De verklaarde variantie verklaard: Een vergelijking van sociologische onderzoeksartikelen in de periode 1975-1998

The R² is often seen as one indicator of how well researchers are able to explain the phenomena they study. The R² is usually not high but there are large differences among studies. In this study, we try to explain these differences by examining the influence of four factors on the value of the R²: the field researchers are working in, the type of models and data they use, and the characteristics of the researchers themselves. To examine these influences empirically, we drew a sample of 285 regression models in 119 research articles, published in major journals in the Netherlands and the United States between 1975 and 1998. Multilevel analyses indicate that there are large differences among fields of study: stratification research and labor market studies have the highest R², studies on norms and values have the lo west R². The type of samples researchers work with appear to have very little impact on the R², but the type of models they estimate do matter, although not as much as is often believed. Finally, we do not find that the R² has increased over the past decades, nor do we find that American researchers are doing better than Dutch researchers. Most surprisingly, we find that fill professors tend to have a lower R² on average than other researchers. Our own R² is 50 percent, indicating that we are able to explain about half of the differences in the R² we observe among studies. [ABSTRACT FROM AUTHOR]; Copyright of Mens en Maatschappij is the property of Amsterdam University Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)