Anda di halaman 1dari 4

SPSS for Windows 15.

0 Two-way ANOVA Non-repeated independent variables In this assignment youll learn how to use SPSS to analyze data from studies where there are two non-repeated independent variables (i.e., each subject only provides data for one level of each independent variable). You will need to know how to test each of the following effects: 1. Interaction between A and B 2. Main effects for A and B 3. Simple Effects for A and B The examples for this assignment will use the data set we talked about in the lab. To access the data file, go to my web site for the course Course Materials Click the link to SPSS data set for two-way ANOVA (both IVs non-repeated). The file should come up automatically in the SPSS spreadsheet. The two independent variables in this study are taskleng (Task Length) and gender. Task Length has two levels (5 minute task = 1; 15 minute task = 2). For Gender, males were coded with values of 1 and females were coded with values of 2. Testing the interaction between two independent variables and the main effects for these two independent variables. The dependent variable in this study is sustatt the percentage of trials in which subjects made errors while performing a sustained attention task).

From the spreadsheet, click Analyze. Click General Linear Model Click Univariate. This should bring up the Univariate window. Select the dependent variable for the analysis. Click sustatt from the variable list at the left side of the window. Click the right arrow key beside the Dependent box. The variable name sustatt should now appear in that box. Select the two independent variables for the analysis. It doesnt matter what order you select the variables in. Lets say you click on the variable name taskleng in the variable list.

Now click the right arrow key beside the Fixed Factor(s) box.

Now lets select the second independent variable, gender. Click the variable name gender in the variable list. Click the right arrow key beside the Fixed Factor(s) box. Youll see the variable name gender in the box.

Click the Options button. Click the open square beside the Descriptive Statistics option to get descriptive statistics for the various groups. Under Factor(s) and Factor Interactions, click the interaction term taskleng*gender. Then click the right arrow beside Display Means For. This will give you a table containing all of the group means. Do the same thing to select taskleng. This will give you the means for the main effect of Task Length. Do the same thing to select gender. This will give you the means for the main effect of gender. Now Click Continue.

Back at the Univariate window, click the Plot button to bring up the Profile Plots window. Put taskleng in the Horizontal Axis box. Put gender in the Separate Lines box. Click Add. This will give you a graph of the interaction.

Click OK after returning to the Univariate window.

In the output window

You first get a table which just tells you how many subjects data were used in the analysis. Next you get a Descriptive Statistics table that gives you the mean and the number of subjects for each group. Next, you get the ANOVA table. There are rows for the two main effects and the interaction between the two independent variables. The interaction gives us an Fratio of 16.166 and a significance level of .000, which reaches significance at the .05 level. The main effect for Task Length has an F-ration of 34.065 and is significant. The main effect of gender has an F-ratio of 30.625 and is also significant.

Testing simple effects using the GLM routine of SPSS for Windows The presence of the significant interaction means that whatever effect of Task Length there is, that effect is not the same for males as it is for females. As you know, a simple effect is the effect of one independent variable at one level of a second independent variable. This, of course, is essentially a one-way ANOVA, but only using the subjects from one level of another independent variable. We need to test the effect of Task Length on sustained attention scores for males and females separately. This means that we need to use the Select Cases option to only make the subjects of one gender available for analysis and then we do a one-way ANOVA with

Task Length as the independent variable and sustained attention scores as the dependent variable. Using the Select Cases option

Lets we want to test the effect of Task Length on sustained attention scores, but only using the data fro males subjects. Remember, males got 1s in the spreadsheet and females got 2s. Minimize the output window so that youre looking at the SPSS spreadsheet. We now need to limit the number of cases that SPSS will use. Click the data pull-down menu. Click Select Cases. Click the If condition is satisfied option. Click the If button. Type gender = 1 in the open box. Click Continue Click Ok. When you look in the spreadsheet, there will now be a slash through the case number of the cases that do not fit the selection conditions. In other words, at this point the only subjects eligible for inclusion in analyses are males. One-way ANOVA

Now, lets perform our One-Way ANOVA. Click Analyze Click Compare Means Click One-Way ANOVA Select sustatt as the dependent variable and taskleng as the independent variable (Factor). Under Options click the descriptives statistics option. Click Continue to return to the One-Way ANOVA window. Click OK

In the output window

You can see from the ANOVA table that the F-ratio is significant (the F ratio is 56.504 and the significance level of .000 is less than .05).

To test the other simple effect in the set Okay, now weve tested the effect of Task Length on sustained attention scores using just the male subjects. What do we have to do to test the effect of Task Length using just the female subject? Of course, just go back to the Select Cases option and change the selection condition from gender = 1 to gender = 2! Then run the one-way ANOVA over again.

When you do this the F-ratio is now only 1.446 and the significance level is .245. The significance level needs to be .05 or lower to be significant, so there is no effect of varying the length of the task on the error rate when only females subjects are used. Text in the results section So, after doing all these analyses, what do I need to write in my results section? For an analysis like this, think of it this way. Whenever you do one test, you learn one thing, and you write one sentence that says what you learned. Weve done a number of different tests, so were going to need to write a number of different sentences. In general, start with the most general pieces of information first. The most general pieces of information are the three overall effects of the interaction and the two main effects. Here we go A significant interaction was observed between Task Length and Gender, F (1, 36) = 16.17, p < .001. A significant main effect of task length was observed, F (1, 36) = 34.07, p < .001. A significant main effect of gender was observed, F (1, 36) = 20.61, p < .001. Okay, because the interaction was significant we needed to report the results from a set of simple effects. Just report them one at a time. A significant simple effect of task length on the percentage of errors on a sustained attention task was observed for male subjects, F (1, 18) = 56.50, p < .001. Male subjects had a significantly higher percentage of trials with error in the 15-minute task condition (M = 5.70, SD = 1.25) than in the 5-minute task condition (M = 1.90, SD = 0.99). No significant simple effect of task length on the percentage of errors was obtained for female subjects, F (1, 18) = 1.45, p = .245. In a results section, these two pieces would go together in one paragraph, as shown below: A significant interaction was observed between Task Length and Gender, F (1, 36) = 16.17, p < .001. A significant main effect of task length was observed, F (1, 36) = 34.07, p < .001. A significant main effect of gender was observed, F (1, 36) = 20.61, p < .001. A significant simple effect of task length on the percentage of errors on a sustained attention task was observed for male subjects, F (1, 18) = 56.50, p < .001. Male subjects had a significantly higher percentage of trials with error in the 15-minute task condition (M = 5.70, SD = 1.25) than in the 5-minute task condition (M = 1.90, SD = 0.99). No significant simple effect of task length on the percentage of errors was obtained for female subjects, F (1, 18) = 1.45, p = .245.