ELSA Index File V3 User Guide - Institute for Fiscal Studies
Contents
1. Wave specific Household Serial Numbers For each Wave there is a randomly generated five digit household serial number IDAHHWO IDAHHW1 IDAHHW2 or IDAHHW 3 These serial numbers do not relate to the serial numbers used during interviewing at HSE or ELSA and the serial numbers were generated anew for each Wave METTE Variable label Original source IdahhwO Analytical Wave 0 household serial number Derived Missing for all those who were not in a responding household at HSE Idahhw1 Analytical Wave 1 household serial number Derived Present for all individuals in responding households at Wave 1 Idahhw2 Analytical Wave 2 household serial number Derived Present for all individuals in responding households at Wave 2 Idahhw3 Analytical Wave 3 household serial number Derived Present for all individuals in responding households at Wave 3 Notes User Guide 4 2 4 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 1 There are a small number of cases who have a value of IDAHHWS beginning with a 2 These cases have values of IDAUNIQ which suggest they were ineligible for interview this is re iterated in their outcome code These cases are ex partners of Refreshment Sample members who have moved to a different address since HSE They will not be followed up for future ELSA interviews as they are ineligible in their own right Please see the documentation associated with Wave 3 from the UK Data2. 410 Refusal to office 420 Refusal of info about occupants of address 431 Refusal at intro before HH module by eligible respondent 432 Refusal at intro before HH module by other 440 Refusal after HH module 450 Broken appointment no recontact 510 Ill at home during survey period 520 Away in hospital all survey period 530 Physically mentally unable incompetent 540 Language difficulties 550 Lost productive 560 Other unproductive 610 Address not attempted 620 Address inaccessible 630 Unable to locate address 680 Moved unable to trace 781 Address out of sample moved to institution 782 Duplicate address 783 Address out of sample moved out of England 791 Ineligible all SM died 792 Ineligible SM ineligible 793 Ineligible other reason 999 Sector not covered Outscw1 Wave 1 self completion outcome Wave 1 variables not archived 2 Not asked interviewed by proxy This variable is an improved 1 Not applicable see User Guide version of SC REC which is in 1 Questionnaire received the main Wave 1 dataset 2 Received but majority of questions not answered 3 Not received Outhhw2 Wave 2 household outcome updated with mortality W2HHOUT from Wave 2 main information data with mortality updates 2 Not in household or ineligible for interview 110 All eligible respondents interviewed in person 120 AII eligible respondents interviewed some in person some by proxy 210 Some but not all eligible respondents
3. W3 C1NP1 unprod new partner found W1 non response at W3 C1NP2 unprod new partner found W2 non response at W3 C1NP3 unprod new partner found W3 non response at W3 C1YP unprod younger partner non response at W3 C1SM unprod sample member non response at W1 amp W2 C3CM refreshment sample core member responded at W3 C3CP refreshment sample member non response at W3 see note C3SM notiss refreshment sample member not issued at W3 C3NP3 refreshment sample new partner found W3 responded at W3 C3OP refreshment sample older partner responded at W3 C3YP refreshment sample younger partner responded at W3 C3SM_unprod refreshment sample member non response at W3 C3NP3_unprod refreshment sample new partner found W3 non response at W3 C3OP unprod refreshment sample older partner non response at W3 C3YP unprod refreshment sample younger partner non response at W3 See ELIGWS was either in the household and ineligible or not in the household at W3 See ELIGWS for further details and FINSTAT2 As for FINSTAT but additionally NP2 indicates the individuals who are eligible but not part of the ELSA panel Note that this variable is the same as that labelled FINSTAT in the main Wave 2 archived data Derived from FINSTAT and OUTINDW3 As for FINSTAT2 but additionally NP3 OP and YP indicates the individuals who are eligible but not part of the ELSA panel Note that this
4. self completion questionnaire weights from Waves 1 2 amp 3 We plan to match on weighting variables from the Wave 2 Nurse Visit in the next version of the file Please see the Wave 3 Core Data User Guide for further information about the Wave 3 weights W3XWGT and W3LWGT especially for clarification about which cases have a value for them For the self completion weights they were created initially only for Core Members whose self completion questionnaire was received with the majority of questions answered i e they had a value of 1 for the self completion outcome code which is named OUTSCWx The Wave 3 cross sectional weight was used in a logistic regression containing a number of predictors of response to ELSA From this regression model a trimmed non response weight was derived This weight was then multiplied by the Wave 3 cross sectional weight and scaled For further details of the derivation of all three self completion weights please see Appendix 1 the derivation for each is very similar METIETST METIETSTTSE ETT Original source Wiwgt Wave 1 non response weight for all core sample members Derived W2wgt Wave 2 non response weight for all core sample members Derived WSlwgt Wave 3 longitudinal weight for core sample members who Derived responded at all Waves W3xwgt Wave 3 cross sectional weight see Wave 3 Core User Derived Guide for details Scw1wgt Wave 1 self completion weight Derived Scw2wgt Wave 2 self completion we
5. 109164 The other was a person that did not take part in an HSE interview IDAUNIQ 108969 OUTINDWO 71 User Guide 13 4 3 4 4 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Sampling Status at HSE The existing variable SAMPSTAT has been updated to include information for the Wave 3 Refreshment Sample If you wish to pick out these cases then their IDAUNIQ s can be used as they will have values of greater than 150000 but less than 700000 or greater than 800000 Note that the partners are designated potential interviewees at the sampling stage and that values of FINSTAT should be used to tell who those cases sampled actually were Variable Sampstat Note Variable label Sampling status at HSE updated for Refreshment Sample 1 Not known at HSE or ineligible SM Sample member YP Younger partner OP Older partner Original source Amalgamation of information from original HSE sampling files and Wave 3 sample file 1 There is one case IDAUNIQ 800220 who was thought to be a potential younger partner at the sampling stage but at the interview was found to have split from their sample member partner and was therefore ineligible for interview in their own right and therefore have ELIGW3 0 Final Status after ELSA Fieldwork FINSTAT and FINSTAT 2 had previously been forward coded so that they each had the same number of valid values In other words FINSTAT contained values f
