Confirmatory factor analysis on separability of planning and insight constructs

Abstract

This study aimed to investigate the separability of planning, a form of noninsight problem solving, from insight problem solving by means of using confirmatory factor analysis (CFA). The relationships of these two types of problem-solving tasks with meta-cognitive awareness were also assessed. Participants performed a set of planning tasks, a set of insight tasks and a self-report inventory on metacognitive ability. The CFA results revealed that planning and insight problem solving were closely related constructs and were not clearly separable. Model comparisons indicated that the fit of the alternative one-factor model was slightly better than the fit of the two-factor model. The correlational results showed that both planning task performance and insight problem-solving performance had no correlations with the metacognitive knowledge or the metacognitive regulation components of the metacognitive awareness inventory.

Keywords:

• Confirmatory factor analysis
• Insight problem solving
• Metacognitive awareness
• Planning

Acknowledgments

The research reported in this paper was carried out under Sevtap Cinan's project on planning, insight problem solving, and metacognition and it is also part of the Master's thesis of GÖ. The HP Touch Smart computer was purchased through Istanbul University Scientific Research Projects Coordination Unit.

Notes

¹The CFAs were conducted again by using only the nonhint score as the insight PST score: two-factor model, χ²(4)=2.88, p=.58, RMSEA=0.0 (0.0–0.13), CFI=1.00, GFI=0.99, AGFI=0.96, SRMR=0.028, AIC=24.88, correlation between the two-factor, .99; one-factor model, χ²(5)=2.86, p=.72, RMSEA=0.0 (0.0–0.10), CFI=1.00, GFI=0.99, AGFI=0.97, SRMR=0.028, AIC=22.86.