Open-ended questions in web panels and web surveys |
|
| Convenor | Professor Matthias Schonlau (University of Waterloo ) |
Near the end of web surveys, respondents are often asked for additional comments. Such final comments are usually ignored, partially because open-ended questions are more challenging to analyze. Final comments have been under-explored, and in particular it is unclear whether final comments are predictive of attrition in survey panels. We categorize a random sample of the final comments in the LISS panel and Dutch Immigrant panel into one of nine categories (neutral, positive, and multiple subcategories of negative) and regress indicator of next-wave attrition on comment category, if any, and socio-demographic variables. A complex answer emerges.
Open-ended questions in web surveys suffer from substantial item nonresponse. For example in a recent survey on a sensitive topic nonresponse reached up to 56%. Our method automatically classifies different types of item nonresponse, allowing us to ask suitable follow-up questions during a web survey and thus reducing final item nonresponse rates substantially. The classification is available for English, German and Spanish and can also be applied to already collected data. The percentage of successfully converting nonresponse into a useful answer varies across questions but is mostly between 20%-40% with exceptions of up to 74%.
This presentation will focus on using a few specific open-ended questions to increase respondent engagement and satisfaction in web surveys. For example the researcher get a possibility to show interest in what the respondent has to say which may have positive effect on the relation between respondent and researcher. Open-ended questions shift some of the power from the researcher to the respondent and may increase satisfaction in terms of giving correct information. I will discuss the benefits of open-ended questions in two different surveys.
An experiment was conducted by random assignment of text box size to open end questions in a web survey of faculty members. We hypothesized that a larger box would result in lengthier responses. While direct effects were not observed, differences were observed with moderating effects of engagement in the activity under investigation and in relation to whether positive or negative reactions were solicited. The implications of the results for web surveys are discussed.
The early dissolution of apprenticeship contracts is an important issue in German labour politics. In search for reasons behind this phenomenon we used a standardized questionnaire. The use of open questions in this survey served two purposes. It should provide the opportunity to respondents to give statements in a more qualified way than just setting crosses in boxes, and we hope to find in these answers information that can enrich the quantitative information. Two questions are investigated: First, what are factors influence the probability of using open answers? Second, do open answers add substantial information?