| Course | PSY 380 Introduction to Probability and Statistics |
|---|---|
| Topic | Topic 7 |
| Paper type | Research article statistics critique |
| Length | About 970 words, 2 pages |
| Format | APA 7 student paper |
| School | Grand Canyon University (GCU) |
| Program | BS in Psychology |
| Updated | October 2026 |
Free sample paper for PSY 380 Topic 7
From Correlation to Recommendation: Reading the Numbers in a Study of Screen Time and Adolescent Wellbeing
[Author Name]
College of Humanities and Social Sciences, Grand Canyon University
PSY-380: Introduction to Probability and Statistics
Topic 7 Research Article Statistics Critique
[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.
From Correlation to Recommendation: Reading the Numbers in a Study of Screen Time and Adolescent Wellbeing
The Article
The article critiqued here is a composite of a common kind of study, constructed for this exercise. Its title claims that daily screen time harms adolescent wellbeing, and its abstract ends with a recommendation that parents limit teenagers' screen time to two hours a day. This critique moves through the article's numbers rather than its abstract, asking what the design entitled the authors to conclude and where the discussion goes beyond the results.
Design and Sample
The study surveyed 1,210 students aged 13 to 17 at four public high schools in one suburban county, recruited through classroom announcements; 1,040 returned usable surveys. Students reported their average daily hours of recreational screen use and completed a ten-item wellbeing scale. All measures were collected once, at the same time. This is a cross-sectional correlational design. A conclusion about adolescents in general rests on how the sample was obtained, and this sample came from one county's volunteers at four schools, so it is not a random sample of adolescents; findings may describe students like these more than adolescents generally.
The Analysis and Whether the Design Entitled It
The authors computed a Pearson correlation between screen time and wellbeing and then ran a multiple regression predicting wellbeing from screen time, controlling for age, gender and school. Both analyses suit the design: they describe how strongly two variables move together, adjusting for a few others. Neither can show that one variable causes the other. Shadish et al. (2002) explained that a causal claim needs the cause to come first in time and rival explanations to be excluded, and a single survey satisfies neither.
The Results Re-Read
The results section reports a correlation of r = minus .12 between screen time and wellbeing, p < .001, and a regression coefficient indicating that each additional hour of daily screen time was associated with a wellbeing score 0.4 points lower on a 50-point scale, after the controls. Re-read closely, these numbers say four things. The direction is negative: more screen time went with lower wellbeing. The relationship is weak: a correlation of minus .12 means screen time accounts for about 1.4 percent of the variation in wellbeing, leaving almost all of it explained by other things. The probability is small because the sample is large, not because the effect is large. And no confidence interval for the coefficient is reported, so a reader cannot see how precisely the 0.4 points is estimated.
Table 1
The Article's Results and What They Support
| Reported result | What it says | What it does not say |
|---|---|---|
| r = minus .12, p < .001 | A weak negative association | That the association is strong or causal |
| Coefficient minus 0.4 per hour | Lower wellbeing by 0.4 of 50 points per hour, on average | That reducing screen time would raise wellbeing |
| p < .001 | Unlikely if there were no association | That the effect is important |
| No interval reported | | How precise the estimate is |
Results Against Discussion
Comparing the results section with the discussion shows where the claims grow. The results describe an association; the discussion says screen time harms wellbeing and that limiting it would help. A correlational finding has acquired causal language. The discussion also states that the effect was especially strong for girls, citing a subgroup analysis in which the correlation among girls was minus .18; the article does not report whether the difference between girls and boys was itself significant, and a subgroup result has become a general claim about girls. Greenland et al. (2016) noted that differences in significance between subgroups do not show that the subgroups differ. Long-standing guidance for psychology journals asks authors to report effect sizes and intervals and to keep causal language out of correlational reports (Wilkinson & Task Force on Statistical Inference, 1999).
Alternative Explanations
The design cannot exclude several explanations for the association. Direction: adolescents with lower wellbeing may spend more time on screens, for example to cope with loneliness or low mood, rather than screens lowering wellbeing. Third variables: family stress, sleep problems, school difficulties or social isolation could raise screen time and lower wellbeing at once, and none was measured. Measurement: both variables are self-reported in one sitting, so a student's mood when completing the survey could color both answers.
Who Did Not Respond
Of 1,210 students recruited, 170 did not return usable surveys, about 14 percent. The article does not describe them. If students with very low wellbeing, or very heavy screen use, were less likely to complete a long survey, the association in the remaining sample could be weaker or stronger than in the full group. A short comparison of responders and nonresponders on age, gender and school, which the authors had, would have let readers judge this.
What a Stronger Design Would Show
A longitudinal design, measuring screen time and wellbeing in the same adolescents at several points, could show whether changes in screen time precede changes in wellbeing or follow them. An experiment, such as randomly assigning some adolescents to reduce social media use for several weeks, could test the causal claim directly, at least for the short term. Either would allow a recommendation of the kind the abstract makes; the cross-sectional survey does not.
What the Study Does Establish
The study does establish something. In a large sample of adolescents from one county, more recreational screen time was weakly associated with lower wellbeing, a pattern consistent with other research and worth knowing. It provides a reason to study the question with designs that follow adolescents over time or test changes in screen use. It does not establish that screen time causes lower wellbeing, that a two-hour limit would help or that the association is big enough to make a practical difference for most families. The abstract's recommendation outruns the numbers.
