Summarizing the Results in Tables
There are several ways to summarize the results of a multiple regression analysis in a paper. Summarized results of the present analyses follow, using table formats that are fairly representative of those appearing in journals of the American Psychological Association (APA, 2001).
The Results of PROC CORR
It is usually desirable to present simple descriptive statistics, such as means and standard deviations, along with the correlation matrix. If reliability estimates (such as Cronbach’s alpha) are available for the predictors, they can be included on the diagonal of the correlation matrix, in parentheses. The results for the present fictitious study are summarized in Table 14.8:
| Intercorrelations | ||||||||
|---|---|---|---|---|---|---|---|---|
| Variable | M | SD | 1 | 2 | 3 | 4 | 5 | |
| 1. | Commitment | 27.71 | 10.20 | (.84) | ||||
| 2. | Rewards | 26.65 | 5.05 | .58[**] | (.75) | |||
| 3. | Costs | 16.58 | 5.50 | .-25 | .-45[*] | (.72) | ||
| 4. | Investment size | 25.33 | 6.08 | .61[**] | .57[**] | .02 | (.83) | |
| 5. | Alternative value | 16.60 | 7.50 | .-72[**] | .-47[**] | .27 | -.45[*] | (.91) |
| Note. N = 48. Reliability estimates (Cronbach's alpha) appear on the diagonal above correlation coefficients. | ||||||||
[**] p < .001
[*] p < .01.,
The Results of PROC REG
Depending on the nature of the research problem, it is often feasible to report standardized regression coefficients (beta weights) and the uniqueness indices in a single table as done in Table 14.9:
| Predictor | B | SE B | β | t | Uniqueness Index |
|---|---|---|---|---|---|
| Rewards | .28 | .27 | 0.14 | 1.02 | .01 |
| Costs | -.11 | .21 | -0.06 | -0.51 | .01 |
| Investment size | .52 | .21 | 0.31 | 2.49[*] | .05[*] |
| Alternative value | -.68 | .15 | -0.50 | -4.64[**] | .18[**] |
| Note. R 2 = .65 (F[4,43] = 19.59), p < .01 for predictor variables. | |||||
[*] p < .05
[**] p < .001