justknimeit-29 - Comparing Distributions between Groups

Challenge 29 - Comparing Distributions between Groups
Level - Easy

Description - Imagine that you want to compare student test scores to find out whether there are any differences between the students' performances in 2020 (group 1) compared to 2019 (group 0). For example, you might want to find out whether there is an unusually high number of very good scores compared to the other year, which could be a sign of cheating. Each student participated in the same three tests and received three test scores (Score 1, Score 2, and Score 3). How similar are the distributions of the three scores between the two groups? Which score distribution differs the most? The output should contain a visualization of the conditional distributions and a statistical test for the equality of mean and variance between the groups.

Hint - Check out the verified components for visualization on the KNIME Hub.

-------------------------------------------------------------------------------------

For this challenge, I have built a viewer that allows the end-user to compare the scores across the two groups (representing the different years), using the box plot and density plot as means of visualization. The component viewer also includes one of the verified components for visualization (Conditional Density Plot).

Findings-
1) Score 1 has a bimodal distribution. Nonetheless, the means between the two groups (years) are significantly different, even though the variances are found to be not significantly different between the two groups.

2) Score 2 has a Gaussian-like distribution. However the means and variances are significantly different between the two groups.

3) Score 3 has a bimodal distribution. However, both mean and variances do not significantly differ between the two groups.

The equality of means was tested using independent t-test, with the assumption of equality of variances depending on the outcome of the Levene's test.

The equality of variances was tested using Levene's test.