Create Variables 1. In SPSS, click on the Variable View tab in the lower left, or View Variables, or ctrl-t 2. Create four variables (one for the pretest, one for the IV and one for each of the DVs) by typing the following information into the appropriate rows and columns of the variable view: Row Name Label Values Measure 1 mnemonic Mnemonic Technique 1 = Elaborative Rehearsal nominal 2 = Mental Imagery 3 = Peg Method 2 pretest List 1 Recall scale 3 lenient List 2 Lenient Recall scale 4 strict List 2 Strict Recall scale To enter the values, click in the Values column of the desired row. Click on the ellipsis button that appears in the cell (circled in red below) Enter the numeric value and label and then click the Add button. Repeat until all the value labels for that variable have been entered. Click OK. The variable view should look like this:
Enter the Data 3. Switch to the data view by clicking on the Data View tab in the lower left corner or clicking View Data or ctrl-t or alt-v, A 4. Value labels may or may not be turned on. If turned on, when you type 1 in the condition column, the label for the value 1 for the condition variable will be displayed instead of the number 1. I usually turn on the value labels option. Click View. If there is a check mark next to Value Labels on the drop down menu, then value labels will be displayed. If desired, toggle the value labels on or off by clicking in the check box next to Value Labels:
5. As we go around the class getting each person s data, type it in the appropriate row and column. All of the data from a given person goes on a single row. The following shows some data from a previous semester. Analyze the Data 6. Determine the appropriate statistical test: The DVs are ratio scaled. The IV has three levels. Because each participant participated in one, and only one condition, this is an independent samples design. This would normally be an independent samples Analysis of Variance (ANOVA). However, we also want to statistically control for the person s overall memory ability. Thus, we will use Analysis of Covariance (ANCOVA). 7. Write the hypotheses: H 0 : μ Elaborative Rehearsal =μ Mental Imagery =μ Peg Method H 1 : not H 0 Where μ is the mean number of words recalled in order after being adjusted for the person s overall ability to recall. We will also repeat the analysis where μ is the mean number of words recalled in any order after being adjusted for the person s overall ability to recall. 8. Do the t-test: Analyze General Linear Model Univariate
9. Select the relevant DV (List 2 Strict Recall) and move it into the Dependent Variable box by clicking on the upper arrow button. 10. Select the IV (Mnemonic Technique) and move it into the Fixed Factor(s) box by clicking on the next arrow button. 11. Select the covariate (List 1 Recall) and move it into the Covariate(s) box by clicking on the arrow next to that box. 12. Click the Options button. Move the IV (mnemonic) to the Display Means for box by clicking on the arrow button. 13. Select Descriptive statistics, Estimates of effect size, and Observed power.
14. Click Continue 15. Click OK Interpret the Output 16. Remember that the data in this example are from a previous semester, so the values in the following output will not be the same as in your output. The Descriptive Statistics part of the output gives us the descriptive statistics of the scores before controlling for the covariate: Descriptive Statistics Dependent Variable:List 2 Strict Recall Mnemonic Technique Mean Std. Deviation N Elaborative Rehearsal 7.0000 3.20713 8 Mental Imagery 8.7500 5.75078 8 Peg Method 12.1250 6.19764 8 Total 9.2917 5.44122 24 The mean number of words recalled in the correct order in the elaborative rehearsal condition is 7.00 with a standard deviation of 3.21 words. The mean number of words recalled in the correct order in the mental imagery condition is 8.75 with a standard deviation of 5.75 words.
17. The next part of the output is the ANCOVA summary table. Dependent Variable:List 2 Strict Recall Tests of Between-Subjects Effects Type III Sum Mean Partial Eta Noncent. Observed Source of Squares df Square F Sig. Squared Parameter Power b Corrected Model 192.819 a 3 64.273 2.633.078.283 7.900.555 Intercept.558 1.558.023.881.001.023.052 pretest 84.236 1 84.236 3.451.078.147 3.451.424 mnemonic 68.770 2 34.385 1.409.268.123 2.818.267 Error 488.139 20 24.407 Total 2753.000 24 Corrected Total 680.958 23 a. R Squared =.283 (Adjusted R Squared =.176) b. Computed using alpha =.05 18. This output will allow us to test the null hypothesis stated in step 7 that the means (adjusted for the covariate) of the ordered recall are equal. Find the row with the independent variable and the column labeled Sig. The value at the intersection of that row and column is the p value. In this case, p =.078. If p α, we should reject H 0. In this case, p > α so we conclude that there is insufficient evidence to suggest that the type of mnemonic had an effect on ordered recall. 19. We would write the results as: The hypothesis states that the ordered recall with the peg method (M = 12.13) should be better than with the mental imagery mnemonic (M = 8.75)which should be better than with the elaborative rehearsal mnemonic (M = 7.00). An analysis of covariance with performance on list one as the covariate failed to reveal an effect of the type of mnemonic on ordered recall, F(2, 24) = 1.409, MS error = 24.407, η 2 =.123, p =.268, α =.05. This is not consistent with the hypothesis. 20. Repeat for the comprehension DV.
Graphing the Results 21. SPSS will not easily produce an APA quality graph. Download the template from the class website at <http://academic.udayton.edu/gregelvers/psy217/labs/lab07/bargraphwithci.xlsx> 22. Find the SPSS output labeled Estimated Marginal Means: Dependent Variable:List 2 Strict Recall Mnemonic Technique 95% Confidence Interval Mnemonic Technique Mean Std. Error Lower Bound Upper Bound Elaborative Rehearsal 7.568 a 1.773 3.870 11.267 Mental Imagery 8.651 a 1.747 5.006 12.296 Peg Method 11.655 a 1.765 7.974 15.337 a. Covariates appearing in the model are evaluated at the following values: List 1 Recall = 14.4583. 23. Copy the three means from the table (e.g. 7.568, 8.651, and 11.655) and enter them in cells B2, B3 and B4 in the Excel worksheet. 24. Copy the three lower bounds from the table (e.g. 3.870, 5.006, and 7.974) and enter them in cells C2, C3 and C4 in the Excel worksheet.