Analysing a binary variable
Reporting
We started with determining the actual percentages we've seen in the sample of the two categories in the binary variable, then determined if the difference was significant and if so how big the difference was.
Usually I'd recommend to add a visualisation but with only two categories this will just be a waste of paper. The report for all of the analysis could simply go something like shown below.
Management was curious if the division of male/female was equal in the company. The result of a small survey showed that out of the 46 respondents that indicated their gender, 12 indicated to be female (26%), and 34 to be male (74%). An exact binomial test indicated that the percentages were significantly different, p = .002. Cohen’s g suggests that the difference can be classified as medium, g = .24. The number of female employees seem to be too low. During a focus session it was discussed what might cause this and how to change it. ..... |
Note the final paragraph explains the some-what technical results into more understandable English, something many readers would often appreciate.
Single binary variable
Google adds