Right statistical test for frequency distribution

Say that after getting data from 5 different years on causes of teenage pregnancy you want to confirm that sexual assault is a common factor contributing to teenage pregnancy.

Dividing the data into two groups 'Sexual assault' and 'Other causes' respectively from the 5 different years data and using their frequencies how can you arrive at the conclusion whether sexual assault is the most common factor contributing to teenage pregnancy or not?

• "Most common" suggests that you are considering other factors. Is that correct?
– Dave
Commented Sep 20, 2021 at 16:05
• How would such data distinguish a true contributing factor from a factor that is only indirectly associated with the outcome?
– whuber
Commented Sep 20, 2021 at 16:15
• @Dave nope. Basically I just use that technique to sort of make an inference. Commented Sep 20, 2021 at 16:33
• @whuba pls suggest the best thing I can do. Commented Sep 20, 2021 at 16:34
• Does your data look like this: year, # of pregnant teenagers, # of pregnancies due to sexual assaults? Commented Sep 21, 2021 at 16:14

In general, when you don't get an answer on this site, it is mainly because people are struggling to understand the question... while you might have a valid point, which, I think, is the case here.

Most of the time, providing data sample contributes a lot to understand better your challenge. In your comment, you indicated that the data looks like this:

year # of pregnant teenagers # of pregnancies due to sexual assaults
2020 500 400
2019 450 250
... ... ...

You indicated that you want to make inference, so I assume that you have only a random sample that represents the population that you want to study. In that case, you might be interested in defining a confidence interval to make inference on the true population proportion. You might also want to test your sample against an assumption you have about the true population parameter. If the data looks like the above, then the factor contributing to pregnancies is already identified since we are talking about "# of pregnancies due to sexual assaults", but you also mentioned "factor contributing to", which might indicate that you are looking for some type of correlation.

Since you answered my comment, I understand this is not an orphan question and therefore I encourage you to edit it and update it with clarifications. You can learn more about how this site works on the help center. I hope it helps!

• The original question is this "sexual assault a common factor contributing to teenage pregnancies" I just assumed that the way to judge this if it's true or not is by using the data on teenage pregnancy for about 5yrs at least to make an an inference whether it's a common factor or not. In your own view how can that question be solved? Commented Sep 24, 2021 at 11:54