| Course | PSYC FPX 3700 Statistics for Psychology |
|---|---|
| Assessment | Assessment 2 |
| Paper type | Hypothesis test write-up |
| Length | About 1,140 words, 5 pages |
| Format | APA 7 student paper |
| School | Capella University |
| Program | BS in Psychology |
| Updated | September 2026 |
Free sample paper for PSYC FPX 3700 Assessment 2
Do Working Students Sleep Less? A Hypothesis Test Write-Up That States Hypotheses in Variable Language, Argues the Test Selection, Checks and Reports Assumptions, Gives One Exactly Formatted Results Paragraph, and Translates the Effect Size Into Magnitude
Student Name
Capella University
PSYC FPX 3700: Statistics for Psychology
Instructor Name
Month Day, Year
The Question and the Data
The composite dataset comes from a survey of 60 undergraduates, 30 who reported working more than 15 hours per week for pay and 30 who reported working 15 hours or fewer, including none. Each student reported average hours of sleep on school nights over the past month. The question is whether the two groups differ in average nightly sleep. The grouping variable is employment, with two levels, and the outcome variable is nightly sleep, measured in hours, a ratio variable.
Hypotheses Written in Variable Language
The hypotheses are written in terms of the variables and the population, not in general terms such as work affects sleep. Null hypothesis: In the population of undergraduates, the mean hours of nightly sleep on school nights is the same for students who work more than 15 hours per week and for students who work 15 hours or fewer. Alternative hypothesis: In the population of undergraduates, the mean hours of nightly sleep on school nights differs between the two groups. The alternative is two-tailed. Although students who work more might be expected to sleep less, the test allows for a difference in either direction, because a one-tailed test chosen on the basis of expectation would not detect the unexpected result that working students sleep more. Writing the hypotheses about population means also clarifies what the test can decide. The test does not ask whether any individual working student sleeps less, nor whether work is the reason for any difference; it asks whether the sample means are far enough apart, relative to the variability in the data, to make equal population means implausible. An alpha of .05 was chosen in advance of any analysis, which fixed the decision rule before the result could influence it.
Test Selection Argued in Prose
The independent-samples t test is chosen for four reasons. The outcome variable, hours of sleep, is continuous, which rules out tests for categorical outcomes such as the chi-square test. The grouping variable has exactly two levels, which makes a t test appropriate and an analysis of variance unnecessary, since with two groups the two give equivalent results. The two groups consist of different students, so the samples are independent, which rules out the paired-samples t test used when the same people are measured twice. And the question concerns a difference in means, which is what the t test compares. A nonparametric alternative, the Mann-Whitney U test, would be preferred if the assumptions below were seriously violated.
Assumptions Checked and Reported
Three assumptions were checked. Independence of observations: each student contributed one response, and students were not paired or grouped in ways that would make their responses dependent. Normality: histograms of sleep in each group were roughly symmetric and unimodal, and Shapiro-Wilk tests were not significant in either group, the working group p = .38 and the nonworking group p = .44, so there was no evidence of serious departure from normality. With 30 students per group, the t test is also fairly robust to moderate departures. Equal variances: the standard deviations were similar, 1.1 and 1.0 hours, and Levene's test was not significant, F(1, 58) = 0.41, p = .52, so the pooled-variance form of the test was used.
One Results Paragraph, Formatted Exactly
The results are reported in one paragraph, formatted following the American Psychological Association (2020). Students who worked more than 15 hours per week reported less nightly sleep (M = 6.4 hours, SD = 1.1) than students who worked 15 hours or fewer (M = 7.1 hours, SD = 1.0). An independent-samples t test showed that this difference was statistically significant, t(58) = 2.58, p = .012, with a mean difference of 0.7 hours, 95% CI [0.16, 1.24]. The effect size was Cohen's d = 0.67. The difference of 0.7 hours comes to about 42 minutes a night, or nearly five hours over a school week.
Effect Size Translated Into Magnitude
Cohen's d was calculated as the mean difference divided by the pooled standard deviation, following Lakens (2013). A d of 0.67 means that the groups differed by about two thirds of a standard deviation. Translated into the units of the study, the difference of 0.7 hours is about 42 minutes per night, or nearly five hours over a school week of seven nights. Translated into overlap, a d of this size implies that about three quarters of students who worked 15 hours or fewer slept more than the average working student, while a substantial minority slept less. Funder and Ozer (2019) argue that effect sizes are better judged by comparing them with familiar benchmarks and by considering their concrete consequences than by applying fixed labels such as small or large. By that standard, a difference of nearly five hours of sleep per week is practically meaningful for students, whose academic work depends on attention and memory. The confidence interval, however, runs from 0.16 to 1.24 hours, from about 10 minutes to about 74 minutes per night, which shows that the size of the true difference is estimated imprecisely with samples of 30.
What the Test Does and Does Not Show
The test shows that, in this composite sample, working students reported less sleep, and that a difference this large would be unlikely if the population means were equal. It does not show that working causes students to sleep less. Students were not assigned to work, and students who work more may differ in other ways, such as family responsibilities or course loads, that also affect sleep. Sleep was self-reported, which may be imprecise. A causal conclusion would require a different design. The test also depends on how the groups were formed. Dividing students at 15 hours per week is a choice, and a different cutoff, such as 10 or 20 hours, might produce a different result. Treating work hours as a continuous variable and examining its correlation with sleep would use more of the information in the data and avoid an arbitrary cutoff. That analysis would answer a slightly different question, about the relationship across the whole range of work hours, and it would be a useful complement to the group comparison reported here.
