Social science relies on two distinct approaches: quantitative research, which analyses large sets of numerical data, and qualitative research, which explores individual experiences and perspectives in depth.
Interviews, the traditional method of qualitative research, require significant human and financial resources and are limited in terms of the number of participants that can be reached. However, large language models (LLMs) offer an opportunity to conduct interviews at scale, thereby creating a bridge between the richness of qualitative data and the statistical power of quantitative data. Specifically, the LLM can act as an interviewer and interact with human respondents recruited on a standard online platform, such as Prolific. This makes it possible to conduct interviews with thousands of respondents in a matter of hours.
We have developed and launched an easy-to-use platform for conducting large-scale qualitative interviews, based on artificial intelligence in just this way. A chat interface allows the respondent to interact with a LLM that collects their responses and generates new questions. The platform is adaptable to researchers’ needs and capable of handling a wide range of topics while maintaining a general structure that ensures the reliability of the interviews.
The conversational agent’s prompt incorporates best practices from sociological literature, such as the importance of conducting interviews in a non-directive manner, encouraging participants to freely express their views, while posing follow-up questions to ensure clarity on each point raised by the respondent. For instance, it gathers “palpable evidence” to deeply understand topics that respondents may sometimes express in abstract terms. Moreover, the conversational agent demonstrates “cognitive empathy,” using follow-up questions to try to understand the respondent’s perspective as closely as they understand it themselves, insofar as doing so is useful to deepen its understanding of the main theme of the interview.
We analyzed its performance against traditional qualitative interview practices, both from the perspective of interviewees and qualitative research experts. First, we asked a team of sociology PhD students from Harvard and the London School of Economics, who specialize in qualitative methods, to assess the quality of interviews based on the interview scripts. The AI-led interviews were rated approximately comparable to an average human expert (under the same conditions). Additionally, during the interviews, respondents were asked to evaluate the quality of the interaction. A vast majority of participants reported enjoying their interaction with the conversational agent and preferred this mode of interview over open text fields. They felt that it captured their thoughts well, tended to write significantly more words compared to open text fields, and found this interview method to be non-judgmental.
We also explored the capacity of artificial intelligence to analyze different kinds of qualitative questions by using it to study three different subjects areas: subjective issues, “meaning in life,” sensitive issues like political opinions, and complex areas such as decision-making for occupational and educational choices.
In the first instance, we aimed to identify key activities that serve as major sources of meaning in life (defined here as a subjective sense that one’s life makes sense, has purpose, and matters to others) and the states individuals associate with meaning in life, with particular interest in differences across socio-demographic groups. AI-led interviews revealed more information than conventional open text fields, with a 142 percent increase in the number of words respondents wrote. They helped uncover significant activities and subjective states that were not easily anticipated, such as the importance of pet care and companionship, which seemed to emerge as a source of meaning in life approximately as frequently as spirituality and religion. The large sample of respondents enabled the documentation of variations between different socio-economic and political groups. For example, Trump voters were more likely to mention religion as a source of meaning compared to Biden voters.
The second application examined political opinions and the reasons behind electoral choices. Here, participants preferred interacting with a conversational agent perceived as a non-judgmental entity, rather than a human interviewer, allowing them to feel more comfortable and to freely express their views, especially on sensitive political issues. Focusing on the French legislative elections, interviews conducted a week before the elections, they show the adaptability of the conversational agent to languages other than English. The study also revealed interesting insights about political polarization. Depending on their political affiliation, voters had different views on political priorities and issues. They are attracted to specific policies as solutions to issues they consider important, rather than being guided by the general ideology of their preferred party.
In the third application, we explored key factors in decision-making using the example of educational and occupational choices, particularly in STEM (science, technology, engineering and mathematics) fields and innovation careers, with a focus on understanding why certain groups are underrepresented. Consistent with previous research on this topic, financial and sociological factors (such as family, mentors, and childhood interests) were crucial, with a major role played by individual interest and passion.
Finally, ethical concerns must be mentioned. To participate in the interviews, respondents must agree to their data being seen by the firms that own the language models, which may raise privacy concerns for some participants. As smaller, local language models become increasingly capable, it is likely that high-quality interviews can be conducted with them, mitigating potential privacy issues.
This approach due to its flexibility and open-source platform, can offer researchers a new tool to conduct large-scale, rigorous interviews on a variety of topics, while allowing researchers to maintain full control over the prompt and interface. The general instructions provided to the LLM are perceived differently in terms of quality, depending on individual researchers’ preferences. The platform’s flexibility allows researchers from different schools of thought to tailor the interviews to their preferences, for example, by focusing on narratives rather than concrete examples. We propose several avenues for improvement and consider the implications of making the conversation, currently conducted via text chat, more fluid by adding a voice option.
