave 2 Cross sectional individual adult response rates by sample origin 20 Table 6 Longitudinal individual re interview rates adults by sample origin Full EIAN AU NV AVG i E cccessvencews lt vunsued A 21 Table 7 Household response rates Wave 3 ceeccccceeeeeeeeeeeeeeeeeeeeeeeeeseeeteeeeeeeeeees 22 Table 8 Wave 3 cross sectional individual adult response rates by sample origin 24 Table 9 Longitudinal individual re interview rates adults by sample origin Full interview at Wave co oat hel cee vel eae vines ve ee ee ei ee een 25 Table 10 Household response rates Wave 4 ooo ccceeeeceeeneeeeee
94. isfaction and modules on environmental behaviours and voluntary work Wave 4 carries for the first time rotating modules on wealth and assets financial attitudes and behaviours and credit and debt Another highlight of the Wave 4 questionnaire is a one off module on leisure participation focussing on the Olympics 2012 which were held in and around London 27th July 12th August 2012 The Wave 4 self completion instrument for adults includes a focus on mental health and well being and gender role attitudes The EMB and related samples have repeat modules on remittances and on ethnic identity and Britishness 38 2 5 4 CHANGES TO THE QUESTIONNAIRE All survey instruments are carefully tested in a pilot so that any issues with question wording and routing interview flow and timings can be identified and fixed before the survey is rolled out to the main sample see Section 2 3 2 2 The survey instruments for a wave of data collection are then fixed for the entire fieldwork period and any question including its routing is repeated each wave of data collection as applicable However under certain circumstances changes to the questionnaire may be undertaken 2 5 4 1 Within a Wave At the end of the first six months of data collection in Wave 1 multiple variables were dropped because of the length of the interview e g cutting of the employment history module At the same time other modifications were made e g in question format Not
95. issued to the field for Wave 3 minus any found to have become ineligible 22 The EMB had the lowest household response rate of all the samples in the study The main reasons for the lower response appear to be a higher level of non contacts and untraced movers in this sample The EMB sample is concentrated in areas of high ethnic minority density which tend to be very urbanized areas particularly in London Residential mobility is higher among those living in these types of area which may contribute to the higher levels of untraced movers Non contact rates also tend to be higher in cities particularly in areas of tower blocks flats or apartments with a common often locked entrance The refusal rates for the EMB are similar to those of the GPS in Britain and only a couple of percentage points higher than the Living in Scotland sample Refusal rates were still higher for the former BHPS samples in Britain than they had been in the later years of the BHPS The refusal rate from Waves 14 18 were around 6 compared to 9 11 at Wave 3 of Understanding Society This may reflect some resistance after the change in fieldwork agency Table 8 shows the cross sectional response rates for adults in Wave 3 Where a household responded we have an individual level outcome for all adults Where a household did not respond we have assigned the household nonresponse outcome to the adults who were issued to that household At Wave 3 the telepho
96. ither log linear or predictive mean matching models For those variables where we have bracketed values rather than point information for example in the case of dividends and interests or when we have a priori information which allows us to bound the missing income variable we use interval regression The type of regression models used to impute missing explanatory variables depends on the level of measurement That is we use log linear regression for continuous variables and binary ordered and multinomial logit models respectively for dummy ordinal and unordered categorical variables The explanatory variables are a set of characteristics collected in the individual personal or household questionnaires The specification of the models varies by income variable but it generally includes most of the following variables e personal socio economic variables age sex self reported ethnic group indicator for respondent born in the UK marital status education level general health current subjective financial situation e personal income variables excluding the one used as the dependent variable e lagged income variables just for Waves 2 and 3 e household characteristics number of children in the household house tenure house type household size e job characteristics log number of hours normally worked per week log number of hours per months in a second job log years of job tenure permanent or temporary job occupation SO
97. ity of each respondent being selected through each sample and to continue being enumerated up to and including Wave 2 of the UKHLS Specifically the following samples are combined Sample 1 BHPS 1991 those living in England Scotland and Wales Sample 2 BHPS 1999 those living in Scotland and Wales Sample 3 BHPS 2001 those living in Northern Ireland Sample 4 GPS 2009 2010 those living in England Scotland Wales and Northern Ireland Sample 5 EMB 2009 2010 those living in HDAs in England Scotland and Wales Each respondent therefore had up to five chances to be selected into UKHLS depending on where they lived at the time that each sample was selected To reflect this for each person selected into any of the five samples we calculate their probability of being sampled in each of the samples above p j 1 5 and add these together to derive the overall inclusion probability pex 5 Pek Pjk j 1 The probability of respondent k entering UKHLS through sample j pjg is calculated as Pjk Sjk x hjk x lik x Vik where Sjx is the selection probability of respondent k in sample j 67 hj is the average household response rate at wave 1 for sample jin the country of residence of respondent k at the time of sample selection ij is the average individual continuous enumeration probability rate between Wave 1 for sample j and Wave 2 of UKHLS specific to respondent Ks country of residence at
98. k Prior to the first wave of the UKHLS Main Survey there were two small pilot studies and a dress rehearsal A cognitive pilot of 70 individuals was conducted March April 2008 to test screening and other questions relevant to the ethnicity strand A translation pilot was conducted in June 2008 50 interviews were carried out using Bengali and Punjabi translations of the questionnaire to see if there were problems with the operation of the translation program or problems with interviewing with the translated instruments A run through of all data collection instruments and procedures in 100 households called a dress rehearsal took place August September 2008 A pilot or run in for Wave 2 tested all instruments and data collection procedures For this wave the data collection also focused on assessing any problems with integrating members of the former BHPS sample component which includes a small segment conducted by telephone interviews In all 237 households were issued Of these 91 were households interviewed in the Wave 1 pilot The BHPS sample component was represented by households that were part of the BHPS between 1997 and 2001 the European Community Household Panel ECHP Households for which we had a telephone number were issued to telephone interview to test the telephone interview instruments and procedures The Wave 2 pilot took place September October 2009 The Wave 3 dress rehearsal pilot took place September November 2010
99. l probabilities are simply the inverse of the design weight a_hhdenus_xd the calculation for which is described above Two inferred probabilities are calculated for respondents actually selected into the BHPS samples the probability of selection through GPS s4x and the probability of selection through EMB ssx 54 iS equal to the number of eligible residential household selected through GPS divided by the number of households in the population according to Census 2011 These probabilities are country specific Ssg depends on the postcode sector of residence in 2009 10 as described in Berthoud et al 2009 and on the ethnic composition of the household and the selected person For BHPS sample members we assigned the EMB postcode sector probability of the postcode of their address at wave 2 of UKHLS 2010 If the sector was in a LDA then Ssg 0 Next for BHPS sample members in high density ethnic minority areas we need information on the ethnic group composition of the household We used information from variables race and racel as recorded in data file xwavedata at the time of BHPS Wave 18 and b_racel as recorded at Wave 2 of the UKHLS For children under 16 for which this information was missing we inferred it from their biological parents The information collected in the BHPS does not perfectly match the ethnic group classification used for EMB selection but after consulting experts on migration 68 research we achieved a
