PSY 335 Week 4 Surveys, Correlation and Measurement Example

Measuring What We Mean: Building a Campus Survey on Bedtime Phone Use and Sleep Quality

[Student Name]

University of Phoenix

PSY/335: Research Methods

Week 4 Assignment

[Instructor Name]

[Date]

The student researcher, the survey and all results are composites written for a model paper; measurement guidance and research findings come from the sources listed.

The experiment designed in Week 3 can test cause and effect in a small group. A survey can show how common a pattern is and how variables relate across a larger population, but only if the measures are sound. This paper builds a campus survey on phones and sleep, with attention to defining, measuring and sampling well.

Why Add a Survey

Marisol's crossover experiment will involve about thirty volunteers. She also wants to know how widespread bedtime phone use is across her university, whether it relates to sleep quality and whether bedtime procrastination, putting off going to bed without a good reason, plays a role. A correlational survey answers these questions, though it cannot establish cause.

Operational Definitions

Each construct needs a precise definition tied to a measure. Bedtime phone use is defined as minutes of phone use in the sixty minutes before lights out on a typical school night. Sleep quality is defined by scores on an established index covering the past month. Bedtime procrastination is defined by scores on a published scale. Demographics include class year, employment hours and whether the student lives on campus.

Choosing Established Measures

Buysse et al. (1989) developed the Pittsburgh Sleep Quality Index, a self-rated questionnaire covering seven components of sleep over the past month, including latency, duration, disturbances and daytime dysfunction. In their validation work, it showed good internal consistency, stable scores over time and the ability to separate good sleepers from poor sleepers diagnosed in a sleep clinic. Using it gives Marisol a measure whose reliability and validity are already documented.

Kroese et al. (2014) introduced bedtime procrastination as a distinct form of self-regulation failure and developed a short scale to measure it. In their survey, people who scored higher on bedtime procrastination reported insufficient sleep more often, even after accounting for general self-control. The scale fits Marisol's question about whether phones are a symptom of delayed bedtimes rather than the cause.

Writing the New Items

For phone use, no single established item fits, so Marisol writes her own. Krosnick (1999) reviewed research showing that answers depend heavily on how questions are asked: respondents often take mental shortcuts, choosing the first reasonable option or agreeing with statements regardless of content, especially when questions are long, vague or tiring. Following this guidance, her items ask about one behavior at a time, use a specific time frame, avoid agree-disagree formats and offer clear response categories.

A weak first draft read: "Do you often use your phone or watch TV late at night?" It is double-barreled and vague. The revision reads: "On a typical school night in the past two weeks, about how many minutes did you spend using your phone during the hour before you turned off the lights?" with options in fifteen-minute steps.

Scales of Measurement

Minutes of phone use form a ratio scale. Sleep quality index scores are summed ratings, usually treated as interval data. Class year is ordinal, and campus residence is nominal. These distinctions determine which statistics fit later.

Sampling

A convenience sample of students in psychology classes would be easy but unrepresentative. Instead, the registrar provides a random list of 400 students, stratified by class year so that first-year through senior students are equally represented. Each receives an email invitation and two reminders. Of the 400, 240 complete the survey, a response rate of 60 percent. Responders and nonresponders are compared on class year and residence to check for nonresponse bias; responders are slightly more likely to live on campus.

A survey can be perfectly consistent and still wrong, recording with great reliability the screen time students believe they use.

Checking Reliability

Internal consistency, whether items on a scale hang together, is checked with Cronbach's alpha. The bedtime procrastination scale reaches .87 and the sleep index's components show acceptable consistency. Test-retest reliability is checked by asking 40 respondents to repeat the phone use items two weeks later; the correlation between occasions is .74, acceptable for a behavior that genuinely varies.

Checking Validity

Reliability does not guarantee validity. To test whether self-reported phone use reflects actual use, 60 respondents agree to share the screen-time totals their phones record for the hour before bed over one week. Self-reports correlate .52 with logged time, a moderate relationship, and most students underestimate their use by about fifteen minutes. This criterion validity check shows that the self-report item captures real differences between students but imprecisely.

Results and Interpretation

Across 240 students, bedtime phone use correlates .28 with poorer sleep quality scores, bedtime procrastination correlates .41 with poorer sleep quality and phone use correlates .36 with bedtime procrastination. All three are significant.

These correlations do not show that phones harm sleep. Three explanations remain. Phone use might worsen sleep directly. Students who already sleep poorly might reach for their phones when they cannot fall asleep, reversing the direction. Or bedtime procrastination, a tendency to delay going to bed, might drive both phone use and poor sleep, a third-variable explanation. The pattern, with procrastination correlating more strongly with sleep than phone use does, makes the third explanation plausible, and Marisol's experiment from Week 3 can test the first.

Ethics Carried Forward

The survey inherits the safeguards from Week 2. Invitations go out from the department rather than from Marisol, participation is voluntary with no course credit attached, the form asks for no names and the screen-time sharing step is a separate, optional consent. Students who report very poor sleep see a closing page listing campus counseling and health center contacts.

What the Survey Adds to the Experiment

The survey and the experiment answer different questions. The experiment can show whether putting a phone away changes sleep for the volunteers who try it. The survey shows how many students use phones at bedtime, how strongly that habit travels with poor sleep across the campus and whether procrastination belongs in the model. Together they make a stronger case than either alone.

Conclusion

A sound survey depends on clear operational definitions, established scales, well-written new items, a representative sample and evidence of reliability and validity. Marisol's survey shows moderate links among phone use, procrastination and sleep quality, while the validity check and the correlational design keep its conclusions modest.

References

Buysse, D. J., Reynolds, C. F., III, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193-213. https://doi.org/10.1016/0165-1781(89)90047-4

Kroese, F. M., De Ridder, D. T. D., Evers, C., & Adriaanse, M. A. (2014). Bedtime procrastination: Introducing a new area of procrastination. Frontiers in Psychology, 5, Article 611. https://doi.org/10.3389/fpsyg.2014.00611

Krosnick, J. A. (1999). Survey research. Annual Review of Psychology, 50, 537-567. https://doi.org/10.1146/annurev.psych.50.1.537