A5078523938- Survey Methodology and Nonresponse
- Focus Groups and Qualitative Methods
- Electoral Systems and Political Participation
- Social Media and Politics
- Health disparities and outcomes
- Economic and Environmental Valuation
- Social and Intergroup Psychology
- Probability and Statistical Research
- Privacy, Security, and Data Protection
- Behavioral Health and Interventions
- Survey Sampling and Estimation Techniques
- Data-Driven Disease Surveillance
- Health and Medical Studies
- demographic modeling and climate adaptation
- Social Policy and Reform Studies
- Advanced Causal Inference Techniques
- Mobile Crowdsensing and Crowdsourcing
- Expert finding and Q&A systems
- Data Quality and Management
- Qualitative Research Methods and Ethics
- Internet Traffic Analysis and Secure E-voting
- Intergenerational Family Dynamics and Caregiving
- Social Capital and Networks
- Health Literacy and Information Accessibility
- Technology Adoption and User Behaviour
GESIS - Leibniz-Institute for the Social Sciences
2015-2025
University of Mannheim
2021-2024
Ludwig-Maximilians-Universität München
2021
University of Bremen
2021
University of Leeds
2021
Leibniz Institute for Neurobiology
2015
Leibniz Association
2013
This study explores how researchers’ analytical choices affect the reliability of scientific findings. Most discussions problems in science focus on systematic biases. We broaden lens to emphasize idiosyncrasy conscious and unconscious decisions that researchers make during data analysis. coordinated 161 73 research teams observed their as they used same independently test prominent social hypothesis: greater immigration reduces support for policies among public. In this typical case...
Identifying inattentive respondents in self-administered surveys is a challenging goal for survey researchers. Instructed response items (IRIs) provide measure inattentiveness grid questions that easy to implement. The present article adds the sparse research on use and implementation of attention checks by addressing three objectives. In first study, we evidence IRIs identify who show an elevated straightlining, speeding, item nonresponse, inconsistent answers, implausible statements...
This study investigates researcher variability in computational reproduction, an activity for which it is least expected. Eighty-five independent teams attempted numerical replication of results from original policy preferences and immigration. Reproduction were randomly grouped into a ‘transparent group’ receiving code or ‘opaque only method description no code. The transparent group mostly verified (95.7% same sign p -value cutoff), while the opaque had less success (89.3%). Second-decimal...
Abstract This article discusses the current challenges of conducting a General Social Survey (GSS) in face-to-face mode and evaluates alternative fielding these surveys self-administered mixed-mode (web, mail) instead. Based on data from Germany, it first illustrates stark decline participation since 2002 reports strong increase cost over same period. It then possibility implementing GSS-type design results experiment implemented German part European Values Study (EVS) 2017/18. The indicate...
Abstract Combining surveys and digital trace data can enhance the analytic potential of both types. We present two studies that examine factors influencing sharing behaviour survey respondents for different types data: Facebook, Twitter, Spotify health app data. Across those types, we compared relative impact four on sharing: method, respondent characteristics, sample composition incentives. The results show rates differ substantially across Two particularly important predicting are...
Interview duration is an important variable in web surveys because it a direct measure of the response burden. In this article, we analyze effects survey design, respondent characteristics, and interaction between these on interview duration. For that purpose, applied multilevel analysis to data set 21 political attitudes behavior. Our results showed factors both levels, individual level, had However, larger share variation explained by characteristics respondents. respect, illustrate impact...
Our study explores the adoption of Facebook and Twitter by candidates in 2013 German Federal elections. Utilizing data from Longitudinal Election Study candidate survey fused with gathered on use candidates, we draw a clear distinction between Twitter. We show that both channels is primarily driven two factors: party money. But impact each plays out differently for While influence money homogenous more resources have, likely they are to adopt, effect stronger Facebook. Conversely, party's...
Satisficing often has been assumed to be a hazard response quality in web surveys because interview supervision is limited the absence of human interviewer. Therefore, devising methods that help mitigate satisficing poses an important challenge survey methodology. The present article examines whether splitting up cognitively demanding grid questions into single items can effective means reduce measurement error and nonresponse resulting from satisficing. Furthermore, we investigate modifying...
Abstract Recent advances in web survey methodology were motivated by the observation that respondents increasingly use mobile devices, such as smartphones and tablets, to participate surveys. Even though we do not doubt this general observation, argue claim is lacking a solid empirical basis. Most research on increasing device surveys covers limited periods of time and/or analyzes data from only one study or panel. There surprising lack comprehensive overviews magnitude In present study,...
Abstract The European Values Study (EVS) was first conducted in 1981 and then repeated 1990, 1999, 2008, 2017, with the aim of providing researchers data to investigate whether individual social values are changing what degree. EVS is traditionally carried out as a probability-based face-to-face survey that takes around 1 hour complete. In recent years, large-scale population surveys such have been challenged by decreasing response rates increasing costs. light these challenges, six...
