Inferential statistics teach us that we need a random probability sample to infer from a sample to the general population. In online survey research, however, volunteer access panels, in which respondents self-select themselves into the sample, dominate the landscape. Such panels are attractive, due to their low costs. Nevertheless, recent years have seen increasing numbers of debates about the quality, in particular about errors in the representativeness and measurement, of such panels (Baker et al., 2010). In this paper, we describe four probability-based online and mixed-mode panels for the general population: the LISS Panel in the Netherlands, the German Internet Panel and the GESIS Panel in Germany, and the ELIPSS Panel in France. We compare them in terms of sampling strategies, offline recruitment procedures, and panel characteristics. Our aim is to provide an overview to the scientific community of the availability of such data sources, to demonstrate to practitioners potential strategies for recruiting and maintaining probability-based online panels, and to direct analysts of the comparative data collected across these panels to methodological differences that may affect comparative estimates.

The Internet is considered an attractive option for survey data collection. However, some people do not have access to it. One way to address this coverage problem for general population surveys is to draw a probabilistic sample and provide Internet access to the selected units who do not have it and accept to participate. This is what the knowledge panel and the Longitudinal Internet Studies for the Social sciences (LISS) panel do. However, a selection effect is still possible. Units without previous Internet access might refuse to participate in a web panel, even if provided with the necessary equipment. Thus, efforts to provide the necessary equipment may not be worth it. This article investigates the gain in terms of representativeness of offering the equipment to non-Internet units in a web panel using tablets: the French Longitudinal Internet Studies for the Social Sciences panel. We find that the number of non-Internet units who accept to participate is low. This is not only due to the fact that their response rates are lower but also to the small proportion of non-Internet units in the French population. In addition, they participate less in given surveys once they become panelists. At the same time, they are very different from the Internet units. Therefore, even if because of the small number of units, the overall gain in representativeness is small, there are a few important variables (e.g., education) on which their inclusion yields a more representative sample of the general population.

In the context of the current “replication crisis” across the sciences, failures to reproduce a finding are often viewed as discrediting it. This paper shows how such a conclusion can be incorrect. In 1981, Schuman and Presser showed that including the word “freedom” in a survey question significantly increased approval of allowing a speech against religion in the USA. New experiments in probability sample surveys (n = 23,370) in the USA and 10 other countries showed that the wording effect replicated in the USA and appeared in four other countries (Canada, Germany, Taiwan, and the Netherlands) but not in the remaining countries. The effect appeared only in countries in which the value of freedom is especially salient and endorsed. Thus, public support for a proposition was enhanced by portraying it as embodying a salient principle of a nation’s culture. Instead of questioning initial findings, inconsistent results across countries signal limits on generalizability and identify an important moderator.