When given a set of raw data that you have been asked to convert into a histogram, it is very beneficial to first organize it into a frequency table. This is even more important if you’ve been given a very large list of data, or if the data is unsorted. Frequency tables are a great way to quickly organize your data into bins.
–
What’s a bin? It’s just a range of values. If we were sorting test scores you might make bins such as 60-69, and 70-79. Those are pretty common bins and they have a bin width of 10. Sometimes you will see bins written as 60-70 and then 70-80. You can do this too if like, but it can be kind of ambiguous on what to do if you have a test score like 70. If you have bins with overlapping numbers, the rule is that if a value lands exactly in the middle, it goes into the higher bin. So in that previous example, if we had a test score of 70, it would go into the 70-80 bin. You can avoid that confusion just be writing your bins like we did in the beginning.
–
How many bins you create to sort your data is up to you. In some questions you will be told a bin width, and in some questions you need to decide on it yourself. I like to have about 4 to 8 bins. Sure we could group all of our test scores from 0-50 and then 51-100, we would only have two bins and they would be horrible bins. We would be much better off making smaller bin widths because then we would get more information.
–