SPSS

For Students in PSY 280, PSY 290 and PSY 40X
Download the full document (PDF, 3MB) or choose the individual section you are interested in from the listing below.

Creating a data file

Provides step by step instructions on creating a SPSS data file. (PDF, 68KM)

When study has one independent variable with two and only two levels, (and the subjects at each level were randomly assigned to that level) and the dependent variable is measured by an interval or ratio scale (PDF, 63KB)

Repeated measures (matched groups) t test

When study has one independent variable with two and only two levels (and the subjects at each level are the same people or, if not the same people, have been in some way linked or paired with one another) and the dependent variable is measured by an interval or ratio scale. (PDF, 104KB)

Pearson's Correlation

Determine if a relationship exists between two continuous variables (variables that are measured by either an interval or ratio scale). Both continuous variables might be dependent variables, one might be conceived of as an independent variable and the other as a dependent variable, or both might be conceived of as independent variables in the study. (PDF, 100KB)

Spearman Rank Order Correlation

Determine if a relationship exists between two sets of data if one or both are in the form of ranks (ordinal scale data) (PDF, 97KB)

Simple regression

Evaluate how well a single independent (predictor) variable predicts a response on a dependent (predicted or criterion) variable. (PDF, 112KB)

Partial correlation

Determine whether two variables are linearly related when the effects of a third (or more) control variables are held constant. (PDF, 100KB)

Multiple regression - standard and stepwise

To assess the relationship between a dependent (predicted) variable and several independent (predictor) variables. (PDF, 125KB)

One-way analysis of variance

When study has one independent variable with three or more levels (and the subjects at each level were randomly assigned to that level) and the dependent variable is measured by an interval or ratio scale. (PDF, 113KB)

Repeated measures one-way analysis of variance

(Also known as treatment-by-subjects design.) When study has one independent variable with three or more levels and the same or matched subjects have been used in each level of the independent variable and the dependent variable is measured by an interval or ratio scale. (PDF, 119KB)

Two-way analysis of variance

When study has two independent variables (regardless of the number of levels of each independent variable), when different subjects have been randomly assigned to the various conditions of the experiment and when the dependent variable is measured by an interval or ratio scale. (PDF, 121KB)

Repeated measures two-way analysis of variance

(Also known as a treatment-by-treatment-by subjects design.) To be used when study has two independent variables (regardless of the number of levels of each independent variable), and when the same subjects are in each condition of the entire experiment and the dependent variable is measured by an interval or ratio scale. (PDF, 137KB)

Two-factor mixed design: repeated measures on one factor

To be used when study has two independent variables (regardless of number of levels) where for one level of one of the independent variables the same subjects are in all levels of the second independent variable. Responses on the dependent variable are measured by either an interval or ratio scale. (PDF, 130KB)

Chi-square analysis

To be used when you wish to determine if a relationship exists between two discrete variables (i.e., between an independent variable with different levels and a dependent variable measured on a nominal scale). See also: Alternative Chi-Square (PDF, 106KB) for large data file and raw data; Chi-square for goodness (PDF, 100KB) of Fit (PDF, 108KB).

Mann-Whitney U

Similar to an independent groups t-test, to be used when study has one independent variable with two levels and a dependent variable measured on an ordinal scale. (PDF, 108KB)

Wilcoxon Matched Pairs Signed Rank Test

Similar to a repeated measures (matched pairs) t-test, to be used when study has one independent variable with two levels and a dependent variable measured on an ordinal scale. The same (or matched) subjects in the study participate in both conditions of the study. (PDF, 102KB)

Kruskal-Wallis H Test

Similar to a oneway between subjects ANOVA, to be used when study has one independent variable with three or more levels and a dependent variable measured on an ordinal scale. (PDF, 110KB)

Friedman Test

Similar to a one-way repeated measures ANOVA, to be used when study has one independent variable with three or more levels and the same subjects are repeatedly measured and the dependent variable is measured on an ordinal scale. (PDF, 110KB)

Cronbach Alpha

To be used in attitude scale construction and determine whether the items in a scale "hang" together and reliably measure the same construct. In order for the global scale to have good reliability, the Cronbach Alpha should minimally have a value of 0.7. (PDF, 129KB)

Graphics

Creating line graphs or bar charts
Independent groups t- test (PDF, 63KB)
Matched t - test (PDF, 60KB)
Pearson Correlation (PDF, 53KB)
Spearman Correlation (PDF, 53KB)
One-way Anova (PDF, 63KB)
One-way repeated Anova (PDF, 60KB)
Two-way Anova (PDF, 56KB)
Mixed two-way Anova (PDF, 81KB)

Psychology