N508 Module 4 assignment: sampling, reliability and qualitative trustworthiness paper, a full sample

Who Was Studied and Can We Trust It? Sampling and Reliability in a Nurse Stress-Management Trial and Trustworthiness in a Qualitative Study of Nurses' Work Stress

Student Name

Master of Science in Nursing Program, Aspen University

N508: Theory and Research

Instructor Name

Month Day, Year

Who Was Studied and Can We Trust It? Sampling and Reliability in a Nurse Stress-Management Trial and Trustworthiness in a Qualitative Study of Nurses' Work Stress

Two questions determine how far a study's findings can travel: who was studied, and how dependable the measurements or interpretations are. In quantitative research the first is a question of sampling and the second of reliability and validity; in qualitative research, both fold into the broader idea of trustworthiness. This paper examines both kinds of research on the same practical problem, stress among nurses. It analyzes the sampling and measurement in a cluster randomized trial of a stress-management program for intensive care nurses, used here in place of the case study provided in the classroom, and then evaluates the trustworthiness of a qualitative interview study of nurses' work-related stress.

The Quantitative Study

Liu et al. (2025) tested an eight-week mindfulness-based stress reduction program for intensive care unit nurses with high levels of presenteeism, meaning working while impaired by illness, stress, or fatigue. ICU nurses with high presenteeism were invited to participate, and wards were randomly assigned by floor to the intervention or a control condition, 40 nurses in each group. The intervention group received the program from a certified mindfulness therapist; the control group received standard psychological counseling covering emotional regulation, psychological adjustment, and sleep management. Presenteeism was measured with the Stanford Presenteeism Scale-6 and mindfulness with the Five Facet Mindfulness Questionnaire at baseline, after the intervention, and 12 weeks later. Using linear mixed models, the authors found lower presenteeism scores in the intervention group at 8 and 12 weeks and higher mindfulness scores, and concluded that the program reduced presenteeism with sustained effects.

Sampling

Sampling determines whom the results describe. The target population here is ICU nurses experiencing high presenteeism, and the accessible population was nurses in the participating hospital's ICUs. Participants were invited rather than randomly selected from all ICU nurses, which makes the sample a form of nonprobability, purposive sample: it includes nurses who met the presenteeism criterion and chose to take part. Nurses who agree to join a stress-management study may be more motivated or more distressed than those who decline, so the results may not generalize to all ICU nurses.

Random assignment is different from random selection, and this trial randomized at the level of wards rather than individuals. Cluster randomization is often the right choice for workplace interventions, because nurses on the same ward talk to each other, and individual randomization could let control nurses pick up mindfulness practices from colleagues, contaminating the comparison. The trade-off is that nurses on the same ward share conditions, such as leadership and workload, so their outcomes are not independent. With few clusters, one ward's circumstances can masquerade as an intervention effect, which is why the authors' use of mixed models, which account for this clustering, matters. A sample of 40 per arm is modest; a power calculation reported in the paper would tell the reader whether it was large enough to detect a clinically meaningful difference.

The comparison condition also shapes interpretation. The control group received standard psychological counseling rather than nothing, which is a strength: it means the trial compares mindfulness training with an active alternative rather than with no attention at all, so the difference cannot be explained simply by nurses receiving support. It also means the trial answers a narrower question, whether mindfulness adds benefit beyond counseling, than a trial with a no-treatment control would. Readers should also consider attrition, the loss of participants over the 12-week follow-up, since nurses who drop out of stress programs may be those under the most strain; reporting how many nurses completed each assessment in each group lets readers judge whether the analyzed sample still resembles the one that was randomized.

Reliability and Validity of the Measures

Reliability is the consistency of a measure: whether it produces similar results under similar conditions. For multi-item scales, internal consistency, usually reported as Cronbach's alpha, shows whether the items measure the same underlying construct, and test-retest reliability shows stability over time when the construct has not changed. Validity is whether the instrument measures what it claims to measure. Both instruments in this trial are established, widely used scales, which is a strength, but a careful reader checks whether the authors reported reliability in their own sample, because an instrument's reliability can differ across languages, cultures, and occupational groups.

