Current Data Issues

This page is updated as necessary with any known issues with the HILDA data.

Release 8.0 Data and Documentation Issues and Corrections

Update 27/04/2010

Following a crash, an update to the notes for the 5 personality scale variables in the wave 5 derived variable coding frameworks was lost. The notes should have been updated for:
epnextrv 'DV: SCQ:B19 Personality scale - Extroversion'
epnagree 'DV: SCQ:B19 Personality scale - Agreeableness '
epnconsc 'DV: SCQ:B19 Personality scale - Conscientiousness'
epnemote 'DV: SCQ:B19 Personality scale - Emotional stability'
epnopene 'DV: SCQ:B19 Personality scale - Openness to experience'.

The notes should say:
In wave 5 respondents were questioned on their personality character traits using a 36-item inventory. The approach used was based on the trait descriptive adjectives approach used by Saucier (1994), which in turn was based on the approach employed by Goldberg (1992), both of which assume a 5-factor structure (as is commonly assumed in the literature). Not all 36 items, however, are used in the five derived scales summarizing the 5 personality factors. First, the ex ante scales were tested for item reliability, with any items omitted if item total correlation was less than 0.3. Second, principal components analysis with a five factor solution was undertaken, with items only retained if their highest factor loading was on the expected factor, exceeded 0.4 and exceeded the second highest factor loading by at least 0.1. A slightly different approach to derivation of these scales, but which obtains identical conclusions, is provided in Losoncz (2009).

References:
Goldberg, L.R. (1992), ‘The Development of Markers for the Big-Five Factor Structure’, Psychological Assessment, vol. 4, pp. 26–42.
Losoncz, I. (2009), ‘Personality Traits in HILDA’, Australian Social Policy, pp. 169–198.
Saucier, G. (1994), ‘Mini-markers: A Brief Version of Goldberg’s Unipolar Big-five Markers’, Journal of Personality Assessment, vol. 63, pp. 506–516

Update 07/04/2010

Replacement General Release Wave 3 cover sheet (“Cover sheet c80c.pdf” on the HILDA DVD contained a copy of the HF).

Update 12/03/2010

There is an incorrect xwaveid in the Eperson and Combined files for wave 8. This was a child that fieldwork treated as a new entrant in wave 8, but she had been become part of the sample in wave 2.
The correct xwaveid was supplied in the master file but the correction was not applied to the Eperson file and, in consequence, the Combined file. The incorrect xwaveid will cause a mismatch when matching the master file to the Eperson or Combined files, or when creating longitudinal files.

SPSS code (apply to Eperson and Combined files):
if hhhrpid ='2328203' and xwaveid = '0800238' xwaveid = '0200519'.

SAS code (apply to Eperson and Combined files):
if hhhrpid ='2328203' and xwaveid = '0800238' then xwaveid = '0200519';

Stata code:
use Combined_h80c, clear
replace xwaveid = "0200519" if hhhrpid=="2328203" & xwaveid=="0800238"
save Combined_h80c, replace
use Eperson_h80c, clear replace xwaveid = "0200519" if hhhrpid=="2328203" & xwaveid=="0800238"
save Eperson_h80c, replace

Update 10/03/2010

The derived variable coding frameworks supplied on the DVD inappropriately excluded the derived History: variables.

A zip of replacement General Release frameworks for each of the 8 waves is available for download in Derived Variable Coding Framework 80 replacements.zip (3Mb).

A zip of replacement In-Confidence frameworks for each of the 8 waves is available for download in In-Confidence Derived Variable Coding Framework 80u replacements.zip (3Mb).

Release 7.0 Data and Documentation Issues and Corrections

Update 10/09/2009

Two wave 7 sample weights need to be overwritten. The supplied versions have a mean greater than 1 and will inflate the sample size.

The two weights are ghhwtes 'DV: Enumerated person sample weight', and ghhwths 'DV: Household sample weight'.

Corrections:
if ghhwte >= 0 ghhwtes = ghhwte * (17281 / 20427816.90).
if ghhwth >= 0 ghhwths = ghhwth * ( 7063 / 8147207.53 ).

Update 01/09/2009

For longitudinal consistency sex should be swapped for these two Wave 7 respondents:
Wave 7 household id 59111 ghgsex1 from 2 to 1 and ghgsex2 from 1 to 2;
xwaveid 0113924 from 2 to 1 and 0113925 from 1 to 2.