This article introduces the evolution, framework, objectives, and design of new data infrastructure “FReDA – The German Family-Demographic Panel Study”, which has been funded by Federal Ministry Education Research (BMBF) since 2020. FReDA is rooted in Generation Gender Survey (GGS) Family (pairfam). was initiated to facilitate research on family demography providing a comprehensive panel study allowing for international comparisons as well dyadic analyses through multi-actor design. survey...
This article addresses the questions of whether paradata can help us to improve models panel attrition and effectiveness propensity score weights with respect reducing biases. The main advantage is that it collected as a by-product survey process. However, still an open question which be used model what extent these are correlated variables interest. Our analysis data from seven-wave web-based had been supplemented by three cross-sectional surveys. split design allowed assess magnitude bias...
This study explores how researchers’ analytical choices affect the reliability of scientific findings. Most discussions problems in science focus on systematic biases. We broaden lens to include conscious and unconscious decisions that researchers make during data analysis may lead diverging results. coordinated 161 73 research teams observed their as they used same independently test prominent social hypothesis: greater immigration reduces support for policies among public. In this typical...
Attention checks detect inattentiveness by instructing respondents to perform a specific task. However, while may correctly process the task, they choose not comply with instructions. We investigated issue of noncompliance in attention two web surveys. In Study 1, we measured respondents’ attitudes toward and their self-reported compliance. 2, experimentally varied reasons given for conducting check. Our results showed that most understand why are conducted, nonnegligible proportion...
Abstract During the COVID-19 pandemic, many survey programs switched to self-administered modes of data collection, often offering respondents both web-based and paper-based questionnaires. However, there is little guidance as when offer which mode, especially aim recruit participants for a panel survey. This study examines effectiveness different mode-sequence designs by analyzing an experiment implemented in recruitment wave German “Family Research Demographic Analysis.” We randomly...
Abstract This data brief introduces the German Family Demography Panel Study (FReDA; https://www.freda-panel.de/), a longitudinal, multi-actor database for family research. Major substantive fields addressed in questionnaire include fertility-related attitudes and behaviours, reproductive health, work-family conflict, couples’ division of labour, gender roles, intimate relationships, separation divorce, parenting intergenerational relations, well-being. FReDA is based on two initially...
Mobile coverage recently has reached an all-time high, and in most countries, high-speed Internet connections are widely available. Due to technological development, smartphones tablets have become increasingly popular. Accordingly, we observed increasing use of mobile devices complete web surveys and, hence, survey methodologists shifted their attention the challenges that stem from this development. The present study investigated whether growing decreased how systematically choice device...
The worldwide spread of the COVID-19 pandemic has disrupted fieldwork surveys. data collection efforts via face-to-face mode have been affected especially, including ongoing surveys that were in field during outbreak and planned scheduled for later 2020. We provide an account how impacted two family studies Germany: “The German Family Panel” (pairfam) “Generations Gender Survey” (GGS) both which will be part “Family Research Demographic Analysis” (FReDA) infrastructure. Based on pairfam, we...
Open-ended questions have a great potential for analyses, but answering them often imposes burden on respondents. Relying satisficing theory as an overarching theoretical framework, we derived several hypotheses about how respondent and survey level characteristics, their interactions, might affect the quality of responses to open-ended attitude question in self-administered surveys. By applying multilevel analyses data from 29 web surveys, examined effects characteristics three indicators...
Survey research is still confronted by a trend of increasing nonresponse rates. In this context, several methodological advances have been made to stimulate participation and avoid bias. Yet, despite the growing number tools methods deal with nonresponse, little known about whether biases show similar trends as rates what mechanisms (if any) drive changes in Our article focuses on cohort distributions U.S. German general social surveys from 1980 2012 one key variables sciences. To supplement...
Abstract In this article, we examine two types of item nonresponse in a face-to-face population survey: ‘don’t know’ (DK) and ‘item refusal’ (REF). Based on the cognitive model survey response, theory satisficing previous research, derive explanatory variables three levels: question, respondent interviewer characteristics. The results our cross-classified show that while levels question respondents’ characteristics affected both nonresponse, only DK answers. Our also confirm REF are...
Knowledge questions frequently are used in survey research to measure respondents’ topic-related cognitive ability and memory. However, self-administered surveys, respondents can search external sources for additional information answer a knowledge question correctly. In this case, the measures accessible procedural Depending on what aims at, validity of is limited. Thus, study, we conducted three experiments using web investigate effects task difficulty, ability, motivation likelihood...
In this article, we present the results of three question design experiments on inter-item correlations, which tested a grid against single-item design. The first and second examined correlations set with five seven items, respectively, third experiment impact follow-up questions. We found that in designs were higher for both items than questions Additionally, if preceding was presented grid. According to findings research, respondents seem apply interpretative near means related heuristic...