# difference in results of hypothesis when checking with histogram and boxplot

I'm not sure why results differ when I'm doing hypothesis with wilcox test for data, the hypothesis results differ from normal boxplot and histogram. Which is the best graph to check the results for the hypothesis if its correct.

whereas hypothesis is rejecting null; which is men and female are equal

First graph says: Men and female has difference

second graph says : men and female are equal

• I don't understand how you infer from the histograms that there is a difference in location. Jan 30, 2023 at 9:51
• by simplying looking at the graph im assuming that they both are different, I'm just trying to understand how to infer the hypothesis. Also, not sure how to understand why in these both graphs are difference. can you tell me ?
– none
Jan 30, 2023 at 10:40
• I don't see a location difference in the histograms. You are not interpreting the plot correctly. Jan 30, 2023 at 11:31
• @Roland are you saying my results should match with histogram? and cos boxplot clearly shows male and female are equal, and histogram says"orange" is more then "purple" which is female and male. I'm not quit understand why, the way I'm looking in histogram is not location I'm looking overall which entry is more and comparing with the hypothesis, maybe I'm interpreting wrong.
– none
Jan 30, 2023 at 11:52
• No, the histogram does not say '"orange" is more then "purple"', not at all. You need to study how to interpret a histogram. You have more orange observations than purple observations but that corresponds to sample sizes. The values, for which you want to compare the location statistics, are on the y-axis. Jan 30, 2023 at 12:09