Wave last interviewed:
Due to a programming error, 223 Wave 7 Continuing Persons were set to not asked (-1) on wave last interviewed (ghgwli) and date last interviewed (ghgdli).
Replacement variables (by xwaveid) are provided in this spreadsheet.

Update 15/07/2009

The flag BHWLINK indicating if household wealth was longitudinally imputed (whether a wave 2 household could be linked to a wave 6 household) was not written into the wave 2 household file. A spreadsheet containing BHWLINK and the wave 2 randomised household id can be downloaded here.

Update 17/03/2009

The imputation flags for the household-level wealth variables for own credit card debt and joint credit card debt only contain missing values.

There was an error in the program which created them (the variable name specified for the unimputed variable was missing the final “t”).

You can create your own versions of these flags:
bhwoccdf =(bhwoccdi <> bhwoccdt).
bhwccdtf =(bhwccdti <> bhwccdt).
fhwoccdf =(fhwoccdi <> fhwoccdt).
fhwccdtf =(fhwccdti <> fhwccdt).

This says the flag (the first variable) equals the logical test of comparing the imputed and unimputed variables, giving a result of 1 if they are not equal and zero if they are equal.

Update 16/03/2009

Replacement marked up wave 1 Household Questionnaire, labelled 80, but no changes from release 70.

Update 23/02/2009

There is a data issue with the 21 wave 3 SCQ "Life Events occurs" variables (got married; separated from spouse etc).
The error was introduced post release 6.0, when, for consistency, a program applicable to later waves was brought back into the wave 3 SCQ cleaning program.

The problem only affects wave 3, the 6 other waves are unaffected.

The following variables need to be corrected:
clemar, clesep, clercl, cleprg, clebth, cleins, cleinf, cledsc, cledrl, cledfr, clevio, clepcm, clejls, clejlf, clertr, clefrd, clejob, cleprm, clefni, clefnw, clemvd

The problem can be corrected in SPSS, STATA or SAS using the code provided below.

These files require correction:
Combined c70c (or Combined c70u for in-confidence users); and Rperson c70c (or Rperson c70u for in-confidence users)

The time period variables for the quarter the wave 3 life event occurred are not affected.

Frequencies or tables should show the following numbers of "2 Yes"

STATA, SPSS, SAS fixes
======================

STATA

replace clemar=2 if ((clemar==-4 | clemar==1) & (clemarq1==1 | clemarq2==1 | clemarq3==1 | clemarq4==1))
replace clesep=2 if ((clesep==-4 | clesep==1) & (clesepq1==1 | clesepq2==1 | clesepq3==1 | clesepq4==1))
replace clercl=2 if ((clercl==-4 | clercl==1) & (clerclq1==1 | clerclq2==1 | clerclq3==1 | clerclq4==1))
replace cleprg=2 if ((cleprg==-4 | cleprg==1) & (cleprgq1==1 | cleprgq2==1 | cleprgq3==1 | cleprgq4==1))
replace clebth=2 if ((clebth==-4 | clebth==1) & (clebthq1==1 | clebthq2==1 | clebthq3==1 | clebthq4==1))
replace cleins=2 if ((cleins==-4 | cleins==1) & (cleinsq1==1 | cleinsq2==1 | cleinsq3==1 | cleinsq4==1))
replace cleinf=2 if ((cleinf==-4 | cleinf==1) & (cleinfq1==1 | cleinfq2==1 | cleinfq3==1 | cleinfq4==1))
replace cledsc=2 if ((cledsc==-4 | cledsc==1) & (cledscq1==1 | cledscq2==1 | cledscq3==1 | cledscq4==1))
replace cledrl=2 if ((cledrl==-4 | cledrl==1) & (cledrlq1==1 | cledrlq2==1 | cledrlq3==1 | cledrlq4==1))
replace cledfr=2 if ((cledfr==-4 | cledfr==1) & (cledfrq1==1 | cledfrq2==1 | cledfrq3==1 | cledfrq4==1))
replace clevio=2 if ((clevio==-4 | clevio==1) & (clevioq1==1 | clevioq2==1 | clevioq3==1 | clevioq4==1))
replace clepcm=2 if ((clepcm==-4 | clepcm==1) & (clepcmq1==1 | clepcmq2==1 | clepcmq3==1 | clepcmq4==1))
replace clejlf=2 if ((clejlf==-4 | clejlf==1) & (clejlfq1==1 | clejlfq2==1 | clejlfq3==1 | clejlfq4==1))
replace clejls=2 if ((clejls==-4 | clejls==1) & (clejlsq1==1 | clejlsq2==1 | clejlsq3==1 | clejlsq4==1))
replace clertr=2 if ((clertr==-4 | clertr==1) & (clertrq1==1 | clertrq2==1 | clertrq3==1 | clertrq4==1))
replace clefrd=2 if ((clefrd==-4 | clefrd==1) & (clefrdq1==1 | clefrdq2==1 | clefrdq3==1 | clefrdq4==1))
replace clejob=2 if ((clejob==-4 | clejob==1) & (clejobq1==1 | clejobq2==1 | clejobq3==1 | clejobq4==1))
replace cleprm=2 if ((cleprm==-4 | cleprm==1) & (cleprmq1==1 | cleprmq2==1 | cleprmq3==1 | cleprmq4==1))
replace clemvd=2 if ((clemvd==-4 | clemvd==1) & (clemvdq1==1 | clemvdq2==1 | clemvdq3==1 | clemvdq4==1))
replace clefni=2 if ((clefni==-4 | clefni==1) & (clefniq1==1 | clefniq2==1 | clefniq3==1 | clefniq4==1))
replace clefnw=2 if ((clefnw==-4 | clefnw==1) & (clefnwq1==1 | clefnwq2==1 | clefnwq3==1 | clefnwq4==1))

