On making causal claims: A review and recommendations

Social scientists often estimate models from correlational data, where the independent variable has not been exogenously manipulated; they also make implicit or explicit causal claims based on these models. When can these claims be made? We answer this question by first discussing design and estimation conditions under which model estimates can be interpreted, using the randomized experiment as the gold standard. We show how endogeneity--which includes omitted variables, omitted selection, simultaneity, common methods bias, and measurement error--renders estimates causally uninterpretable. Second, we present methods that allow researchers to test causal claims in situations where randomization is not possible or when causal interpretation is confounded, including fixed-effects panel, sample selection, instrumental variable, regression discontinuity, and difference-in-differences models. Third, we take stock of the methodological rigor with which causal claims are being made in a social sciences discipline by reviewing a representative sample of 110 articles on leadership published in the previous 10 years in top-tier journals. Our key finding is that researchers fail to address at least 66 % and up to 90 % of design and estimation conditions that make causal claims invalid. We conclude by offering 10 suggestions on how to improve non-experimental research.

Interaction analysis and psychology: A dialogical perspective

Interaction analysis is not a prerogative of any discipline in social sciences. It has its own history within each disciplinary field and is related to specific research objects. From the standpoint of psychology, this article first draws upon a distinction between factorial and dialogical conceptions of interaction. It then briefly presents the basis of a dialogical approach in psychology and focuses upon four basic assumptions. Each of them is examined on a theoretical and on a methodological level with a leading question: to what extent is it possible to develop analytical tools that are fully coherent with dialogical assumptions? The conclusion stresses the difficulty of developing methodological tools that are fully consistent with dialogical assumptions and argues that there is an unavoidable tension between accounting for the complexity of an interaction and using methodological tools which necessarily "monologise" this complexity.

Conviction Statistics as an Indicator of Crime Trends in Europe from 1990 to 2006

Convictions statistics were the first criminal statistics available in Europe during the nineteenth century. Their main weaknesses as crime measures and for comparative purposes were identified by Alphonse de Candolle in the 1830s. Currently, they are seldom used by comparative criminologists, although they provide a less valid but more reliable measure of crime and formal social control than police statistics. This article uses conviction statistics, compiled from the four editions of the European Sourcebook of Crime and Criminal Justice Statistics, to study the evolution of persons convicted in European countries from 1990 to 2006. Trends in persons convicted for six offences -intentional homicide, assault, rape, robbery, theft, and drug offences- and up to 26 European countries are analysed. These trends are established for the whole of Europe as well as for a cluster of Western European countries and a cluster of Central and Eastern European countries. The analyses show similarities between both regions of Europe at the beginning and at the end of the period under study. After a general increase of the rate of persons convicted in the early 1990s in the whole of Europe, trends followed different directions in Western and in Central and Eastern Europe. However, during the 2000s, it can be observed, throughout Europe, a certain stability of the rates of persons convicted for intentional homicides, accompanied by a general decrease of the rate of persons convicted for property offences, and an increase of the rate of those convicted for drug offences. The latter goes together with an increase of the rate of persons convicted for non lethal violent offences, which only reached some stability at the end of the time series. These trends show that there is no general crime drop in Europe. After a discussion of possible theoretical explanations, a multifactor model, inspired by opportunity-based theories, is proposed to explain the trends observed.

Análisis bibliométrico sobre el Aprendizaje Basado en Problemas (ABP) en la educación superior a partir de la base de datos ERIC

[spa]El estudio bibliométrico descriptivo que sigue presenta tendencias de investigación sobre el aprendizaje basado en problemas entre 1974 y 2009. Se sirve de la base de datos ERIC y trabaja con una muestra de 1007 documentos. Se analizan los registros a tenor de cuatro variables: el año de publicación o realización, la titulación, el área de conocimiento y la tipología de investigación. En cuanto a la producción científica, pueden diferenciarse tres fases: la primera, de 1974 a 1989, supone el inicio de las publicaciones y muestra escasa relevancia estadística; la segunda, de crecimiento, durante la década de los 90 del pasado siglo; y la última, de maduración, desde el año 2000 hasta el año 2009. La distribución de los registros por sectores de conocimiento destaca Ciencias de la Salud, Ciencias Sociales y Enseñanzas Técnicas. Las titulaciones en las que más ha proliferado el ABP son Medicina, Económicas, Empresariales, Pedagogía, Formación del Profesorado y el conjunto de las ingenierías.

C.S. Peircen salapoliisimetodologian ja merkkiteorian relevanssi yhteiskuntatieteelle

summary: C.S. Peirce's model for research and its relevance to the social sciences

Densité urbaine et psychose - est-ce que vivre en ville rend schizophrène [Urban density and psychosis - does living in a city cause schizophrenia?].

