Pearson Correlation and Spearman Correlation

LabWriteUp deals with Spearman and Pearson correlations in similar ways. Once you have selected either of these tests, the above window appears. Begin by entering the level of statistical significance obtained in the test and the number of participants.

You may enter a value (e.g., .034, or whatever value you obtained) or select one of the significance levels, ranging from 'p > .05' to 'p < .001'.

Click the 'Next' button to continue. When you do this the following window appears (Pearson example):

Enter the names of the variables and the nature of the association, choose from:
positive or linear - when the correlation is greater than zero (e.g., .124)
negative or inverse - when the correlation is less than zero (e.g., -.251)

The choice between positive or linear and negative or inverse depends upon how you wish to express the correlation (e.g., "there was a positive correlation between IQ and motivation" or "there was a linear relationship between IQ and motivation" - they both mean the same thing, though!).

When these selections have been made, click 'Save and Preview' to examine the textual description of the analysis. In this example the description is:

For the Pearson test only, LabWriteUp can calculate the variance that both variables share, so that it is possible to state that one of the variables can account for x% of the variance in the other variable:

In this example, IQ is said to account for a percentage of the variance in motivation. However, we could easily state the opposite - that motivation can account for a percentage of the variance in IQ. Which ever way it is expressed depends on a theoretical viewpoint or just common sense (e.g., a correlation between ice-cream sales and outside temperature can be explained by the fact that the sun causes people to eat ice cream, rather than ti gets warmer the more people eat ice-cream!).

Next >> One Way Repeated Measures ANOVA