SPSS

if any(clemar,-4,1) and any(1,clemarq1,clemarq2,clemarq3,clemarq4) clemar = 2.
if any(clesep,-4,1) and any(1,clesepq1,clesepq2,clesepq3,clesepq4) clesep = 2.
if any(clercl,-4,1) and any(1,clerclq1,clerclq2,clerclq3,clerclq4) clercl = 2.
if any(cleprg,-4,1) and any(1,cleprgq1,cleprgq2,cleprgq3,cleprgq4) cleprg = 2.
if any(clebth,-4,1) and any(1,clebthq1,clebthq2,clebthq3,clebthq4) clebth = 2.
if any(cleins,-4,1) and any(1,cleinsq1,cleinsq2,cleinsq3,cleinsq4) cleins = 2.
if any(cleinf,-4,1) and any(1,cleinfq1,cleinfq2,cleinfq3,cleinfq4) cleinf = 2.
if any(cledsc,-4,1) and any(1,cledscq1,cledscq2,cledscq3,cledscq4) cledsc = 2.
if any(cledrl,-4,1) and any(1,cledrlq1,cledrlq2,cledrlq3,cledrlq4) cledrl = 2.
if any(cledfr,-4,1) and any(1,cledfrq1,cledfrq2,cledfrq3,cledfrq4) cledfr = 2.
if any(clevio,-4,1) and any(1,clevioq1,clevioq2,clevioq3,clevioq4) clevio = 2.
if any(clepcm,-4,1) and any(1,clepcmq1,clepcmq2,clepcmq3,clepcmq4) clepcm = 2.
if any(clejls,-4,1) and any(1,clejlsq1,clejlsq2,clejlsq3,clejlsq4) clejls = 2.
if any(clejlf,-4,1) and any(1,clejlfq1,clejlfq2,clejlfq3,clejlfq4) clejlf = 2.
if any(clertr,-4,1) and any(1,clertrq1,clertrq2,clertrq3,clertrq4) clertr = 2.
if any(clefrd,-4,1) and any(1,clefrdq1,clefrdq2,clefrdq3,clefrdq4) clefrd = 2.
if any(clejob,-4,1) and any(1,clejobq1,clejobq2,clejobq3,clejobq4) clejob = 2.
if any(cleprm,-4,1) and any(1,cleprmq1,cleprmq2,cleprmq3,cleprmq4) cleprm = 2.
if any(clefni,-4,1) and any(1,clefniq1,clefniq2,clefniq3,clefniq4) clefni = 2.
if any(clefnw,-4,1) and any(1,clefnwq1,clefnwq2,clefnwq3,clefnwq4) clefnw = 2.
if any(clemvd,-4,1) and any(1,clemvdq1,clemvdq2,clemvdq3,clemvdq4) clemvd = 2.

