Background: One way to broaden the participation of women in engineering beyond the commonly reported 20% proportion of degrees awarded is through providing outreach (e.g., enrichment programs) for young learners. Yet, we do not know the full impact of outreach, especially how it impacts persistence and engineering identity (eID) among girls, because these enrichment programs often happen in silos. Therefore, with the fast propagation of engineering education (EnEd) research, there is a need to quickly evaluate relevant research to identify gaps in our knowledge of eID development via outreach.
Purpose: A traditional (i.e., by hand) thematic literature review was conducted as a part of an ongoing study on Middle School outreach, eID, and persistence for women in engineering. However, we wanted to understand the viability and accuracy of a computer-driven analysis, Linguistic Inquiry and Word Count (LIWC), as a resource for fast, reliable analysis of literature.
Scope/Method: The program LIWC was used as an analysis tool to quickly gather data on a set of six literature review papers, with both user-defined and built-in dictionaries, as well as a topic modeling procedure, to refine the methodology for this novel approach.
Results: The use of LIWC to conduct a thematic literature review on a subset of articles confirmed the same themes that arrive via traditional coding methods , yet the novel computational method took less time and offered a few surprises. Thus, a priori codes using traditional LIWC analysis, with both the standard dictionary and our custom dictionary, and in vivo codes using LIWC meaning extraction method (MEM analysis), allowed us to quickly analyze how many papers used the same terms.
Conclusions: While the available computational tools allow us to quickly focus on the most salient of themes in the literature and come to inter-rater consistency faster, its use does not replace the need to read. Novel tools like LIWC might be the future for rapidly understanding the language of EnEd research and could help researchers more easily categorize prior research in their areas.
Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.