100. le for 4 household level all enumerated individuals 3 Adult proxy and main interview 2 Adult main interview only no proxy 1 Adult or youth self completion interview 50 For example if in one cross sectional model for Wave n you use questions from the proxy and full interview as well as from the self completion then the correct weight will be n_indscus_xw the weight for the self completion questionnaire as its level 1 is lower than the level for proxy and full interview 3 The following tables list the weight variables The list has been broken into separate tables so the user can go quickly to the data source for the planned analysis and then select the particular relevant weight for example cross sectional vs longitudinal Each table focuses on a major data source and has the weight variables used for cross sectional and longitudinal analyses Please also note that weights are defined for particular sample components Wave prefixes refer to a specific wave a_ or b_ or to waves in general n_ Here is an example a longitudinal weight for BHPS EMB GPS represents waves 2 and 3 but the first time one can find it is in Wave 3 release Thus waves representing is 2 3 but the wave it starts from is 3 We would indicate that the above weight represents waves 2 all waves starting from 2 Where is this useful If a user wants to analyse longitudinal data starting with Wave 1 then they cannot use the combined longitud
101. les samples or Wave 8 i e for the NIHPS all collected in 2008 Once more we see that there is a higher re interview rate in the Northern Ireland samples than in Great Britain The lowest re interview rate is in the EMB sample largely due to a higher level of non contacted households or households who moved but could not be traced Interestingly the re interview rate was higher in the General Population GB sample than in the three samples that made up the former BHPS GB samples Overall in the Waves 1 and 2 data pairs of observations are available for 45 836 adults If proxy and telephone interviews are included this increases to 47 282 adults For more detail please see the working paper on nonresponse and attrition Lynn Burton et al 2012 19 Table 5 Wave 2 Cross sectional individual adult response rates by sample origin UKHLS GP sample EMB Former BHPS UKHLS UKHLS Living in Living in Living in GB NI Britain Seatiand Wales es Total Full interview 32 381 1 770 4 978 6 140 1 461 1 651 2 008 50 389 60 8 62 3 46 3 61 6 58 1 59 6 71 7 59 4 Proxy interview 2 22 87 615 253 49 86 58 3 870 5 1 3 1 5 7 2 5 2 0 3 1 2 1 4 6 Telephone interview 202 66 58 326 7 a 7 2 0 2 6 2 1 7 0 4 Other non interview 1 184 126 472 200 58 64 107 2 211 2 2 4 4 4 4 2 0 2 3 2 3 3 8 2 6 Refusal 2 104 218 511 341 92 114 133 3 513 4 0 7 7 4 8 3 4 3 7 4
102. mpletion Adult self completion BHPS 3 Adult self completion BHPS GPS and EMB 2 Adult self completion b_indscus_xw b_indscbh_xw n_indscub_xw n_indscus_lw Table 29 Weights for analysis using youth self completion Wave s Years representing 1 2 2 Wave starts from Data source Youth self completion Youth self completion Youth self completion BHPS 3 Youth self completion BHPS GPS and EMB 55 Analysis Weight a_ythscus_xw b_ythscus_xw b_ythscbh_xw n_ythscub_xw Table 30 Weights for analysis using nurse health assessment data Wave s Years representing Wave starts from Data source 3 Nurse visit with Wave 2 to Wave n full interviews GPS BHPS Nurse visit with 1991 to Wave n full interviews BHPS Analysis Weight n_indnsub_lw n_indns91_lw Note Other weights for use with the health assessment data that were collected at waves 2 and 3 are available with the separate health assessment data release see Section 4 3 2 and are fully documented in the associated user guide see McFall Petersen et al 2014 These two weights are documented here as they will be released with the main data release each year Table 31 Design and inclusion weights Wave s Analysis Years level representing Household R s Wave starts from Data source
103. n conditional on household response the post stratification adjustment The individual nonresponse correction conditional on household nonresponse is modelled at three levels e For adult respondents age 16 or older who either completed the main interview or for whom a proxy interview was completed for a_indpxus_xw e For adult respondents age 16 or older who completed the main interview only for a_indinus_xw and a_ind5mus_xw e For respondents aged 10 or older who completed and returned the self completion questionnaire for a_indscus_xw and a_ythscus_xw Note that the same model was used for respondents regardless of whether they were selected into GPS or EMB that the response propensity is assumed to not depend on whether respondents received the Extra 5 minutes or not and that conditional on age present in the model the response to self completion is assumed to have the same predictors for adults and youth this assumption allowed modelling the response in each country separately which wouldn t otherwise be possible for youth sample The individual level response conditional on household response was modelled using backward stepwise logistic regression separately for England Wales Scotland and Northern Ireland The four models were implemented for each of the three levels described above The predictors used in the models include all the predictors used for the household level nonresponse models and individu
104. n panel Data from the IP has been released through the UK Data Service SN6849 Understanding Society Innovation Panel Waves 1 5 2008 2013 http www esds ac uk findingData snDescription asp sn 6849 5 CITATIONS AND ACKNOWLEDGEMENTS Any publication whether printed electronic or broadcast based wholly or in part on the Understanding Society data collection provided by the UK Data Service must be accompanied by the correct citation and acknowledge the Institute for Social and Economic Research as the data provider and the UK Data Service as the data distributor The acknowledgement which gives credit to sponsors or distributors is not a replacement for a proper citation 5 1 CITATION OF THE DATA The format for bibliographic references is as follows University of Essex Institute for Social and Economic Research and National Centre for Social Research Understanding Society Wave 1 4 2009 2013 computer file 6 Edition Colchester Essex UK Data Service distributor December 2014 SN 6614 http dx doi org 10 5255 UKDA SN 6614 6 5 2 CITATION OF THE USER MANUAL The User Manual is to be cited as follows Knies Gundi ed 2014 Understanding Society UK Household Longitudinal Study Wave 1 4 2009 2013 User Manual Colchester University of Essex 86 5 3 ACKNOWLEDGMENTS People who participated in writing sections of the documentation for this or prior releases include in alphabetical order Randy Banks
105. n zero only for the sample reflecting the respondent s country of residence at that time and 70 is zero for all other countries for that time point Thus for a person who has always lived in England the non zero selection probabilities will be those for England in 1991 and for England for 2009 2010 For those who immigrated to England in 2008 all selection probabilities will be zero except for the probabilities for England in 2009 2010 The total probability pex is therefore the sum of all the above probabilities It reflects multiple possible ways of selection through different samples and continuous enumeration of a respondent up to and including Wave 2 of the UKHLS The inclusion weight b_psnenub_li is the inverse of the total probability This weight is not scaled and will serve as a base weight for all the weights that combine BHPS GPS and EMB samples The weight is defined for all OSM respondents enumerated at Wave 2 of the UKHLS 3 7 3 8 W2 Cross sectional Weights for Total Sample BHPS GPS and EMB Components Two cross sectional weights are created for Wave 2 of the UKHLS for the combined samples of BHPS GPS and EMB The first is the cross sectional individual enumeration weight which is created through the weight share technique where the inclusion weight b_psnenub_li is weight shared to TSMs and those OSMs that have missed at least one wave between selection and Wave 2 of the UKHLS The weight share is done in a standar
106. naire GHQ See for example C_ypsqda to c_ypsday on data file c_youth or d_scghq2_dv on data file d_indresp Most added value variables i e variables that are produced post field are clearly marked in the data by suffixes weights are shown by the suffixes _Iw or xw most derived variables are shown by the suffix _dv and pointers to other members in the household typically end on pno or pid The prefix ff_ following the wave prefix shows variables that were fed forward from previous waves to route respondents appropriately in the script We have attempted to keep the names of variables that came from the BHPS the same for the convenience of analysts but this has not always been possible given overruling naming conventions in the UKHLS To identify corresponding variables in the BHPS analysts should consult the BHPS documentation https www iser essex ac uk ohps documentation volb index html 42 Note that the variable name does not change over time so long as the underlying question does not change substantially Analysts are advised however to carefully read the variable notes in the online documentation to keep track of any definitional changes or changes in the code frame that may impact study results An example is the derived variable w_hiqual_dv which from Wave 2 onward includes information from the BHPS and where the code frames on the BHPS and Understanding Society do not perfectly align