Friedrich Geiecke (pictured) is assistant professor of computational social science at the Department of Methodology at the London School of Economics. Xavier Jaravel is professor of economics at the Department of Economics at LSE.
A 2024 report by the National Academies explores the latest advances in artificial intelligence (AI) technology and their potential effects on economic productivity, job stability, and income inequality. It also highlights key research opportunities and data needs to help workers and policymakers adapt to the evolving AI landscape.
During the final stages of editing the proofs for Artificial Intelligence and Work: Transforming Work, Organizations, and Society in an Age of Insecurity, […]
Thanks to a collaboration between the Inter-American Foundation (IAF) and the Social Science Research Council (SSRC), applications are now being accepted for […]
Solicitations are now being sought for the National Science Foundation’s Scholarships in Science, Technology, Engineering, and Mathematics program, and in an unheralded […]
What are the three biggest challenges Australia faces in the next five to ten years? What role will the social sciences play in resolving these challenges? The Academy of the Social Sciences in Australia asked these questions in a discussion paper earlier this year. The backdrop to this review is cuts to social science disciplines around the country, with teaching taking priority over research.
Canada’s Social Sciences and Humanities Research Council, the Canadian Institutes of Health Research, and the British Columbia-based Michael Smith Health Research BC […]
Sage (the parent of Social Science Space) and Surviving Society’s collaborative podcast series, Social Science for Social Justice, has returned for a […]
Sage (the parent of Social Science Space) and the Surviving Society podcast have launched a collaborative podcast series, Social Science for Social […]
In this month’s issue of The Evidence newsletter, Josephine Lethbridge examines new research into prehistoric gender dynamics – and what this tells […]
Decisions taken now around how generative AI is used by academics and universities will shape the future of research. Mark Carrigan argues whilst optimistic scenarios are possible, generative AI stands ready to feed into an existing productivity oriented framing of academic work.
‘What Do We Know and What Should We Do About the Irish Border?’ is a new book from Katy Hayward that applies social science to the existing issues and what they portend.
Brexit seems likely to extend the hostility of the UK immigration system to scholars from European Union countries — unless a significant change of migration politics and prevalent public attitudes towards immigration politics took place in the UK. There are no indications that the latter will happen anytime soon.
A new report from the Royal Society about the effects on Brexit on science in the United Kingdom has our peripatetic Daniel Nehring mulling the changes that will occur in higher education and academic productivity.
In this blog post, co-authors Catherine Brentnall and David Higgins reflect on their interest in how educators change themselves and their practice […]
In this article, co-authors Stefanie Beninger, Alex Reppel, Julie Stanton and Forrest Watson reflect on the inspiration behind their research article, “Facilitating Generative AI […]
In this article, co-authors Hanan Afzal, Sameer Deshpande, and Joan Carlini reflect on the history and inspiration behind their new research article, “Glowing Beyond Shades: […]
The origin of the phrase “publish or perish” has been intriguing since this question was first raised by Eugene Garfield in 1996. Vladimir Moskovkinl talks about the evolution of the meaning of this phrase and shows the earliest use known at this point.
Social psychologist Felice Levine, who has served as executive director of the American Educational Research Association for more than 22 years, will step down in 2025.
The National Academies’ Committee on National Statistics seeks nominations for members of an ad hoc consensus study panel — sponsored by the U.S. Census Bureau — to review and evaluate the quality of the 2020 Census.
Over the past decade, furtive commercial entities around the world have industrialized the production, sale and dissemination of bogus scholarly research, undermining […]
Bravery takes many forms, and since 2006 the International Publishers Association has honored publishers who have upheld the standards and justice and […]
You’ve likely heard the hype around artificial intelligence, or AI, but do you find ChatGPT genuinely useful in your professional life? A free course offered by Sage Campus could change all th
The School of Information Sciences at the University of Illinois Urbana-Champaign seeks nominations for the 2024 Robert B. Downs Intellectual Freedom Award. […]
Over the past decade, furtive commercial entities around the world have industrialized the production, sale and dissemination of bogus scholarly research, undermining […]
Research into pressing societal challenges increasingly depends on data coming from across different disciplines and research contexts. Gordon Blair argues that to create a research culture that makes the best use of available data, the 2016 FAIR principles need to be extended in ways that address issues that have emerged in the decade following their creation.
Thanks to a collaboration between the Inter-American Foundation (IAF) and the Social Science Research Council (SSRC), applications are now being accepted for […]
What is the best strategy for finding someone missing in the wilderness? It’s complicated, but the method known as ‘Lost Person Behavior’ seems to offers some hope.
