Purpose: The purpose of this article is to describe a performance assessment to measure computational thinking for middle school students. A second purpose involved determining potential reasons for variations of results across students in an effort to strengthen this assessment in engaging K-12 students in CT.
Findings: 1. Across the three tasks for the 311 students taking the Fairy assessment, the following mean scores were reported. Each task was out of 10 possible points. Task 1: Mean= 6.03, SD= 3.87; Task 2: Mean = 4.54, SD=4.73; Task 3: Mean= 5.93, SD= 4.34. Mean of 16.50 with SD of 9.95 was the total of all 3 task reported. 2. Parent education and use of computers at work as well as frequent computer use and higher grades on last report card were positively associated with higher assessment score. 3. The student's confidence in computers was a predictor of higher performance
Recommendations: One recommendation stated that more research is needed to determine the all the benefits of pair programming for middle school students. Several limitations were reported in this study that may help future work in this area. Some of the limitations in this study to consider for the future include: gather measures of construct validity of the Fairy assessment to ascertain that the assessment is measuring the CT aspects of thinking algorithmically and making effective use of abstraction and modeling. Secondly, the data collected from pair programming condition should be independent as to not violate statistical assumptions. Thirdly, reporting associations does not provide information about relative importance of each factor.
Sample Size: 311
Participant Type: 311 students from 7 public schools in central California coast. This was a two year study were 78 students participated in after-school program in year 1, and 37 participated in after-school and 196 participated from an elective technology course for year 2. Of the 311 students, 36% were females with an age range of 10-14 years (mean=12).
Notes: Reported students were 52% white and 37% Latino/a