CHRISTINA VILLALON, PHD
Templates for Writing Statistical Analyses
Below you can find some general APA templates for ways to write up interpretations from various statistical analyses.
Types of Statistical Analysis
Correlation
Using Excel to calculate a Bivariate Correlation
Using SPSS to calculate a Bivariate Correlation
​
Bivariate
Correlation coefficients were computed to examine the relationship(s) between __________________and _____________________. The results of the corrrelational analyses (see Table __), show that ____ of ____ were statistically significant at p < _____, and were of a magnitude greater than or equal to _____. The correlation between _______ and _____ was ________ (direction) and _________ (magnitude). Therefore, as _________ increased/decreased, ________ increased/decreased.
Regression
Using Excel to calculate a Regression
Using SPSS to calculate a Regression
​
Linear Regression
A linear regression analysis was conducted to determine the prediction of _____________ on _____________. The regression equation for predicting (dependent variable) was/was not significant, accounted for ____% variance, and is presented below:
y = ______ + ____X
The 95% CI for the slope _____ to _____ does/does not contain the value of zero; therefore, the reading score is/is not significantly related to their _____________. On average, _____________ increased/decreased _______ for every ______ increase in ____________.
​
Multiple Regression
A multiple regression analysis was conducted to assess how well _____, ______, and _____ predicted individuals' ___________. The linear combination of these predictors resulted in a significant/not significant prediction model for _____________ (F(__,__) = ___, p< ___). The sample multiple correlation coefficient was ___, which indicates approximately ___% of the variance of the _______ in the sample can be accounted for by the linear combination of these (number of) predictor variables. _________ was not a significant predictor of _____________ on ______________ (b= _____, p = ____). However, __________ (b= _____, p = ____) and ______________ (b= _____, p = ____) were found to be significant (direction) predictors of _________________________. Thus, holding ______ and _______ constant, ____________ predicted, on average, to increase/decrease ________ for every _______ increase on __________. Meanwhile, ___________ is predicted, on average, to increase/decrease for every __________ increase on ___________.
In table ___, the bivariate and partial correlations of the (number of predictors) with the _________ are presented. The correlations are/are not all significant (p < ______), and in the expected directions. The partial correlations for ________ and ________ are ______ and ______, respectively; therefore, each uniquely shares approximately _______% variance with ________.
​
Multiple Stepwise Regression
...
t-test
Using Excel to calculate an independent t-test
Using Excel to calculate a dependent/paired samples/repeated measures t-test
Using SPSS to calculate an independent t-test
Using SPSS to calculate a dependent/paired samples/repeated measures t-test
​
Independent
An independent t-test was conducted to evaluate the hypothesis that _________ would be significantly greater/less than ___________. The alpha level was set to ________. The test was/was not significant, t(___) = ____, p = ______. This result supports that __________ did/did not have a significant effect on _______________. The effect size (d = _____) also supported that there was/was not a meaningful effect from ________________ with respect to ______________.
​
Dependent
A __________-samples t-test was conducted to evaluate the hypothesis that _________ would be significantly greater/less after ___________. The alpha level was set to ________. The test was/was not significant, t(___) = ____, p = ______. This result supports that __________ did/did not have a significant effect on _______________. The effect size (d = _____) also supported that there was/was not a meaningful effect from ________________ with respect to ______________.
ANOVA
Using Excel to calculate an ANOVA
Using SPSS to calculate an ANOVA
​
One-way ANOVA
A one-way analysis of variance (ANOVA) was conducted to evaluate the relationship between ___________ and ___________. The independent variable (______________) included __#__ levels/groups: ___(insert names of groups)___. The dependent variable was the ____________________. The alpha level for the ANOVA was set, a priori at .05. The omnibus F test was significant, F(__,__) = ____, p = _____. Based upon the Levene’s test yielding a _________ result (p = _______), the Brown-Forsythe statistic was also conducted, and found to be ____________ (p = _______). The strength of the relationship between the _________________ (treatment) and ____________ as assessed by n2, was ___(strength)___, with the _____________ (grouping) accounting for ______% of the _____________ (dependent variable).
Follow-up tests were conducted to evaluate the pairwise differences among the group means. Dunnett’s C test was conducted to assess pairwise group mean differences, since Dunett’s C does not assume homogeneous variances. There were/were not significant pairwise group mean differences in ____________. [If significant] The _____ group had significantly more/less ______, on average, than the ____ group (95% CI(Group 1) [ _____ , _______], 95% CI(Group 2) [ _____ , _______]). The _____ group had significantly more/less ______, on average, than the ____ group (95% CI(Group 1) [ _____ , _______], 95% CI(Group 3) [ _____ , _______]). The _____ group had significantly more/less ______, on average, than the ____ group (95% CI(Group 2) [ _____ , _______], 95% CI(Group 3) [ _____ , _______]). Therefore, the ______________ resulted in _________ than ______________; however, __________ resulted in the best/worst/most/least _________.