Statistics Info – Psychology Educational Assistants

# Psychology Educational Assistants

Statistics Review

Note: These instructions describe how to run basic statistical analyzes on SPSS that are particularly utilized in psychology courses/research. Therefore, these instructions simply involve generating basic output. Please note that statistical tests require that certain assumptions are met, which are not described below.

Independent Samples T-test

A t-test compares a typical outcome for one group with what is typical for another group. Means are found for each group separately. Then these means are compared taking standard deviation into account.

In a t-test, the independent variable is nominal (with a maximum of two levels) and the dependent variable is interval/ratio.

There are two main types of t-tests—independent and dependent. In an independent t-test, groups are mutually exclusive. This means that participants are either in one group or the other (they aren’t in both). For example, you would use an independent samples t-test if you were comparing two groups (males and females) on their life satisfaction. Participants are either in the male or female group.

Procedural Steps on SPSS

Analyze –> Compare Means –> Independent Samples t-test

Test Variable = DV

Grouping Variable = IV

Paired Samples T-test

Dependent samples t-test most commonly refers to a paired samples t-test. In a paired samples t-test, the same participants are in both groups. For example, suppose you followed a group of participants and collected “before” and “after” data in relation to an intervention. If you were going to compare how they performed on one variable (i.e. working memory performance) before and after an intervention, you would use a repeated measures t-test.

Procedural Steps on SPSS

Analyze –> Compare Means –>Paired samples t-test

Unlike the independent samples t-test, the paired samples t-test doesn’t utilize a grouping variable. Therefore, put each group’s outcome on the DV in the separate variable columns.

I.E. if I collected reaction time scores on the same participants as a pre and post test, I would put all the pre-test reaction time scores in one column in SPSS, and this column goes under “Variable 1.” The post-test reaction times scores would go in another column in SPSS, and this column goes under “Variable 2.”

One-Way ANOVA:

A one-way ANOVA is used for determining if significant differences in mean scores (of a dependent variable) exist across 2 or more groups. Basically, it compares variation between groups with variation within groups. In performing an ANOVA, we are therefore asking, ‘Do groups differ significantly from each other?’

A one-way ANOVA will tell if you if significant differences exist between groups. However, it won’t tell you where these differences occur. That’s why we use post hoc tests. They help determine where these differences exist.

In a one-way ANOVA, the independent variable is nominal and the dependent variable is interval/ratio. A one-way ANOVA only has ONE independent variable. This variable defines the groups that are being compared.

Generally, you would only use an ANOVA if your independent variable has 3 or more levels. If the independent variable only had 2 levels, the ANOVA would just be doing the same thing as a t-test.

An ANOVA will perform an F-test. The “F” is simply the between groups variation divided by the within groups variation. Therefore, if it is large, it means there is more variability between groups than within groups. This would lead to a significant difference. If F is small, it means there is about the same (or less) variability between groups than there is within groups. This would lead to no significant differences.

Overall, a significant finding means that there are some differences between your groups and that the differences between groups are larger than differences within groups. A non-significant finding means that the groups don’t really differ from each other.

Procedural Steps on SPSS:

Analyze –> General Linear Model–> Univariate

Dependent variable = DV

Fixed Factor = IV

To get graph: “Plots”– > IV goes on Horizontal Axis –> Click “Add” –> “Okay”

Post Hoc: The overall ANOVA will tell if you if significant differences exist. Post hoc tests will tell you between which groups do these significant differences exist.

Options: It’s helpful to get descriptive statistics at a minimum.

Factorial ANOVA

A Factorial ANOVA is used when you have two or more independent variables. A factorial ANOVA can examine if there is an interaction between your independent variables, or if the outcome on the dependent variable for one independent variable depends on the value of the other independent variable. Factorial ANOVAs can be reported by a _ x_ format, where each _ signifies an independent variable. For example, a 2 x 2 ANOVA has 2 independent variables. The numbers in each _ (in this case, both are 2), refer to the number of levels in that respective independent variable. So in this example, both of our independent variables have 2 levels each.

Procedural Steps on SPSS:

Analyze –> General Linear Model –> Univariate

Dependent variable = DV

Fixed Factor = All of your IVs

To get graph: “Plots”– > one IV goes on “Horizontal Axis”, the other goes on “Separate Lines” à Click “Add” à “Okay”

Post Hoc: The overall ANOVA will tell if you if significant differences exist. Post hoc tests will tell you between which groups do these significant differences exist.

Options: It’s helpful to get descriptive statistics at a minimum.

Repeated Measures ANOVA

Repeated Measures ANOVA is used when the same participants are measured on each independent variable. This type of ANOVA accounts for the decrease in within groups variation that will naturally occur when participants are the same between independent variables.

Procedural Steps on SPSS:

Analyze –> General Linear Model–> Repeated Measures

Within-Subjects Factor Name: IV

Number of levels: number of levels of your IV

–> Define

Within-Subjects Variables

Move IVs over to each respective _?_(1), _?_(2), etc.

To get graph: “Plots”– > IV goes on Horizontal Axis –> Click “Add” –> “Okay”

Post Hoc is not applicable for repeated measures ANOVA. That’s why you’re not able to access it. However, you can still get some post hoc analyses by going to “Options” à “Display Means for” (move over IV),” –> Click “Compare Main effects” –> Click on the appropriate Confidence interval adjustment

It’s also helpful to get descriptive statistics.

Chi Square

Chi Square test of independence is used when both of your variables are categorical (nominal).

Procedural Steps on SPSS:

Analyze –> Descriptive Stats –>Crosstabs

Put one variable under “Rows”

Put the other variable under “Columns”

“Statistics”–> Click on “Chi Square”

“Cells” –> “Percentages” –> at least “Row”

–> Okay

Correlations (see handouts below)  Reference sources:

Dr. Amel’s Psych 212 class notes

Morling, B. (2015). Research Methods in Psychology: Evaluating a World of Information (2nd ed.). S.L. Snavely (Ed.). New York: W.W. Norton & Company, Inc.