Satisficing Among Reluctant Respondents in a Cross-National Context

Abstract

Declining response rates have led to extended efforts to convert reluctant respondents. If respondents with low motivation are likely to use cognitive shortcuts in survey responding (to engage in satisficing), bringing reluctant respondents into the respondent pool may result in an increase of measurement error. In a cross-national setting, the differential efforts of converting reluctant respondents across countries may result in differential data quality, potentially leading to non-equivalence. This study examines the relationship between reluctance and respondents' likelihood to engage in satisficing, controlling for cognitive ability in Belgium, Germany, the Netherlands, Norway, Spain and Sweden. The study is based on the European Social Survey (ESS), third round, conducted in 2006 and 2007. Latent class analysis is used to develop two main constructs of interest, including reluctance and satisficing. The results of the latent class analysis indicate a presence of a single satisficing factor for all the countries with the exception of Germany and Spain. Thus, Structural Equation Modeling is used for Belgium, the Netherlands, Norway and Sweden, and regression models are used for Germany and Spain. Through an overview of the separately developed literature on reluctance and satisficing a number of common factors are identified. As hypothesized, a positive relationship between reluctance and satisficing is found in Belgium, the Netherlands, Norway and Sweden, and in a number of instances in Germany and Spain. Nevertheless, the results suggest that in Belgium, the Netherlands, Norway and Sweden the relationship between reluctance and satisficing is spurious, and is completely explained by cognitive ability. Similarly, in Germany and Spain, the effect of reluctance on satisficing decreases or becomes statistically nonsignificant after controlling for cognitive ability. The cross-country comparison suggests that the direction and the magnitude of the relationship among satisficing, reluctance, and cognitive ability are generally invariant across the countries. However, this does not guarantee the equivalence of estimates. Potential impacts of differences in cognitive ability, reluctance and in the overall likelihood to satisfice on equivalence are discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 154 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.