107. ncome where we convert amounts reported net to gross where gross is not reported using a deterministic model based on the tax and national insurance system In computing total personal income it is assumed that all other sources are reported gross or are not subject to taxation The imputation of the missing income sources in the individual questionnaire permits the computation of total earnings i e the sum of income from the first and second job and total income the sum of earnings plus investment income pensions and benefits for all adult non proxy respondents 3 8 4 ITEM NONRESPONSE FOR INCOME VARIABLES IN THE PROXY QUESTIONNAIRE The only income variables reported in the proxy questionnaires are the total gross earnings and total gross income We impute missing values for these two variables again using hot deck methods The imputation is based on the sample of persons responding to the individual questionnaire together with the sample of individuals for whom a proxy questionnaire is available Since for non proxies total earnings cannot be higher than total income we impose additional restrictions to the brackets such that this relationship also holds in the case of the imputed earnings and income for proxies The explanatory variables used for the imputation of gross earning and gross income for proxies are those used in the imputation of the income variables for non proxies and listed above 3 8 5 INDIVIDUAL NON RESPONDENTS WI
108. nd Refusals are expected to be higher at the second wave of a longitudinal study than at subsequent waves The higher than expected refusal rate for the former BHPS sample particularly those in Great Britain may be due to the aforementioned change in the name and logo of the study as well as the change in fieldwork agency and thus for most households a change of interviewer Table 5 shows the Wave 2 cross sectional response rates for adults Where a household responded we have an individual level outcome for all adults Where a 18 household did not respond we have assigned the household nonresponse outcome to the adults who were issued to that household From this we can see for example that we were not able to interview 7 229 adults in the UKHLS GPS in Great Britain because they were residing in households who refused to participate at Wave 2 In the Great Britain samples of the former BHPS there is a relatively small group of households who only give telephone interviews On a longitudinal study such as the UKHLS researchers are typically interested in having pairs of observations on the same individual to investigate individual level change over time Table 5 takes as the baseline all those who gave a full interview at the previous wave and shows their outcome at Wave 2 For the former BHPS samples the previous wave was Wave 18 i e for the Living in Britain sample Wave 10 i e for the Living in Scotland and Living in Wa
109. ne interview was conducted using the same instrument as the face to face instruments but with slight changes to reflect the aural rather than visual delivery of the questionnaire e g there were no showcards Apart from these changes the content of the telephone questionnaire was the same as the face to face questionnaire and so these interviews are classified as full interviews below In a longitudinal study such as the UKHLS researchers are typically interested in having pairs of observations on the same individual to investigate individual level change over time Table 9 takes as the baseline all those who gave a full interview at the previous wave and shows their outcome at Wave 3 Once more we see that there is a higher re interview rate in the Northern Ireland samples than in Great Britain The lowest re interview rate is in the EMB sample largely due to a higher level of refusal households or households who moved but could not be traced Unlike at Wave 2 the re interview rate was lower in the General Population GB sample than in the three samples that made up the former BHPS GB samples The lower re interview rate at Wave 2 for these BHPS sample households may reflect a blip caused by the changes in fieldwork agency the name and branding of the survey Overall in the Waves 2 and 3 data pairs of observations are available for 45 903 adults If proxy interviews are included this increases to 49 708 adults 23 Table
110. ned in your own life that has been especially important to you Can you please tell me anything that has happened to you or your family over the past year that has stood out as important The respondent could give up to four answers The answers were recorded verbatim and manually coded for type of event and its subject 29 2 5 DOCUMENTATION OF THE SURVEY INSTRUMENTS The text of the questionnaires in PDF format is part of the documentation provided through the UK Data Service Questionnaires can also be found https www understandingsociety ac uk documentation mainstage questionnaires There are household and individual questionnaires and the adult and youth self completion instruments The instruments are an important source of information about the wording of individual questions who was asked and what questions precede and follow Most of the interview is conducted with a computer assisted personal interview CAPI The CAPI instrument governs the flow of questions and recording of answers but it is not convenient for documentation On the study website we present the questionnaire in PDF format Similar to other PDF documents the text of the questionnaire can be searched for specific words such as variable names or words in questions The PDF self completion instruments correspond to the way they appeared to participants except they have been annotated with variable names The principal adult questionnaires are organize
111. ng benefit 23 council tax benefit offset against council tax 30 foster allowance guardian allowance 31 rent rebate NI only 32 rate rebate NI only offset against rates 33 employment and support allowance 34 return to work credit 36 in work credit for lone parents 37 other disability related benefit or payment 39 income from any other state benefit not asked in wave 1 Deduction component 9 council tax w_dep9ctax Calculated from council tax band and local authority district GB Estimates for Northern Ireland rate charges have not been included The variables w_dep9ctax and w_hhnetinc3 are only available in the Special Licence version of the data set 3 9 EXAMPLE CODE FOR MATCHING FILES We are including six examples of common data management tasks useful in analysing the data Each task is illustrated with code for Stata Because Stata is case sensitive we have not displayed file and variable names in upper case but in lower case To run these programmes on UKHLS data files please remove the _ip suffix from the data file names Statements beginning with are comments 80 hold Longitudinal Study Wave 1 4 2009 2013 The first task is distributing household level information to individual level We can do this by merging household level file such as w_hhresp with an individual level file such as w_indresp within the same wave see Figure 3 Figure 3 Stata Code Distributing household level information
112. nnaires and the module and variable views in the online documentation system Many of the basic non derived variables can be learned about directly from the questionnaires As was shown in Figure 2 the questionnaire has much useful information Please note that in the questionnaire the variable name does not have the wave prefix It also shows the brief variable label text of the question source of the question and value labels Show cards to help the respondent in answering are also marked as part of the questionnaire You can go back and forth from the question view to the variable view 3 2 2 VARIABLE NAMING AND LABELLING CONVENTIONS Most variables have a mnemonic name Variables begin with a prefix designating the wave of data collection a_ for the first wave b_ for the second wave in this user guide we have used w_ to denote waves in general To ease identification of groups of variables a number of additional general naming conventions have been applied For instance following the wave prefix information from the self completion interview with adults starts with the prefix sc information from the interview with young adults with the prefix ya and information from the child development module with the prefix cd Similarly we have attempted to include in the variable name the acronym of well known instruments such as the Strengths and Difficulties Questionnaire SDQ or the General Health Question
