Shakir, U., & Ovink, S., & Katz, A. (2022, August), Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race  Paper presented at 2022 ASEE Annual Conference & Exposition, Minneapolis, MN. 10.18260/1-2--41779

\bibitem{asee_peer_41779}
  Shakir, U., \& Ovink, S., \& Katz, A. (2022, August), \emph{Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race}
  Paper presented at 2022 ASEE Annual Conference \& Exposition, Minneapolis, MN.
  10.18260/1-2--41779

Umair Shakir, Sarah Ovink, and Andrew Katz.  "Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race".  2022 ASEE Annual Conference & Exposition, Minneapolis, MN, 2022, August.  ASEE Conferences, 2022.  https://peer.asee.org/41779 Internet.  19 Apr, 2024

\bibitem{asee_peer_41779}
  Umair Shakir, Sarah Ovink, and Andrew Katz.
  "Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race".
  \emph{2022 ASEE Annual Conference \& Exposition, Minneapolis, MN, 2022, August}.
  ASEE Conferences, 2022.
  https://peer.asee.org/41779 Internet.  19 Apr, 2024

@INPROCEEDINGS{asee_peer_41779
author = "Umair Shakir, Sarah Ovink, and Andrew Katz"
title = "Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race"
booktitle = "2022 ASEE Annual Conference \& Exposition"
year = "2022"
month = "August"
address = "Minneapolis, MN"
publisher = "ASEE Conferences"
note = {https://peer.asee.org/41779}
number = {10.18260/1-2--41779}
}

TY  - CPAPER
AB  - Groups within the engineering education community are enabling an environment for generating transformative scholarship and pedagogical practices to challenge existing systematic educational barriers for traditionally marginalized ethnic and socio-political identities. For example, the engineering education research community is creating research opportunities to compare perspectives of minority and majority students about institutional culture issues, e.g., equitable access to and success in engineering education spaces for minority students. Unfortunately, quantitative research methodologies may lose the subtle nuances of students’ experiences. On the other end of the methodological spectrum, qualitative research methodologies can capture the rich description of students’ experiences, yet those methods can be resource-intensive and have issues related to scalability and transferability. These logistics and intrinsic quality control issues in the qualitative research paradigm may be addressed by recent state of the art developments in the natural language processing (NLP). This study aims to use NLP tools to explore students’ perspectives of the role of social class, gender, and race in shaping their (and other students') college experiences

	In this study, we sought to answer the following research question: How do undergraduate students describe the role of social class, gender, and race in shaping students’ college experiences in a higher education institution? This study publication was based on a research project that used a longitudinal qualitative research methodology. The research site was an R1, Mid-Atlantic university. For the first qualitative phase of the research project, in-depth semi-structured interviews were conducted with 113 undergraduate students at the university - 41 students were from engineering majors, and 72 were outside of engineering majors. Of the total 113, 5 % (6 out of 113), and 10% (12 out of 113) of the study sample reported their annual combined family income less than 25 thousand dollars, or more than 250 thousand dollars, respectively. This study publication analyzed a portion of those transcribed interviews that were pertinent to the role of social class, gender, and race in shaping students’ college experiences. With the help of an NLP, human-in-the-loop workflow, we took, embedded that text corpus, using a pre-trained transformer (a specific kind of neural network architecture trained to encode inputs and decode outputs), and perform a sequence of dimension reduction techniques capped with a final clustering step. The research team then utilized these groupings to perform a thematic analysis this interview is with a more nuanced understanding that only a human could do. Informed by an intersectionality lens, the results identified a wide variation (or complexity) in students’ perspectives of the role of social class, gender, and race in shaping their (and or other students') college experiences. In addition to findings about student beliefs and perspectives, this study also demonstrated the utility of NLP as a first pass on the raw text since it can help group similar responses together expeditiously. A researcher can then utilize these groupings to perform further analysis with more nuanced  understanding that only a human could do.  

AU  - Umair Shakir
AU  - Sarah Ovink
AU  - Andrew Katz
CY  - Minneapolis, MN
DA  - 2022/08/23
PB  - ASEE Conferences
TI  - Using Natural Language Processing to Explore Undergraduate Students’ Perspectives of Social Class, Gender, and Race
UR  - https://peer.asee.org/41779
DO  - 10.18260/1-2--41779
ER  -