For a given set of data, what statistical test should I use to see if those data are related?

Question

And please provide the formula/links for the statistical test

It is just a bunch of data taken against time, and I want to know if those numbers just occur randomly, or they have some kind of relationship

But we don't know the true mean for those numbers

Answer 1

You can simply plot the data against time and see if any pattern emerges. You can try time and log time for a starter. If there are different parameters in the data that you collected, use different color points. for example, if I collect the data of people's heights and weights over time of a few years, I will find that the height reaches a constant value and the weight keeps increasing (in an obese society).

Pl. feel free to discuss your data further here or through email.

Answer 2

Test 71: The Rank Correlation Test For Randomness of a Sample

Method:

Order your observations in increasing order of magnitude.

The correlation between that order and the observed order is tested using Spearman Rank Correlation Test (#58 http://en.wikipedia.org/wiki/Spearman's_rank_correlation_coefficient).

So if your data is like
Obs: 1,2,3,4,5,6,7
Data: 9,1,2,6,7,5,2

Sorted (based on data):
Obs:2,3,7,6,5,1
Data:1,2,2,5,6,7,9

You are comparing

1,2,3,4,5,6,7
to
2,3,7,6,5,1

By way of the Spearman Rank Correlation (see link)

Answer 3

Well, it really depends on the categories of data, how many variables you are comparing and how many factors are there to comare them. You have to look at the whole set of data in order to see what test. It is hard to tell what kind of tests though. You may have to google "statistical test" to look at formulas but make sure it is the fomula for what your actually testing for.

Answer 4

you can use Chi squared statistical test to determine if there is a correlation.

Or you can also use Pearsons R.

Either one will work to determine if there is a correlation in your set of data.