Table 7 Performance of o1-preview and human participants on Merk et al.’s Data literacy assessment.

From: Comparative evaluation of OpenAI O1 and human performance in higher order cognition

Data literacy

Human pre-test

Human post-test

o1-Preview

Pre-test

Post-test

Use data (9 items)

3.28 ± 1.84

4.17 ± 2.02

8.60 ± 0.70

2.89

2.19

Transform data into information (5 items)

2.96 ± 1.21

4.04 ± 1.34

4.80 ± 0.42

1.52

0.57

  1. Item counts for the Merk et al. subscales reflect the reduced text-only subset used in this study. Items requiring graphical inspection or interactive components were excluded; scores represent proportion correct within each retained subscale
Back to article page