Web Survey Bibliography
Decreasing participation in surveys leads to increasing survey costs or smaller precision because of reduced sample size, and may increase non-response bias in the likely case that noncontacts and refusals differ from respondents. Online panels have been promoted as the solution to this problem as online interviewing is relatively inexpensive, large samples can easily be drawn from panels comprising hundreds of thousands willing respondents and – as internet penetration increases – even specific groups such as the elderly will have internet access and belong to a panel. Sadly, and despite the rapidly increasing market share of web research, the rising star of online panels is rapidly on the wane. Firstly, response rates of samples drawn from access panels are not always easy to compute as part of the panel may be deadwood – people who have long stopped to participate – and others may be a member under different aliases. This also casts doubt on the size of the panel as advertised. Moreover due to self-selection, selection probabilities are completely unknown, and this makes it impossible to compute unbiased estimates of population characteristics. Secondly, response rates from panel members are decreasing at an astonishing speed. Where in the recent past the high response rates of access panels were the unique selling proposition, response rates of 20% from these access panel members are no exception. In addition, even as internet access becomes almost universal in some western countries, members of access panels are likely to have a number of characteristics – heavy internet user, being online regularly, eager to express one’s opinion, experienced with filling in forms, fond of incentives – that may well be related to survey questions and that may cause considerable bias. Finally, there seems to be a high degree of panel overlap. Recent estimates in the US show that an average panel member belongs to 7 other survey panels. Similar evidence from the Netherlands indicates that 80% of the total number of panel questionnaires is completed by 20% of the total number of panel members. Some survey researchers claim that the problems mentioned above can be reduced by applying some kind of weighting adjustment procedure, e.g. using weighting variables measured in a reference survey. We argue this is a too optimistic point of view.
The presentation will outline why access panels do not solve the problems caused by non-response, will discuss access panels in the framework of total survey error and will give examples of when access panels can be useful for academic and governmental research.
European survey research associaton conference 2007 (abstract)
Web Survey Bibliography - Bethlehem, J. (33)
- Web Panels for Official Statistics? ; 2013; Bethlehem, J., Cobben, F.
- Web Panels; 2012; Bethlehem, J., Biffignandi, S.
- Use of Response Propensities; 2012; Bethlehem, J., Biffignandi, S.
- Weighting Adjustment Techniques; 2012; Bethlehem, J., Biffignandi, S.
- The Problem of Self-Selection; 2012; Bethlehem, J.,Biffignandi, S.
- The Problem of Undercoverage; 2012; Bethlehem, J., Biffignandi, S.
- Mixed-Mode Surveys; 2012; Bethlehem, J., Biffignandi, S.
- Designing a Web Survey Questionnaire; 2012; Bethlehem, J., Biffignandi, S.
- Web Surveys: Methodological Problems and Research Perspectives; 2012; Biffignandi, S., Bethlehem, J.
- About Web Surveys ; 2012; Bethlehem, J., Biffignandi, S.
- Errors in Web Surveys; 2012; Bethlehem, J., Biffignandi, S.
- Sampling for Web Surveys.; 2012; Bethlehem, J., Biffignandi, S.
- The Road to Web Surveys; 2012; Bethlehem, J., Biffignandi, S.
- Web Surveys and Other Modes of Data Collection; 2012; Bethlehem, J., Biffignandi, S.
- Handbook of Web Surveys; 2012; Bethlehem, J. Biffignandi, S.
- Handbook of Nonresponse in Household Surveys ; 2011; Bethlehem, J., Cobben, F., Schouten, B.
- Can web surveys provide an adequate alternative to phone and face to face surveys?; 2011; Bethlehem, J.
- How Representative Are Online Panels? Problems of Coverage and Selection and Possible Solutions; 2010; Bethlehem, J., Scherpenzeel, A.
- Selection Bias in Web Surveys; 2010; Bethlehem, J.
- Can we make official statistics with self-selection web surveys?; 2009; Bethlehem, J.
- The rise of survey sampling; 2009; Bethlehem, J.
- New developments in survey methodology for official statistics; 2009; Bethlehem, J.
- Indicators for the representativeness of survey response; 2009; Schouten, B., Cobben, F., Bethlehem, J.
- Use of Web surveys in Official Statistics; 2009; Bethlehem, J.
- Applied Survey Methods: A Statistical Perspective (Wiley Series in Survey Methodology); 2009; Bethlehem, J.
- Nonresponse Bias in Surveys; 2009; Bethlehem, J., Vehovar, V., Stoop, I., Schouten, B., Shlomo, N., Skinner, C., Montaquila, J.
- Representativity of web surveys – an illusion?; 2008; Bethlehem, J.
- How accurate are self-selection web surveys?; 2008; Bethlehem, J.
- Online Panels - A Paradigm Theft? ; 2007; Bethlehem, J., Stoop, I.
- Access Panels and Survey Nonresponse: Making It Better or Worse?; 2007; Stoop, I., Bethlehem, J., de Bie, S.
- Blaise – Alive and kicking for 20 years; 2006; Bethlehem, J., Hofman, L.
- Internet Survey Developments At Statistics Netherlands; 2005; Bethlehem, J.
- TADEQ: A Tool for the Documentation and Analysis of Electronic Questionnaires; 2004; Bethlehem, J., Hundepool, A.