Out of 60 rolls, i roll 15 1's, then the second time i roll 8 1s. say that i wanted to analyze how likely it is that i rolled the proportion of 1s that i did. would it be appropriate to use a normal model to figure something like this out? and why?