113. or the first three components are the same except the EMB sample and the GPC sample have an Extra 5 minutes of questions specifically relevant to ethnic minority communities e g ethnic identity and remittances In Waves 2 and 3 Understanding Society augmented survey questions with direct health assessments and the collection of blood samples The Health Assessment data can be accessed through the UK Data Service SN7251 In addition there is a separate survey the Innovation Panel IP which is fielded in the year before the main survey It tests varying measurement issues and its instruments are somewhat different from the main survey The IP can be accessed through the UK Data Service SN 6849 2 2 SAMPLE DESIGN The Understanding Society sample consists of a new large General Population Sample GPS plus four other components the Ethnic Minority Boost EMB sample the General Population Comparison GPC sample the former BHPS sample and the UK Innovation Panel IP sample The design of all five components is described in more detail in an Understanding Society working paper see Lynn 2009 The Innovation Panel is prepared as a separate study which can be accessed via the UK Data Service SN 6849 The GPS is based upon two separate samples of residential addresses of England Scotland and Wales and for Northern Ireland The England Scotland and Wales sample is a proportionately stratified equal probability clustered
114. ot divided into modules From Wave 3 onwards the self completion content was carried as CASI modules where the participant would answer the questions using the lap top Table 14 summarizes the content in waves 1 to 4 Table 14 Summary of adult self completion content Wave 1 Wave 2 Wave 3 Wave 4 GHQ 12 X X X X Sleep X Environmental X X attitudes and beliefs Neighbourhood X X belonging and participation Trust Life Satisfaction gt lt gt lt gt lt x lt x lt x lt Short Warwick Edinburgh Mental Well Being Scale Attitudes to risk X Partnership X X relationship quality activities happiness SF 12 in adult X X X interview Health and Disability module Identity Alcohol consumption Feelings of control Social support XIXI X X X lt Gender role opinions Big 5 X Personality Sex Orientation X Identity X age 16 21 Family Support X age 16 21 and Activities Bullying X age 16 21 Family X age 16 21 Relationships 36 Smoking X age 16 21 Behaviour Alcohol related X age 16 21 Behaviour Drug Use X age 16 21 Social Networks X 3 Best Friends Non Co X X resident Relationship Child X about X about Development child age 3 5 child age 3 5 or8 ors Parenting Style X refersto X
115. oud Fumagalli et al 2009 e A wide range of indicators from Census 2001 and the most updated version of neighbourhood statistics as of summer 2011 linked separately for England Wales Scotland and Northern Ireland see below The household nonresponse correction weight was calculated as the inverse of probability from the above model This weight was multiplied by the household design weight to create the Wave 1 household level weight The design effect was estimated using this weight No truncation was necessary The obtained weight was scaled to a mean of 1 and was named a_hhdenus_xw 60 Neighbourhood Statistics For England and Wales the information was linked at Middle Layer Super Output Area MSOA or Lower Layer Super Output Area LSOA levels and was obtained from http neighbourhood statistics gov uk The examples of linked information obtained from Census 2001 include the proportions in the MSOA of employed retired outright property owners travellers to work using different types of transport single household members households with one car people with different types of qualification and professional occupation among others Other linked information includes 2010 information on multiple deprivation indexes on crime instances 2009 information on inflow and net change of neighbourhood population the proportion of different allowance claimants and 2008 information on hospital admissions and energy consumption For S
116. oxy interview BHPS GPS and EMB since 2010 2011 Note The weights listed in Table 25 through 53 Analysis Weight a_indpxus_xw b_indpxus_xw b_indpxbh_xw n_indpxub_xw n_indpxus_lw n_indpxub_lw Table 30 all apply to analysis at the individual level Table 26 Weights for analysis using adult main interviews Wave s Years representing 7 2 2 n 1 1991 2001 2 Table 27 Weights for analysis using adult Extra 5 minutes interview Wave s Years representing 1 1 Wave starts from Data source Adult main interview Adult main interview Adult main interview BHPS 3 Adult main interview BHPS GPS and EMB 2 Adult main interview 2 Adult main interview BHPS Great Britain 2 Adult main interview BHPS UK Adult main interview 3 BHPS GPS and EMB since 2010 2011 Wave starts from Data source Adult extra 5 minutes interview 3 Adult extra 5 minute interview 2 Adult extra 5 minutes interview Analysis Weight a_indinus_xw b_indinus_xw b_indinbh_xw n_indinub_xw n_indinus_lw n_indin91_lw n_indinO1_lw n_indinub_lw Analysis Weight a_ind5mus_xw n_indS5mus_xw n_ind5mus_lw Table 28 Weights for analysis using adult self completion Analysis Weight Wave starts Data source Adult self completion 54 a_indscus_xw 1 Adult self co
117. pidp Cross wave person identifier pid Cross wave person identifier for continuing BHPS sample members w_hidp Household identifier w_pno Within wave person number in the household mpid fpid Cross wave identifier of natural mother father w_ppid w_sppid Cross wave identifier of current cohabitee spouse w_ivfio Individual response outcome w_psu Primary sampling unit w_strata Sampling strata w_hhorig Sample origin w_psnenus_xw Weight for household grid and household interview w_indinus_xw Weight for individual interview w_xtr5min_dv Identifier for the Extra 5 minutes analytical sample w_country Country of the UK w_gor_dv Government Office Region w_hhsize Household size w_hhtype_dv Household type w_tenure_dv Housing tenure w_sex Respondent sex also see w_sex_cr w_dvage Respondent age also see w_age_cr w_marstat Legal marital status w_marstat_dv De facto marital status w_ukborn Born in the UK and UK country of birth w_racel_dv Ethnicity w_jbstat Current economic activity employment status w_jbhas Did paid work last week w_finhmngrs_dv Gross household income in past 30 days w_jonssec8_dv Social class NS SEC 8 category version w_jbsoc00 Current occupation SOC2000 w_fenow Still in further education w_qfhigh Highest educational qualification 44 w_health Long standing illness or impairment
118. ps all members of target ethnic groups were deemed to be members of the EMB sample including children All persons of other ethnic groups are not EMB sample members They will be interviewed as temporary sample members for so long as they remain co resident with at least one EMB sample member The overall sampling fractions combine a the probability of sampling the sector b the fraction of addresses selected within the sector and c the probability of a household being retained following the application of the random selection mechanism described above 2 2 4 FORMER BHPS SAMPLE The sample issued at Wave 2 consisted of all members from the BHPS sample who were still active at Wave 18 of the BHPS and who had not refused consent to be issued as part of the Understanding Society sample It should be noted that the BHPS sample contains different components including the original sample first selected in1991 boost samples in Scotland and Wales first selected in 1999 and a Northern Ireland sample selected in 2001 For further details of the BHPS sample see section IV of the BHPS User Guide http www iser essex ac uk bhps documentation vola vola html 2 2 5 SAMPLE STATUS AND FOLLOWING RULES There are three possible sample statuses Original Sample Members OSMs Temporary Sample Members TSMs and Permanent Sample members PSMs The definitions are as follows 2 2 5 1 Original Sample Members OSMs All members of Understan
119. psnenub_lw For each instrument a logistic regression is run to predict response in both Waves 2 and 3 with predictors from the Wave 3 household questionnaire and household grid The models are restricted to adults The estimated probabilities were then inversed multiplied by c_psnenub_Iw and scaled to a mean of one 71 From Wave 4 onwards each of these longitudinal weights are based on the equivalent weight from the previous wave adjusted by the reciprocal of the predicted value from a logistic regression model of wave on wave response and scaled to a mean of one For example a logistic regression model of enumeration at Wave 4 was based on sample members with a non zero value of c_psnenub_Iw and the reciprocal predicted values were multiplied by c_psnenub_Iw and then scaled to produce d_psnenub_Iw 3 7 3 10 Cross sectional Weights for Waves 3 and 4 Starting at Wave 3 cross sectional weights were created for the combined sample that includes the BHPS GPS and EMB components The cross sectional enumeration weight n_indenub_xw was created based on the longitudinal enumeration weight n_indenub_lw via the weight share method The weight was shared from OSMs with nonzero longitudinal enumeration weights to TSMs and PSMs except those who were selected at Wave 1 and OSMs with a longitudinal weight of zero For the household cross sectional weight n_hhdenub_xw the lowest cross sectional enumeration weight among adults n_indenub_x
