DBA 8100 Module 2 Qualitative Sampling Paper: Purposive Selection, Saturation and Information Power Example

How Many Supervisors Are Enough? Sampling for Depth in a Capstone Interview Study

Student Name

Doctor of Business Administration, Thomas Edison State University

DBA-8100: Qualitative Research Methods and Applications

Mentor Name

Month Day, Year

How Many Supervisors Are Enough? Sampling for Depth in a Capstone Interview Study

Introduction

Qualitative studies do not aim for statistical representativeness, so the logic of sampling differs from that of surveys. The aim is to recruit people whose experience lets them speak in depth about the phenomenon and to collect enough data to answer the research question well. Committee members and reviewers nevertheless ask how many participants are needed, and a capstone proposal must give a defensible answer. This paper designs the sample for the supervisor study and justifies its size.

Purposive Sampling

The study will use purposive sampling, selecting participants because they have experienced the phenomenon. Eligible participants are employees promoted from hourly roles to first-line supervisor within the past three years. To capture both outcomes in the research question, the sample will include supervisors who stayed beyond twelve months and those who left the role within twelve months, whether they returned to hourly work or left the firm. Within these groups, the sample will seek variation in warehouse location, shift and gender, so that findings are not limited to one site or schedule. This is sometimes called maximum variation sampling within a defined group.

Evidence on Saturation

Saturation, the point at which new data produce little new information, is the most common justification for sample size. Guest et al. (2006) tested it empirically. Working through 60 in-depth interviews collected in two West African countries, they tracked when new codes appeared. By the twelfth interview nearly all codes had appeared, and the outline of the main themes was already visible by the sixth. They cautioned that their results applied to fairly homogeneous participants answering structured questions on a focused topic.

Hennink and Kaiser (2022) systematically reviewed 23 studies that tested saturation empirically. Across these studies, saturation was reached in 9 to 17 interviews or 4 to 8 focus groups, particularly when the study population was relatively homogeneous and the objectives were narrow. Studies with broader aims or more diverse participants needed more.

Information Power

Malterud et al. (2016) argued that saturation, borrowed from grounded theory, does not fit every study and proposed information power as an alternative guide. The more information relevant to the study a sample holds, the fewer participants are needed. They named five factors: a narrow study aim needs fewer participants than a broad one; a sample with highly specific experience needs fewer than a sparse one; the use of established theory reduces the number needed; strong dialogue in interviews holds more information than weak dialogue; and in-depth analysis of a few cases needs fewer than an exploratory cross-case analysis.

Planned Size

Combining the evidence, the study plans an initial sample of 16 supervisors: 8 who stayed and 8 who left. This falls within Hennink and Kaiser's range for a focused study, allows comparison of the two groups and accounts for the weaker dialogue that insider status may produce. If analysis after 16 interviews still yields new codes relevant to the research question, up to 4 more will be added.

A Stopping Rule

To make the judgment of saturation transparent, the study will track new codes after each interview in a simple table. After the first 12 interviews, if three consecutive interviews add no new codes relevant to the research question, and the main themes appear in both groups, data collection will stop. The decision and the evidence will be reported in the methods section.

Recruitment

Participants will be identified from HR records and invited by a letter from a university address rather than an HR email, to signal that participation is voluntary and separate from employment. Former employees will be reached through the contact details they provided at exit, with permission confirmed by the firm's counsel. A modest gift card will thank participants for their time.

Risks to the Sample

Supervisors who left on bad terms may decline. Current supervisors may worry that criticism will reach their managers. These risks could tilt the sample toward satisfied participants. The study will monitor who agrees and who declines, and will report any imbalance.

Why Not a Larger Sample

Some committee members expect larger samples, reasoning that more is always better. In qualitative research, more participants can dilute analysis, since each interview must be read, coded and interpreted in depth. With hour-long interviews producing 25 to 30 transcript pages each, 30 participants would mean close to 900 pages, which would push the analysis toward surface coding. A smaller, well-chosen sample analyzed carefully serves the research question better.

Should Managers Be Interviewed?

The research question centers on supervisors' own accounts, so managers are not part of the sample. Their views would add a second perspective but would also shift the study toward evaluating supervisors. Managers' perspectives may be a topic for a later phase of the firm's work.

Timeline

Recruitment will run for six weeks, with interviews scheduled as consent forms arrive. Analysis will begin after the first interview rather than waiting for all of them, so that the code tracking table needed for the stopping rule is current when each new interview is added. If recruitment of supervisors who left runs slowly, the window will extend by two weeks rather than closing with an unbalanced sample, since comparison between the two groups is central to the question.

Limits

Saturation evidence comes mostly from health research and may not transfer exactly to workplace studies. Sixteen participants from one firm cannot represent all supervisors, which is consistent with the qualitative aim of depth rather than breadth.

Conclusion

Purposive sampling of supervisors who stayed and left, a planned size of 16 justified by Guest and colleagues' and Hennink and Kaiser's saturation evidence and Malterud and colleagues' information power model and a transparent stopping rule give the capstone a defensible sample.

References

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18(1), 59-82. https://doi.org/10.1177/1525822X05279903

Hennink, M., & Kaiser, B. N. (2022). Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Social Science & Medicine, 292, Article 114523. https://doi.org/10.1016/j.socscimed.2021.114523

Malterud, K., Siersma, V. D., & Guassora, A. D. (2016). Sample size in qualitative interview studies: Guided by information power. Qualitative Health Research, 26(13), 1753-1760. https://doi.org/10.1177/1049732315617444