6. Archive for further clarification of eligibility for interview 2 Itis unknown from the Wave 3 interview whether IDAUNIQ 801791 was still in IDAHHWS 13571 at the time It has been assumed that they were not for the purposes of this file 3 There are 2281 cases which have a missing value either 2 or 5 for all four of the IDAHHWx s These cases have either never been issued for ELSA were issued for Wave 3 but were unproductive or were ineligible but in the household with the latter group 4 We estimate that at least 1500 of the above 2281 cases who are from the Wave 3 Refreshment Sample and were in responding households at HSE will have an IDAHHWO assigned in future when the HSE data for these cases is archived as an addition to the existing Wave 0 data For now the cases who we think fall into this group have a value of 5 for IDAHHWO 5 IDAHHWS s 15385 and 11192 have had identical addresses since HSE but are considered separate households as the latter is an annex to the former property 6 The cases that have a 2 for any of the IDAHHWX s were not present in a responding household at that wave This could be because they were in the household at the time of the HSE interview but had moved out by the time of the ELSA interview or because they had moved out between waves or because they had died PREV IDA and PREV HHW3 As mentioned in Section 2 1 there are some cases which have an IDAUNIQ beginning with a 7 who were eli
7. HSE 2001 that have had their outcome changed as they asked for their data to be deleted These cases are IDAUNIQ 2113965 and IDAUNIQ 113373 2 There are two cases who were eligible for a full in person main interview at Wave 2 who were accidentally routed through the by proxy questions IDAUNIQ 116787 and IDAUNIQ 1 17471 Both of these respondents live alone and have outcomes showing full in person interview and both were eligible for a nurse visit 3 The cases which have a value of 1 for the self completion outcomes were not productive at that Wave 4 There are just over 200 cases who were in the household at the time of the HSE interview but do not have an outcome code This could be for a number of reasons one of which is that only two children per household were selected for the HSE interview so any more than two children will not have an outcome code This is also the case for the HSE self completion outcome where additionally any children aged less than 8 were not given a self completion questionnaire 4 Status Variables These variables describe the status of the individuals in terms of how they were classified for sampling then after each interview and any related fieldwork status information Eligibility for Interview Eligibility for interview at Wave 3 is shown by ELIGWS with equivalent variables for the earlier Waves This variable reflects the information found at the Wave 3 interview rather than what was t
8. and over and their partners living in private households in England The samples have been drawn from households that had previously responded to the Health Survey for England HSE between 1998 and 2004 with the exception of HSE 2000 Households from HSE 2000 have been sampled for piloting each wave of ELSA and have not been included in the Index File Fieldwork for ELSA Wave 1 was conducted in 2002 3 Wave 2 in 2004 5 Wave 3 in 2006 7 and Wave 4 in 2008 9 The fieldwork for Wave 5 is scheduled to run from 2010 to 2011 The Index File contains key information such as analytical serial numbers status at sampling and response outcomes as well as mortality data Some of this information has already been archived with other ELSA datasets whereas some is only available in this file This file contains all age eligible individuals within households identified from HSE 1998 1999 and 2001 2004 over 24 700 people It also contains all ineligible individuals living in households with interview eligible individuals around 6 400 for the original sample and around 1 800 for the Wave 3 Refreshment Sample The file contains a total of 32 900 individuals This file has been created by combining data from a number of sources including HSE files ELSA sample files ELSA interview data and NHSCR records from ONS It is indicated next to each variable in the following sections of this User Guide where the variable originated from The aim of ELSA is
9. interview The variables examined included Age group Sex Urban rural indicator Marital status Government Office Region Index of Multiple Deprivation 2004 quintiles Ethnic group white non white Financial unit type Tenure Highest educational qualification Whether has living child living with respondents not with respondents Number of people in household Current status i e retired employed permanently sick or disabled looking after home or family Equivalised income quintile Self assessed health O OO D 0 T 0 00000 O O User Guide 21 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing e Anon response weight to compensate for potential response bias to the self completion questionnaire was based on a logistic regression model of response 1 response 0 non response with the predictor variables entered by a forward stepwise selection procedure The model was run using the Wave 3 cross sectional weight The variables remaining in the optimal model were Age by sex Marital status Region Financial unit type Self assessed health Whether has living child or not White non white Highest educational qualification Equivalised income quintile Current status Tenure Oe Oe OOF e O O OF Or OVO e The predicted response probabilities from the final optimal model were converted into a non response weight by obtaining their reciprocal e g a response probability of 0 8 correspond
10. original and Refreshment samples However 103 of the potential eligible sample members born between 1 March 1952 and 28 February 1953 the missing year of birth were in fact successfully interviewed in Wave 3 Originally such individuals were classified as younger partners if in the original sample or older partners if in the Refreshment sample These have now been reclassified as core members belonging to the Refreshment Sample The diagram below clarifies the dates of birth that were eligible for each of the samples Original Sample Refreshment Sample It is hoped that with the next version of the Index File the following additional variables will be deposited Weighting variables from the Wave 2 Nurse Data available on the Wave 2 Nurse Data Variables showing levels of consent to administrative data linkage and extraction and storage of genetic material HSE clustering and stratification variables updated for the Refreshment Sample versions for the Core Sample are available on the previous version of the Index File as well as wave specific archived datasets Variables from the ELSA Life History interview available on the Life History data Variables from at least one ELSA End of Life interview For more information about ELSA please see the User Guides for Waves 0 1 2 and 3 and the Technical Reports for Waves 1 2 and 3 all of which are available from the UK Data Archive User Guide 2 1 UK Data Arc
