Development and validation of the Conceptual Survey on Wave Optics

A new diagnostic instrument, the Conceptual Survey on Wave Optics (CSWO), was developed and validated on 224 high school students (aged 18–19 years) in Croatia. The process of test construction, which included the testing of 61 items on the total of 712 students is presented. The final version of the test consists of 26 multiple-choice items which cover basic conceptual aspects of interference, diffraction, and polarization of light at the high school level. The construction of the test and the evaluation of its functioning and validity were conducted using the Rasch model. The test construct was based on the increasing levels of cognitive complexity in accordance with the Webb’s depth of knowledge model. The theoretical construct was empirically confirmed, and it suggests that the explanations of wave optics phenomena and their application to real-life situations are the most difficult aspects of high school students’ understanding of wave optics. The results of the Rasch analysis suggest that the CSWO is a reliable diagnostic instrument suitable for administration as a post-test to high school students after instruction on wave optics and possibly also for university level students in some introductory physics courses.

Figure 1

Item Q22 from the CSWO.

Figure 2

Wright map (item-person map) for the CSWO. Each “x” represents one student. The distribution of students on the left-hand side and the distribution of the CSWO items on the right-hand side are shown along the same logit axis in the middle, where $M$ stands for the mean of each distribution, and $S$ and $T$ mark points 1 or 2 standard deviations away from the mean.

Figure 3

Bubble chart of the CSWO item difficulties vs Infit MNSQ values in logit. Each item is represented by a circle whose size is proportional to its calibration uncertainty (standard error). The maximum range of item misfit that is acceptable for measurement is shown (0.5–1.5 logit). The central line represents the model value of Infit of 1.

Figure 4

Scatter plot of item difficulties, obtained from separate analyses of male and female students, with confidence bands.

Figure 5

Empirical validation of the theoretical construct of the CSWO. Difficulties of the five groups of items belonging to the different levels of the construct are shown with error bars indicating three standard errors for each group.