Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries

Olteanu, Alexandra; Castillo, Carlos; Dı́az, Fernando; Kıcıman, Emre

doi:10.3389/fdata.2019.00013

Cited by 395 publications

(220 citation statements)

References 245 publications

Supporting

Mentioning

180

Contrasting

Unclassified

Order By: Relevance

“…We must often repurpose born-digital data (e.g., Twitter was not designed to measure public opinion), but data biases may lead to spurious results and limit justification for generalization (Olteanu et al, 2019). In particular, data collected via black box APIs designed for commercial, not research, purposes are likely to introduce biases into the inferences we draw, and the closed nature of these APIs means we rarely know what biases are introduced, let alone how severely they might impact our research (Morstatter et al, 2013;Tromble et al, 2017).…”

Section: Data Qualitymentioning

confidence: 99%

How We Do Things With Words: Analyzing Text as Social and Cultural Data

Nguyen

Liakata

DeDeo

et al. 2020

Front. Artif. Intell.

View full text Add to dashboard Cite

In this article we describe our experiences with computational text analysis involving rich social and cultural concepts. We hope to achieve three primary goals. First, we aim to shed light on thorny issues not always at the forefront of discussions about computational text analysis methods. Second, we hope to provide a set of key questions that can guide work in this area. Our guidance is based on our own experiences and is therefore inherently imperfect. Still, given our diversity of disciplinary backgrounds and research practices, we hope to capture a range of ideas and identify commonalities that resonate for many. This leads to our final goal: to help promote interdisciplinary collaborations. Interdisciplinary insights and partnerships are essential for realizing the full potential of any computational text analysis involving social and cultural concepts, and the more we bridge these divides, the more fruitful we believe our work will be.

show abstract

Section: Data Qualitymentioning

confidence: 99%

How We Do Things With Words: Analyzing Text as Social and Cultural Data

Nguyen

Liakata

DeDeo

et al. 2020

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Data-driven domains in computer science, including many applied artificial intelligence/machine learning domains, might profit from psychology's rich tradition of conducting carefully designed studies with human subjects, and psychologists' focus on reliable and valid data gathering methods, therefore giving opportunities for stronger empirical scientific foundations and better data quality (e.g., Lipton & Steinhardt, 2019). Furthermore, psychologists have studied biases and fairness issues in a variety of settings for decades (e.g., in the personnel decision-making process, see for instance Harvey, 1938), which is a current and highly relevant topic for computer scientists (e.g., Olteanu, Castillo, Diaz, & Kıcıman, 2019). For psychologists, potentials unfold when employing novel data gathering tools, cleaning unstructured data from various sources for further analyses, and using alternative data analysis approaches that are still uncommon within psychological practice and research (e.g., decision trees, and deep learning approaches).…”

Section: Why Should You Be Interested In Collaborating Withmentioning

confidence: 99%

Introduction to the Special Issue on Advanced Technologies in Assessment: A Science-Practice Concern

König

Demetriou

Glock

et al. 2020

PAD

View full text Add to dashboard Cite

This article is based on conversations from the project “Big Data in Psychological Assessment” (BDPA) funded by the European Union, which was initiated because of the advances in data science and artificial intelligence that offer tremendous opportunities for personnel assessment practice in handling and interpreting this kind of data. We argue that psychologists and computer scientists can benefit from interdisciplinary collaboration. This article aims to inform psychologists who are interested in working with computer scientists about the potentials of interdisciplinary collaboration, as well as the challenges such as differing terminologies, foci of interest, data quality standards, approaches to data analyses, and diverging publication practices. Finally, we provide recommendations preparing psychologists who want to engage in collaborations with computer scientists. We argue that psychologists should proactively approach computer scientists, learn computer scientific fundamentals, appreciate that research interests are likely to converge, and prepare novice psychologists for a data-oriented scientific future.

show abstract

“…On the one hand, their contribution can provide important insights in the interpretation of data. On the other hand, self-selection issues which apply to standard experimental settings (Henrich et al, 2010;Olteanu et al, 2019) are to be considered through different lenses when participation is enhanced.…”

Section: Social Impactmentioning

confidence: 99%

Reformulating computational social science with citizen social science: the case of a community-based mental health care research

Bonhoure¹,

Cigarini²,

Vicens³

et al. 2019

Preprint

View full text Add to dashboard Cite

“Citizen Social Science” is a fast-growing term within Citizen Science world but not many experiences have been shared in the literature to better shape its meaning. This article discusses novel possibilities of these participatory practices through the critical analysis of a concrete community-based project. We here embrace under the “Social” tag of Citizen Social Science both Computational Social Science methodologies and the Social concerns expressed by the mental health community. The interpretation place persons with a mental condition, as well as caregivers and relatives, at the center of the research cycle by taking the role of Competent Experts In-The-Field. A synergetic relation between Citizen Science, Mental Health research and Computational Social Science also imply a conceptual shift in comparison to standard approaches such as the inclusion of research subjects as active co-designers or the consideration of communitarian spaces as most natural experimental spaces. We here describe how these concepts are put into practice during the whole duration of a research project that has studied the social interactions inside the Community Mental Health Care ecosystem. Important steps entail the creation of a Knowledge Coalition, including a diversity of relevant actors with diverse knowledge and expertise, that is later involved in co-designing the research and in embedding experimental settings in communitarian spaces. The experience allows us to open a wider discussion on the possibilities and limitations of Citizen Social Science practices. Having in mind the ethical debate raised by Citizen Science Public Health and Patients’ driven research, we propose a set of values and practices to be agreed on. We also analyse the participation of the scientists in these kind of projects, which forces them to adopt the role of “camaleons” when executing diverse and versatile tasks. We additionally advocate for a more extended collective interpretation of the results in order to produce socially robust knowledge and enhance actions and policies grounded on these results. Exemplified with the experience herein presented, a more horizontal process that include the enhancement of participation and the revision of the notion of experiment offer new opportunities for Social Sciences from a multidisciplinary perspective.

show abstract

Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries

Cited by 395 publications

References 245 publications

How We Do Things With Words: Analyzing Text as Social and Cultural Data

How We Do Things With Words: Analyzing Text as Social and Cultural Data

Introduction to the Special Issue on Advanced Technologies in Assessment: A Science-Practice Concern

Reformulating computational social science with citizen social science: the case of a community-based mental health care research

Contact Info

Product

Resources

About