# Confused about type of analysis between categorical variable (degree of severity of accident) and continuous variable (number of car passengers)

Good afternoon, As indicated in the title, for a project I am interested in finding any (if at all) relation between the number of passengers in a car and the severity of an accident.

I am using JMP software and I am not great at the program by any means, but still I know a fair amount about certain analyses, but this one has me stumped.

As of yet, I have ran two different tests via JMP. The first was a simple Y by X distribution, which yielded me these results:

The second, which honestly was more of a guess, was using fit model Y by Construct, which yielded me this:

Now, when looking at the distributions, I can pretty much see that the overwhelming majority of all degrees of severity were with 0 passengers.

But for my second analysis, I don't quite understand the output, I don't know if it's the right test, and most importantly whether it is even necessary. It's for a project so I want to be thorough but I'm at wits end.

Any help would be appreciated! -Sam

Create a 3 x 3 table like following:

                      Number of Passengers
----------------------------------
Severity    0         1          2     total
---------------------------------------------
Fatal
Serious
Slight
Total
---------------------------------------------


Then perform a Pearson Chi-Square test or Fisher's Exact test to find any (if at all) relation between the number of passengers in a car and the severity of an accident.