Version 1: A combination of a box and whisker plot and a density curve is called a violin plot. In version 1 with only 1 dataset.
This combination displays a dataset's distribution in a more detailed way than a traditional box plot.
Violin plots show both the summary statistics (the box and whiskers)and a probability density of the data (the shape of the plot.
This allows for a richer visualisation.
Version 2: Plot with 2 different datasets. Each plot with individual scaling of the dataset.
When individual scales might be considered:
1. When comparing distribution SHAPE only - if you only care about skewness, spread pattern, or outliers, regardless of magnitude
2. With extremely different variances - where one dataset's range makes the other unreadable
3. When the research question is specifically about internal distribution patterns
4. Individual scales can be misleading. Both might look "nice with normal distributions" dispite one being 3x larger than the other.
Version 3: Plot with 2 different datasets. Both plots with synchronized dimensions of the datasets.
Why sxnchronized dimensions might be better for comparisation:
1. Preserves relative scale and relationshipd
2. The magnitude difference becomes visually obviou
3. You can visually estimate the performance difference
4. The golden rule is: If you want to compare data sets, they must be represented in the same coordinate system.
5. Version 3 allows to activate or deactivate the numbering around the graphics if needed for any reason
6. Input fields can be cleared to simplify the input with own datasets.