Conclusion
An independent-samples t test on composite data found that students working more than 15 hours per week reported about 0.7 fewer hours of nightly sleep than those working less, t(58) = 2.58, p = .012, d = 0.67. The difference is about five hours per week, though its size is imprecisely estimated, and the design does not support a causal conclusion.
References
American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.). https://doi.org/10.1037/0000165-000
Funder, D. C., & Ozer, D. J. (2019). Evaluating effect size in psychological research: Sense and nonsense. Advances in Methods and Practices in Psychological Science, 2(2), 156-168. https://doi.org/10.1177/2515245919847202
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
Reading the PSYC-FPX3700 Assessment 2 instructions
Assessment 2 in PSYC FPX 3700 has you test a hypothesis and write up the results. You state the research question and hypotheses, choose a statistical test and explain why it fits the design and data, check the test's assumptions, such as normality and equal variances, run the test, report the results in APA format including the test statistic, degrees of freedom, p value, effect size and confidence interval, and interpret the finding in plain language. Non-significant results should be reported as fully as significant ones. A statistics text and the APA Publication Manual support the write-up.
How the PSYC-FPX3700 Assessment 2 example is put together
Six short parts, in the order the analysis actually happened. The question comes first, stated as something about people rather than about numbers. The hypotheses follow, null and alternative, directional or not, with the decision about direction explained since it changes the test. The third part justifies the test in prose: the variables, their measurement levels, the number and independence of groups, and what those facts rule out as well as what they rule in. The fourth part reports the assumption checks and what was done about any that failed, which is the section weak write-ups omit entirely. The fifth part is the results paragraph, formatted exactly, with the statistic, degrees of freedom, p value and effect size embedded in prose rather than listed. The sixth part interprets, cautiously: what was found, at what magnitude, and what the design does not license.
Reading the PSYC FPX 3700 Assessment 2 scoring guide
Marks go to the hypotheses, the choice of test, the assumption checks, the results paragraph, the effect size and the interpretation. A Distinguished paper writes the hypotheses in terms of the variables, such as mean nightly sleep hours for working and non-working students. The test scores higher when its choice is argued from the design, two independent groups and a continuous outcome pointing to an independent-samples t test. Assumptions earn credit when each is checked and reported, with any violation addressed, for example by using Welch's correction for unequal variances. The results paragraph is judged on exact APA formatting, including the test statistic, degrees of freedom, p value and confidence interval. Effect size should be reported and interpreted in plain terms. Interpretation should avoid causal claims, since students chose whether to work. Distinguished write-ups are precise and honest about what the test shows.
Common PSYC FPX 3700 Assessment 2 mistakes, and how to avoid them
Hypothesis write-ups often state hypotheses vaguely, like "working affects sleep," which cannot be tested as written. Another weakness is choosing a test without explaining why it fits the design. Assumptions are not checked, or they are checked but the results are never reported. The results paragraph has formatting errors, such as missing degrees of freedom or an exact p value written as zero. Effect size is omitted, so a statistically significant but tiny difference looks important. A non-significant result is described as proving that no difference exists. Causal language appears, as if working caused students to sleep less. Writing the hypotheses, the reason for the test and the assumption checks before running anything keeps the write-up disciplined and prepares the critique in Assessment 3.
Get a PSYC FPX 3700 Assessment 2 example written to your instructions
Hand us the Assessment 2 instructions and scoring guide from your PSYC FPX 3700 section, with the dataset, the output or the scenario your question rests on. The example comes back written to those criteria, the test justified, assumptions reported and results in APA form, inside 24-48h. No cost on a first custom sample. The paper above is an original model document written by our desk, not a submitted student paper and not an official Capella University document.
More PSYC FPX 3700 papers and BS in Psychology samples
- PSYC FPX 3700 Assessment 1: Descriptive Statistics Report
- PSYC FPX 3700 Assessment 3: Statistical Critique
- PSYC FPX 3770 Assessment 2: Performance Case Analysis
- PSYC FPX 3520 Assessment 2: Group Behavior Analysis
- PSYC FPX 4101 Assessment 2: Comparative Systems Essay
- PSYC FPX 3501 Assessment 2: Scenario Analysis
PSYC FPX 3700 Assessment 2 questions, answered
What if an assumption is violated?
Report it, say what you did, and explain why. Options include a nonparametric alternative, a transformation, or proceeding with a note on how robust the test is to that particular violation, and any of them can be defensible when argued. Silence is the only choice that always costs marks, because a grader cannot tell whether you checked and decided or never looked.
How do I write up a result that is not significant?
As a finding, in the same format, with the same components. Report the statistic, degrees of freedom, exact p and effect size, then say that the data did not provide evidence of a difference of the size you could detect. Avoid writing that there is no difference, and consider mentioning what sample size would have been needed to detect a small one.
Is an effect size always required?
In practice yes, and in most sections explicitly. Current APA reporting expects one alongside every test, and the interpretation criterion in this course is largely about translating it into plain magnitude. Report the appropriate measure for your test, give the conventional benchmark if you cite one, and then say what a difference of that size would look like in the behavior being studied.
Where can I find a free PSYC FPX 3700 Assessment 2 sample paper?
This page presents the full working students sleep write-up, stating hypotheses in variable language, arguing the test selection, checking and reporting assumptions and giving one exactly formatted results paragraph.
How do I write up a result that is not significant in PSYC FPX 3700 Assessment 2?
Report it in full with the same statistics, and say the data did not provide evidence of a difference. A non-significant result does not prove no difference exists.