The President’s Management Agenda Learning Agenda: Public Participation & Community Engagement Evidence Challenge is dedicated to forming a strategic, evidence-based plan that federal agencies and external researchers can use to solve big problems.
Decisions taken now around how generative AI is used by academics and universities will shape the future of research. Mark Carrigan argues whilst optimistic scenarios are possible, generative AI stands ready to feed into an existing productivity oriented framing of academic work.
Biomedical research in the U.S. is world-class in part because of a long-standing partnership between universities and the federal government. On Feb. […]
Decisions taken now around how generative AI is used by academics and universities will shape the future of research. Mark Carrigan argues whilst optimistic scenarios are possible, generative AI stands ready to feed into an existing productivity oriented framing of academic work.
In his new book, Queering the Asian Diaspora: East and Southeast Asian Sexuality, Identity and Cultural Politics, the University of Nottingham’s Hongwei […]
Donald Trump’s executive actions have to a surprising extent focused on education and knowledge production. Philip N. Cohen argues beyond their short-term implications, these measures represent a concerted effort to undermine scientific enquiry across all fields of research.
Over the past decade, furtive commercial entities around the world have industrialized the production, sale and dissemination of bogus scholarly research, undermining […]
Qualitative data analysis is a way of creating insight and empathy. Strategies for data analysis and interpretation are tools for meaning-making and […]
The word censorship might bring to mind authoritarian regimes, book-banning, and restrictions on a free press, but Cory Clark, a behavioral scientist at […]
In recent years there has been an increased focus on how research papers and supplemental data can be preserved openly. Andy Tattersall, Liz Such, Joe Langley and Fiona Marshall argue equal attention should also be paid to curating communication outputs aimed at engaging non-academic audiences.
The negative consequences of relying too heavily on metrics to assess research quality are well known, potentially fostering practices harmful to scientific research such as p-hacking, salami science, or selective reporting. To address this systemic problem, Florian Naudet, and collegues present six principles for assessing scientists for hiring, promotion, and tenure.
Candace Jones, Mark Lorenzen, Jonathan Sapsed , eds.: The Oxford Handbook of Creative Industries. Oxford: Oxford University Press, 2015. 576 pp. $170.00, […]
Biomedical research in the U.S. is world-class in part because of a long-standing partnership between universities and the federal government. On Feb. […]
The U.S. National Science Foundation and the American Association for the Advancement of Science have teamed up present a 90-minute online session examining how to balance public access to federally funded research results with an equitable publishing environment.
Five organizations representing knowledge networks, research libraries, and publishing platforms joined the Federation of Humanities and Social Sciences to review the present and the future of open access — in policy and in practice – in Canada
As the U.S. Congress debates the reauthorization of the Higher Education Act, a new paper in Policy Insights from the Behavioral and Brain Sciences urges lawmakers to focus on provisions aimed at increasing the numbers of black and Latinx teachers.
To help in decisions surrounding the effects and aftermath of the COVID-19 pandemic, the the journal ‘Policy Insights from the Behavioral and Brain Sciences’ offers this collection of articles as a free resource.
Psychologist Susan Fiske was the founding editor of the journal Policy Insights from the Behavioral and Brain Sciences. In trying to reach a lay audience with research findings that matter, she counsels stepping a bit outside your academic comfort zone.
SAGE Publishing — the parent of Social Science Space – will hold its Third Annual Critical Thinking Bootcamp on August 9. Leaning more and register here
On May 13, the American Academy of Political and Social Science hosted an online seminar, co-sponsored by SAGE Publishing, that featured presentations […]
On Friday, April 23rd, join the Population Association of America and the Association of Population Centers for a virtual congressional briefing. The […]
After viewing the the televised version of the The Decameron, our Robert Dingwall asks what the farce set during the Black Death says about a more recent pandemic.
Kaye Husbands Fealing, an economist who has done pioneering work in the “science of broadening participation,” has been named the new leader of the U.S. National Science Foundation’s Directorate for Social, Behavioral and Economic Sciences.
Annie Pilote, dean of the faculty of graduate and postdoctoral studies at the Université Laval, was named chair of the Federation for the Humanities and Social Sciences at its 2023 virtual annual meeting last month. Members also elected Debra Thompson as a new director on the board.
