## What Test?!

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

liv965
Posts: 2
Joined: Wed Feb 24, 2016 11:05 am

### What Test?!

I need to know if there's a significant difference between the number of cases I have each year.
I was thinking of doing a linear regression with year as the dependent variable - but what would I use for the independent.

GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: What Test?!

how are you cases set up? do you have a unique id number for each person, indicating whether or not they were prestent at t1 and t2? or are they completely different subjects at t1 and t2?
liv965
Posts: 2
Joined: Wed Feb 24, 2016 11:05 am

### Re: What Test?!

Each case does have a unique identifier. But I need to know how many diagnoses there were each year so its new cases each year.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: What Test?!

It is difficult to say because it is unclear what the population here is that you base your test on, and what your data look like.
I mean: is 4 more than 3? yes. but is it significant? well, then it depends on the population you drew your 4 and 3 from, and how you determined what "4" is.
It looks like you have only 2 observations, namely number of diagnoses at T1 and T2. You cannot infer anything from 2 observations.
Or maybe you can, and you have more than 2 observations. for instance, you have 100 hospitals, each have a number of diagnosis in year 1 and year 2, and you want to see if the average number of observations differs.
But to know whether or not you can actually do anything with statistics, we need to know more about your data.

### Who is online

Users browsing this forum: No registered users and 3 guests