SAS

if ((clemar=-4 or clemar=1) and (clemarq1=1 or clemarq2=1 or clemarq3=1 or clemarq4=1)) then clemar = 2;
if ((clesep=-4 or clesep=1) and (clesepq1=1 or clesepq2=1 or clesepq3=1 or clesepq4=1)) then clesep = 2;
if ((clercl=-4 or clercl=1) and (clerclq1=1 or clerclq2=1 or clerclq3=1 or clerclq4=1)) then clercl = 2;
if ((cleprg=-4 or cleprg=1) and (cleprgq1=1 or cleprgq2=1 or cleprgq3=1 or cleprgq4=1)) then cleprg = 2;
if ((clebth=-4 or clebth=1) and (clebthq1=1 or clebthq2=1 or clebthq3=1 or clebthq4=1)) then clebth = 2;
if ((cleins=-4 or cleins=1) and (cleinsq1=1 or cleinsq2=1 or cleinsq3=1 or cleinsq4=1)) then cleins = 2;
if ((cleinf=-4 or cleinf=1) and (cleinfq1=1 or cleinfq2=1 or cleinfq3=1 or cleinfq4=1)) then cleinf = 2;
if ((cledsc=-4 or cledsc=1) and (cledscq1=1 or cledscq2=1 or cledscq3=1 or cledscq4=1)) then cledsc = 2;
if ((cledrl=-4 or cledrl=1) and (cledrlq1=1 or cledrlq2=1 or cledrlq3=1 or cledrlq4=1)) then cledrl = 2;
if ((cledfr=-4 or cledfr=1) and (cledfrq1=1 or cledfrq2=1 or cledfrq3=1 or cledfrq4=1)) then cledfr = 2;
if ((clevio=-4 or clevio=1) and (clevioq1=1 or clevioq2=1 or clevioq3=1 or clevioq4=1)) then clevio = 2;
if ((clepcm=-4 or clepcm=1) and (clepcmq1=1 or clepcmq2=1 or clepcmq3=1 or clepcmq4=1)) then clepcm = 2;
if ((clejlf=-4 or clejlf=1) and (clejlfq1=1 or clejlfq2=1 or clejlfq3=1 or clejlfq4=1)) then clejls = 2;
if ((clejls=-4 or clejls=1) and (clejlsq1=1 or clejlsq2=1 or clejlsq3=1 or clejlsq4=1)) then clejlf = 2;
if ((clertr=-4 or clertr=1) and (clertrq1=1 or clertrq2=1 or clertrq3=1 or clertrq4=1)) then clertr = 2;
if ((clefrd=-4 or clefrd=1) and (clefrdq1=1 or clefrdq2=1 or clefrdq3=1 or clefrdq4=1)) then clefrd = 2;
if ((clejob=-4 or clejob=1) and (clejobq1=1 or clejobq2=1 or clejobq3=1 or clejobq4=1)) then clejob = 2;
if ((cleprm=-4 or cleprm=1) and (cleprmq1=1 or cleprmq2=1 or cleprmq3=1 or cleprmq4=1)) then cleprm = 2;
if ((clemvd=-4 or clemvd=1) and (clemvdq1=1 or clemvdq2=1 or clemvdq3=1 or clemvdq4=1)) then clefni = 2;
if ((clefni=-4 or clefni=1) and (clefniq1=1 or clefniq2=1 or clefniq3=1 or clefniq4=1)) then clefnw = 2;
if ((clefnw=-4 or clefnw=1) and (clefnwq1=1 or clefnwq2=1 or clefnwq3=1 or clefnwq4=1)) then clemvd = 2;

Update 06/02/2009

We have identified a problem with one case in the enumerated person file and combined file for waves 6 and 7. The cross-wave identifier for this case is incorrect so when you match files across waves or to the master file using xwaveid, this person will not be matched correctly. To fix the problem, the following code should be run on the enumerated file and the combined file for both wave 6 and wave 7:

SAS: if xwaveid='0600941' then xwaveid='0112961';

SPSS: if xwaveid='0600941' xwaveid='0112961'.

Stata: replace xwaveid = "0112961" if xwaveid == "0600941"

Data managers – please make the HILDA users in your organisation aware of this problem.

Release 6.0 Data and Documentation Issues and Corrections

Update 04/04/2008

Current data and documentation issues
Replacement Household Form e60c.pdf