Scores on five constructs of the Motivated Strategies for Learning Questionnaire (MSLQ) for 300 simulated respondents. Each score is the mean of the construct's Likert items, on a scale from 1 to 7. The responses were generated by a large language model (GPT), not collected from students: they come from a study of how language models answer psychological surveys, and are used here because they are a well-structured, openly shareable cross-sectional example. They should not be read as evidence about students.
Format
A data frame with 300 rows and 5 columns:
cognitive_strategiescognitive strategy use
intrinsic_valueintrinsic value of the course
self_efficacyself-efficacy
self_regulationself-regulation
test_anxietytest anxiety
Source
The SRL_GPT data of the psychnets package, with descriptive
column names. Saqr, M. (2025). Delving into the psychology of machines:
Exploring the structure of self-regulated learning via LLM-generated
survey responses. Computers in Human Behavior, 173, 108769.
doi:10.1016/j.chb.2025.108769
Details
The five constructs are strongly correlated (about 0.8, except test anxiety), which makes the data a good illustration of how the choice of covariance structure shapes a latent profile analysis.
Examples
descriptives(srl, names(srl))
#> variable n n_missing mean sd min max
#> 1 cognitive_strategies 300 0 5.707179 0.5408047 4.153846 7.000000
#> 2 intrinsic_value 300 0 6.116296 0.5226347 4.222222 7.000000
#> 3 self_efficacy 300 0 5.926296 0.5870661 4.000000 7.000000
#> 4 self_regulation 300 0 5.544074 0.5000432 3.777778 6.777778
#> 5 test_anxiety 300 0 3.700000 0.6443088 2.000000 5.750000
#> n_distinct
#> 1 34
#> 2 22
#> 3 25
#> 4 25
#> 5 16