References
Greenland, S., Senn, S. J., Rothman, K. J., Carlin, J. B., Poole, C., Goodman, S. N., & Altman, D. G. (2016). Statistical tests, P values, confidence intervals, and power: A guide to misinterpretations. European Journal of Epidemiology, 31(4), 337-350. https://doi.org/10.1007/s10654-016-0149-3
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.
Wilkinson, L., & Task Force on Statistical Inference. (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54(8), 594-604. https://doi.org/10.1037/0003-066X.54.8.594
What the PSY 380 Topic 7 instructions ask for
Topic 7 in PSY-380 asks students to critique the statistics in a published study. Strong critiques record the design, the sample and how it was obtained, identify the analyses run and ask whether the design entitled the authors to them. Students should re-read the reported results closely, noting direction, exact probabilities, whether magnitudes and intervals were reported and how much variation an association explains. The critique should compare the results with the discussion, marking where correlational findings acquire causal language or subgroup results become general claims, name alternative explanations the design cannot exclude and state what the study does establish. A critique that only lists flaws, or only praises, misses the evaluation the topic asks for.
How the PSY-380 Topic 7 example is put together
The finished critique moves through the article's numbers rather than its abstract. It records the design, who took part and how many, since a conclusion about a population rests on how that sample was obtained. It identifies the analysis run and asks whether the design entitled the authors to it. Reported results are re-read closely: the direction of the effect, the exact probability where given, whether a magnitude was reported and whether an interval accompanied it. The critique then compares the results section with the discussion, marking any place where a correlational finding acquires causal language or a subgroup result becomes a general claim. Alternative explanations the design cannot exclude are named. The closing passage states what the study does establish, which is usually something.
PSY 380 Topic 7 rubric: what Excellent looks like
Faculty reward a study read through its numbers, not its abstract. An Excellent critique records the cross-sectional design and county volunteer sample and judges correlation and regression fit only for association, after Shadish, Cook and Campbell. Credit goes to re-reading r = minus .12 as about 1.4 percent of variance and noting the missing interval, in a table. The rubric values marking causal language and an untested subgroup claim, with Greenland and colleagues, and asking about nonresponders. Strong critiques name reverse causation and third variables. Top marks go to stating what the study does establish.
PSY 380 Topic 7 help: mistakes that cost marks
Critiques lose most by summarizing the article instead of assessing it, which produces an accurate account that judges nothing. Criticism of writing style, journal choice or topic relevance sits outside what this rubric measures. Declaring a sample too small, with no reference to the effect the study was looking for, is an assertion rather than an analysis. Missing the causal language in a discussion of correlational data leaves the article's main overreach untouched, and it is the one instructors most expect to see caught. Critiques that find nothing supportable are as unbalanced as those that find no fault. Quoting the abstract as evidence of what was found skips the results section entirely.
Get a PSY-380 Topic 7 example written to your instructions
Send the PSY-380 Topic 7 assignment instructions, the rubric from your classroom and the article your section assigned or the one you selected. We write a custom example against those criteria, with the design recovered, the analysis tested against it, the reporting checked for gaps and the discussion held to the results, in 24 to 48 hours. The first one is free.
More PSY 380 papers and BS in Psychology samples
- PSY 380 Topic 1: Descriptive Statistics Exercise
- PSY 380 Topic 2: Variability Analysis
- PSY 380 Topic 3: Sampling Distribution Problem Set
- PSY 380 Topic 4: Hypothesis Test Write-up
- PSY 380 Topic 5: P Value Interpretation
- PSY 380 Topic 6: Effect Size Report
- PSY 380 Topic 8: Analysis Selection Report
- PSY 357 Topic 7: Late Adulthood Case Analysis
- PSY 102 Topic 7: Criteria Application Paper
- PSY 352 Topic 7: Intervention Evaluation
PSY 380 Topic 7 questions, answered
How do I critique statistics I have not been taught?
Work with what the course covers, which reaches further than it seems. Whether the design permits a causal claim, whether the sample supports the population being described, whether magnitudes and intervals were reported, and whether the discussion stays inside the results are all answerable without knowing the mechanics of an unfamiliar procedure. Say plainly which parts you assessed and which you did not.
What if the article looks sound throughout?
Then say so and show the checking that led there. A critique is an evaluation, not a hunt for faults, and a well conducted study read carefully still yields limits the authors themselves would acknowledge, usually about the sample or the population the findings extend to. Manufacturing objections to fill a section is more visible than reporting a study that held up.
Where does causal language usually creep in?
In the discussion, after a results section that was careful. Papers reporting associations between measured variables frequently shift to words like leads to, produces or reduces once the numbers are behind them. Quoting the sentence, naming the design that cannot support it and stating what would have been needed is one of the most reliable ways to earn marks on this topic.
Where can I find a free PSY 380 Topic 7 sample paper?
The statistics critique is shown here complete: the article, design and sample, the analysis, results re-read, results against discussion, alternative explanations, nonresponse, a stronger design and what the study establishes. Forward the PSY 380 Topic 7 prompt you received for a free critique.
How much does screen time explain in PSY 380 Topic 7?
A correlation of minus .12 means screen time accounts for about 1.4 percent of the variation in wellbeing. Almost all of the variation comes from other things.