Survey Design and Measurement
SDMI events
11 October 2011 - Mode Effects on Survey Measurement, one day conference
5 April 2011 - New Perspectives on Survey Nonresponse and Attrition, one day conference
21 September 2010 - New Measurement Issues in Survey Research, one day conference
11 October 2011 - Mode Effects on Survey Measurement
The Royal Statistical Society, Errol Street, London
Programme
Causes of mode effects on survey measurement (PDF)
How and when does the mode of data collection affect survey measurement? (PDF)
The use of cognitive interviewing methods to evaluate mode effects (PDF)
The role of visual and aural stimuli in producing mode effects (PDF)
The role of the interviewer in producing mode effects (PDF)
Is it a good idea to optimise question format for mode of data collection? (PDF)
Designing better mixed mode studies (PDF)
Handout (PDF)
5 April 2011 - New Perspectives on Survey Nonresponse and Attrition: Findings from the ESRC Survey Design and Measurement Initiative
The Royal Statistical Society, Errol Street, London
A serious and endemic threat to valid inference from surveys is differential nonresponse. Where sample members differ on analysis variables from those who are not contacted or who choose not to participate, estimators are biased and less precise than they would be under complete response. Yet, despite a growing literature on the topic, the causes of nonresponse are still rather poorly understood. This one day conference presents findings from four projects funded under the ESRC Survey Design and Measurement Initiative which adopt a range of research designs to shed new light on this important topic.
Programme
10.00 Introduction: Patrick Sturgis, University of Southampton (Chair)
10.15 The role of contact data in weighting to adjust for nonresponse bias (PDF): Annelies Blom, University of Mannheim and Peter Lynn, University of Essex
10.45 Experiments with methods to reduce attrition in longitudinal surveys (PDF): Peter Lynn, University of Essex
11.15 Coffee/tea
11.45 Response propensity weighting for nonresponse using multilevel models (PDF): Chris Skinner, University of Southampton
12.15 Discussant comments by Patten Smith, Ipsos-MORI
12.45 Lunch
1.45 Is skill in obtaining contact and cooperation correlated with the magnitude of interviewer variance? (PDF): Patrick Sturgis, University of Southampton
2.15 Factors affecting attrition and nonresponse in longitudinal surveys (PDF): John Bynner, Longview | Accompanying paper (PDF)
2.45 Coffee/tea
3.15 Panel attrition: How important is it to keep the same interviewer? (PDF): Olena Kaminska, University of Essex
3.45 Discussant comments by Jelke Bethlehem, University of Amsterdam
4.15 General discussion
4.45 Close
21 September 2010 - New Measurement Issues in Survey Research: Findings from the ESRC Survey Design and Measurement Initiative
The Royal Statistical Society, Errol Street, London
As social and technological change apply increasing downward pressure on response rates and upward pressure on costs, new and innovative methods are increasingly required to maintain and improve the quality of survey measurement. It seems inevitable that surveys in the future will rely increasingly on mixed mode strategies, drawing together responses elicited face-to-face, online, via telephone interview, and other forms of administration. Similarly, ‘routine’ and administrative information will be obtained via automated procedures and through linking to administrative and other forms of commercial and ‘transactional’ data. At the same time, changing patterns of cohabitation and the fracturing of traditional family units will increasingly problematise the conventional household as the appropriate unit of data collection and analysis. This one day conference draws together findings from projects funded under the ESRC Survey Design and Measurement Initiative which, collectively, focus on these key issues of survey measurement and quality in a rapidly changing social and technological landscape.
Programme
10.45 Welcome and Introduction: Patrick Sturgis, University of Southampton
11.00 Designing questions for mixed mode data collection: What have we learnt? (PPT): Gerry Nicolaas, National Centre for Social Research
11.45 Using scanner technology to collect expenditure data (PDF): Andrew Leicester, Institute for Fiscal Studies
12.30 Lunch
1.30 Whose household? The fuzziness of a critical concept in household surveys (PPT): Ernestina Coast, London School of Economics
2.15 Experimental thinkaloud protocols: a new method for evaluating the validity of survey questions (PPT): Patrick Sturgis, University of Southampton
3.00 Discussant comments (PDF): Fred Conrad, Survey Research Centre, University of Michigan
3.30 Discussion and questions from the audience
4.00 Close