11. survey period 530 Physically mentally unable incompetent 540 Language difficulties 550 Lost productive 560 Other unproductive 561 Productive but respondent requested deletion 610 Address not attempted 620 Address inaccessible 630 Unable to locate address 680 Moved unable to trace 781 Address out of sample moved to institution 782 Duplicate address 783 Address out of sample moved out of England 791 Ineligible all SM died 792 Ineligible SM ineligible 793 Ineligible other reason 799 Died updated outcome unprod 999 Sector not covered Wave 1 individual outcome updated with mortality Wave 1 variable INDOC information updated with mortality data 2 Not in household or ineligible for interview 11 Full interview in person 13 Full proxy interview 21 Partial interview in person 23 Partial proxy interview 31 No contact 43 Refusal before interview 44 Refusal during interview 45 Broken appointment no recontact 51 Ill at home during survey period 52 Away ill in hospital during survey period 53 Physically mentally unable incompetent 54 Language difficulties 55 Lost productive 56 Other unproductive 79 Died ineligible 99 Died updated outcome unproductive 310 No contact with anyone at address 330 Contact made but not with responsible resident 340 Contact made but not with eligible resident User Guide UK Data Archive Study Number 5050 English Longitudinal Study of Ageing
12. to study a sample of people aged 50 and over As the study progresses all of our respondents get older and our sample effectively ages therefore the youngest people need to be replaced as they are no longer represented In order that our sample continues to be representative new people from HSE 2001 2004 were sampled for Wave 3 The people selected for this additional sample were those aged 50 to 53 and their older younger partners The original sample also known as Cohort 1 consisted of people born before 1 March 1952 and their partners while the Refreshment Sample also known as Cohort 3 consisted of people born between 1 March 1953 and 29 February 1956 and their partners Note that there is an overlap between the original sample and the Refreshment Sample as a number of young partners in the original sample from HSE 2001 moved into their 50 s in Wave 3 and so were potential refreshment sample core members i e born after 29 February 1952 If they responded at Wave 3 these cases have changed status see FINSTAT2 and FINSTAT3 from C1YP Cohort 1 Younger Partner to C3CM Cohort 3 Core Member User Guide 1 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing It is important for users to note that the algorithm used to select the Wave 3 Refreshment Sample excluded potential eligible sample members born between 1 March 1952 and 28 February 1953 This has resulted in a gap of one year s birth between the
13. variable is the same as that labelled FINSTAT in the main Wave 3 archived data Some values of FINSTAT now do not refer to response at consecutive Waves e g C1CM C1CP Please see the interview outcome codes for the outcome of Wave 2 1 All known new partners are included in the variables above but note that some will not have been enumerated because of missing information from non responding households This implies that response rates for new partners cannot be calculated because the denominator is unknown User Guide 15 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 2 Note that the core sample and the refreshment sample overlap as a number of young partners in the original sample from HSE 2001 moved into their 50 s in Wave 3 and so were potential refreshment sample core members i e born after 29 February 1952 If responding at Wave 3 these cases changed status from C1YP to C3CM 3 C3CP Respondents who have this FINSTAT3 were sampled as Refreshment Sample Core Members but did not respond at Wave 3 However someone else in their household is a C3CM These cases are similar to C1CP s at Wave 1 4 IDAUNIQ 151361 was sampled as a potential core member but due to discrepancy with their date of birth recorded at the HSE interview they have a FINSTATS of C3YP 5 Weighting Variables This version of the Index File has been expanded to include Wave 3 weighting variables as well as
14. 9 8 1 8 2 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing HSE Year The HSE year from which the respondent s household was originally sampled is given by the variable HSEYR All household members present at the time of the HSE interview will have a value Those who do not have a value are either new partners who have joined the household at either Wave 1 Wave 2 or Wave 3 see the FINSTAT variables for more information or other household members ineligible for interview in their own right who have joined the household since HSE The variable NOSAMP in addition shows which of the new partners have responded to at least one wave of ELSA Variable Variable value labels Original source HSE source survey year Included in original HSE 3 Ineligible for interview and joined the household after HSE sampling files 2 New partner who joined the household after HSE Otherwise gives the HSE year Nosamp Responding New Partner no sample design information as Derived not in household at HSE 1 Not applicable not in HSE and has not responded includes ineligibles 0 In HSE sample 1 Notin HSE but responded to at least one Wave of ELSA Notes 1 Full sample design variables are only available for cases that were in HSE and responded to at least one ELSA Wave Those who responded to at least one ELSA wave but were not in HSE have NOSAMP equal to 1 2 There are three cases which have a value of 1 for NOSA
15. Away in hospital during survey period 53 Physically or mentally incompetent 54 Language difficulties 56 Other unproductive 60 Untraced 68 Moved unable to attempt contact at new address 71 In institution 78 Moved out of Britain 79 Outcome unknown 90 Died 99 Died updated outcome unproductive Wave 2 self completion outcome OUTSCW 2 from Wave 2 main 2 Not asked interviewed by proxy data 1 Not applicable see User Guide 1 Questionnaire received 2 Received but majority of questions not answered 3 Not received Wave 3 household outcome updated with mortality W3HHOUT from Wave 3 main information data with mortality updates 2 Not in household or ineligible for interview 110 All eligible respondent s interviewed in person 120 All eligible respondent s interviewed some by proxy 210 Some but not all eligible respondents interviewed 310 No contact with anyone at address 330 Contact made at address but not with responsible adult 340 Contact made but not with eligible respondent 410 Office refusal 420 Complete refusal of info about occupants of address 430 Refusal at introduction before HH module 440 Refusal after HH module 450 Broken appointment no re contact 510 Ill at home during survey period 520 Away from home in hospital during survey period User Guide Outindw3 Outscw3 Outschw3 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 530 Physically mentall
16. MP but do have a value of HSEYR This is because these cases were in the household at HSE but were either not the partner of a sample member at the time but became one at Wave 3 or they had moved out not having responded to ELSA and were found to have moved back in again at Wave 3 HSE Interview Status Variable VEEL CEELI Original source Hseint Whether interviewed at HSE Outcome of HSE interview 1 Yes 2 No User Guide 20 9 1 10 1 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 9 General Notes Order of Variables The variables have been ordered in the file according to their type An alternative order would be to have those of each Wave clustered together especially the outcome codes and Wave specific identifiers for example IDAHHW1 ISSUEW1 ELIGW1 FINSTAT OUTINDW1 OUTHHW1 OUTSCW1 10 Appendices Derivation of Self Completion Weights ELSA WAVE 3 e The analysis focused on ELSA core sample members only from Cohorts 1 and 3 i e the Core Sample and the Refreshment Sample e Non respondents to the self completion questionnaire were defined by values of OUTSCW3 2 or 3 ELSA core members who were interviewed by proxy were not included in the analysis and were not issued with the self completion questionnaire e Using the Wave 3 cross sectional weight W3XWGT the nature of response to the self completion questionnaire was examined using variables collected in the Wave 3 main
