East Carolina University
Department of Psychology

APA-Style Presentation of Results

     Summary Statements. When you test hypotheses for this class, I want you to state your conclusion as it would be stated in an APA (American Psychological Association) style journal. Your summary statement should include each of the following:

• who or what the research units were (sometimes called “subjects” or “participants”)
• what the null hypothesis was (implied, not explicitly stated)
• descriptive statistics such as sample sizes, means and standard deviations
• whether or not you rejected the null hypothesis (implied, not explicitly stated)
• if you did reject the null hypothesis, what was the observed direction of the difference between the obtained results and those expected under the null hypothesis
• what test statistic (such as t) was employed
• the degrees of freedom
• if not obtainable from the degrees of freedom, the sample size
• the computed value of the test statistic
• the p value (use SPSS or SAS to get an exact p value)
• an effect size estimate
• and a confidence interval for the effect size parameter.

     For the t-test that we did earlier, here is a APA-style summary statement: The mean math SAT of my undergraduate statistics students (M = 535, SD = 93.4) was significantly greater than the national norm (516), t(113) = 2.147, p = .034, d = .20. A 95% confidence interval for the mean runs from 517 to 552. A 95% confidence interval for d runs from .015 to .386.

     Suppose that the sample mean we obtained was not 534.78 but 532. Our summary statement would read: The mean math SAT of my undergraduate statistics students (M = 532, SD = 93.4) was not significantly different from the national norm (516), t(113) = 1.83, p = .07, d = .17. A 95% confidence interval for the mean runs from 515 to 549. A 95% confidence interval for d runs from -.014 to .356. Note that I did not indicate a direction of difference with this “nonsignificant” result -- to do so would imply that I was testing directional rather than nondirectional hypotheses.

     Suppose that I was testing directional hypotheses, with the alternative hypothesis being that the mean is greater than 516, and the obtained sample mean being 532. Now my summary statement would read: Employing a one-tailed test, the mean math SAT of my undergraduate statistics students (M = 532, SD = 93.4) was significantly greater than the national norm (516), t(113) = 1.83, p = .035, d = .17. A 90% confidence interval for the mean runs from 517 to 547. A 90% confidence interval for d runs from .016 to .326. Notice that I shifted to a 90% confidence interval, because with the one-tailed test I put all of alpha in one tail rather than splitting it into two tails – but confidence intervals are, IMHO, naturally bidirectional, so I put 5% in both tails for the confidence interval. If I did not make this change in the confidence coefficient, the confidence interval would include the null value, which would be in disagreement with the prior decision to reject the null hypothesis.

     Suppose that the mean was only 530, still testing the directional hypotheses. Now my summary statement would read: Employing a one-tailed test, the mean math SAT of my undergraduate statistics students (M = 530, SD = 93.4) was not significantly greater than the national norm (516), t(113) = 1.60, p = .057, g = .15. A 90% confidence interval for the mean runs from 515 to 545. A 90% confidence interval for d runs from -.005 to .305. Even though the result is not “significant,” I use the phrase “not significantly greater than” rather than “not significantly different from” because the test was directional.

Dr. Karl L. Wuensch, January, 2018.