Intelligent AR tutoring produces significantly higher post-test performance than non-intelligent AR training for motherboard assembly¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1–r2 · q2 · i3 large
Currency · newest study 2015 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The experimental group using the intelligent AR tutor scored significantly higher on the written post-test than the control group using the traditional AR tutor. → Giles Westerfield 2015 q2 i3 Normalized learning gains differed significantly between groups, with a printed effect size of Cohen's d = 0.981. → Giles Westerfield 2015 (2)

Evidence¶

Giles Westerfield 2015¶

Giles Westerfield, Antonija Mitrovic & Mark Billinghurst. (2015). Intelligent Augmented Reality Training for Motherboard Assembly. Int J Artif Intell Educ 25:157–172. https://doi.org/10.1007/s40593-014-0032-x

q2 · i? · causal · r2

In the evaluation study, 16 university students were randomly allocated to the intelligent AR tutor or the traditional AR tutor. Post-test performance of the experimental group was "significantly higher than that of the control group (t(14) = 3.374, p<.005)". Means were 9.13 (SD 1.13) versus 6.63 (SD 1.77) out of 10.

"The post-test performance of the experimental group is significantly higher than that of the control group (t(14) = 3.374, p<.005). This is less than the Bonferroni-corrected value of .0083 (.05/6), so the intelligent AR tutor produced a significantly better learning outcome than the non- intelligent AR tutor."

Giles Westerfield 2015 (2)¶

Giles Westerfield, Antonija Mitrovic & Mark Billinghurst. (2015). Intelligent Augmented Reality Training for Motherboard Assembly. Int J Artif Intell Educ 25:157–172. https://doi.org/10.1007/s40593-014-0032-x

q2 · i3 · causal · r1

Normalized learning gains were 0.66 (SD 0.26) for the experimental group and 0.40 (SD 0.21) for the control group. The article reports "a significant difference between the normalized learning gains of the two groups (t(14) = 2.198, p<.05)" with an effect size of "Cohen 'sd " = 0.981, a large effect.

"There is also a significant difference between the normalized learning gains of the two groups (t(14) = 2.198, p<.05). The effect size (Cohen 'sd ) is 0.981, which is a significant improvement."

Discussion¶

Related Claims¶

• Participants trained with the intelligent AR tutor completed the unassisted physical assembly task significantly faster, with no significant difference in errors — related
• Intelligent tutoring systems improve achievement over non-individualised instruction — a broader claim this one bears on
• AI-enabled personalized STEM education yields a statistically significant overall effect of g = 0.455, which its authors class as small, compared with traditional non-AI instruction in K-12 schools — a broader claim this one bears on