17. NH were not asked of proxy respondents A test of an interaction between age and sex was not statistically significant User Guide 24
18. UK Data Archive Study Number 5050 English Longitudinal Study of Ageing s NatCen National Centre for Social Research ELSA Index File V3 User Guide Author Susan Nunn UK Data Archive Study Number 5050 English Longitudinal Study of Ageing T OVON EE 1 2 Identifiers 2 1 IDAUNIQ 2 2 Person NUmb er o ae esed SEAE A EEEE AEE EAA ESSA SEEE A EEAO Na EE EAEAN EMEA TAA 4 2 3 Wave specific Household Serial Numbers seeessessseeeseeeeeeene nennen nennen nnne nnne nnne nns 4 2 4 PREV IDAand PREV IW uiii hte iti Er ciat eR E PRESA EHE HEELS EE EHE EESEREFLO EEEE ILEESE MSTA 5 4 Oulcoms Variables iini d Gd pisce CEP dsU C pa Leu ERE AMI Q2 eens 5 CEMECCUICATIILOIID 4 1 jEligibility TON IMOT VIEW ss crore ceri teneo rh s are rb a YE ER EE ka eb Pee e erbe ESETE SaR Ear Poena detnr resa seta HE Ke ara Ree saad 4 2 Reason Tor hob being ISSUBG 6 tuo ecran riori merken E khe ieri ees Uo eb a EUER A RN eR ka F V Esu eR AER APER KEEN RAS 4 3 Sampling Status at HSE sss nennen nennen nrennetnre treten trennen nne tne erret et rennen 4 4 Final Status after ELSA Fieldwork SER IJ RE d c 16 6 Mortality Variables rniseceecissese kis cut anande naaa ENPXR ERO EX RR aeaaaioii 17 7 Demographic Variables aaaea aa Eaa ra nnne nnns nnn nn ansia sana snnm nasa 19 8 Sampling V
19. a There are at least the 1500 cases mentioned in Note 4 under Section 2 3 above who currently have a value of ELIGWO 1 but will have a value denoting eligibility in the future Reason for not being issued The variable ISSUEWS shows whether a case was issued for interview at Wave 3 and if not the reasons for this This variable has been calculated differently to previous Waves Previous versions have included new partners found at that Wave as being issued whereas now new partners discovered at the Wave 3 interview have a value of 9 There is a new category for those who were not issued but did have a productive Wave 3 interview which includes of a small number of cases who were not issued as we were mistakenly informed that they had died prior to Wave 3 There is another new category of 7 for a small number of cases who did not fit into any of the existing categories Those who are not applicable because they are ineligible for interview i e children in the household who will never be eligible for interview have a value of 10 The information to compile ISSUEWS came from pre Wave 3 fieldwork updates including those of deaths and moves abroad User Guide 12 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Variable VEEL CEE LI Original source Issuew1 Reason for not being issued at Wave 1 Derived 10 Not applicable as ineligible for interview 9 Not known at time 5 Part of Wave 3 Refreshment Samp
20. ariables iones scien cece suus bios uio GPs ne a di DER DL An ra aiU G RAIDER A Mida R E RIdE 19 ts m E AEE E I 8 2 HSE Interview Status 9 General gU Tm s 21 9 1 Order of Variables RRRRREEEEEEEEEMEEMEEMMEEMEEMMMMMMMMMMNFM 21 10 rop m 10 1 Derivation of Self Completion Weights 11 Goantact Detalle ies uidusmicnzudeatttuxE a0 tei ac oam Cor cU Dp I dab m oe n EDD CUM EU MNT deber RU PE 24 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 1 Overview This User Guide contains information about the third release of the English Longitudinal Study of Ageing ELSA Index File This file may be considered to be a master file that provides links between the different Waves of ELSA and currently gives information relating to Waves 0 1 2 and 3 It is envisaged that equivalent information will be added for subsequent Waves and that a new version of the file will be made available in 2010 which will contain information from Wave 4 as well as additional variables from earlier Waves This section of this User Guide gives an overview of what the file is for and how it was created The sections that follow give detailed information about the variables in the file which have been grouped under the broad headings of Identifiers Outcomes Status Weighting Mortality Demographics and Sampling ELSA is a study of people aged 50
21. d 99 Died updated outcome unproductive Wave 3 core self completion outcome OUTSCWS from Wave 3 main 2 Not asked interview by proxy data 1 Not applicable see User Guide 1 Questionnaire received 2 Received but majority of questions not answered 3 Not received Wave 3 health self completion outcome OUTSHWS from Wave 3 main 3 Respondent was not part of the sub sample selected to data receive that questionnaire 2 Not asked interview by proxy User Guide 10 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 1 Not applicable see User Guide 1 Questionnaire received 2 Received but majority of questions not answered 3 Not received Outscww3 Wave 3 work self completion outcome OUTSWWS from Wave 3 main 3 Respondent was not part of the sub sample selected to data receive that questionnaire 2 Not asked interview by proxy 1 Not applicable see User Guide 1 Questionnaire received 2 Received but majority of questions not answered 3 Not received Notes 1 18 cases from HSE 1998 had missing or incorrect outcomes for OUTINDWO IDAUNIQ 100355 100456 100677 100725 100771 100838 101744 101842 103612 100821 100362 100636 100710 Two of these cases 100710 and 100636 have been changed because they had been classified as a productive interview but in fact they were not This was due to an error with the interview program itself There are also two additional cases from
22. echnical Reports which are available from the UK Data Archive User Guide 5 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Note that there are differences within the HSE household and individual outcome codes as different HSE years use different codes They have not been assimilated either within themselves or to the ELSA outcome codes but this may be carried out for a later version of the file METTE Variable label Original source OuthhwO HSE survey household outcome HSE Wave 0 survey data 2 Not in household or ineligible for interview variable FINOUTC 41 All interviewed in person 42 Atleast one person interviewed but not all 44 No interviews all refused broken appointments 45 No interviews other reasons combinations 110 Complete interview by desired respondents 212 Partial non contact with other household members 213 Partial refusals from other household members 214 Partial partial for other combination of reasons OutindwO HSE survey individual outcome HSE Wave 0 survey data 2 Not in household or ineligible for interview variable INDOUT 1 Was in household but does not have an outcome code 51 Full interview 52 Partial interview 53 Full interview in translation 70 No contact 71 Personal refusal 72 Proxy refusal 73 Broken appointment 74 Ill at home 75 Ill in hospital 76 Away from home 77 Senile incapacitated 78 Inadequate English 79 Other reasons no interview 110 Full CAPI interv
