The Iris dataset, shown in a loreplot below, is visualized using a few different options further down this page. Note how using stacked bar plots in some cases can distort the data.
In the plot below, individuals are separated into "large" and "small" groups based on an arbitrary threshold for sepal width. This approach can obscure how sepal width is distributed within species, particularly for the virginica species.
Here, individuals are divided into six equal segments (bins) based on sepal width. The plot below emphasizes the small number of setosa specimens with small sepals, showing how they disproportionately influence the plot.
The plot below slices the data into percentile ranks, which leads to bins of varying widths. For instance, the largest bin covers a range of ~1 cm, while others span just 1-2 mm. This can distort the perception of data distribution.