120. rates for the former BHPS samples were higher than the Understanding Society samples The household response rate was lower in the former Living in Scotland than the other BHPS samples Table 10 Household response rates Wave 4 UKHLS GP EMB Former BHPS sample Living Living See a n von in NIHPS Total Britain Wales Fully 12 897 603 1 505 2 579 594 666 789 19 633 responding 62 0 56 9 46 9 68 9 63 2 62 7 67 9 61 4 Partially 3 958 233 819 600 146 216 209 6 181 responding 19 0 22 0 25 5 16 0 15 5 20 3 18 0 19 3 All 16 855 836 2 324 3 179 740 882 998 25 814 responding 81 0 78 9 72 4 84 9 78 7 83 0 85 9 80 7 Non 858 63 244 113 41 39 45 1 403 contact 4 1 5 9 7 6 3 0 4 4 3 7 3 9 4 4 Untraced 754 50 200 87 33 39 46 1 209 mover 3 6 4 7 6 2 2 3 3 5 3 7 4 0 3 8 Refusal 2 134 96 392 325 100 94 53 3 194 10 38 9 1 12 2 8 7 10 6 8 9 4 6 10 0 Other non 202 15 50 38 26 8 20 359 interview 1 0 1 4 1 6 1 0 2 8 0 8 1 7 1 1 Total 20 803 1 060 3 210 3 742 940 1 062 1 162 31 979 Base is all households issued to the field for Wave 4 minus any found to have become ineligible The EMB continues to have the lowest household response rate of all the samples in the study The main reason for the lower response continues to be a higher level of non contacts
121. re given a design weight of 0 while non EM persons in the GPS are given the household design weight The weights for EM persons are adjusted for their dual probability of being part of GPS and EMB Individual level design weights for those eligible to answer the Extra 5 minutes is similar to the above design weight but differs in the following ways It adjusts for the fact that the GPS Comparison Sample is only 1 45 of the GPS original sample that all EM members in low density areas were administered the Extra 5 minutes and that EM members in high density areas had a chance to be selected into either the GPS Comparison sample or the EMB Similar to the above weight non EM persons were assumed to have a chance to be part of the GPS Comparison Sample only and not part of the EMB Additionally we provide GPS design weights a_hhdengp_dw and a_indengp dw These weights are valid only for sample members selected through the GPS and adjust for oversampling in Northern Ireland and for subsampling within households from multiple dwellings per address or multiple households per dwelling Household level Nonresponse Adjustment Household level nonresponse adjustment is more complex than in other surveys given the large number of households which were selected as part of the EMB with unknown eligibility Households who were selected as part of the EMB sample were screened on whether they contain at least one member of a relevant ethnic minori
122. re with questions on total earnings and total income as well as other variables Finally for individuals in responding households for whom neither the personal nor the proxy questionnaire is available we impute only the total personal income This 73 is not directly included in the data set but is used in the imputation of total household income Based on these imputations we can compute total personal and household income for all individuals belonging to responding households For each income variable for which amounts are imputed there is a separate imputation flag variable with a suffix _if instead of _dv indicating whether the variable is imputed In most cases this takes the value 1 if imputed and 0 if not but in the case of the following variables it shows the proportion of total income imputed w_fimngrs_if w_fibenothr_if and w_fihhmngrs_if In the income data file there may be multiple receipts of income from the same source For example a respondent may have multiple pensions from a previous employer These are summed and imputed as such and the imputed values are in the variable w_frmnthimp_dv As a consequence the variable w_frmnthimp_dv for the first income receipt from a given source is equal to the total value of all receipts from that source while it is set to inapplicable for the second and subsequent receipt 3 8 2 IMPUTATION PROCEDURES Missing income values in Understanding Society are replac
123. reasonable fit to the EMB classification While the selection into EMB depended on whether a person belongs to the group of interest or not see section on EMB the selection probability depends on the household ethnic group composition at the time of selection considered 2010 for this purpose If the household did not have any ethnic minority member of interest we set ss 0 If the household had some ethnic minority members of interest the probability was calculated as the product of postcode sector selection probability and the largest screening probability among ethnic group members All ethnic minority members were assigned this selection probability and members of any other groups either British or those not of selected in EMB were assigned the value of zero Additionally postcode sectors from which the Bangladeshi boost was selected were identified and those of Bangladeshi origin were assigned a probability reflecting the additional chance of being selected into Bangladeshi boost see Berthoud Fumagalli et al 2009 Average household response rate at Wave 1 h We treat nonresponse correction as having three components the first of which is the household response probability at the first wave when the sample is selected Here real and inferred probabilities are calculated in the same way We divide the number of responding households by the number of eligible residential households in the sample these are country and time
124. rnings interests and dividends and their predictors are imputed jointly using chained equations ICE Hot deck is used to impute income sources for proxies and non respondents missing values in the variables defining the categories are set equal to their median All variables are imputed as reported except for wages and self employment income where we convert amounts reported net to gross where gross is not reported using a deterministic model based on the tax and national insurance system In computing total personal income it is assumed that all other sources are reported gross or are 74 not subject to taxation Net income estimates are also included in the data set see Section 3 8 6 In what follows we outline briefly the characteristics of the main cross sectional imputation methods used Chained equations ICE It is a multivariate imputation method used to impute a set of variables jointly We used it to impute the main income variables for respondents plus their predictors ICE allows for interdependence between income and auxiliary variables by considering univariate models estimated separately and sequentially through stochastic imputation see van Buuren Boshuizen et al 1999 and Ragunathan Lepkowski et al 2001 This method has been already used in some major household panel surveys such as the ECHP The ICE starts by considering the following recursive triangular system of imputation equations Yi Qo
125. robability of main response at Wave 2 conditional on enumeration at Wave 2 The adjustment is the inverse of the response propensity predicted by a separate logistic regression model based just upon all adults and inferred to rising 16 year olds using covariates from the Wave 2 household questionnaire and household grid The base weight for rising 16 year olds correction is continuous enumeration since 1991 b_psnen91_Iw for the BHPS 1991 main response weight b_indin91_Iw and is the BHPS 2010 longitudinal enumerated person weight b_psnenbh_lIw see next section below for the BHPS 2001 main response weight b_indin01_Iw The main 64 response weight for each rising 16 year old is then scaled by a constant factor so that the ratio of rising 16 year olds to older adults among main questionnaire respondents equals the equivalent proportion among all enumerated respondents The weights b_psnen91_lIw b_psnen01_Iw b_indin91_lw and b_indin01_lIw are calculated by multiplying the respective BHPS Wave 18 weight and the adjustment and are scaled to one Starting at Wave 3 longitudinal weights for BHPS are created based on the previous wave longitudinal weights For longitudinal enumeration weights n_psnen91_ Iw and n_psnen01_Iw enumeration in Wave nis predicted among adults having positive longitudinal weight in previous wave Enumeration is modelled using logistic regression with covariates from the household questionnaire and household
126. rom ISER to all sampled addresses addressed to The Occupier together with a small leaflet outlining the purpose of the survey Then the interviewer called within a week of the mailing At the end of the first interview all participating households received a more detailed brochure giving further information about the survey and thanking respondents for participating 14 A minimum of six calls is made at each sampled address before it is considered a non contact Interviewers are encouraged to make further calls if possible If there is a potential for success a special conversion letter is sent to households which had refused to participate or had not been contacted Post interview quality control is carried out with a telephone recall on 10 of all completed interviews Interviewers upload their work daily including information about all the calls they have made whether or not there was any response This information is collated by NatCen to construct a weekly field progress monitor report for ISER During the second year of Wave 3 2012 a telephone mop up was introduced This was started in April but also covered the sample from January March The aim of this was to contact adults who could not be contacted by face to face interviewers during the main fieldwork period Adults in households that were non responding in the main fieldwork period except those who had adamantly refused or were deemed to be mentally or physically
