Abstract
This issue of the BMS includes two research articles, one each in English and in French, and three research notes, of which two are in English and one in French.
The first research article, in English, “Controlling Variables in Social Systems - A Structural Modelling Approach”, by Michel Mouchart, Guillaume Wunsch and Federica Russo, uses a structural modelling approach to present a consistent framework for determining a coherent set of guidelines when deciding what variables should be controlled simpler and more comprehensive approach than alternatives based on Pearl’s back-door criterion.
The second research article, in French, “Finite Mixture Models (FMM) Applied to the Segmentation of the Bogota Labor Market”, by Thibaud Deguilhem and Suneha Seetahul, shows that the semi-parametric FMM approach is a real alternative to clustering procedures to explain social phenomena.
The first ongoing research note, in English, “Methodological Aspects of Central Left-Right Scale Placement in a Cross-national Perspective”, by Evi Scholz and Cornelia Zuell, concerns design effects on central placement on the left-right scale across countries and shows that the best indicator of central placement was non-response to the probe questions.
The second ongoing research note, in French, “How Many Demonstrators? The Cafés of Statistics - Evening of 10 November 2015”, by Groupe “Statistique et enjeux publics”, looks at different methods of counting persons during public demonstrations and that resolving differences requires establishment of trust between stakeholders.
The third ongoing research note, in English, “Social Desirability Bias and Mode Effects in the Case of Voting Behavior”, by Pei-shan Liao, compares face-to-face interviews (PAPI) and computer-assisted telephone interviews (CATI) on voter turnout of the 2004 presidential election in Taiwan, showing that from the PAPI sample respondents are partial toward the elected officials, but the CATI results are not.