Prism can perform correlation analyses from XY, Column, and Multiple Variables data tables. Click the Analyze button and choose correlation.

Compute correlation between which pairs of columns?

Compute the correlation between two specific columns, between all columns (correlation matrix), or between each column and a control data set (which is X, if you are analyzing an XY table).

How to handle missing data?

When selecting to compute r for every pair of Y data sets (correlation matrix), Prism offers an option on what to do when data are missing. By default, the row containing the missing value is only omitted from the calculation of the correlation coefficients for the variable/column containing the missing value. Other values on this row (i.e. values in other variables) are included in calculations for the variables that they belong to.

Prism offers the option to exclude the entire row of values when any value is missing on a row. This ensures that all correlation coefficients are computed from data in the same set of rows.

Assume data are sampled from a Gaussian distribution?

Prism offers two ways to compute correlation coefficients:

•Pearson correlation calculations are based on the assumption that both X and Y values are sampled from populations that follow a Gaussian distribution, at least approximately. With large samples, this assumption is not too important.

•Spearman nonparametric correlation makes no assumption about the distribution of the values, as the calculations are based on ranks, not the actual values.

One- or two-tailed P values?

Prism can compute either a one-tailed or two-tailed P value. We suggest almost always choosing a two-tailed P value. You should only choose a one-tail P value when you have specified the anticipated sign of the correlation coefficient before collecting any data and are willing to attribute any correlation in the “wrong” direction to chance, no matter how striking that correlation is.