Two further measurement issues deserve attention. Both outcomes were self-reported, and nurses who knew they had received the mindfulness program might report more favorable scores, an expectation effect that blinding could not fully prevent in a behavioral intervention. And presenteeism is a construct with an objective side, reduced productivity, that self-report captures only partly. Pairing the scales with an observed measure, such as medication errors or sick days, would have strengthened the evidence.

The Qualitative Study

A Swedish interview study explored the work-related stress that registered nurses in municipal aged care facilities experienced during the COVID-19 pandemic (Arén et al., 2022). Using a convenience sample, they interviewed 12 registered nurses from six facilities, seven women and five men aged 25 to 70, with 1 to 40 years of experience, who had worked in aged care for at least a year. Semi-structured interviews used eight open questions with follow-up probes and took place at the nurses' workplaces. The data were analyzed with qualitative content analysis following Graneheim and Lundman: two authors listened to the interviews independently, transcribed them, divided the text into meaning units, and condensed and coded them into categories. Themes included the constant need to prioritize, heavy workload, and the link between stress, dissatisfaction, and leaving the profession, captured in one nurse's words: "We are nurses we are not superhumans."

Evaluating Trustworthiness

Trustworthiness, in the framework Lincoln and Guba (1985) set out, rests on four criteria. Credibility asks whether the categories the researchers built are faithful to what the nurses actually said. The study supports credibility through independent listening and coding by two authors, systematic analysis following a published method, and quotations that let readers judge the fit between data and categories. I did not find member checking described, in which participants review the findings, and its absence is a modest limitation.

Transferability concerns whether findings may apply elsewhere, which readers can judge only if the setting and the people in it are described richly. The report describes the facilities, the pandemic context, and participants' age, gender, and experience, allowing readers to judge similarity to their own settings. The convenience sample from one Swedish municipality, where aged care is publicly organized, may differ from U.S. long-term care, and the authors acknowledge that participants' shared working conditions could limit variation. Dependability, the consistency of the research process, is supported by a clear description of interview procedures and analytic steps, and confirmability, the degree to which findings reflect participants rather than researchers, is supported by the audit-like presentation of meaning units, condensations, and codes. A statement about the researchers' own backgrounds and assumptions would have strengthened confirmability further.

The study's sample size, 12 nurses, is appropriate for qualitative content analysis, where the aim is depth rather than statistical representation. The more useful question is whether the interviews captured enough variation, and the diversity in age, gender, and experience across six facilities suggests that they did, even if all participants shared one municipal system.

What the Two Studies Offer Together

The trial and the interview study answer different questions about nurse stress. The trial suggests that a structured mindfulness program can reduce presenteeism among ICU nurses, within the limits of its volunteer sample, cluster design, and self-reported measures. The interview study explains what makes the work stressful in the first place: relentless prioritization, heavy workloads, and the sense of never doing enough for patients and families. Read together, they caution against treating stress as an individual problem to be solved only with individual programs. A nurse leader informed by both would offer evidence-based stress-reduction training while also addressing the workload and prioritization pressures that the qualitative study describes.

The pairing also suggests a better study. A mixed-methods evaluation could measure presenteeism and stress with validated scales before and after a program, while interviewing a subset of nurses about which parts helped and which work conditions undermined them. That design would combine the trial's ability to estimate an effect with the interview study's ability to explain it.

Conclusion

Sampling and reliability determine how far a quantitative study's findings can be trusted to travel, and trustworthiness does the same for qualitative research. The mindfulness trial used an appropriate cluster design and established instruments, but its volunteer sample, modest size, and self-reported outcomes limit generalization. The interview study demonstrated credibility and dependability through systematic, transparent analysis, with transferability limited by its setting. For nurses, the lesson is to read every study with both questions in mind: who was studied, and how much can the findings be trusted.

References

Arén, C., Jaçelli, A., Gesar, B., & From, I. (2022). The work-related stress experienced by registered nurses at municipal aged care facilities during the COVID-19 pandemic: A qualitative interview study. BMC Nursing, 21(1), Article 296. https://doi.org/10.1186/s12912-022-01059-x

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Sage.

Liu, X., Luo, S., Wen, X., Huang, X., Wu, J., Chen, M., & Jia, P. (2025). The effect of mindfulness-based stress reduction on presenteeism among ICU nurses: A cluster randomized controlled trial. PLOS ONE, 20(10), Article e0334825. https://doi.org/10.1371/journal.pone.0334825