Introduction

This report uses a case study on well-differentiated thyroid cancer to find the best graphs for showing data. By looking at different types of graphs, we aim to see which ones clearly show important patterns in risk, recurrence rates, and patient recovery. This helps us choose the right graphs for different kinds of data, making the information easy to understand and accurate.

Graph #1

The initial graph compares the total risk of cancer for males and females. This bar graph effectively highlights the differences in risk between genders.

A grouped bar graph, as suggested by ChatGPT shows separate bars for male and female categories with risk levels on the x-axis. However, the first bar graph is more effective for comparing risks between genders, while the grouped bar graph can be confusing.

I have tried another representation using the 100% stack graph because the ratio of male and females is 271: 71. This graph can be misleading as it shows an equal count for males and females, despite the actual ratio. Therefore, it is not suitable for overall representation.

Another representation where I used a stacked graph, which can be a second option if there more categories on the x-axis.

Graph #2

This graph shows the recurrence of cancer at each stage, with more categories on the x-axis

A stacked graph, as suggested by ChatGPT, provides a good representation when there are multiple categories.

Another 100% stacked graph was used for the same representation. However, it is not suitable because the count for each stage of cancer is different. For example, it shows a 100% chance of recurrence for stage IVA cancer, but the actual count for stage IVA is just 3, making this graph misleading.

Graph #3