Suppose I am doing a Case control study. Lets say Group 1 is a clinical population (N=30), Group 2 is a healthy control population (N=30). I have measured various variables (Continuous data) in both the groups, and using t-test I have found the difference between the two groups. Now, suppose I want to find relationship between the two variables, can both the groups be clubbed together (N=60), or do I do separate correlation analysis for each group?
For Example: If "satisfaction with life" and "quality of life" are research variables in two groups, specifically Patients with anxiety vs Healthy control. I can get continuous data for both these variables using a questionnaire, and I can do a t-test and establish if there is a difference in satisfaction with life and quality of life between these two groups. Now, if I want to know the association between satisfaction with life and quality of life, can I club both patient group and healthy control group together? If yes, is it applicable always, or, are there some conditions? Please explain as my research question is different, and I have just given an example here.
No need to clubbed the data. find the correlation between the variables not the groups. For example in group one variables height and weight and similarly in group two. so we will apply the correlation analysis between the variables height and weight. we will check how much correlated height and weight to each other.
if the data is normal then use Pearson correlation otherwise Spearsman. After getting the correlation value ( r or rho) then simply check in which group the variables are much correlated ( in control or case group).
No need to clubbed the data. find the correlation between the variables not the groups. For example in group one variables height and weight and similarly in group two. so we will apply the correlation analysis between the variables height and weight. we will check how much correlated height and weight to each other.
if the data is normal then use Pearson correlation otherwise Spearsman. After getting the correlation value ( r or rho) then simply check in which group the variables are much correlated ( in control or case group).
Yes, you can. However, if the correlation between the two variables is different in one group compared to the other, then your calculated correlation coefficient may not be accurate.
Depends on the scientific question you want to study. Is it the relationship between those variables in general - then combine all the data, if you want to test to see if the strength of the relationship between X and Y differs between those two groups - then separate out? I would make a scatterplot of all 60 subjects, then also make a scatterplot that shows groups - different symbols or colors and a trend line for each group.
You could even do a regression analysis of some sort: QoL = Group + Satisfaction. Here you would see the relationship of Satifcation and QoL (or vice versa) while controlling for Group.
Dear Sahithya,
Firstly, we do correlation to find out how closely the two variables are related. So, a correlation coefficient necessarily show us the degree of association between two "Variables".
Now in your example, if you do correlation study, you will find the degree of association between Var1 (sayisfaction with life) and Var 2 (quality of life). If you combine response of both case and control group in respect of var1 and var2, you ll find association between these 2 variables. But what would that result mean to you? I think nothing really.
Rather,if you run correlation i.r.o. the said 2 variables, separately for case and control group and find 2 different r value, you ll be able to compare the degree of association between 2 groups. I guess that rather means something.
Best,
Praheli
I suppose
For Correlation,
There may be Two Outcome ,
For Individual Group,
You can get correlation between satisfaction and quality of life .It represent only for each group.
If you mixed two group,
This will represent generalization for whole population if your sample is randomizied and followed normal distribution. You need to test your sample is followed for Normality Test .
In a case control study design, the focus will be on the variables.
Yes, you can - if your focus is on the correlation only, and not the differences between cases and controls. However, you may need to verify your assumption that the cases and controls behaves the same way with respect to the variables under investigation!
Before selecting the proper statistical test, you need to check the type of the variables in your study, and as you mentioned, your variables are continuous then the proper Statistical test is Simple Correlation Coefficient, assuming that its normally distributed. After that, see the the directions whether its positive or negative and also the strength of the association.
The other point here is that, it's not necessary to be clubbed the cases and controls. Just drawing the scatterplots (plots will represent the cases and controls here) will give you the final decision whether it's direct, indirect or no association and also see the strength, whether its Perfect, Strong, Moderate or weak.
No, it should not be clubbed/ added (together). Again, correlation analysis may not also be required. What is perhaps expected to study, is, effect/ impact... so the significance difference tests should suffice.
- Badges
- Science topic