23. el CP NP1 and YP indicates the individuals are eligible but not part of the ELSA panel Derived from FINSTAT and OUTINDW2 please note the differences in the labels of the categories between FINSTAT User Guide Finstat3 Notes UK Data Archive Study Number 5050 English Longitudinal Study of Ageing C1NP1 new partner found W1 responded at W2 C1NP2 new partner found W2 responded at W2 C1YP younger partner responded at W2 C1CM_unprod core sample member responded at W1 non response at W2 C1CP_unprod core sample member partner of a C1CM non response at W1 amp W2 C1NP1 unprod new partner found W1 non response at W2 C1NP2 unprod new partner found W2 non response at W2 C1YP unprod younger partner non response at W2 C1SM unprod sample member non response at W1 amp W2 See ELIGW2 was either in the household and ineligible or not in the household at W2 See ELIGW for further details Final status after Wave 3 fieldwork C1CM core sample member responded at W1 amp W3 C1CP core sample member partner of a C1CM non response at W1 responded at W3 C1NP1 new partner found W1 responded at W3 C1NP2 new partner found W2 responded at W3 C1NP3 new partner found W3 responded at W3 C1YP younger partner responded at W3 C1CM_unprod core sample member responded at W1 non response at W3 C1CP_unprod core sample member partner of a C1CM non response at W1 amp
24. es The Index File contains demographic information relating to the year of birth and the sex of the respondent This information is in addition to the information relating to sex year of birth and age present in each of the archived main data sets The year of birth has been collapsed for those aged 90 or over as at 29 February 2006 This is in line with what is done on the main archived core data for confidentiality reasons METIETST Variable label Original source Sex of respondent HSE SEX W1 INDSEX W2 8 Information not known INDSEX and W3 INDSEX 1 Male survey data 2 Female Dobyear Year of birth collapsed for those aged 90 or over for HSE DOBDAY DOBMON confidentiality reasons DOBYEAR W1 DHDOB W2 8 Information not known DHDOB W3 DHDOB 7 Respondent aged 90 or over as at 29 02 2006 Otherwise gives year of birth Notes 1 The reason that the sex or date of birth of some respondents is not known is that they were not the person who answered the household demographics module at either HSE or ELSA and the person who did answer did not know their details Alternatively their details could be missing as they were not selected for the HSE interview If someone s date of birth was not known or not given then their age usually was 8 Sampling Variables The Index File contains a number of variables from HSE which were used in sampling for both the ELSA Core Sample and the Refreshment Sample User Guide 1
25. ew and the time of the Wave 3 interview but one respondent was thought to have died in between Wave 1 and Wave 2 and so was not issued for the latter 4 There are 263 cases that have died but we do not have any information from NHSCR records for them They can be identified as having a value of 2 in MORTFIN but a value of 2 in YRDEATH or AGDEAD2 5 There are five cases which have a year and age of death from the NHSCR records but do not have cause of death ICD information They have a value of 2 in MAINCOD 6 Further one of these five cases IDAUNIQ 100277 has a date of death in the NHSCR records which is before their HSE interview We are currently checking these details with ONS but for the time being the variables relating to the time and age of their death have been set to 3 Details being checked with ONS Additional note about MORTWAVE and the sources of information referred to Sampling refers to checks conducted during the sampling procedure for that wave Data from a number of sources are checked such as outcome from the previous interview ONS mortality status and ad hoc feedback from panel maintenance through activities such as pre advance letters and Christmas cards Fieldwork refers to information received by interviewers and the wider survey team whilst conducting fieldwork This may be from other sample members who knew the person concerned External update refers to updated NHSCR records being rece
26. focused on the ELSA core sample members only e Non respondents to the self completion questionnaire included people who responded to the main interview but didn t fill in the questionnaire By this definition non respondents to the self completion included 158 ELSA core sample members who were interviewed by proxy e Using the W1WGT weight the nature of response to the self completion questionnaire was examined The variables examined included Age group Sex Urban rural indicator Marital status Government Office Region Index of Multiple Deprivation 2000 quintiles Ethnic group white non white Type of benefit unit Tenure Education qualifications Current employment status Equivalised income quintile 1 These two measures were used as proxy measures of general health Unlike these two measures the general health variables themselves e g HEGENH were not asked of proxy respondents A test of an interaction between age and sex was not statistically significant User Guide 23 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Whether took part in 1 paid work 2 self employed work 3 voluntary work 4 caring for a sick or disabled adult 5 looked after home or family or 6 training or education course in the last month Eyesight condition Hearing condition A non response weight to compensate for potential response bias to the self completion ques
27. gible for interview This is due to the discovery during the process of updating the Index File for Wave 3 that a small number of cases from the Refreshment Sample were living in households that had already been sampled for and responded to at least Wave 1 of ELSA These cases have had their IDAUNIQ s changed to those originally assigned to them when they were not eligible for interview in their own right There are two new variables in the file which show the identifiers that a case had in previous versions of the Index File PREV IDA shows the IDAUNIQ while PREV HHWS shows the IDAHHWS We will be clarifying this issue further with a new release of the ELSA Wave 3 Core Data and User Guide in the near future 3 Outcome Variables The Index File includes the key outcome variables for each wave household individual and self completions We hope to add further outcome variables with the next version of the file With the Core Data for each wave only outcome codes for productive interviews are archived but information about unproductive interviews is archived with the Index File Broadly the outcome codes and status variables for individuals can be classified under the headings of response non response or ineligible Further details are not given here however as the calculation of response rates for ELSA is beyond the scope of this User Guide Users of the data are advised to seek further information about these rates from the ELSA T
28. hive Study Number 5050 English Longitudinal Study of Ageing 2 Identifiers IDAUNIQ All the ELSA data files deposited in the archive contain a unique individual analytical serial number IDAUNIQ to enable users to link the different files if necessary Everyone in the Index File has a unique value for this variable The values of IDAUNIQ which were assigned to those new to the Index File at Wave 3 have been grouped a little differently to those assigned for Waves 0 1 and 2 Analysts wishing to categorise Wave 3 only cases into the same strata as for earlier Waves are advised to use the variables HSEYR FINSTAT3 and OUTINDW3 in this instance Variable EUELA ET Notes Idauniq Unique individual Derived for Index file within strata below individuals were randomly ordered serial number 100001 100034 new partner at W1 non responder at W1 100035 100106 new partner at W1 responded at W1 100107 101026 HSE 1998 not issued at W1 101027 103711 HSE 1998 issued non responder at W1 103712 108862 HSE 1998 issued responded at W1 108863 109411 HSE 1999 not issued at W1 109412 110841 HSE 1999 issued non responder at W1 110842 113098 HSE 1999 issued responded at W1 113099 114009 HSE 2001 not issued at W1 114010 116680 HSE 2001 issued non responder at W1 116681 121300 HSE 2001 issued responded at W1 121301 121338 new partner at W2 responded at W2 121339 121358 new
29. hought at the time of issuing the Wave 3 sample Respondents who were found to have died at the User Guide 11 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Wave 3 interview and also through the mortality update are coded 1 for ELIGWS This is also the case for ELIGWx for earlier waves ELIGWS3 was derived from a combination of sampling information eligibility for earlier interviews interview outcome information and information fed forward from HSE plus previous ELSA interviews Variable Variable label Original source EligwO Eligibility for Wave 0 interview Derived 1 Not in household at Wave 0 or unknown at HSE 0 In household and ineligible for Wave 0 1 In household and eligible for Wave 0 Eligibility for Wave 1 interview Derived 1 Not in household at Wave 1 or unknown at Wave 1 0 In household and ineligible for Wave 1 1 In household and eligible for Wave 1 Eligibility for Wave 2 interview Derived 1 Not in household at Wave 2 or unknown at Wave 2 0 In household and ineligible for Wave 2 1 In household and eligible for Wave 2 Eligibility for Wave 3 interview Derived 1 Not in household at Wave 3 or unknown at Wave 3 0 In household and ineligible for Wave 3 1 In household and eligible for Wave 3 Notes 1 The updating of ELIGWO for the refreshment sample is more complex and will not be completed until the HSE data for these cases is archived as an addition to the existing Wave 0 dat
30. iew 210 Partial CAPI interview 340 No contact 431 Refusal before interview personal 432 Refusal before interview proxy 440 Refusal during interview 450 Broken appointment no recontact 510 Ill at home during survey period 521 Away during survey period 540 Language difficulties OutscwO HSE self completion outcome HSE survey data variables not Not asked was third child in household archived elsewhere Not asked was aged less than 8 Not applicable see User Guide Fully completed Partially completed Not completed Outhhw1 Wave 1 household outcome updated with mortality Wave 1 variable RTHHOUT information updated with mortality data 2 Not in household or ineligible for interview 110 Fully productive all in person User Guide 6 Outindw1 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 121 Fully productive part in person part by proxy 122 Fully productive all by proxy 210 Partially productive some but not all eligible resp interviewed 310 No contact with anyone at address 330 Contact made but not with responsible resident 340 Contact made but not with eligible resident 410 Refusal to office 420 Refusal of info about occupants of address 431 Refusal at intro before HH module by eligible respondent 432 Refusal at intro before HH module by other 440 Refusal after HH module 450 Broken appointment no recontact 510 Ill at home during survey period 520 Away in hospital all
31. ight Derived Scw3wgt Wave 3 core self completion weight Derived Notes 1 The weighting variables are the only ones in the Index File where missing values have not been assigned and therefore those cases that do not have a value are system missing in the file This is to ensure that no values are erroneously used if the weights are applied to the data User Guide 16 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 6 Mortality Variables Information about whether respondents have died has been obtained from a number of sources Collected by interviewers during fieldwork Communication from relatives and others between Waves of fieldwork From the Department for Work and Pensions DWP as part of our procedure to try and trace respondents who have moved We only obtain information about those respondents who have given us permission to link their data to DWP records From the National Health Service Central Register NHSCR held by the Office of National Statistics ONS Again we only obtain information about those respondents who have given us permission to link their data to NHSCR records METTE Variable label Original source Mortfin Current mortality status from all sources Derived from NHSCR records 1 00 Alive not known to have died Wave outcomes and panel 2 00 Dead maintenance contact data Yrdeath Year of death Derived from NHSCR records 3 Details being checked with ONS 2 Info
32. interviewed 310 No contact with anyone at address 330 Contact made but not with responsible adult 340 Contact made but not with eligible respondents 410 Office refusal 420 Complete refusal of info about occupants of address 430 Refusal at intro before HH module 440 Refusal after HH module completed 450 Broken appointment 510 Ill at home during survey period 520 Away from home in hospital during survey period 530 Physically mentally unable incompetent 540 Language difficulties 560 Other unproductive 610 Address not issued to interviewer 630 Unable to locate address 682 Moved address obtained but not attempted 700 Address does not exist User Guide 8 Outindw2 Outscw2 Outhhw3 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 720 Address demolished derelict 730 Address vacant empty 780 All eligible respondents moved out of Britain 781 All eligible respondents moved into institution 791 All eligible respondents died 910 No follow up address obtained Wave 2 individual outcome updated with mortality W2INDOUT from Wave 2 main information data with mortality updates 2 Not in household or ineligible for interview 11 Full interview in person 13 Full proxy interview 21 Partial interview in person 23 Partial interview by proxy 31 No contact 43 Refusal before interview 44 Refusal during interview 45 Broken appointment 46 Office refusal 51 Ill at home during survey period 52
33. ived from ONS and fed back into the data So using the time periods associated with the sampling of Waves 1 2 and 3 as an example User Guide 18 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Pre Wave 1 Issue this refers to all dates prior to being issued for the Wave 1 interview The sample for Wave 1 was issued in three batches with a deadline assigned for each batch If someone had died before this deadline then the time period was labelled as pre Wave 1 issue The date when the batch was issued was assumed to be the first day of that month to make the calculations easier Pre Wave 1 Interview this refers to the period between being issued for the Wave 1 interview and the interviewer trying to make contact Post Wave 1 Issue and Pre Wave 2 Issue this refers to the period between the interviewer completing the Wave 1 interview and the respondent being issued for the Wave 2 interview The Wave 2 sample was issued in monthly batches with a deadline assigned for each batch Pre Wave 2 Interview this refers to the period between being issued for the Wave 2 interview and the interviewer trying to make contact Post Wave 2 Issue and Pre Wave 3 Issue this refers to the period between the interviewer completing the Wave 2 interview and the respondent being issued for the Wave 3 interview The Wave 3 sample was issued in two batches with a deadline being assigned for each batch 7 Demographic Variabl