While it has often been stated that prevalence of schizophrenia is the same around the world, many publications have shown this illness is twice more frequent in urban areas. Although many hypotheses have been proposed, the mechanisms explaining this phenomenon are still unknown. Besides potential biological explanations, a certain number of hypotheses emerging from social sciences have recently enriched the debate. This article reviews the literature related to this issue and describes the development of a research projects conducted in collaboration between the Institut of Geography at the University of Neuchâtel, the Department of Psychiatry at the Lausanne University and the Swiss branch of ISPS, a society promoting the psychological treatment of schizophrenia and other psychoses.

La función social de la investigación

Este artículo reflexiona y analiza la función social de la investigación. Evidencia la importancia y complementariedad de los tres ámbitos del conocimiento científico que son: la Tecnología como componente presente en casi la totalidad de investigaciones. La Ciencia, elemento clave y básico para explicar los hechos, datos o acontecimientos que se investigan, así como la Innovación, que es la que garantiza que siga avanzando el conocimiento. El artículo se detiene a analizar los aspectos clave de la función social de la investigación y sus especificidades, tanto en la investigación básica como en la aplicada. Concluye el texto con el análisis de la responsabilidad social de la investigación así como la importancia de la transferencia de los resultados de la misma a la sociedad, animando a la construcción social de una comunidad investigadora que siga trabajando para repercutir socialmente los resultados de sus investigaciones.

Social reciprocity measures based on dyadic discrepancies: Methodological review and software development

This paper examines statistical analysis of social reciprocity at group, dyadic, and individual levels. Given that testing statistical hypotheses regarding social reciprocity can be also of interest, a statistical procedure based on Monte Carlo sampling has been developed and implemented in R in order to allow social researchers to describe groups and make statistical decisions.

Being Female: a handicap for researchers in the competion for NCCR.

The Swiss National Science Foundation made a call for National Centers fo Competence in Research (NCCR) for the first time in 1999 and 2004. Together, these announcements concerned all disciplines and led to 126 preproposals, which were put forward by 2134 men and women researchers. It can be assumed that this operation mobilised Swiss researchers who regarded themselves as particularly well qualified to conduct high-level research in their field. The article uses network analysis and regression analysis methods to examine to what extend women had a lower success rate than men in the two selection rounds because of their sex. On the whole, the findings attest the gender neutrality of the National Science Foundation's selection procedures. However, they also confirm the well-known fact that women scientists are less represented in the higher echelons of academia and concentrated in the social sciences and humanities, as well as showing that this concentration reduces women's chances of success in scientific competition. The article shows that unequal gender-specific success rates prior to the NCCR funding contest play a fairly significant role.

Hacia una definición de la investigación en didáctica de las ciencias sociales

El objetivo de este articulo es apuntar algunas cuestiones centrales en la definición de lo que puede entenderse por investigación en el área de conocimiento de didáctica de las ciencias sociales. Se plantean problemas de carácter epistemológico, metodológico y teórico. Se apunta la posibilidad de un ámbito autónomo de investigación en la enseñanza de la geografía y la historia. La investigación en didáctica de las ciencias sociales debe considerarse como un campo de investigación educativa, situado en la encrucijada de diversas disciplinas sociales entre las que están, por un lado, las que se ocupan tradicionalmente del aprendizaje y la enseñanza y, por otro, las que constituyen la base de los conocimientos que se pretenden enseñar. Palabras clave. Epistemología, historia, geografía, metodología, investigación, ciencias sociales.

Aspectes metodològics i técnica d'analisi en l'estudi dels grups i les classes socials

En aquest article es presenta el procés metodològic, com també les tècniques d'anàlisi de dades, que s'ha utilitzat en l'anàisi dels grups socials mitjançant la informació d'una gran enquesta sobre hàbits i condicions de vida. L'anàlisi dels grups socials des d'una perspectiva globalitzadora i estructuradora de les condicions de vida, tenint en compte com aquests sorgeixen mitjançant I 'associació i la coherència entre múltiples característiques. Les tècniques d'anàlisi de dades que s'han utilitzat han estat l'anàlisi de correspondències múltiples i l'anàlisi de classificació.

Tratamiento multivariado de subpoblaciones en una gran encuesta social: la construcción de zonas sociales

Cuando se realiza una encuesta social en un amplio territorio queda siempre el deseo de aplicar análisis similares a los realizados en la encuesta a poblaciones o territorios más reducidos, evidentemente utilizando los propios datos de la encuesta. El objetivo de este articulo consiste en mostrar cómo cada estrato de una muestra estratificada puede constituir una base muestral para llevar a cabo dichos análisis con todas las garantías de precisión o, al menos, con garantías calculables y aceptables sin aumentar el número muestral para la encuesta general.