Insights into preservice nurses' attitudes to autism service provision training : text analysis
Authors
Advisor
Issue Date
Keyword
Degree
Department
Other Identifiers
CardCat URL
Abstract
The purpose of this study is to explore how text mining and topic modeling methods can be useful in analyzing open-ended short survey responses using data from a simulation lab done at Ball State’s nursing department to teach nursing students how to successfully care for a person with autism. Nursing students underwent a simulation in which there was a “patient” role playing a person with autism. After the simulations, students were asked open-ended questions about their prior coursework and the simulation in regard to the effectiveness of the care they were able to provide during the simulation. Using “R” statistical software, the data was analyzed with various text mining and topic modeling procedures and R packages. It was found that overall topic modeling performed well on short, open-ended survey responses. In the case of the nursing data, the overall themes showed that the nursing students found the simulation to be a positive and realistic learning experience. Emphasis was placed on working as a team and the need for more training on how to successfully take vital signs for a patient with autism spectrum disorder. Topic modeling can be very useful to qualitative researchers in hopes of discovering more about their data, even in instances of having small textual data sets. However, topic modeling should be maintained a tool alongside traditional qualitative methods to ensure quality of data analyses.