ESRA logo
Tuesday 14th July      Wednesday 15th July      Thursday 16th July      Friday 17th July     

Wednesday 15th July, 11:00 - 12:30 Room: HT-102

Modelling unit nonresponse and attrition processes 1

Convenor Ms Carina Cornesse (GIP, Mannheim University )
Coordinator 1Dr Gabriele Durrant (School of Social Sciences, University of Southampton)
Coordinator 2Professor Annelies Blom (GIP, Mannheim University)

Session Details

This session focuses on analysing the processes leading to unit nonresponse in cross-sectional and attrition in longitudinal data. Unit nonresponse and attrition are major issues affecting data quality in surveys. Their importance has increased over the past decades as response rates in the US and Europe have been decreasing across survey modes and nonresponse rates may be related to nonresponse bias.

When modelling the fieldwork processes leading to nonresponse, research can draw on auxiliary data sources. These may include paradata, such as call record data, interviewer observations, time stamps during the interview, or variables from external data sources, such as administrative, register and census data.

In recent years, the statistical techniques that have been developed to model unit nonresponse and attrition and applied to survey data have become increasingly sophisticated. In addition, both ex-post modelling to learn from previous fieldwork outcomes and real-time modelling to inform adaptive and responsive survey designs have found its way into the survey methodological realm.

For our session we invite submissions from researchers who model unit nonresponse and attrition processes. We specifically encourage submissions of papers that use auxiliary data to model unit nonresponse and attrition processes and papers that use complex statistical models for this purpose.

Paper Details

1. Using Call Record and Previous Wave Information to Predict Final Outcome and Length of Call Sequences in a Longitudinal Survey
Dr Gabriele Durrant (School of Social Sciences, University of Southampton)

This paper models call record data predicting final call outcome and length of a call sequence early on in the data collection process taking account of previous wave information and previous and current call record and interviewer observation data. Separate binary logistic and joint multinomial models for the two outcomes are considered, where the models account for the clustering of sample cases within interviewers. Of particular interest is to identify good explanatory variables that predict final outcome and length of a call sequence, in particular characterising long unsuccessful call sequences.

2. Explaining attrition. Whether it occurs, how and when.
Dr Peter Lugtig (Utrecht University/ University of Essex)

This paper discusses statistical models to estimate 1. whether 2. how and 3. when attrition takes place in panel surveys. Only, when these 3 models are well explained, can we start thinking on potentially targeting respondents by using adadptive survey designs. Data from the British Household Panel Study are used to illustrrate the models.

3. Call and Response: Modelling Longitudinal Contact and Cooperation using Lagged Contact Records Data
Mr Carlos Lagorio (Institute for Social and Economic Research - University of Essex)

In longitudinal survey literature, there is little discussion on how call record data (and particularly, specific call sequences) are able to account for household-level response propensities in subsequent waves. This paper uses call records as well as observed data from Understanding Society’s Wave 1 to model Wave 2 and Wave 3 household contact and cooperation propensities. Single- and multi-level logistic models are used to account for the nested structure of the data (households within interviewers) and to explore effects of the interviewer, household traits, and aggregates of individual and call record data from the preceding wave.