Grouped Scatter Plot of Medical Error Recognition Performance

551 Words2 Pages
Grouped Scatter Plot of Medical Error Recognition Performance 1.Figure 1 in the publication by March and colleagues (2013) is a grouped scatter plot. This chart depicts the distribution of scores for each study participant within each group. 2.The purpose of Figure 1 (March et al., 2013) is to show the distribution of individual clinicians' performance in error recognition when reviewing a simulated patient's record using an electronic healthcare record (EHR) interface. The authors wanted to show how the level of experience impacted error recognition performance and therefore grouped the distribution by fellows, residents, and intersn. The figure also shows three lines per grouping, with the middle line representing the mean. The other two lines are not mentioned, but they should represent standard error. At the top of the graph is a horizontal line depicting a statistically significant difference between the two groups with the biggest difference in experience: fellows and interns. This was an appropriate graphical choice for presenting this data. 3.Without having personal insight into the author's rationale for choosing a grouped scatter plot to present this data, I have to assume that the authors wished to visually portray the distributions and a scatter plot was their choice. An alternative and equally valid choice would have been the box plot, but the granularity of the distribution would have been lost by using a box plot. In other words, the box plot would have

More about Grouped Scatter Plot of Medical Error Recognition Performance

Open Document