Elementary school children monitor and control their answers to open-ended text comprehension questions more accurately than answers to true-false questions¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1–r2 · q3 · i2–i3
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i3 Monitoring accuracy (gamma correlations) was significantly higher for open-ended than true-false questions at both measurement points. → Steiner 2020 q3 i2 Control accuracy (gamma correlations between maintenance and performance) was significantly higher for open-ended than true-false questions at both measurement points. → Steiner 2020 (2)

Evidence¶

Steiner 2020¶

Steiner, M., van Loon, M. H., Bayard, N. S., & Roebers, C. M. (2020). Development of Children's monitoring and control when learning from texts: effects of age and test format. Metacognition and Learning, 15, 3–27. https://doi.org/10.1007/s11409-019-09208-5

q3 · i3 · causal · r1

Repeated measures ANOVA with test format as within-subject factor, collapsed across both grades (N = 302), comparing gamma correlations between confidence judgments and performance for the two formats. Monitoring was significantly more accurate for open-ended than true-false questions at both measurement points, with ηp2 = .15 at T1 and ηp2 = .11 at T2.

"The difference between the gamma correlations for monitoring accuracy for open-ended and true-false questions was significant at T 1, F (1, 251) = 44.89, p <. 0 0 1 ,ηp2 =. 1 5 , a n d T2, F (1, 274) = 35.31, p <. 0 0 1 ,ηp2 = .11"

Steiner 2020 (2)¶

Steiner, M., van Loon, M. H., Bayard, N. S., & Roebers, C. M. (2020). Development of Children's monitoring and control when learning from texts: effects of age and test format. Metacognition and Learning, 15, 3–27. https://doi.org/10.1007/s11409-019-09208-5

q3 · i2 · causal · r2

Repeated measures ANOVA comparing gamma correlations between maintenance decisions and performance across test formats in the same sample. Control was more accurate for open-ended questions at both T1 (ηp2 = .10) and T2 (ηp2 = .13), as visible in Table 3.

"The difference between control accuracy (Gamma correlations between maintenance and performance) for open-ended and true-false questions was significant at T 1, F (1, 206) = 22.43, p < .001, ηp2 =. 1 0 , a n d T2, F (1, 231) = 35.75, p < .001, ηp2 = .13"

Discussion¶

Learner Variables¶

• Metacognition — outcome: instruction changes it

Related Claims¶

• Fourth graders monitor and control answers to open-ended questions more accurately than second graders, but no significant age difference was found for true-false questions — related
• Monitoring-based control (the relation between confidence judgments and maintenance decisions) is stronger for fourth graders than second graders and strengthens over time for open-ended questions — related
• Children of both age groups improved monitoring and control accuracy from T1 to T2 for both test formats, mainly for incorrect rather than correct answers — related
• Children are considerably more overconfident about incorrect answers to true-false questions than about incorrect answers to open-ended questions — related
• Children's text comprehension test performance improved from T1 to T2, driven mainly by second graders, with fewer omission errors over time — related
• Children's test performance is stable over six months but monitoring and control accuracy shows little inter-individual stability — related
• Monitoring discrimination and overconfidence improve with age between kindergarten and second grade — related
• Calibration (judgment accuracy) measures such as the Goodman-Kruskall gamma coefficient and signal detection theory's d' can be empirically compared, and calibration may vary across knowledge domains — related