Prompt Engineering as a Tool for Developing Scientific Text Understanding at Higher Education Institutions

Abstract

Purpose of the research. This article aims to identify the potential of PROMPT engineering as one of the ways to help students comprehend what they read, including scientific texts, in the context of digitalization of higher education. The purpose of this study is to explore the potential of PROMPT engineering when working with large-scale artificial intelligence language models to improve students’ comprehension of scientific texts. Research materials and methods: the study design involved obtaining indexes of scientific text comprehension before and after the implementation of PROMPT engineering. The study utilized testing based on T. Borzov’s “Understanding scientific text” methodology, which relies on student self-reflection, and a questionnaire, which allows for a further analysis of scientific text comprehension at a deeper, qualitative level, as opposed to self-reflection. To create a prompt, students were given a brief outline, emphasizing that prompt writing was a creative activity. Mathematical statistics were used to analyze the data. A total of 57 students were surveyed, including 25 first-year students and 32 second-year students. The distribution of each group corresponded to a normal distribution (Shapiro-Wilk, p = 0.548). Results. To evaluate the results before and after the application of prompt engineering based on T. Borzov’s “Understanding scientific text” method; we used a paired samples t-test to compare two measurements of the same group: Measure 1 (before) and Measure 2 (after). The purpose of the test is to determine whether there is a significant mean difference between the two related data sets. 25 participants completed the study (p = 0.373): a two-tailed p-value indicating no statistically significant difference between the mean values before and after the study at the α = 0.05 level. Furthermore, a qualitative analysis of the questionnaires indicates that students’ responses after the prompt were characterized by greater accuracy and clarity of presentation, relevance, conciseness, and depth. Conclusion. The authors concluded that students experience difficulty understanding scientific texts, particularly in identifying key points and conclusions. These results of the pilot study generally suggest positive predictions regarding the use of prompt engineering to develop students’ deeper understanding of scientific texts.