If you’re going to look at the frequency of scores between 0 and 100, you could have 100 bins, one for each possible score. The problem is that these may be arbitrary. You need to decide on the “bins” that your frequency counts will be sorted into. If you wanted to compare the frequency distributions between two groups on a single variable, you’d need multiple histograms. For example, if you only wanted to look at the weight distribution of a certain age group or gender, you should only include data for that group. Be careful not to mix the data from groups you don’t want to measure together into one histogram. For example, if you have the weights of a group of people, you’d have each measured weight recorded in your dataset. The first requirement is fairly straightforward. In order to make a histogram, you need a few things:
Those tests still use histograms as a basis though and creating and observing a histogram is a crucial first step in showing you roughly what sort of distribution you may be dealing with. Of course, if you really want to determine whether your frequency distribution is normal or not, you’d run a normality test in Excel on your data.
