Closed llaarraa closed 10 years ago
Will check. Preliminary analysis: it appears there are 13* median = 0, 2_median=1 and 1_median=2 in both the graph and in the table. I suspect the Symptom labels are wrong/switched in either the graph or in the table. The same for IQR - if ignoring Symptom labels, the graph and table seem to have the same values for IQR, at first glance.
The table had the wrong ordering of symptom labels (labels were ordered via decreasing proportions of symptoms, while the data was not). Fixed in 1b1db8c90aec1565164ad29a9f42b49685b28fb3.
The data about medians and IQR are not consistent in the graph and table. For example: fatigue has a median of 2 in the table, 0 in the graph - also iqr not consistent