2.4. Data Analysis
For the analysis of the quantitative data, one-way ANOVA, Chi-Square test of independence and descriptive statistics were applied. For ANOVA, in order to determine which groups caused the significant difference, Tamhane T2 test, one of the most common post hoc (multiple comparison) tests, was used. For the normal distribution of data, the skewness and kurtosis coefficients were examined. For the normality test skewness coefficient of a distribution taken in the range of -1.5 to +1.5 and the kurtosis coefficient of a distribution taken in the range of -1.5 to +1.5 according to Tabachnick and Fidell (2013). For ordinal data, Chi-Square test of independence was performed. Besides ANOVA and Chi-Square test of independence, …show more content…
A significant relationship between video lecture watching behaviors of learners and lecturers’ gender was found (X2 = 29.31, N = 10282, p < .001). 62% of completely watched events belonged to female lecturers’ videos.
A Chi-square test of independence was calculated to compare the viewing behaviors of learners and lecturers’ age in “40 and below” and “below 40”. There was a significant relationship between viewing behaviors of learners and lecturers’ age, (X2 = 14.46, N = 10282, p < .001). 57% of completely watched events belonged to videos of lecturers, who were 40 years old and below.
3.2.4 Video Events with respect to Learner Profile
In order to determine whether the video events of learners differed significantly depending on learner profile, Chi-square test of independence test was applied. A significant relationship between doSeek event and learners’ gender was found (X2 = 58.67, N = 10282, p < .001). Female students (58%) preferred to watch video lectures without seeking rather than male learners (42%). A significant relationship was also found between FullScreen event and gender (X2 = 11.02, N = 10282, p < .001). 54 % of FullScreen watch made by male
For the analysis of the quantitative data, one-way ANOVA, Chi-Square test of independence and descriptive statistics were applied. For ANOVA, in order to determine which groups caused the significant difference, Tamhane T2 test, one of the most common post hoc (multiple comparison) tests, was used. For the normal distribution of data, the skewness and kurtosis coefficients were examined. For the normality test skewness coefficient of a distribution taken in the range of -1.5 to +1.5 and the kurtosis coefficient of a distribution taken in the range of -1.5 to +1.5 according to Tabachnick and Fidell (2013). For ordinal data, Chi-Square test of independence was performed. Besides ANOVA and Chi-Square test of independence, …show more content…
A significant relationship between video lecture watching behaviors of learners and lecturers’ gender was found (X2 = 29.31, N = 10282, p < .001). 62% of completely watched events belonged to female lecturers’ videos.
A Chi-square test of independence was calculated to compare the viewing behaviors of learners and lecturers’ age in “40 and below” and “below 40”. There was a significant relationship between viewing behaviors of learners and lecturers’ age, (X2 = 14.46, N = 10282, p < .001). 57% of completely watched events belonged to videos of lecturers, who were 40 years old and below.
3.2.4 Video Events with respect to Learner Profile
In order to determine whether the video events of learners differed significantly depending on learner profile, Chi-square test of independence test was applied. A significant relationship between doSeek event and learners’ gender was found (X2 = 58.67, N = 10282, p < .001). Female students (58%) preferred to watch video lectures without seeking rather than male learners (42%). A significant relationship was also found between FullScreen event and gender (X2 = 11.02, N = 10282, p < .001). 54 % of FullScreen watch made by male