Køn og metodevalg blandt samfundsvidenskabelige specialeskrivende

Forfattere

  • Rasmus Munksgaard
  • Oskar Enghoff

DOI:

https://doi.org/10.22439/dansoc.v29i1.5741

Nøgleord:

digitale metoder, akademisk produktion, topic modelling, køn.

Resumé

Feministisk teori og forskning har argumenteret for to sammenhænge mellem køn og forskningsmetoder: Kvinder benytter oftere kvalitative metoder, og køn påvirker valget af forskningsområder. Tidligere forskning baseret på fagfællebedømte publikationer understøtter disse foreslåede sammenhænge, men anerkender bias som følge af homogeniserende mekanismer såsom akademisk professionalisering og fagfællebedømmelse. Vi komplementerer disse studier gennem en analyse af de »nedre lag af akademisk produktion«, specifikt 1.103 socialvidenskabelige specialer, hvilket giver en alternativ vinkel på studiet af køn og forskningsdesign. Vi benytter nylige innovationer indenfor digital tekstanalyse og estimerer en structural topic model for at modellere korpussets latente tematiske struktur. Ud fra denne model tester vi empirisk de foreslåede sammenhænge mellem køn, forskningsmetoder og forskningsområder. Vi finder, at de kvindelige specialestuderende er mere tilbøjelige til at benytte kvalitative metoder, og at nogle forskningsområder er kønnede. Topic modelling bliver demonstreret som et effektivt redskab til at analysere akademiske tekster. ENGELSK ABSTRACT: Rasmus Munksgaard and Oskar Enghoff: Gender and choice of method among social science masters students Feminist theory and research have argued that gender and research methods are related in two ways: women are more likely to employ qualitative methods, and gender affects choice of research area. While previous research on peer-reviewed publications supports these claims, the authors acknowledge that the data is biased due to the homogenizing mechanisms of academic professionalization and peer-review. We complement these previous studies with an analysis of ”lower-level academic production”, specifically 1,103 master’s theses, providing an alternate angle to the study of gender and research design. We employ recent innovations in digital text analysis, and estimate a structural topic model of the corpora to model the latent thematic structure. Using this model, we test the proposed links between gender, research methods and research area. We find that female students are more likely to employ qualitative methods than men, and that some research areas are gendered. Topic modeling is shown to be an efficient tool in the analysis of academic texts. Keywords: Digital methods, academic production, topic modelling, gender.

Referencer

Abend, G. (2006). Styles of sociological thought: Sociologies, epistemologies, and the Mexican and U.S. quests for truth. Sociological Theory 24(1), 1-41. https://doi.org/10.1111/j.0735-2751.2006.00262.x

Bache, S.M., & Wickham, H. (2014). magrittr: A Forward-Pipe Operator for R. Tilgængelig påhttps://CRAN.R-project.org/package=magrittr

Baumer, E.P.S., Mimno, D., Guha, S., Quan, E., & Gay, G.K. (2017, 4). Comparing grounded theory and topic modeling: Extreme divergence or unlikely convergence? Journal of the Association for Information Science and Technology. In press.

Blei, D.M. (2012). Probabilistic topic models. Communications of the ACM, 55(4), 77-84. https://doi.org/10.1145/2133806.2133826

Blei, D.M., & Lafferty, J.D. (2007). A Correlated Topic Model of Science. The Annals of Applied Statistics, 1(1), 17-35. https://doi.org/10.1214/07-AOAS114

Blei, D.M., Ng, A.Y., Jordan, M.I., & Lafferty, J. (2003). Latent Dirichlet Allocation. Journal of Machine Learning Research, 3(4/5), 993-1022.

Brewer, R.M. (1989, 3). Black women and feminist sociology: The emerging perspective. The American Sociologist, 20(1), 57-70. https://doi.org/10.1007/BF02697787

Chang, J., Gerrish, S., Wang, C., Boyd-Graber, J.L., & Blei, D.M. (2009). Reading tea leaves: How humans interpret topic models. In Proceedings of the 22Nd International Conference on Neural Information Processing Systems, NIPS ‘09, 288-296.

Cook, J.A., & Fonow, M.M. (1986). Knowledge and Women’s Interests: Issues of Epistemology and Methodology in Feminist Sociological Research. Sociological Inquiry, 56(1), 2-29. https://doi.org/10.1111/j.1475-682X.1986.tb00073.x

Denny, M. & Spirling, A. (2017). Text Preprocessing for Unsupervised Learning: Why It Matters, When It Misleads, and What to Do about It. Tilgængelig på https://ssrn.com/ abstract=2849145. https://doi.org/10.2139/ssrn.2849145

DiMaggio, P., Nag, M., & Blei, D. (2013). Exploiting affinities between topic modeling and the sociological perspective on culture: Application to newspaper coverage of US government arts funding. Poetics, 41(6), 570-606. https://doi.org/10.1016/j.poetic.2013.08.004

Dunn, D., & Waller, D. (2000, 4). THE METHODOLOGICAL INCLINATIONS OF GENDER SCHOLARSHIP IN MAINSTREAM SOCIOLOGY JOURNALS. Sociological Spectrum, 20(2), 239-257. https://doi.org/10.1080/027321700279974

Erola, J., Reimer, D., Räsänen, P., & Kropp, K. (2015). No crisis but methodological separatism: a comparative study of Finnish and Danish publication trends between 1990 and 2009. Sociol