Kantar Public are proud to be presenting five methodological papers at the European Survey Research Association (ESRA) Conference in Ljubljana, Slovenia, taking place from the 15th to the 19th of July 2013.

Based on recent research projects, these illustrate TNS BMRB’s collaboration with clients towards improving methodology and driving value. The following papers will be presented:

Are ‘better’ interviewers more successful at engaging reluctant respondents?: Evidence from an experiment conducted on the 1970 British Cohort Study

Presented by Hannah Carpenter (Kantar Public) with Lisa Calderwood (Centre for Longitudinal Studies) and Matthew Brown (Centre for Longitudinal Studies) on 17th July

This paper evaluates the success of targeting particular interviewers at reluctant respondents by comparing the response rates achieved by two groups of interviewers and examining survey process data containing call records to assess the impact on fieldwork.

Developing an online calendar for life event histories

Presented by Emily Pickering (Kantar Public) with Matthew Brown (Centre for Longitudinal Studies) and Carrie Harding (Kantar Public) on 17th July

The functionality of the web can facilitate the use of visual and interactive event history calendars, which have previously been shown to increase the quality of reporting of events in longitudinal studies. The calendars collect histories across multiple domains and aid recall by providing immediate visual feedback across these domains.
This paper will discuss the development of the interactive calendar and describe the iterative programme of usability testing. It will emphasise the importance of allowing for multiple stages of testing by showing how significantly the research instrument changed during the development process as well as sharing videos that were recorded during the usability testing.

Using a visual calendar to improve the accuracy of event histories: Evidence from the 1970 British Cohort Study

Presented by Matthew Brown (Centre for Longitudinal Studies) with Emily Pickering and Hannah Carpenter (Kantar Public) on 17th July

By making comparisons with the data from prior follow-ups, this paper evaluates the extent to which a new visual calendar has affected the quality of the data collected. We anticipate that missing data, seam effects and the reporting of illogical dates will be reduced and that consistency between transitions across domains will be increased.

Using administrative data for tracking in longitudinal surveys: evidence from the UK Millennium Cohort Study and the 1970 British Cohort Study

Presented by Hannah Carpenter (Kantar Public) with Lisa Calderwood (Centre for Longitudinal Studies), Matthew Brown (Centre for Longitudinal Studies) and Angela Thompson (Ipsos MORI) on 19th July

This paper will present evidence on using administrative data to track participants on longitudinal studies, rather than effective but more expensive field-tracking. The paper looks at the most recent waves of the UK Millennium Cohort Study (MCS) and the 1970 British Cohort Study (BCS70) in 2012. Health records were used for tracking on both studies, and education records were also used on MCS. This paper will compare effectiveness of these two different types of administrative data for MCS, and will compare the effectiveness of tracking through health records for two different study populations; 11-year-old children in the case of MCS and 42-year-old adults in the case of BCS70. It examines the proportion of mover cases for which new addresses were found, and uses survey process data to ascertain whether the participants were successfully located and interviewed at these new addresses.

General population surveys on the web: new findings from the UK

Presented by Joel Williams (Kantar Public) on 19th July

This paper summarises findings from an experimental general population survey in England that used web as its primary data collection mode for one part of the sample, and traditional face-to-face interviewing for the other part. The paper highlights the implications for research commissioners when considering switching data collection modes. The work has parallels with work carried out in the US by Dillman and in the UK by the Office for National Statistics, albeit on a larger scale than any previous single study.