Picture bias: graphs made students more likely to confirm problem statements, irrespective of whether the statements were true¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1–r2 · q2
Currency · newest study 2017 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Presentation condition significantly predicted giving a confirmatory answer, with more confirmations in the multimedia condition. → Ögren 2017 q2 i? A significant interaction between presentation condition and statement truth: performance was better for true statements with a graph present, while true/false made no difference without a graph. → Ögren 2017 (2)

Evidence¶

Ögren 2017¶

Ögren, M., Nyström, M., & Jarodzka, H. (2017). There's more to the multimedia effect than meets the eye: is seeing pictures believing? Instructional Science, 45, 263–287. https://doi.org/10.1007/s11251-016-9397-6

q2 · i? · causal · r2

Multi-level logistic regression predicting confirm/reject answers; the 'withoutgraph' predictor was significant (z = -2.226, p = 0.0260), indicating the picture bias the authors hypothesized.

"The sign of the estimate for 'withoutgraph' tells us that more confirmatory answers were given in the multimedia condition"

Ögren 2017 (2)¶

Ögren, M., Nyström, M., & Jarodzka, H. (2017). There's more to the multimedia effect than meets the eye: is seeing pictures believing? Instructional Science, 45, 263–287. https://doi.org/10.1007/s11251-016-9397-6

q2 · i? · causal · r1

Regression including answer truth as factor found the interaction significant (z = -2.624, p = 0.008); true answers were answered more correctly with a graph, but truth had no influence without a graph. A post-hoc comparison between conditions when the answer was true was marginally significant (p = 0.056).

"there is a significant interaction between presen- tation condition and whether the answer is true or false"

Discussion¶

Related Claims¶

• A first evaluation of 3B confirms its conformance to critical usability requirements including safety and comfort, believability, simplicity, acceptability, extensibility, affordability, and mobility — related
• No overall multimedia effect on problem-solving performance in vector calculus: graphs did not significantly improve correctness — related
• Solving problems with graphs increased mental effort, shown by a higher proportion of silence in think-aloud data — related
• Graphs drew visual attention away from text: proportion of dwell time on input and problem statement was significantly shorter when a graph was present — related
• Contrasting problems show a graph can help on one problem and harm on another: verbal data reveal both more active graph use and more shallow, confident confirmation when graphs were present — related