127. s which are not available to us from the questionnaire It is hoped that a later release of the data will provide further is estimates of some of these other components A technical working paper providing more information the derivation of these net income variables is in preparation Data are included at the household and individual level The individual level the total estimated net income is w_netinc1 At the household level there are two variables w_hhnetinc1 which is the sum of net incomes from all household members including proxies and within household non respondents and w_hhnetinc3 which w_hhnetinc1 less council tax liability Council tax benefit which is included in w_hhnetinc1 is netted off from the council tax liability and not included in total income In addition to these summary variables there are estimates of the different income components following the structure used by HBAI These are as follows Component 1 Labour income w_inc1lab w_paynu_dv net usual pay w_inc1labem w_seearnnet_dv net self employment income w_inc1labse w_jb2pay_dv gross pay in second job less estimated tax and national insurance w_inc1labj2 Component 2 Miscellaneous income w_inc2misc Receipts reported in income record where w_ficode equals 24 educational grant not student loan or tuition fee loan 27 payments from a family member not living here 38 any other regular payment not asked wave 1 Component 3 private bene
128. sample of addresses selected from the Postcode Address File Northern Ireland has an unclustered systematic random sample of addresses selected from the Land and Property Services Agency list of domestic addresses 2 2 1 GENERAL POPULATION SAMPLE The sample for England Scotland and Wales was selected in two stages The first stage was to select a sample of postcode sectors as the primary sampling units PSU s The second stage was to select addresses within each sampled sector Prior to selection any postcode sector with fewer than 500 residential addresses was grouped with an adjacent sector and thereafter treated as a single sector The list of all sectors was then sorted into twelve geographical strata consisting of ten regions in England plus Scotland and Wales as separate strata Within each of the twelve strata sectors were sorted into three sub strata based upon the proportion of household reference persons classified as non manual workers from 2001 Census data Within each of the 36 sub strata sectors were then sorted into three further sub divisions based on population density households per hectare and within each of the 108 resultant sub divisions sectors were listed in order of ethnic minority density From the sorted list a systematic random sample of 2 640 sectors was selected with probability proportional to the number of residential addresses in the sector These sectors were then allocated systematically to 24 monthly sampl
129. scbh_xw each consist of the cross sectional individual enumerated weight with an additional adjustment for nonresponse to the relevant instrument conditional on household response These adjustments were based on logistic regression models with both 66 individual level and household level covariates taken from responses to the UKHLS Wave 2 household grid and household questionnaire The BHPS cross sectional household weight b_hhdenbh_xw is set equal to the minimum cross sectional person enumerated weight b_psnenbh_xw amongst adults in the household 3 7 3 6 Combined Sample BHPS GPS and EMB Weights Starting at Wave 3 we provide both cross sectional and longitudinal weights for combined analysis of respondents in all samples including BHPS GPS and EMB The weights are based on the inclusion enumeration weight which accounts for combined probabilities of being selected in any of the continuing samples of BHPS GPS and EMB at the time each was selected and continuously being enumerated up to and including Wave 2 of UKHLS We first explain the calculation of the inclusion enumeration weight the development of cross sectional weights for Wave 2 based on it then the calculation of longitudinal weights and finally the cross sectional weights that are created from Wave 3 onwards 3 7 3 7 BHPS GPS and EMB Inclusion Enumeration Weight To combine samples from the BHPS GPS and EMB sample components we calculate the joint probabil
130. sets of weights are not identical for these three analysis bases reflecting differences in data collection Considering the complexity of the study design weights should be selected carefully following the advice provided below 49 The first part of this section covers the purpose of the weights and how to use the naming conventions for the weight variables to interpret and select the different weight variables from among a complex assortment This is the most important section for a user wanting to select the appropriate weight for a planned analysis This section is followed by the technical details of how weights were calculated If your aim is to generalise to the UK population unweighted analyses should be avoided For advanced users who want to model nonresponse in their own way we provide design weights see below which adjust the sample for unequal selection probabilities Note that adjusting for the first wave nonresponse is different from adjusting for attrition and requires variables which have values for both responding households and never responding households Note that a number of longitudinal weights are provided corresponding to the year of sample selection BHPS since 1991 BHPS since 2001 GPS and EMB since 2010 2011 and combined BHPS GPS and EMB since 2011 2012 Cross sectional weights starting at Wave 3 will be based on combined BHPS GPS and EMB samples 3 7 1 SELECTING THE CORRECT WEIGHT FOR YOUR ANALYSIS
131. specific For the EMB sample the household response rate was calculated using the design weight to account for unknown eligibility Average individual continuous enumeration probability i The second nonresponse component is an average individual continuous enumeration probability This probability is calculated as the ratio of the number of people who were enumerated continuously in all waves since selection up to and including Wave 2 of the UKHLS to the number of people who were enumerated in the wave of selection minus those known to have become ineligible deceased or out of scope In other words this is the response rate where nonresponse is defined as missing at least one wave Individual specific variation from the mean continuous enumeration probability v The third nonresponse component reflects the variability of enumeration propensities among different individuals We obtain it by the following procedure first we invert the longitudinal enumeration weight for Wave 2 of UKHLS b_psnenus_lIw for the GPS and EMB samples and b_psnenbh_Iw for the BHPS sample second we divide this inversed weight by three probabilities described above selection probability s average household probability of respond in wave 1 hj and average individual enumeration probability i The remainder is then scaled to a mean of 1 00 within each country and each time point of selection reflecting the enumeration probability for each person
132. sport Accessibility Statistics 2009 2011 at the LSOA level with the UKHLS household identifier is available through the UKDS SN7533 Understanding Society Waves 1 3 2009 2012 Special Licence Access Geographical Accessibility For further information on this file see Knies and Menon 2014 4 1 3 SECURE ACCESS Postcode grid references and full date of birth SN 6676 can only be accessed under secure settings Further information about data available under this access route and about how to apply for access can be found on the UKDS see http discover ukdataservice ac uk catalogue sn 6676 amp type Data 20cataloque and the Secure Data Service see http ukdataservice ac uk use data secure lab aspx 4 2 REVISIONS TO PREVIOUS RELEASES We release the preceding waves of data when we make a new edition available Users should refer to the document UKHLS 2014 Revisions see http doc ukdataservice ac uk doc 6614 mrdoc pdf 6614 ukhls 2014 revisions pdf We request that researchers using the data notify us about errors inconsistencies and other problems with the data identified during their use of the data We make use of this information in improving the data Please raise any issues relating to data or data analysis with our User Support service at https www understandingsociety ac uk support projects support We communicate information to members of the Understanding Society users group Please register for the group using
