AI evidence extraction

To use or not to use ChatGPT in higher education? A study of students’ acceptance and use of technology

Artur Strzelecki · Interactive Learning Environments · 2023

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

7/10

Relevance

0/4

Quality (LMQS)

Evidence

691

Citations

356.19

FWCI

Top 10%

Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.1080/10494820.2023.2209881

Methodology & findings

Study design

Cross-sectional survey with structural equation modeling (partial-least squares SEM) analysis of self-reported data

Sample

N = 534, 2 groups

Primary method

Partial least squares (PLS) method of structural equation modeling (SEM) for data analysis

Main result

The study found that "Habit was found to be the best predictor of behavioral intention, followed by performance expectancy and hedonic motivation. The dominant determinant of use behavior was behavioral intention, followed by personal innovativeness." Additionally, "Nine out of ten proposed hypotheses were confirmed by the results."

Reports effect sizes.

Research paradigm

Positivist/Quantitative

Author conclusions

The authors conclude that "The research highlighted the need for further examination of how AI tools could be adopted in learning and teaching." The model demonstrates that technology adoption among students is driven primarily by habit formation and performance expectations, with implications for educational technology implementation.

Risk of bias

Self-report bias (all data self-reported); Selection bias (single Polish university, may not represent diverse student populations); Cross-sectional design (cannot establish causality); Potential social desirability bias in technology adoption responses; Lack of temporal separation between predictor and outcome measurement; Self-reported data from a single institution (Polish state university); Cross-sectional design limits causal inference; Potential selection bias from voluntary participation; Single geographic location and cultural context; Self-reported data susceptible to social desirability bias; Single institution sample (Polish state university) limits generalizability; Cross-sectional design prevents causal inference; Potential common method variance from self-report methodology

Limitations

The study was based on "self-reported data of 534 students from a Polish state university," which may limit generalizability
The authors note that further examination is needed regarding how these findings apply across different educational contexts and populations.

Open questions raised

Need for further examination of how AI tools could be adopted in learning and teaching
The abstract suggests future research should explore actual implementation of ChatGPT in educational contexts
The need for further examination of how AI tools could be adopted in learning and teaching contexts
The authors identify the need for further examination of how AI tools could be adopted in learning and teaching contexts, and suggest investigation of these technology adoption patterns across different educational settings and student populations.

Data: not_statedCode: not_statedExtracted from: pdfAgreement 60%

Explore related topics

Productivity Effects Research Productivity Survey