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

Tuesday 14th July, 11:00 - 12:30 Room: HT-102

Comparing the individual name, address and household samples in comparative surveys
Convenor	Dr Piotr Jabkowski (Institute of Sociology, University of Poznan, Poland )
Coordinator 1	Professor Ryszard Cichocki (Quality of LIfe Research Centre, University of Poznan, Poland)

Session Details

Existing literature on survey methodology is not rich in studies of the relationship between the type of sample-frames and the patterns of their fieldwork execution. Most studies focus on the impact of the within-household selection of respondent (characteristic of the address and household samples) on the imperfect coverage of the individuals comprising the population; much less attention has been devoted to the problem of what effects the different sample-frames have on the response-rates and the post-survey structure of the set of respondents and non-respondents.

On the one hand, given the necessity of multi-stage selection of respondents in address and household samples, one would expect lower response and cooperation rates than in the individual-name samples. On the other hand, due to the fact that in the address and household samples there are limits of the control that researchers actually have over the respondent selection, the interviewers may be inclined to select respondent with a higher readiness to cooperate. In other words, if there were to be systematic irregularities committed in the selection of individuals within address and household samples, the contact rates should be higher and refusal rates lower than in individual samples.

The present session would focus on theoretical issues and practical challenges connected with:
- the process of intra-unit selection of respondents;
- within-unit coverage errors;
- statistical consequences of random selection of individuals within address / household samples;
- data sources for individual and address-based / household samples;
- fieldwork based on address and individual-based / household samples;
- response rates in address and individual-based / household samples;
- cooperation and refusal rates in address / household and individual-based samples.

Paper Details

1. Accuracy and Coverage Assessment of House units Frame using Post-Frame update survey (PFS) Methods
Dr Elmogiera Elawad (Social and Economic Survey Research Institute - Qatar University)
Mr Mohamed Agied (Social and Economic Survey Research Institute - Qatar University)

The Post-Frame update survey (PFS) is a method for evaluating the results of the frame house units update, as the results of the frame are used more surveys, it is important to examine the quality and accuracy of the frame data, the PFS is an independent survey that repeats frame update. The survey results are compared with house units frame update results, permitting estimates to be made of coverage and content errors, basically this new method similar to PCS post census survey, with some differences in practices.

2. RDD versus List Samples. CATI Interviewers’ Experience
Dr Wojciech Jablonski (University of Lodz)

This presentation investigates the issue of differences in interviewers’ opinion and attitudes connected with two different sampling procedures used by survey organizations – Random Digit Dialing (RDD) in Poland and list samples in Norway and Iceland. As in Poland no reliable directory of telephone numbers exists, it is necessary to implement RDD approach. On the contrary, in Nordic countries it is usually possible to merge random files of individuals (drawn from the population registers) with the telephone number database. The interviewers encounter different difficulties, depending on the sampling schemes used in surveys.

3. Nonresponse and Bias in the German Sample of European Social Survey
Mr Michael Weinhardt (Bielefeld University)
Professor Stefan Liebig (Bielefeld University)

This paper looks at systematic patterns of nonresponse and the possibility of biased estimates in the German sample of the European Social Survey 2012. The sample is drawn from official registers of local municipalities which are usually assumed to be the best sample frame available in Germany. Overall, the realised sample is a good depiction of the population, at least for the characteristics for which data is available from official statistics. One exception is education: respondents in the sample show a much higher maximum degree than people in the German population based on census data.