Thread: histograms
View Single Post
  #2   Report Post  
 
Posts: n/a
Default histograms

"Stacy U" wrote:
I am confused on the 'bins'.. Does the program automatically
fill in these or do you have to input the numbers? Also how
do you know what numbers to input? Confused on histograms
in general.


You are not alone. First, one way to learn what will happen
is simply to try it. Second, if you do not use the "bins" option,
Excel will select a default. But you will very probably not like
the default.

As to what numbers to put into "bins" -- and how many bins
to have -- that is controversial. It depends a great deal on
the distribution of your data and the purpose of the histogram.

A very general guideline is 5-15 bins. If you do a google
search, you will learn that some scholars in the past have
tried to fomalize the choice. There are two commonly-used
formulas. I don't remember them off-hand. They might be
a good starting point.

But as you know, Excel wants to know the "right-most" limit
of the bins, not the number of bins :-(. Again, this is
controversial. What you chose can dramatically alter the
shape of the histogram and, hence, the interpretation of it.

Generally, consider the range of your data, divide by the
number of bins you want, then set the "bins" values to the
highest limit of each bin. You will want to fudge these limits
to ensure that all data fits into some bin.

People will point to free tools that automate this process.
Bewa since the decisions are controversial, you might
not agree with the implementer's choice.