On-line statistics

Key for the selection of a test for correlations, trends or relationships

Are your data to be selected at random from the population you are studying?

Has every data point available got the same chance of being selected and independent from each other?

These two points should be set in stone! If not the test results will be biased in one form or another and the conclusions that you make may be completely irrelevant. For example, by sampling a river bed with a net of mesh size 0.5mm, for animals of between 0.25 and 1.3mm, you would be including bias in your sampling by excluding those individuals smaller than 0.5mm.

Get into the habit of drawing a scatter graph of your data before analyses are carried out. You should be able to judge whether a relationship exists from a simple graph. Linear relationships will have the points scattered in a straight(ish) line. Non-linear relationships that cannot be transformed to linear should be dealt with by regression rather than correlation.

Use the key by answering the questions in the most relevant way. It is advised that you double check through a good reference guide.

1. Is the relationship between the data sets linear?

Yes...go to 2

No....Try transforming, otherwise use non-linear regression

2. Are your data ordinal (categorised and in order)?

No.....go to 3

Yes....Spearman's rank

3. Are your data sets normally distributed or can they be transformed to normal?

No....Spearman's rank

Yes...Pearson's Product moment