Page 141 [141]
OCR
140 12 ALGORITHM VISUALIZATION ENVIRONMENTS: CAN AN OPTIMAL... The preliminary guestionnaire, the pre-test, and the post-test were conducted in the Socrative online classroom application. The preliminary guestionnaire consisted of 8 questions. In order to test participants’ CT, the pre-test included 8 questions based on programming-free tasks selected from the website of the Bebras (2020) contest. The learning phase was built around the AlgoRythmics animation of the shell sort algorithm (on 10-length sequence) since only 7% of the participants had heard about this sorting strategy and none of them had studied it. The experiment included the animation in all three modes: viewing, responding, and constructing. The post-test questionnaire consisted of the following 12 questions (potential range: 0-12): — Q1-6: Considering a “3-1 Shell sort strategy” (for gap values 3 and 1) on the sequence x[0]...[6] = {1,19,7,8,12,11,9}, which are the first three steps (indicate the operation and the pair of elements to be swapped or compared)? FIRST/SECOND/THIRD (I?], x[?]) — Q7-8: Consider the “3-1 Shell sort strategy” for the sequence x[0]...[6] = {1,19,7,8,12,11,9}. After the compare (x[3], x[6]) / compare(x[5], x[6]) operation, which is the next pair of elements to be compared? — Q9-12 (generalization): How many compare/swap operations are performed by a “3-1 Shell sort” algorithm for a 7-length increasing/decreasing ordered sequence? The experiment lasted a total of 2 hours. The pre-test was held for all three groups together in an amphitheatre of the university. As a preparation for the learning phase, we presented the AlgoRythmics online learning tool because many of the students were encountering this online learning environment for the first time (15 minutes). After this, each group (G1, G2, and G3) was assigned to a computer laboratory, where the learning phase began. During this, students were required to register for the online learning environment, after which they started to complete the designated course. For each group, the learning phase included two viewings of the animation. As a first step, all groups watched the animation in viewing mode (the core form of engagement) (Naps et al., 2002). During the second viewing, groups G1, G2, and G3 watched the animation with no, half, and full interactivity respectively. At the end of the experiment, participants from all groups were invited to answer the post-test questions. 12.4 Results and discussion The pre-test performances (see Figure 12.3) were analysed with one-way analysis of variance (ANOVA). The independent variable was the instructional condition (no, half, full interactivity), and the dependent variable was the pre-test score (Levene’s test showed that equal variances could be assumed: p = 0.93 > 0.05). As we expected, no significant differences were found (F(2,130) = 0.007, p = 0.99 > 0.05).
