Randomize the Nurse, the Night or the Unit? Choosing a Research Design for a Scheduled Nap Study on Two Medical-Surgical Units
[Student Name]
University of Phoenix
NSG/456: Research Outcomes Management for the Practicing Nurse
Week 3 Assignment
[Instructor Name]
[Date]
The units and the study are a composite written for a model paper.
The question from Week 1 concerns one change, a protected half-hour nap for nurses working nights, and two outcomes, how sleepy they feel as the shift ends and how often medication errors and near misses are reported. Week 2 set the protections the study must keep. This paper chooses a design. The choice turns on a practical question that sounds simple: what gets assigned to the nap, the nurse, the night or the unit?
What the Question Needs From a Design
The question is about an intervention and its effect, which points toward a quantitative design that compares nurses who have the nap opportunity with nurses who do not. An ideal design would make the two groups as similar as possible in every way except the nap, so that any difference in sleepiness could be attributed to it. The design also has to work on real units, where naps require coverage by colleagues and approval by managers, and where error reports come from the whole unit rather than from individuals.
Option 1: A Randomized Controlled Trial
In a randomized trial, each nurse or each night would be randomly assigned to a nap opportunity or to usual breaks. Randomization is the strongest protection against differences between groups, because it spreads unknown differences evenly (Polit & Beck, 2021). Smith-Coggins et al. (2006) used this design with 49 physicians and nurses in an emergency department and could therefore attribute the lower end-of-shift fatigue in the nap group to the nap.
On a medical-surgical unit, randomizing individual nurses creates problems. A nurse assigned to no nap would work alongside a colleague who napped and might cover her patients, so the groups would not be independent. Randomizing nights, so that all nurses on some nights have a nap opportunity and on other nights do not, is more workable, but the event data for errors are too sparse to compare night by night. A randomized trial is the strongest design in principle, but on a working unit the nurses in each group share patients, coverage and a medication room, and the design's main advantage begins to leak away.
Option 2: A Quasi-Experimental Design With a Comparison Unit
In a quasi-experimental design, one unit would introduce the nap opportunity and a similar unit would continue usual practice, with sleepiness and event reports measured on both units before and after the change. The design matches how nurse managers actually make decisions, one unit at a time, and it matches the implementation project by Geiger-Brown et al. (2016), in which naps were offered unit by unit. It also fits the Week 2 decision to report error data only at the unit level.
Its weakness is that the two units will differ in ways the study cannot control, such as patient acuity, staffing and culture. Measuring both units before the change helps, because the analysis can compare the change on each unit rather than the raw values.
Option 3: A Qualitative Study
A qualitative descriptive study would interview night-shift nurses and managers about fatigue and napping. It could explain why managers resist naps, what nurses do now to stay alert and what would make a nap program acceptable. It could not answer the question of whether naps reduce sleepiness, because it does not measure an effect. It would be a useful companion study, not a substitute.
The Choice
I would choose the quasi-experimental design, a nonequivalent control group pretest-posttest design, with my unit as the intervention unit and a similar medical-surgical unit in the same hospital as the comparison. Both units would collect sleepiness ratings from consenting nurses for eight weeks before the change and twelve weeks after, and event reports would be counted for the same periods. A short qualitative component, interviews with six nurses and both managers after the intervention, would add explanation to the numbers, making this a mixed methods study with a quantitative core.
Sample and Duration
The two units together employ about 60 nurses who work nights, and a realistic goal is that 40 consent. If each works about three study shifts a week, the eight-week baseline and twelve-week intervention periods would yield several hundred sleepiness ratings on each unit, enough to detect a one-point change on the nine-point Karolinska scale if one exists. Error counts would be much smaller. At the unit's recent rate of reported medication errors and near misses on nights, a twelve-week period would produce perhaps a dozen reports per unit, which is too few to test a difference with confidence. That is why errors are a secondary, exploratory outcome in this design. The twenty-week total length is also chosen deliberately: long enough to see whether nurses keep using the nap window after the novelty fades, and short enough to finish before the hospital's annual staffing changes in the spring.
Threats to Validity and How to Reduce Them
Selection: the units may differ at baseline. Comparing the change on each unit, rather than the post-intervention values alone, reduces this threat.
History: other events, such as a new staffing policy, could affect sleepiness during the study. The study would record any such changes on either unit.
The Hawthorne effect: nurses who know they are being studied may report differently. This affects both units, which limits its effect on the comparison.
Contamination: nurses who float between the two units might nap on the comparison unit. Floating nurses would be excluded from the analysis.
Statistical conclusion validity: medication errors are uncommon enough that the counts may be too small to show a difference. Sleepiness is the primary outcome for this reason, and the error findings would be reported as exploratory.
Conclusion
A randomized trial would provide the strongest evidence, but the realities of a medical-surgical unit make a quasi-experimental design with a comparison unit the better fit for this question and for the ethical protections already set. Week 4 will look at how the data would be collected and how the published numbers on napping should be read.
References
Geiger-Brown, J., Sagherian, K., Zhu, S., Wieroniey, M. A., Blair, L., Warren, J., Hinds, P. S., & Szeles, R. (2016). Napping on the night shift: A two-hospital implementation project. American Journal of Nursing, 116(5), 26-33. https://doi.org/10.1097/01.NAJ.0000482953.88608.80
Polit, D. F., & Beck, C. T. (2021). Nursing research: Generating and assessing evidence for nursing practice (11th ed.). Wolters Kluwer.
Smith-Coggins, R., Howard, S. K., Mac, D. T., Wang, C., Kwan, S., Rosekind, M. R., Sowb, Y., Balise, R., Levis, J., & Gaba, D. M. (2006). Improving alertness and performance in emergency department physicians and nurses: The use of planned naps. Annals of Emergency Medicine, 48(5), 596-604. https://doi.org/10.1016/j.annemergmed.2006.02.005
How this NSG 456 Week 3 example is structured
The NSG/456 description names understanding the elements of research design as a core topic, and many sections place design in the middle of the course. This paper compares three designs against the same question, using published nap studies as examples of each, and chooses the design that fits the question and the ethical protections set in Week 2, then names the threats to validity that remain. Students search this week as NSG 456 Week 3, NSG456 Wk 3 or NSG/456 Wk 3; all three are the same assignment.
NSG/456 Week 3 questions, answered
What does NSG/456 Week 3 usually ask for?
The course description lists research design as a core topic. Many sections ask students to compare quantitative, qualitative and mixed designs and choose the one that fits their research question, with reasons.
What is a quasi-experimental design?
A design that tests an intervention without randomly assigning participants, often comparing a group that receives the intervention with a similar group that does not. It is weaker than a randomized trial at ruling out other explanations.
What is a threat to internal validity?
Anything other than the intervention that could explain the result, such as differences between groups at the start, other changes happening at the same time or people changing their behavior because they know they are being studied.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official University of Phoenix document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.