# Thread: One sample study design - Multivariate methods?

1. ## One sample study design - Multivariate methods?

Good Day,

I am planning a one sample study of 230 patients. I am thinking of using the one sample t-test and the binomial test to perform univariate analysis of the continuous and dichotomous variables, against a dichotomous outcome (death following surgery). I am considering using a binary logistic regression to test significant variables from the univariate analysis. Would it be possible to do this with a one sample study or is there another method which is used?

Thank you.

2. ## Re: One sample study design - Multivariate methods?

I am planning a one sample study of 230 patients. I am thinking of using the one sample t-test and the binomial test to perform univariate analysis of the continuous and dichotomous variables, against a dichotomous outcome (death following surgery).
I suppose that you want to perform two-sample tests,
in order to compare deceased versus undeceased with
respect to your continuous (independent samples t-test,
or Mann-Whitney U-test) or dichotomous (Chi² test)
variables.

With kind regards

K.

3. ## Re: One sample study design - Multivariate methods?

As Karabiner assumed, you are just looking to examine variables related to a binary outcome, correct? If so, his suggestions should help.

So you may have a "cohort". Is your data, retrospective, cross--sectional, or prospective?

4. ## Re: One sample study design - Multivariate methods?

Originally Posted by Karabiner
I suppose that you want to perform two-sample tests,
in order to compare deceased versus undeceased with
respect to your continuous (independent samples t-test,
or Mann-Whitney U-test) or dichotomous (Chi² test)
variables.

With kind regards

K.
Thanks Karabiner. I was forced to go with a one sample study due to funding constraints. I want to compare my results with that of a "control" population already described in the literature. I know there are one sample versions of the t-test and chi-square tests, which I consider univariate tests. But is there any method I can use to determine independent predictors of death in the one sample study or is this not possible for the one sample study design I am planning?

5. ## Re: One sample study design - Multivariate methods?

Originally Posted by hlsmith
As Karabiner assumed, you are just looking to examine variables related to a binary outcome, correct? If so, his suggestions should help.

So you may have a "cohort". Is your data, retrospective, cross--sectional, or prospective?

Thanks hlsmith. The data will be prospectively collected. I am trying to identify independent predictors of death in my study, but I wasn't sure if a one sample design would lend itself to a logistic regression model.

6. ## Re: One sample study design - Multivariate methods?

I want to compare my results with that of a "control" population already described in the literature. I know there are one sample versions of the t-test and chi-square tests, which I consider univariate tests. But is there any method I can use to determine independent predictors of death in the one sample study or is this not possible for the one sample study design I am planning?
You actually seem to have 2 different studies. 1) Compare your sample
data to a population (using one-sample tests), and 2) Find out
predictors of death versus survival, using your sample data.
The Mann-Whitney test referred to the latter; you have 2 (sub-)samples,
dead versus surviving, and you can compare them with respect to
predictor variables, using U-test, independent samples t-tests,
and Chi² test of association. You can also use models with multiple
predictors, such as the multiple logistic regression you mentioned.

With kind regards

K.

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts