A2: Who participates in online panels?
Determinants of access-panel participation: Recent experiences from the recruitment of members for a mixed-mode access panel using random telephone samples
University of Bremen, Germany
The data quality of access panels can be threatened by self-selection processes into the panel and mode/response effects. Self-selection is likely to lead to biased sample estimates, while mode effects and mode-specific response effects preclude any generalisation of outcomes produced by one survey mode to another. To study both types of effects, we built up a large access panel for the adult population of Germany using probability sampling for the recruitment of people by phone (landline and cell phones). Possible access panel modes are landline, cell phone and Internet. The project is part of the Priority Programme 1292 on ‘Survey Methodology’ (see www.survey-methodology.de)
The present paper is focused on an analysis of panel participation. In a first step, the propensity to response to the initial survey request (participation in the recruitment interview) is modelled using an extensive set of paradata and mixed-effects logistic regression analysis, controlling for interviewer effects. Then latent variable modelling is used to estimate determinants of follow-up cooperation using an extensive set of paradata, metadata, and survey items as well as the response propensity modelled at step 1. There are two major dependent variables: a) the readiness to participate in the access panel as expressed by the respondent in the concluding part of the recruitment interview. b) Subsequent panel participation when the target persons were re-contacted some weeks later for realizing the actual panel-initializing interviews. As in step 1, here again multilevel analyses were carried out to control for interviewer effects in attempts to realize these panel-initializing interviews. The model of panel participation differentiates between participation in the online and telephone track of the panel.
Asking sensitive questions in a recruitment interview for an online panel: the income question
Relevance & Research Question: It is a well-known fact that sensitive questions such as the income question show an increased item nonresponse compared to other survey questions. Less is known about how such questions affect the decision process of respondents to join a panel for follow-up surveys. Hence, the study examines in a split experiment whether there are differences concerning the recruitment success for a panel when the income question is asked or omitted during the recruitment interview. Furthermore, it investigates if those respondents, who reject to answer the sensitive question, tend to reject the panel request more often. The main question is: Are respondents affected by the sensitive question itself or is there a general lower trustworthiness among some respondents which influences both unit nonresponse in sensitive questions and the rejection of panel participation?
Methods & Data: The interview for panel recruitment is conducted via telephone with a probability sample. In a randomized split experiment half of the respondents are asked about their net household income in the recruitment interview and the other half are not asked any income information. At the end of the interview all respondents are asked whether they would be willing to join an online panel and answer questions about life in Germany via online questionnaires on a monthly basis. For this, respondents have to provide an e-mail address in the end of the telephone interview. The dependent variable is the willingness to participate in the online panel.
Results: Our previous research shows that 47% of those who answer the income question are willing to participate in an online-follow-up survey. In contrast, this figure plunges to only 22% for those who decline to provide income information (total n=818). Data collection for the current study starts in December 2010.
Added Value: The anticipated results will give insights in the (offline) recruitment process for an online panel which is based on a probability sample. The knowledge about how sensitive questions can affect later decisions of participation is crucial for designing a recruitment process, especially when sensitive questions are an important predictor in estimations of nonresponse bias.
The influence of personality traits and motives for joining on participation behavior in online panels
WU Wien, Austria
(a) Relevance & Research Question:
Due to the dropping penetration rate of landline telephones and a general decline in the willingness to participate in telephone surveys, data collection using online methods – especially online access panels – are becoming more popular all over the world. Though widely adopted among marketing researchers critics still fear that the new sampling methodology leads to biased results produced by a breed of survey-savvy volunteer respondents who are solely interested in monetary incentives and therefore cannot be compared to the general population. This study seeks to give in-depth insight into the personality of online panelists by analyzing their motives for joining the pool as well as their personality traits (Big Five, materialism).
(b) Methods & Data:
In a survey among 1,729 members of an Austrian online access panel participants were asked to answer standardized scales measuring materialism (Richins 1987) and the Big Five personality traits (Rammstedt & John 2005). In addition they had to name the reasons for joining the online panel. The data was then analyzed for its influence on participation behavior in the pool during the last 13 month.
The results show that money is a relevant motive but not the dominating reason for becoming an online panel member. It is also found that psychographic characteristics have rather little influence on participation behavior within the panel.
(d) Added Value:
The use of personality traits to predict participation behavior in surveys in general and online panels in particular is virtually unknown in the literature. This study is the first that examines the influence of those factors in an online panel environment.