Improving Survey Measurement of Income and Employment, 2001-2003

<P>Abstract copyright UK Data Service and data collection copyright owner.</P><br>The <i>Improving Survey Measurement of Income and Employment</i> (ISMIE) survey was undertaken to analyse issues of data validation and dependent interviewing. The end of funding for the <i>European Community Household Panel</i> (ECHP) study in 2001 presented a rare opportunity to collect data for this purpose. The validation exercise had two parts: comparisons of survey reports of social security benefit income with administrative records, and of survey reports of employment characteristics (pay, hours, status, etc.) with employer records. The survey also contained an experiment to test alternative dependent interviewing strategies, and compare them with traditional independent interviewing, in terms of impact on validity and accuracy. The data currently deposited are the household survey and the validation data collected from employers.<br>
<br>
The documentation for this study describes the survey design, methodological work and data structure, in so far as it differs from standard practices for the <i>British Household Panel Survey</i> (BHPS). Dataset users are referred to the BHPS documentation for background information on the ECHP subsample and general survey processes and data characteristics. The BHPS documentation is available from the <a href="http://www.iser.essex.ac.uk/bhps/documentation" title="BHPS User Documentation and Questionnaires"> Institute for Social and Economic Research</a> (ISER) web site.<br>
<br>
For the second edition, files 'lincom' and 'lincpay' were re-deposited due to previous double reporting problems, and the user guide was updated accordingly.<br><br><B>Main Topics</B>:<BR><br>The dataset covers a broader range of thematic areas including household composition, housing conditions, residential mobility, education and training, health, socio-economic values, income from employment, benefits and pensions. Furthermore, detailed information was collected about respondents' economic activities and activity spells since the previous interview.<br>
<br>
With regard to the structure of the records, one set contains information at the household level, another set includes information at the individual level, and a third set is organised by substantive contents, for example job spells or income sources. In addition, for the sections subject to the dependent interviewing experiment, there are separate files for each treatment group.