which statistical test do i use?

i have undergone an experiment and do not know which test to use. i want to see if there is a relationship between age, gender and height with grip strength.

[dpb]

Backwards...design and run the experiment after you develop the null hypothesis and decide upon the appropriate test statistic. Then the experiment will be able to produce data suitable for that test.

[Shanise Bolat]

I have got the data and have developed a null and research hypothesis. i cant decide the appropriate test to use?

Answer 1

[Star Strider]

Doing a study without a statistical design first will almost guarantee a useless result. Please, in the future, give some thought to how you will analyse your data before you do the study! At the very least, you will determine how many observations are needed in order to produce a statistically valid result.

First go to the literature in PubMed to see what previous research has been done and the statistical techniques others have used to analyse their data. (This is not a new problem.)

Second, plot grip strength individually against age, gender, and height. This will give you some idea of how the data are related (linear or nonlinear).

Third, as a preliminary exploration, I would use the regress function. It should provide you with the basic information you need.

Fourth, use the most appropriate statistical techniques in the published literature (that you have discovered in your PubMed search) to complete the analysis of your data. You will also want to compare your results to those.