133. stage dataset documentation The online documentation has extensive links between questions and detailed views of variables and data files There is also a search facility for searching questions variables modules and data files e The example Stata code for matching variables from different data files Section 3 8 In assembling the documentation we have drawn upon the documentation for the BHPS see Taylor 2010 and http www iser essex ac uk bhps 2 UNDERSTANDING SOCIETY STUDY DESIGN 2 1 OVERVIEW Understanding Society is a panel survey of households with yearly interviews Data collection for a single wave is scheduled across 24 months The study began with a representative probability sample of households There is an extended discussion of sample design in Section 2 2 and in Lynn 2009 Adult household members age 16 or older are interviewed and the same individuals are re interviewed in successive years to see how things have changed Household members aged 10 15 years are asked to complete a short self completion youth questionnaire Children become eligible for a full interview once they reach the age of 16 We refer to them as rising 16 s The overall study has multiple sample components In the Main Survey there is the a General Population Sample GPS with its subset the General Population Comparison GPC sample b the Ethnic Minority Boost EMB sample and c participants from the BHPS The instruments f
134. substantive research reasons because of the additional contextual information they may provide for the analysis of OSMs At present there is only one category of PSM but others may be defined in the future Any TSM father of an OSM child born after Wave 1 and observed to be co resident with the child at the survey wave following the child s birth is a PSM PSMs remain potentially eligible for interview for the life of survey 2 3 DATA COLLECTION AND RESPONSE OUTCOMES 2 3 1 OVERVIEW The UKHLS is issued to field as 24 monthly samples There is some variation in this pattern The Northern Ireland and the former BHPS sample components are issued over the first 12 months of the wave Table 1 shows the timing of the sample issue for the data included in this release Most of the data collection is conducted face to face via computer aided personal interview CAPI There are also self completion instruments for youth and adults The youth instruments are administered on paper The adult self completion questionnaire shifted from paper to computer administered self interview CASI in Wave 3 From Wave 3 onwards there was also a telephone mop up at the end of the fieldwork period for each sample month 11 Table 1 Timing of data collection start Year Quarter Survey Q1 Q2 Q3 Q4 2008 Q1 Q2 Wave 1 Q3 year 1 Q4 2009 Q1 Q2 Wave 1 Wave 2 Q3 year 2 year 1 Q4 2010
135. tervals often every two to three years The long term content plan summarizes the pattern that has been collected or planned https Awww understandingsociety ac uk system uploads assets 000 000 018 original Long Term Content Plan Nov2011 3 pdf 1355920157 Table 13 shows that some modules are annual such as Current Employment We start to see the repetition of rotating modules on alternating waves for example Family Networks Parents and Children and Harassment Table 13 Summary of questionnaire modules Module Wave 1 Wave 2 Wave 3 Wave 4 Demographics X X NE X NE X Initial Conditions X X NE X NE X NE Family X X NEornot X NE or not background interviewed interviewed Ethnicity and X X NE X NE X NE National Identity Language X see Childhood language Religion X X NE in X NE in X EMB LDA EMB LDA GPC orNI GPC or Nl Religious X EMB Practice LDA GPC recent immigrant Migration X History Partnership X X NE X NE X NE History Fertility History X X NE X NE X NE Health and X see General see General Disability Health Health module module Disability Disability module module Caring X X X X Employment X asked of see Own First see Own First Status History first 6 Job module Job module months of first year Current X X X X Employment Employees X X X X Self X X X X employment Job Satisfaction X X X X 32
136. to individuals li open the household level file use a_hidp a_hhsize using a_hhresp_ip clear lI sort it on the household identifier w_hidp sort a_hidp lI save this temporary file save hhinfo replace lI open the individual level file use pidp a_hidp a_marstat using a_indresp_ip clear li sort it on the household identifier w_hidp sort a_hidp lI merge it with the earlier saved file on w_hidp The output shows how many cases matched merge m 1 a_hidp using hhinfo li drop this variable essential step drop _merge save final1 replace li clean up unwanted files erase hhinfo dta The second task is summarising individual level information at the household level We will summarise individual level information within a household number of 18 24 year olds in the household and then match that onto the household level file see Figure 4 Figure 4 Stata Code Summarizing individual level information to household level open household level file keeping only relevant variables and saving use a_hidp a_hhsize using a_hhresp sort a_hidp save hhinfo replace open enumerated individual level file keeping only relevant variables use pidp a_hidp a_dvage using a_indall_ip clear i create a variable that counts the number of 18 24 sear zh household bysort a_hidp egen n1824 sum a_dvage gt 18 amp a_ovage lt 24 keep only first observation for every household bysort a_hidp keep if_n 1 keep only hous
137. to their new address and those living with the OSM child are eligible for interview If the OSM child moves into an institution where normally just the OSM PSM would be interviewed and not co residents a split off household is created containing only the OSM child and the household enumeration grid completed The child OSM is an eligible sample member even if they are not eligible for interview because of their age 2 2 5 2 Temporary Sample Members TSMs Any members of an enumerated household eligible for inclusion in the EMB sample at Wave 1 who are not from a qualifying ethnic minority are Temporary Sample Members TSMs at Wave 1 This was the only category of TSM at Wave 1 In all parts of the sample any new person found to be co resident in an OSM or PSM household after Wave 1 is a TSM This would include any child born to an OSM father after Wave 1 but not an OSM mother and observed to be co resident with the father or any other OSM at the survey wave following the child s birth TSMs remain eligible for interview as long as co resident in an OSM PSM household TSMs who are not co resident in an OSM PSM household are not followed and become ineligible for interview TSMs are identified as re joiners if they are subsequently found in an OSM PSM household and then become eligible for interview 2 2 5 3 Permanent Sample Members PSMs PSMs are TSMs who are followed for interview after they no longer live with an OSM This is done for
138. ty group Berthoud Fumagalli et al 2009 Given the low proportion of eligible households in the EMB sample it is unrealistic to assume that all non responding 59 households would be eligible that is contain at least one EM member To take this into account we modelled eligibility and used this information in household nonresponse adjustments such that households which were more likely to be eligible had a higher influence on the nonresponse correction Note that the predicted eligibility multiplied by the design weight is released for all the EMB sample households of unknown eligibility as part of a hhdenus_xd This will enable an advanced user to model Wave 1 household nonresponse taking into account the chance to be eligible among households of unknown eligibility To model eligibility we used predictors from the sampling frame and administrative neighbourhood data linked at a geographical level for detailed description see below After excluding ineligible addresses like businesses or demolished and non existent addresses the eligibility was modelled using only EMB households with known eligibility status either screened out or screened in This prediction was then extrapolated onto EMB households of unknown eligibility not contacted Given the limited number of selected addresses in Wales and Scotland and differences between countries in the available auxiliary variables see below we predicted eligibility using two models
139. umber of siblings in the household using the egoalt file The resulting file can be merged with any individual level file Figure 6 Stata Code Using the egoalt file to create household composition variables Load the data file that stores relationship information se b_hidp b_epno b_relationship using b_egoalt_ip clear create a variable that counts the number of siblings in the household bysort b_hidp b_epno egen nsiblings sum b_relationship gt 14 amp b_relationship lt 17 ab var nsiblings number of siblings in household keep one observation per person 1 save final4 replace 82 To match individual level files across two waves into a long format do the following for more waves add wave specific prefix in the foreach statement Figure 7 Stata Code Merging individual files across waves into long format oreach winab li open the individual level file use pidp W_johas using wW_indresp_ip clear i drop the wave prefix from all variables renpfix w _ ll create a wave variable gen wave strpos ab w lI save one file for each wave save temp w replace open the file for the first wave wave a_ use tempa clear oreach win b lI append the files for second wave onwards append using temp w save the long file ave final5 replace erase temporary files oreach winab erase temp w dta To match individual level files across two waves into a wide format
140. versity of Essex the University of Warwick and the London School of Economics Professor Nick Buck is the principal investigator Fieldwork for waves 1 to 4 was conducted by the National Centre for Social Research NatCen with collaboration with the Central Survey Unit of the Northern Ireland Statistics and Research Agency NISRA in Northern Ireland The overall purpose of Understanding Society is to provide high quality longitudinal data about subjects such as health work education income family and social life to help understand the long term effects of social and economic change as well as policy interventions designed to impact upon the general well being of the UK population To this end the study collects both objective and subjective indicators and offers opportunities for research within and across multiple disciplines such as sociology and economics geography psychology and health sciences The study also provides a platform for additional data collections 1 2 HOW TO NAVIGATE THIS USER MANUAL This release has data for the Understanding Society Main Survey which collects information from the UK General Population Sample GPS and the Ethnic Minority Boost EMB sample From Wave 2 onward the Main Survey also includes information collected from continuing participants of the British Household Panel Survey BHPS a household panel survey of around 8 000 households in the UK which has completed 18 annual waves of data collection
141. w was selected The weight was then scaled to a mean of one Cross sectional weights for proxy and main n_indpxub_xw main n_indinub_xw and self completion n_indscub_xw were created based on the Wave n cross sectional enumeration weight Three logistic regressions were run to predict the response for the relevant instrument with predictors from the household grid and household questionnaire in Wave n conditional on enumeration in Wave n The models were restricted to adults The estimated probabilities were inversed multiplied by n_indenub_xw and were scaled to a mean of one The cross sectional weight for youth n_ythscub_xw was also created based on enumeration at the same Wave n n_indenub_xw Among eligible youth members between 10 and 15 a logistic regression was run with predictors from the household grid and household questionnaire from Wave n to predict response to the youth questionnaire The inverse of the probability was multiplied by n_indenub_xw and was scaled to a mean of one 3 8 IMPUTATION OF INCOME VARIABLES Understanding Society collects detailed information each wave on personal income All individuals aged 16 or more are asked to report wages self employment earnings second job earnings interest and dividends pensions National Insurance state retirement pension pension from a previous employer pension from a spouse s previous employer private pension annuity widow s or war widow s pension w
142. w TSM participants in successive waves as long as they live in the household of an OSM A male TSM who fathers a child with an OSM female becomes a Permanent Sample Member PSM PSMs are treated in the same way as OSMs in the following rules In sum TSMs are not followed for interviews when they leave the household but OSMs and PSMs are For panel maintenance ISER maintains a database of information on respondents sO we can send communications to them and to allocate interviewers This information is vital for minimising attrition The data base builds on contact information collected during the survey interviews and is updated throughout the year There are for example new addresses household splits and moves out of the country or into an institution Change of address cards were also returned to ISER in cases where a whole household moved or a new resident returned the card giving the forwarding address It is possible for ISER to be notified of some deaths through these means A between wave mailing is also used to help maintain contact with participants and update addresses The mailing has a report of research findings an address confirmation slip and materials to encourage registration with the participant website The participant website can be seen at https www understandingsociety ac uk participants 2 3 4 RESPONSE OUTCOMES 2 3 4 1 Wave 1 The Wave 1Main Survey fieldwork started on 8 January 2009 and ended on the 7 Marc
143. w_scsfi2mcs_dv SF 12 Mental health component score w_scsf12pcs_dv SF 12 Physical health component score w_nchild_dv Number of respondent s natural children in household 3 3 NOTES ON USING THE BHPS SAMPLE The continuing sample from the British Household Panel Survey BHPS joined the Understanding Society sample in Wave 2 Both samples can be used for cross sectional and longitudinal analyses For the appropriate weights select from Table 24 to Table 31 The cases in the two samples can be distinguished using the variable w_hhorig The variable also allows the identification of different components of the BHPS sample see below The questionnaires used for the two samples are the same There are however a few differences in the data collected One important issue is that the date of previous interview for GPS sample members who were interviewed at the previous was approximately 12 months earlier while for the former BHPS sample the gap was between 13 and 27 months for sample members interviewed at Wave 18 of BHPS This means that the reference period for the history of events since the last interview will be longer for the BHPS sample This variation in the reference period applies only to Wave 2 For longitudinal analysis of the GPS sample cases may be matched to Wave 1 data available as part of this release from the UK Data Service using the variable pidp the Understanding Society cross wave person identifier However for t
144. weight as this is the wave at which the Scotland and Wales boost samples were added so all of the members of those samples who entered UKHLS were enumerated at that wave as were the vast majority of members of the original BHPS Wave 1 sample who entered the UKHLS This component therefore encompasses a design weight Wave 1 post stratification and adjustments for nonresponse at each of the Waves 1 to 9 of the BHPS The second component is derived from a model of the propensity to be issued at UKHLS Wave 2 conditional on being enumerated in BHPS Wave 9 This adjusts for all the stages of dropout between BHPS Wave 9 in 1999 and UKHLS Wave 2 in 2010 Model covariates were taken from the Wave 9 household grid and household questionnaire BHPS OSM newborns since Wave 9 England Scotland or Wales or Wave 11 Northern Ireland whose parents are both OSMs were then assigned a base weight equal to the smaller BHPS inclusion weight of their OSM parents in the child s 2010 issued to UKHLS household This reflects the idea that the probability of the child entering the UKHLS sample equals the probability of at least one of his or her parents entering the sample which in turn is equal to or greater than the probability of the parent who has the greatest probability of entering the sample BHPS OSM newborns born to one OSM parent and one TSM parent were assigned a base weight equal to half of the OSM parent s weight in the child s 2010 issued to U
145. ze from the previous wave Wave 1 The variable b_ff_plbornc is the country of birth of the respondent fed forward from the previous wave Note the use of the prefix ff_ Some of the fed forward variables were not used in the wording of a question but were used by the CAPI script to route respondents appropriately based on information from the previous wave Information collected using dependent interviewing is merged with the respective information collected using independent interviewing e g when a respondent did not provide the information in the previous interview or when they are new to the study and stored in the data file under the variable name used for the latter i e the variable stem name from Wave 1 See for example the socio economic classification of the current job w_jbsoc00 and the standard industrial classification w_jbsic07 We use look up files between SOC 2000 and other classifications to derive additional occupational classifications For further information and to obtain look up files see http Awww cf ac uk socsi CAMSI S occunits distribution html UK We provide the following classifications International Standard Classification of Occupations ISCO88 Registrar General Social Class RGSC National Statistics Socio economic Classification NS SEC Employment Status ES and Socio economic Group SEG These are computed for the respondent s current job and last jobs only SOC 2010 is provided but SOC 9

Understanding Society: Waves 1-4, User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents