PSY 380 Topic 6 effect size report example

Introduction to Probability and Statistics Grand Canyon University Free custom sample in 24 to 48h

This PSY 380 Topic 6 sample paper reports the effect size behind a significant result: an online study skills course raised 400 students' biology exam scores by 2 points, t(398) = 2.00, p = .046. It was written for GCU PSY 380, Introduction to Probability and Statistics, catalog code PSY-380, part of Grand Canyon University's BS in Psychology. Cohen's d is 0.20 with an interval of about 0.00 to 0.40, following Lakens and Cumming. Beyond Cohen's labels, the effect is shown as points, 58 percent above the control mean and a 56 percent chance of outscoring, and the course is judged a modest, low-cost addition.

CoursePSY 380 Introduction to Probability and Statistics
TopicTopic 6
Paper typeEffect size report
LengthAbout 990 words, 2 pages
FormatAPA 7 student paper
SchoolGrand Canyon University (GCU)
ProgramBS in Psychology
UpdatedOctober 2026

Free sample paper for PSY 380 Topic 6

1

Significant but Small: Reporting the Effect Size of an Online Study Skills Course on Exam Scores

[Author Name]

College of Humanities and Social Sciences, Grand Canyon University

PSY-380: Introduction to Probability and Statistics

Topic 6 Effect Size Report

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

What this page is doingThe title states the report's central point. The program and data are constructed for coursework; the sources are real.
2

Significant but Small: Reporting the Effect Size of an Online Study Skills Course on Exam Scores

The Finding

A composite university evaluated a four-week online study skills course by randomly assigning 400 first-year students, 200 to take the course and 200 to a waiting list, and comparing their scores on a common final exam in introductory biology, scored out of 100. Students who took the course scored a mean of 74.0 (SD = 10.0); the waiting-list group scored 72.0 (SD = 10.0). An independent-samples t test gave t(398) = 2.00, p = .046. The university's newsletter announced that the course significantly improves exam performance. This report carries two numbers about the same finding, the test result and the effect size, and treats them as answering different questions.

Two Questions

The test result answers one question, whether chance alone could plausibly produce the gap: if the course had no effect, a difference of 2 points or more would occur in under 5 percent of studies this size. It does not say how large the difference is. A difference can be statistically significant because it is large, or because the sample is large enough to detect even a small difference precisely. With 400 students, this study could detect small differences, so significance alone says little about whether the course matters.

Computing the Effect Size

The effect size used here is Cohen's d, the difference between means divided by the pooled standard deviation. With equal group sizes and equal standard deviations of 10.0, the pooled standard deviation is 10.0, so d = (74.0 minus 72.0) / 10.0 = 0.20. Lakens (2013) recommended reporting such standardized effect sizes for t tests, along with the sample sizes they rest on, so that results can be compared and combined across studies. Here, n = 200 per group.

An approximate 95 percent confidence interval for d, using the standard error for the difference between two independent means, runs from about 0.00 to 0.40. Cumming (2014) argued that such intervals should accompany estimates because they show how precisely the effect is known. This interval says the data are consistent with an effect from essentially zero to small-to-moderate.

Beyond the Labels

Cohen (1988) suggested that d = 0.20 be called small, 0.50 medium and 0.80 large, but Cohen offered those benchmarks as rough guidance for when nothing better is available, and their meaning differs across areas of research. A small effect on a life-or-death outcome can matter enormously; a medium effect on a trivial one may not. The label small does not settle whether this course is worthwhile.

The magnitude is more useful in concrete terms. On the exam's own scale, the course raised the average score by 2 points out of 100, about one fifth of a letter grade band if bands are ten points wide. In terms of overlap, with d = 0.20, about 58 percent of students who took the course scored above the average of the waiting-list group, compared with 50 percent expected if there were no effect. Put another way, if one student were picked at random from each group, the student who took the course would score higher about 56 percent of the time, only a little better than a coin flip.

Table 1

The Course Effect Expressed in Several Ways

ExpressionValueMeaning
Raw difference2.0 points out of 100Course group averaged 2 points higher
Cohen's d0.20Difference is one fifth of a standard deviation
95 percent interval for dabout 0.00 to 0.40Effect could be near zero or moderate
Percent of course group above control meanabout 58 percentCompared with 50 percent if no effect
Probability a course student outscores a control studentabout 56 percentSlightly better than chance

Where Significance and Importance Part

The newsletter's claim that the course significantly improves exam performance is correct in the statistical sense and misleading in the ordinary sense, because readers hear significantly as substantially. The data show that the course probably has some positive effect, and that the effect is small and imprecisely estimated. If the same 2-point difference had been found with 20 students per group instead of 200, the test would not have been significant, even though the effect would have been identical.

Why the Interval Matters for the Decision

The confidence interval matters for the decision as much as the point estimate. At its lower end, near zero, the course would do essentially nothing; at its upper end, around 0.40, it would raise scores by about 4 points, a gain many instructors would notice. A decision maker who saw only d = 0.20 might treat the effect as settled. The interval shows that the evidence is compatible with both a negligible and a useful course, which is a reason to collect more data before investing heavily, not a reason to dismiss the course.

What Further Evaluation Would Add

A further evaluation could narrow the interval and test where the course helps most. Doubling the sample would shrink the interval's width by about 30 percent. Recording how much of the course each student completed would show whether those who engaged fully gained more, which an average across all assigned students hides. And measuring outcomes beyond one exam, such as grades across the first year, would show whether the small gain persists or fades.

Should Anything Change?

Whether the university should act depends on costs and alternatives as well as the effect. The course is online and inexpensive to run; if it costs little and takes students only a few hours, a small average gain might be worth offering. But it should not be presented as a major intervention, required of all students in place of other support, or credited with large improvements in pass rates, since a 2-point average gain moves few students across a grade boundary. A practitioner should treat the course as a modest, low-cost addition, and the university should report the effect size and interval alongside the significance test in any future description.

References

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.

Cumming, G. (2014). The new statistics: Why and how. Psychological Science, 25(1), 7-29. https://doi.org/10.1177/0956797613504966

Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, Article 863. https://doi.org/10.3389/fpsyg.2013.00863

PSY 380 Topic 6 instructions, in plain terms

Topic 6 in PSY-380 asks students to report the size of an effect alongside its test result. Strong reports treat the two numbers as answering different questions, compute the effect size with its formula shown and report the sample sizes it rests on. Students should interpret the magnitude beyond conventional small, medium and large labels, expressing it in concrete terms such as points on the scale, overlap between groups or variance explained, and include a confidence interval where possible. The report should explain where statistical significance and practical importance part and end by saying whether a practitioner should change anything. Calling an effect small, medium or large without saying what it means for people earns little credit.

Inside the PSY 380 Topic 6 example

The finished report carries two numbers about the same finding and treats them as answering different questions. The test result establishes that the difference is unlikely to be an accident; the effect size says how large it is, computed with its formula shown and reported with the sample sizes it rests on. Interpretation goes beyond the conventional small, medium and large labels, since those benchmarks were offered as rough guidance and mean different things across areas of research. The magnitude is expressed in something concrete: points on the scale used, overlap between the groups, or the proportion of variance the variable accounts for. A confidence interval accompanies the estimate where the assignment allows. The report ends by saying whether a practitioner should change anything.

Where the marks sit in the PSY 380 Topic 6 rubric

Faculty reward magnitude made concrete. An Excellent report separates what the t test answers from what d answers. Credit goes to computing d = 0.20 with its formula and sample sizes, as Lakens recommends, and an interval from about 0.00 to 0.40, as Cumming urges. The rubric values treating Cohen's benchmarks as rough guidance and expressing the effect as 2 points, 58 percent above the control mean and a 56 percent chance of outscoring, in a table. Strong reports show how the newsletter's wording misleads. Top marks go to a proportionate recommendation.

PSY 380 Topic 6 help: mistakes that cost marks

The first losses are computational: the wrong effect size for the design, a pooled standard deviation assembled incorrectly, or a value reported without the sample sizes behind it. After that the failures are interpretive. Quoting the conventional labels as though they were fixed thresholds ignores that they were offered as rough guidance and vary by area. Reporting an effect size and then discussing only whether the test was significant wastes the calculation. A statistically significant finding presented as important, when the magnitude is trivial, is the exact error the topic exists to prevent. Reports that never translate the number into anything a reader could act on have left the interpretation to somebody else.

Get a PSY-380 Topic 6 example written to your instructions

Send the PSY-380 Topic 6 assignment instructions, your classroom rubric and the output or study you were given to work from. We write a custom example against those criteria, with the correct effect size computed and shown, the magnitude interpreted against comparable research and translated into practical terms, back in 24 to 48 hours. The first one is free.

More PSY 380 papers and BS in Psychology samples

PSY 380 Topic 6 questions, answered

Why report effect size when the test was significant?

Because significance depends heavily on sample size and magnitude does not. With enough participants, a difference too small to matter to anyone will cross the threshold, and with too few, a substantial effect will not. Reporting the size tells a reader which situation they are in, which is why psychology now expects it alongside the test rather than instead of it.

Are the standard benchmarks reliable?

They are useful starting points and were described that way when proposed. What counts as a large effect in a tightly controlled laboratory task differs from what counts as large in an intervention delivered across schools, where small effects can be worth having. Comparing your value with what similar studies report gives a better reading than applying the labels mechanically.

Which effect size goes with which analysis?

It follows from the design and the test. Comparisons of two group means usually take a standardized mean difference; relationships between continuous variables take the correlation and its square; analyses of variance take a proportion of variance explained. Choosing one that does not match the analysis is a checkable error, so name the design and let it decide.

Where can I find a free PSY 380 Topic 6 sample paper?

The effect size report is printed here in full: the finding, two questions, computing d, beyond the labels, where significance and importance part, the interval and the decision, further evaluation and a recommendation. A paper written to your PSY 380 Topic 6 prompt costs nothing.

How can a 2-point gain be significant in PSY 380 Topic 6?

With 200 students per group, the study estimates the difference precisely enough to rule out chance even though the effect is small. The same gap with 20 per group would not reach significance.