Historian Timothy Snyder, whose work exploring the “bloodlands” between Western Europe and the Russian empire has proven remarkably timely over and over, […]
The School of Information Sciences at the University of Illinois Urbana-Champaign seeks nominations for the 2024 Robert B. Downs Intellectual Freedom Award. […]
Nominations are open for the 2025 John Maddox Prize, an international award that recognizes researchers who have defended scientific evidence and advanced […]
A 2024 report by the National Academies explores the latest advances in artificial intelligence (AI) technology and their potential effects on economic productivity, job stability, and income inequality. It also highlights key research opportunities and data needs to help workers and policymakers adapt to the evolving AI landscape.
To address racial and ethnic inequalities in the U.S. criminal justice system, the National Academies of Sciences, Engineering and Medicine just released “Reducing Racial Inequality in Crime and Justice: Science, Practice and Policy.”
The ECPR-IPSA World of Political Science Survey 2023 assesses political science scholar’s viewpoints on the global status of the discipline and the challenges it faces, specifically targeting the phenomena of cancel culture, self-censorship and threats to academic freedom of expression.
The social and behavioral sciences supply evidence-based research that enables us to make sense of the shifting online landscape pertaining to mental health. We’ll explore three freely accessible articles (listed below) that give us a fuller picture on how TikTok, Instagram, Snapchat, and online forums affect mental health.
With research-based evidence increasingly being seen in policy, we should acknowledge that there are risks that the research or ‘evidence’ used isn’t suitable or can be accidentally misused for a variety of reasons.
Over a 10-year period Carol Tenopir of DataONE and her team conducted a global survey of scientists, managers and government workers involved in broad environmental science activities about their willingness to share data and their opinion of the resources available to do so (Tenopir et al., 2011, 2015, 2018, 2020). Comparing the responses over that time shows a general increase in the willingness to share data (and thus engage in Open Science).
Nominations are open for the 2025 John Maddox Prize, an international award that recognizes researchers who have defended scientific evidence and advanced […]
When scientists make important discoveries, both big and small, they typically publish their findings in scientific journals for others to read. This […]
When scientists make important discoveries, both big and small, they typically publish their findings in scientific journals for others to read. This […]
Kate Winslet’s biopic of Lee Miller, the pioneering woman war photographer, raises some interesting questions about the ethics of fieldwork and their […]
In his new book, Queering the Asian Diaspora: East and Southeast Asian Sexuality, Identity and Cultural Politics, the University of Nottingham’s Hongwei […]
Qualitative data analysis is a way of creating insight and empathy. Strategies for data analysis and interpretation are tools for meaning-making and […]
Xenotransplantation is a fascinating subject – for obvious reasons, as it involves transplanting organs or other body parts across species boundaries (in […]
In the first post from a series of bulletins on public data that social and behavioral scientists might be interested in, Gary Price links to an analysis from the Transactional Records Access Clearinghouse.
The next in SAGE Publishing’s How to Get Published webinar series focuses on promoting your writing after publication. The free webinar is set for November 16 at 4 p.m. BT/11 a.m. ET/8 a.m. PT.
At a time when there are so many concerns being raised about always-on work cultures and our right to disconnect, email is the bane of many of our working lives.
The interactional skill of large language models enables them to carry out qualitative research interviews at speed and scale. Demonstrating the ability of these new techniques in a range of qualitative enquiries, Friedrich Geiecke and Xavier Jaravel, present a new open source platform to support this new form of qualitative research.
The Accelerator For Innovation and Research Funding Experimentation (AFIRE) is a new tool dedicated to boosting and revitalizing the design, synthesis, and implementation of experiments through innovation and research funding.
A new database houses more 250 different useful artificial intelligence applications that can help change the way researchers conduct social science research.
In his new book, Queering the Asian Diaspora: East and Southeast Asian Sexuality, Identity and Cultural Politics, the University of Nottingham’s Hongwei […]
This March, the Sage Politics team launches its first Politics Webinar Week. These webinars are free to access and will be delivered by contemporary politics experts —drawn from Sage’s team of authors and editors— who range from practitioners to instructors.
The U.S. National Science Foundation and the American Association for the Advancement of Science have teamed up present a 90-minute online session examining how to balance public access to federally funded research results with an equitable publishing environment.
In recent years there has been an increased focus on how research papers and supplemental data can be preserved openly. Andy Tattersall, Liz Such, Joe Langley and Fiona Marshall argue equal attention should also be paid to curating communication outputs aimed at engaging non-academic audiences.
In recent years there has been an increased focus on how research papers and supplemental data can be preserved openly. Andy Tattersall, Liz Such, Joe Langley and Fiona Marshall argue equal attention should also be paid to curating communication outputs aimed at engaging non-academic audiences.
Xenotransplantation is a fascinating subject – for obvious reasons, as it involves transplanting organs or other body parts across species boundaries (in […]