30 July 2020

Merlin Schaeffer receives runner-up prize for best article

Award

Associate Professor Merlin Schaeffer has been awarded the second prize in the ‘ESR Prize for the Best Article of the Year’, which is awarded to stimulate rigorous empirical–theoretical sociological research in Europe.

Merlin Schaeffer’s article 'Why You Should Always Include a Random Slope for the Lower-Level Variable Involved in a Cross-Level Interaction' has been awarded one of the two the runner-up prizes in the 2020 ‘ESR Prize for the Best Article of the Year’.

Merlin Schaeffer
Merlin Schaeffer

The prize is open to all papers published in the European Sociological Review, one of the leading sociological journals, during the previous year. It awards articles that stimulate rigorous empirical–theoretical sociological research in Europe.

The article, co-written with Jan Paul Heisig from the WZB Berlin Social Science Center, focuses on some of the statistical challenges and potential flaws when using so-called multilevel models involving cross-level interactions. Their article clarifies how to specify such models correctly – models that sociologists use to identify how human attitudes and behaviour are shaped by their social environments.

By analysing articles published in the European Sociological Review between 2011 and 2016, the two authors not only question the statistical significance of results from studies that do not use the ‘random slope’ in their modelling; they go a step further to raise the question of how much robust evidence of cross-level interactions sociology has actually produced over the past decades.

Overall winners of the 2020 prize were Nan Zhang, Amelie Aidenberger, Heiko Rauhut, and Fabian Winter for the article ‘Prosocial Behaviour in Interethnic Encounters: Evidence from a Field Experiment with High- and Low-Status Immigrants’.

Read more about the award and find links to the articles on the website of the European Sociological Review. The awarded article are all free to read online for a limited time.

Direct link to the article: Why You Should Always Include a Random Slope for the Lower-Level Variable Involved in a Cross-Level Interaction.