34. le 1 Dead 2 Refused re interview 3 Details incorrect insufficient 8 Other ineligible 9 Other non response 10 Issued Issuew2 Reason for not being issued at Wave 2 Derived 10 Not applicable as ineligible for interview 9 Not known at time 8 No CM respondent in HH 7 CM non respondent 5 Part of Wave 3 Refreshment Sample Dead Refused re interview Details incorrect insufficient Moved out of GB Moved into institution 0 Issued of WD Issuew3 Reason for not being issued at Wave 3 Derived 10 Not applicable as ineligible for interview 9 Not known at time of issuing 8 No OM respondent in HH 7 CM non respondent 1 Dead 2 Refused re interview 3 Details incorrect insufficient 4 Moved out of GB 6 Not issued but did have productive interview at Wave 3 7 Other reason 8 Other ineligible 9 Other non response 10 Issued Notes 1 18 people from 11 households were mistakenly classified as not having an HSE 1999 interview at the sampling stage Their IDAUNIQ values are 108937 109287 109055 108293 109334 108983 109395 109335 108981 108880 109282 109005 109136 108986 109065 109124 109255 and 108906 These cases have a Wave 0 outcome code of 53 full interview in translation If these people had been correctly categorised they would have been issued for interviewing Also note that residing within these 11 households are two additional people One was a person who was known to have died prior to fieldwork IDAUNIQ
35. or people such as NP2 s This has not been repeated for FINSTAT3 Users of the data are advised to look at the variables denoting eligibility for interview as well as those showing whether someone was issued for each Wave for the reasons why cases are missing FINSTAT s at any wave iE TOES Finstat Finstat2 Variable label Final status after Wave 1 fieldwork C1CM core sample member first responded at W1 C1NP1 new partner found W1 first responded at W1 C1NP2 new partner found W2 responded at W2 C1YP younger partner first responded at W1 C1SM notiss unconfirmed sample member not issued at W1 C1SM_unprod sample member first non response at W1 C1CP_unprod core sample member first non response at W1 but partner of a C1CM C1YP_notiss unconfirmed younger partner not issued at W1 C1YP_unprod younger partner first non response at W1 C1NP1_unprod new partner found W1 first non response at W1 C1NP2 unprod new partner found W2 non response at W2 See ELIGW1 was either in the household and ineligible or not in the household at W1 See ELIGW1 for further details Final status after Wave 2 fieldwork C1CM core sample member responded at W1 amp W2 C1CP core sample member partner of a C1CM non response at W1 responded at W2 Original source Derived C1 indicates the individuals were sampled as part of cohort 1 at Wave 1 CM indicates the individuals are considered to be in the ELSA pan
36. partner at W2 non responder at W2 121360 121384 new partner at W3 responded at W3 121385 121399 new partner at W3 non responder at W3 150001 151609 W3 refreshment sample issued responded at W3 151610 152470 W3 refreshment sample issued non responder at W3 152471 152491 new partner of W3 refreshment sample responded at W3 152492 152503 new partner of W3 refreshment sample non responder at W3 152504 152858 W3 refreshment sample issued non responder at W3 152860 152974 HSE 2001 not issued at W3 refreshment sample 152975 153087 HSE 2002 not issued at W3 refreshment sample 153088 153249 HSE 2003 not issued at W3 refreshment sample 153250 153344 HSE 2004 not issued at W3 refreshment sample 700001 706440 non eligible individuals in core sample households 800001 801826 non eligible individuals in W3 refreshment sample households Notes 1 IDAUNIQ 121359 was assigned to someone thought to be new partner but who was then found to be an existing sample member 2 IDAUNIQ 150589 was assigned to someone thought to be a responding refreshment sample member when in fact they were a non responder 3 IDAUNIQ 152589 is a non responding new partner of a Wave 3 refreshment sample member 4 IDAUNIQ 705145 became eligible for interview at Wave 3 by becoming the partner of the ELSA sample member in the household but their IDAUNIQ has been left as originally assigned when
37. rmation not available or not applicable Otherwise gives year of death Agedead2 Age at death with those aged 90 or over re coded to avoid Derived from NHSCR records disclosure 3 Details being checked with ONS 2 Information not available or not applicable 99 Aged 90 or over Otherwise gives age at death Maincod Main cause of death using ICD 9 and 10 grouped Derived from NHSCR records 2 ICD information not available 1 Alive or no permission given to check NHSCR records 1 Cancer 2 CVD 3 Respiratory 4 Other Mortwave Data source used and when died in relation to survey Derived from NHSCR records Wave see further note below wave outcomes and panel 5 Joined household after HSE and ineligible for interview maintenance contact data Will not have mortality data 3 Details being checked with ONS 0 00 A Alive not known to have died 11 00 Pre Wave 1 issue Wave 1 sampling 12 00 Pre Wave 1 Wave 1 fieldwork 13 00 Pre Wave 1 issue external update 21 00 Post Wave 1 issue amp pre Wave 2 issue Wave 2 sampling 22 00 Pre Wave 2 Wave 2 fieldwork 23 00 Post Wave 1 issue amp pre Wave 2 issue external update 31 00 Post Wave 2 issue amp pre Wave 3 issue Wave 3 User Guide 17 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing sampling 32 00 Pre Wave 3 Wave 3 fieldwork 33 00 Post Wave 2 issue amp pre Wave 3 issue external update 43 Post Wave 3 issue amp pre Wave 4 issue exte
38. rnal update Notes 1 We have checked the quality of the mortality data by looking at sex and date of birth For every case sex matches but for 86 of the cases the date of birth is different either day month year or more than one of these components For the cases that did not have the same date of birth the name of the person has been compared in the mortality data against HSE data Occasionally there were small differences between the names that were probably typographical errors 2 There were a number of respondents where NHSCR records did not indicate that they had died but they are flagged as being found to be dead found during fieldwork or other communications 3 After Wave 1 of ELSA we introduced the End of Life Interview for which we approach a close friend or relative of an eligible ELSA respondent who has died since Wave 1 to do an interview about the deceased During Wave 3 fieldwork we found that a small number of respondents who we had been informed were dead via one of the sources listed above were in fact still alive These respondents have had their mortality information corrected in this Index file Any other cases discovered after this release of the data has been deposited will not be updated until we produce a second release of the Index File We expect the number of such cases to be very small if any at all The majority of these respondents were thought to have died in between the time of the Wave 2 intervi
