Surveys using an online access panel: methodological findings from an employee survey on the prevention of common mental disorders
Standardised surveys are increasingly being carried out using the internet. In addition to "open" online surveys, use can also be made of online access panels (OAPs) from marketing and opinion research institutes. These pools of voluntary respondents, who receive a small fee for taking part in surveys, offer good access to samples with guaranteed sample size and high data quality, but must also be given critical consideration with regard to possible bias. The purpose of this article is to report on the lessons learnt from the survey of an employee sample.
The inter-company survey of an OAP sample with a priori defined job types took place in 2016 and related to experience of and attitudes to mental and psychosomatic illnesses, their causes and prevention. Study dropouts were analysed for personal characteristics.
The desired target group size (600 complete data sets) was achieved within two weeks with a rate of oversampling that was almost double the planned rate. No cases of job type-related dropout bias were found.
Inter-company access to employees by means of an OAP has methodological limitations (e.g. lack of representativeness, volunteer bias, absent characteristics of company populations), but it also has advantages compared to an open internet survey (e.g. targeted sector-specific recruitment, dropout analysis possibilities) or company-specific surveys (data protection concerns, absence of permission from company management). Less effort and expenditure for scientists and a higher level of planning reliability are important advantages compared to company-based approaches.
Online surveys using access panels have a good cost-benefit ratio for scientific surveys, not least due to guaranteed response rates. A critical discussion of methodological limitations in the interpretation of results is essential.