Evaluation of cognitive load and scoring procedures of concept map-based tests

Using concept map-based tests as assessment tools is one of the new assessment methods that can be an alternative for traditional assessment methods. So, the goal of the present study was evaluating cognitive load and scoring procedures of these tests. The research sample consisted of students of four biology classrooms from four different high schools (totally 109 students) who were selected via a multi-stage cluster sampling method. The research tools were a multiple-choice test (researcher-made), four concept map-based tests (researcher-made) and the NASA-TLX scale (Hart & Staveland, 1988). The study was done with some type of evaluation design. The statistical methods included Pearson's correlation coefficient, independent samples t-test and one-way ANOVA (Welch test and Games-Howell post hoc). Findings showed that construct-a-map tests imposed significantly more cognitive load on the subjects than fill-in-a-map tests. Also, relational, structural and similarity to expert map scoring procedures had acceptable and significant criterion validity, discriminant validity and reliability. All in all, relational, structural and similarity scoring procedures can be used appropriately for scoring the concept map-based tests. Also, using the fill-in-a-map tests can be an appropriate strategy to overcome the cognitive overload during the concept map-based tests. Thus, the results of the present research can be useful to design tests based on the cognitive rules and to bridge the gap between academic evaluation and cognitive sciences.