39. s to a non response weight of 1 25 Two extreme weights at the top end of the distribution were found These were trimmed In addition to avoid other large weights having an undue influence on the estimates the non response weight was trimmed at the 97 5 percentile in a similar fashion to the Wave 2 self completion weight e The final weight to analyse the Wave 3 data SCW3WGT was obtained by multiplying the initial weight W3XWGT with the trimmed non response weight The weight was then scaled to sum to the number of respondents to the self completion questionnaire 7 406 ELSA WAVE 2 e The analysis focused on the ELSA core sample members only e Non respondents to the self completion questionnaire were defined by values of OUTSCW2 2 or 3 ELSA core members who were interviewed by proxy were not included in the analysis and were not issued with the self completion questionnaire e Using the Wave 2 interview weight W2WGT the nature of response to the self completion questionnaire was examined The variables examined included Age group Sex Urban rural indicator Marital status Government Office Region ndex of Multiple Deprivation 2004 quintiles Ethnic group white non white A Financial unit type Tenure Education qualifications Whether has living child Number of people in household Current status i e retired employed permanently sick or disabled looking after home or family Equi
40. they were not eligible for interview User Guide 3 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing 5 There are 24 values of IDAUNIQ beginning with a 7 excluding 705145 which belong to individuals that have been eligible for at least one Wave of ELSA These cases were identified as being among those from HSE 2001 where the same household was selected twice for ELSA once as part of the original sample and once as part of the Refreshment Sample Please see the User Guide of the latest version of the Wave 3 Core Data for further clarification of this issue and section 2 5 of this User Guide for the variables which show the identifiers that these cases were previously known as Person Number Each person within the household was given a number starting from 1 at the time of the HSE interview PERSNO These person numbers were fed forward into the ELSA interview and are intended to remain the same over time The numbering was continued for new people that entered the household after the HSE interview e g if there were three people in the household at HSE and someone else was found to have joined the household at ELSA Wave 3 they would be assigned a person number of 4 METIETST ETETIEEETTI Original source Persno Person number within household Included in original HSE sampling files or filled in from household grid at first Wave they were interviewed Present for all individuals in the Index File
41. tionnaire was based on a logistic regression model of response 1 response 0 non response with the predictor variables entered by a forward stepwise selection procedure The variables remaining in the optimal model were Age group Government Office Region ndex of Multiple Deprivation quintiles Ethnic group white non white Type of benefit unit Education qualifications Whether took part in voluntary work or looked after home or family in the last month Eyesight condition Hearing condition The predicted response probabilities from the final optimal model were converted into a non response weight by obtaining their reciprocal e g a response probability of 0 8 corresponds to a non response weight of 1 25 A number of extreme weights at the top end of the distribution were found To avoid extreme weights having a large influence on the estimates the non response weight was trimmed at the 97 5 percentile The final weight to analyse the data SCW1WGT was obtained by multiplying the initial weight W1WGT with the trimmed non response weight The weight was then scaled to sum to the number of respondents to the self completion questionnaire 10 332 11 Contact Details Susan Nunn ELSA Senior Data Manager E mail susan nunn natcen ac uk Telephone 020 7250 1866 3 These two measures were used as proxy measures of general health Unlike these two measures the general health variables themselves e g HEGE
42. valised income quintile User Guide 22 UK Data Archive Study Number 5050 English Longitudinal Study of Ageing Eyesight condition Hearing condition e Anon response weight to compensate for potential response bias to the self completion questionnaire was based on a logistic regression model of response 1 response 0 non response with the predictor variables entered by a forward stepwise selection procedure The variables remaining in the optimal model were Sex Age group Marital status Whether has living child Index of Multiple Deprivation quintiles Ethnic group white non white Type of financial unit Tenure Education qualifications ncome quintile Current activity Self reported eyesight e The predicted response probabilities from the final optimal model were converted into a non response weight by obtaining their reciprocal e g a response probability of 0 8 corresponds to a non response weight of 1 25 A number of extreme weights at the top end of the distribution were found To avoid extreme weights having a large influence on the estimates the non response weight was trimmed at the 97 5 percentile e The final weight to analyse the data SCW2WGT was obtained by multiplying the initial weight W2WGT with the trimmed non response weight The weight was then scaled to sum to the number of respondents to the self completion questionnaire 7 803 ELSA WAVE 1 e The analysis
43. y unable incompetent 540 Language difficulties 550 Respondent requested data deletion 560 Other unproductive 612 OFFICE APPROVAL ONLY Address not attempted 620 OFFICE APPROVAL ONLY Address inaccessible 630 OFFICE APPROVAL ONLY Unable to locate address 710 In an institution unproductive 780 All eligible respondents moved out of Britain 781 NO follow up for institution address obtained 790 OFFICE APPROVAL ONLY All of the sample are ineligible 791 All eligible respondents died updated 910 Moved no follow up address obtained 920 Moved follow up address outside area Wave 3 individual outcome updated with mortality W3INDOUT from Wave 3 main information data with mortality updates 2 Not in household or ineligible for interview 11 Full interview in person 13 Full interview by proxy 21 Partial interview in person 24 Institutional interview in person 25 Institutional interview by proxy 31 No contact 43 Refusal before interview 44 Refusal during interview computed 45 Broken appointment no re contact 46 Office refusal 51 Ill at home during survey period 52 Away ill in hospital during survey period 53 Physically or mentally unable incompetent 54 Language difficulties 55 Respondent requested data deletion 56 Other unproductive 60 Untraced 71 Ina institution unproductive 77 Ineligible partner who is now not living with Core Member 78 Out of Britain 79 Ineligible issued in error 95 Die
Download Pdf Manuals
Related Search