After a two-to-three-week computer vision professional development program, 13 of 18 (72%) of middle school teachers' computer vision constructs shifted clusters, which the authors read as restructured perceptions (pre/post, no comparison group)¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · newest study 2026 · checked against its sources 2026-10-29 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q2 i? After computer vision PD, 13 of the 18 constructs (72%) on which 12 teachers rated computer vision shifted clusters, which the authors read as participants restructuring how they distinguished computer vision constructs; a pre/post design with no comparison group cannot show the PD caused the shift. → Kurz 2026

Evidence¶

Kurz 2026¶

Kurz, T. L., Jayasuriya, S., Swisher, K., Mativo, J., Pidaparti, R., & Robinson, D. T. (2026). Computer vision versus human vision: analyzing middle school teachers' construct restructuring following computer vision professional development. Education Tech Research Dev. https://doi.org/10.1007/s11423-026-10594-2

q2 · i? · causal · r1

Pre/post repertory grid study (no comparison group) of 12 in-service middle school teachers (six in Arizona, six in Georgia) using hierarchical cluster analysis (Ward's Method) of construct ratings. The results section reports that computer vision constructs shifted clusters far more often than human vision constructs, with "72% (n = 13) of the constructs shifted clusters" for computer vision versus 6% (n = 1) for human vision. The counts are of the 18 constructs in the repertory grid, not of teachers.

"Regarding human vision, only 6% (n = 1) of the constructs shifted clusters; regarding computer vision, 72% (n = 13) of the constructs shifted clusters."

Discussion¶

Related Claims¶

• All computer vision cluster means increased post-PD and the most descriptive cluster became more detailed, indicating teachers viewed computer vision constructs as more descriptive after PD — a broader claim this one bears on
• Teachers' perceptions of human vision remained largely stable following computer vision PD, with only one construct (6%, n = 1) shifting clusters — related
• After PD, teachers rated computer vision as more similar to human vision, with computer vision cluster 1 matching human vision cluster 1 on seven constructs post-PD versus one pre-PD — a narrower finding that bears on this claim
• Teachers' construct for computer vision 'sensing the surroundings' was initially viewed as not descriptive of computer vision but became more descriptive after PD — a narrower finding that bears on this claim
• Individual computer vision constructs showed notable mean increases and cluster shifts, including sensing the surroundings (26% increase, cluster 4 to 2) and able to detect objects (18% increase, cluster 3 to 1) — a narrower finding that bears on this claim
• Short-duration TPACK-based professional development improved teachers' ML knowledge and interest (ML4STEM) — related
• An online PD program using unplugged explainable machine-learning activities increased K–12 educators' AI self-efficacy from below-moderate to above-moderate levels — related