This primer systematizes the emerging literature on causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction to building and optimizing custom deep learning models and shows how to adapt them to estimate/predict heterogeneous treatment effects. It also discusses ongoing work to…

In this paper, we propose a method for constructing an occupation-based socioeconomic index that can easily incorporate changes in occupational structure. The resulting index is the occupational percentile rank for a given cohort, based on contemporaneous information pertaining to educational composition and the number of workers at the occupation…

Distances derived from word embeddings can measure a range of gradational relations--similarity, hierarchy, entailment, and stereotype--and can be used at the document- and author-level in ways that overcome some of the limitations of weighted dictionary methods. We provide a comprehensive introduction to using word embeddings for relation…

This paper explores research in the always already colonized spaces of academia with people in what is now the United States. In research projects, Wilson (2008) reminds us to begin with community relationships. Through ethnographic work, we trouble the idea of beginning research via local powerbrokers, who may privilege particular narratives and…

The promise of process tracing methods is that they can help us better understand how things work in real-world cases. Despite the many advances in the past two decades, we contend that existing accounts result in either under-socialised accounts in which the moves made by actors are studied while neglecting the social dimension of action, or…

The growing popularity of generative artificial intelligence (AI) tools presents new challenges for data quality in online surveys and experiments. This study examines participants' use of large language models to answer open-ended survey questions and describes empirical tendencies in human versus large language model (LLM)-generated text…

Balancing speed and quality during crises pose challenges for ensuring the value and utility of data in social science research. The COVID-19 pandemic in particular underscores the need for high-quality data and rapid dissemination. Given the importance of behavioural measures and compliance with measures to contain the pandemic, social science…

This article considers the multiple relations that emerge from and between Facebook commenters, as well as between commenters, researchers, and the research project during recruitment. To do so, we draw on our experiences of recruiting individuals who have concerns about or are opposed to a range of recent social and legal changes in…

Generative artificial intelligence (AI) has the potential to revolutionize social science research. However, researchers face the difficult challenge of choosing a specific AI model, often without social science-specific guidance. To demonstrate the importance of this choice, we present an evaluation of the effect of alignment, or human-driven…

Considerations related to generalization of a study's findings are often interconnected to researchers' judgements regarding the 'quality' of the methodology and methodological pluralism. Too often, researchers consider generalization as only possible with respect to quantitative studies with large numbers of randomly selected participants…

Theory is a core component in the teaching of many social sciences, including criminology-related programmes in higher education, but both students and lecturers can find it challenging. Given the growing popularity of criminology programmes, scholarship on the teaching and learning of criminological theory is increasingly important. This article…

Background: A systematic review is a type of literature review that uses rigorous methods to synthesize evidence from multiple studies on a specific topic. It is widely used in academia, including medical and social science research. Social science is an academic discipline that focuses on human behaviour and society. However, consensus regarding…

Including a large number of predictors in the imputation model underlying a multiple imputation (MI) procedure is one of the most challenging tasks imputers face. A variety of high-dimensional MI techniques can help, but there has been limited research on their relative performance. In this study, we investigated a wide range of extant…

Large language models (LLMs), through their exposure to massive collections of online text, learn to reproduce the perspectives and linguistic styles of diverse social and cultural groups. This capability suggests a powerful social scientific application--the simulation of empirically realistic, culturally situated human subjects. Synthesizing…

Rating scales are ubiquitous in the social sciences, yet may present practical difficulties when response formats change over time or vary across surveys. To allow researchers to pool rating data across alternative question formats, the article provides a generalization of the ordered logit model that accommodates multiple scale formats in the…