Program at a glance


Ciutadella Campus
Ramon Trias Fargas, 25-27
08005 Barcelona
Conference Registration and the Conference Opening/Keynotes are located in the underground space between the Jaume I building (Building 20) and the Roger de Llúria building (Building 40). There will be signage and volunteers to direct you.

Full program online    Floor plans    

Download the BigSurv18 App

All sponsor exhibits will be in Room 30.S02 S. Expo on Friday 26th and Saturday 27th

Big Data Enhancements to Surveys: Social Issues

Chair Dr Nicholas Biddle (Australian National University)
TimeSaturday 27th October, 11:00 - 12:30
Room: 40.004

How Does Research Productivity Relate to Gender? Analyzing Gender Differences for Multiple Publication Dimensions

Dr Sabrina J Mayer (University of Duisburg-Essen)
Mr Justus M K Rathmann (University of Zurich) - Presenting Author

Measures of research productivity have become widely used for obtaining tenure, third-party funding, and additional resources from universities. However, previous studies indicate that men might have a higher research output than women, with mixed conclusions about the factors that drive these differences. This study explores to what extent the research productivity of psychology professors in Germany is related to gender and, furthermore, how any gender gaps can be reduced by controlling for individual and organizational factors. In addition, three publication dimensions (publications in top 10 % journals, journal articles, and book and collection chapters) are distinguished to determine the effect of gender on research productivity as precisely as possible. A unique data set based on all full professors in psychology in Germany and their publication record in 2013 and 2014 is used (N (Prof) = 294; N (Articles) = 2,650).

Our approach combines several modes of data collection for obtaining a comprehensive data set. First, we conducted curriculum vitae (CV) analysis for all full professors at German universities that offer undergraduate degrees in Psychology. We collected available data from the universities' web pages, personal web pages and career networks for gender, career age, and sub-discipline. We collected a subset from the Web of Science database (that contains more than 68 million entries) for the publications of every professor in 2013 and 2014. Furthermore, we crawled the web site of the German psychology-specific publication data base PSYNDEX for records of publications in edited volumes. In addition, we add a further level of information by appending university-level data from the European Tertiary Education Register (ETER) to our data set.

Thus, this study combines Big Data and additional data sources, and provides a current overview of the state of research productivity in an entire discipline after researchers receive tenure and external restrictions are lessened. We find significant gender differences for research productivity in academic journals, even after we control for the most important individual and organizational factors, but no relationship between gender and publications of chapters in edited volumes. Yet, gender accounts for less than 1 percent of the variance of research output. Overall, we conclude that additional research that focuses on the motives and beliefs of researchers is needed, both to improve gender equality in academia and to give women better chances to gain recognition and prestige.

Social Diffusion of Xenophobic Attacks in Germany – An Application of Web Crawling

Dr Thomas Hinz (University of Konstanz) - Presenting Author
Mr Johannes Laufer (University of Konstanz)
Ms Sandra Walzenbach (LMU Munich)
Ms Franziska Weeber (University of Konstanz)

Download presentation

In 2015, Germany experienced a massive migration of refugees. Over the year, about 800.000 refugees newly arrived which equals around one percent of the total population. In the German public, this migration process was labelled as refugee crisis which was accompanied by a contested political debate. While there was a huge and widespread willingness to voluntarily help arriving refugees the number of xenophobic attacks against refugees and asylum seekers drastically increased. Our paper will focus on a specific form of xenophobic violence with a strong symbolic meaning: We will analyze how arson attacks against collective accommodation facilities spread.

Using a web chronicle, we collected temporal and spatial data about arson attacks perpetrated on accommodations or facilities for refugees in Germany between 2015 and 2017. We assigned each incident location to its county with the aid of Google’s Geocoding API. In addition, we extracted media data from digital archives. Besides classical media (newspapers’ online content), we were able to use data from a hoax data base to cover local fake news on refugees. Then we arranged an event data set as balanced panel data with the county identifier as individual dimension, and a period of 14 days as time dimension. Subsequently, we further enriched this panel with time-constant and time-variant county-level data, and with time-variant nation data.

Preliminary results indicate that spatial patterns indeed drive the diffusion process of violence as well as the potential of support for a right-wing party. Interestingly, local media coverage seems to be irrelevant for the diffusion process.

Potentials of Linking Administrative Data and Survey Data for Inequality Research

Dr Rudolf Farys (University of Bern)
Dr Oliver Hümbelin (Bern University of Applied Sciences)
Professor Ben Jann (University of Bern) - Presenting Author

Our project deals with economic inequality and poverty in the Canton of Berne, Switzerland. It is a pilot project with the medium-term goal of delivering reliable results for the distribution of income and wealth among the population in Switzerland as a whole. It is a continuation of a project funded by the Swiss National Science Foundation to investigate economic inequality based on tax data. Building on the work carried out so far, the data quality is now to be significantly improved by supplementing the tax data of the Canton of Berne with personal information from further register data and survey data. This includes data from the housing register, official vital statistics (births, marriages, divorces, deaths), social security data, and survey data. The combination of all data sources is intended to remedy the shortcomings of previous approaches, such as missing information in administrative data and sampling problems in survey data, and to open up new opportunities for inequality research. The main objective of this project is to clarify methodological questions and overcome technical and legal obstacles that arise when linking the various data sources, in order to then investigate, which new findings on the distribution of economic resources can be gained from the linked data bases. This includes the examination of results from pure survey-based analyses (e.g. estimates of poverty rates and inequality measures) as well as methodological questions such as whether heterogeneity among meso-level units within Canton can be exploited together with additional statistics to draw inference about inequality and poverty for the whole of Switzerland. In addition, more detailed questions are on the research agenda. These are, primarily, decomposition analyses of inequality indicators, which show the contribution of individual income components to the increase or reduction of income inequality, poverty, and redistribution (e.g. effects of wealth income, social benefits and direct taxes). In our talk we will report on the experiences with the data linkage (a mix of legal and technical problems and solutions) and will present first results based on the linked data. The results will include a systematic comparison of our calculations of inequality levels and poverty with previous findings from survey-based studies. In addition, we will examine whether our enriched tax data are suitable for estimating inequality and poverty figures for the whole of Switzerland by reweighting Bernese municipalities with respect to key characteristics that are available for the totality of the Swiss municipalities.

The Four Faces of Political Participation in Comparative Perspective

Mr R. Michael Alvarez (California Institute of Technology)
Mr Gabriel Katz (University of Exeter) - Presenting Author
Mrs Ines Levin (University of California, Irvine)
Mr Lucas Nuñez (California Institute of Technology)

The literature on political participation has generally focused attention on the act of voting, and has payed relatively less attention to the myriad other avenues of political expression available to those living in democratic societies. Here we study both conventional and unconventional forms of political participation, using survey data from twenty-three Latin American and Caribbean countries. We simultaneously examine people’s involvement in multiple political activities, employing a methodology, multilevel latent class analysis, that allows us to study the "four faces of political participation." Our methodology allows us to classify citizens in democratic societies along conventional and unconventional dimensions of participation, and to examine how different types of expression on each dimension cluster into four types of participants: conventionals, outsiders, activists, and agitators. Using a multi-level approach, we control for country characteristics that might influence participation decisions. This approach accounts for the possibility that survey questions about political phenomena have different meanings across the countries included in our study, and produces country-level estimates that are comparable across the countries in our sample. Following the implementation of the classification procedure, we test a variety of hypotheses about what factors are associated with different types of participation, across these various nations.