Chapter Three: Research Methodology

Introduction

This chapter describes the research methodology employed to investigate whether psychological safety influences employee retention in private educational institutions in Uganda. It describes the research design and philosophical underpinnings, population and sampling procedures, data collection instruments, validity and reliability measures, data analysis techniques, and ethical questions governing the study. The methodology seeks to answer the research questions and objectives outlined in Chapter One through quantitative as well as qualitative inquiry.

Research Philosophy

This work rests on the pragmatist philosophical model in which knowledge has practical value (Creswell & Plano Clark, 2018). Pragmatism is the best philosophical position for this study since it permits the integration of quantitative and qualitative techniques with no rigid attachment to one ontological or epistemological worldview (Morgan, 2007). Not seeing reality as objective or subjective, pragmatism accepts that multiple perspectives are necessary for a better understanding of organizational phenomena such as psychological safety and employee retention.

It also adopts some features of interpretivism in the qualitative component, noting that employees’ impressions and lived experiences of psychological safety are socially constructed and context-dependent (Bryman, 2016). For the quantitative component, post-positivist assumptions are made, accepting that objective measurement of variables is possible but subject to error and requires triangulation.

Research Design

This study employs a concurrent mixed methods research design, also known as a converged parallel design (Creswell & Plano Clark, 2018). In this design, quantitative and qualitative data are collected simultaneously, independently analyzed, and merged at the interpretation stage to derive an overview of the investigated phenomenon. Hence, this design is suited for studies that require statistical generalizability and contextual depth, both essential for understanding the complex dynamics of psychological safety and its relation to employee retention in private educational institutions in Uganda.

The quantitative strand employs a cross-sectional survey to measure psychological safety and its statistical relationship with employee retention intentions. The qualitative strand consists of semi-structured interviews/focus group discussions to explore employees ‘subjective experiences/perceptions and organizational practices associated with psychological safety and their decision to stay/leave. Integrating these strands permits triangulation of results (Bryman, 2016).

Study Population

The study population is academic and non-academic staff of private educational institutions based in Kampala/Wakiso districts in Uganda. They were selected because these districts contain most of the private universities, secondary schools, and primary schools in Uganda (Uganda National Council for Higher Education, 2022). Its population includes teachers and lecturers, administrative officers and support staff, and institutional leaders such as principals, directors, and human resource managers.

These private rather than public educational institutions were selected because they are characterized by contractual employment arrangements, competitive labor markets, and variable working conditions that make psychological safety a relevant issue for employee retention decisions (Ssali, Onen, and Gyaviira, 2019). Academic and non-academic staff are included to ensure psychological safety is understood across functional roles within these institutions.

Sample Size and Sampling Procedures

Quantitative

For the quantitative strand, 220 academic and non-academic staff will be drawn from private educational establishments in Kampala and the Wakiso districts. This sample size is based on Krejcie and Morgan (1970) sample size determination table and is adequate for correlation and regression analyses with adequate statistical power (Cohen 1992). To ensure proportionate representation of academic and non-academic staff across selected institutions, stratified random sampling will be used.

The strata will be differentiated according to staff category academic/non-academic and institution type: university, secondary school or primary school. Within each stratum, simple random sampling will be used to select respondents. This reduces sampling bias and ensures representative findings of the study population.

Qualitative

In the qualitative strand, 20 participants will be selected by purposive sampling for semi-structured interviews and 3 focus group discussions with 6-8 participants each. This requires selective sampling because the researcher wants information-rich cases that shed light on lived experiences of psychological safety and retention decisions (Patton, 2015). Participants will be selected on the basis of role (teachers/administrators / HR officers / institutional leaders), service length (to capture varied retention perspectives), and willingness to participate.

Focus group participants will represent different staff categories to allow diverse interactions and collective sense-making about workplace experiences. Because their positions are sensitive, institutional leaders will be interviewed individually to encourage candid disclosure about organizational practices.

Data Collection Instruments

Structured Questions

For quantitative data collection a structured self-administered questionnaire will be the principal instrument. The questionnaire will have 3 sections:

Section A: Demographic information: gender, age, educational level, staff category, years of service, and type of institution.

Section B: Psychological Safety: Edmondson’s (1999) Validated Seven-Item Psychological Safety Scale. This scale measures comfort with speaking up, mutual respect, openness to new ideas, and tolerance for interpersonal risk-taking among other dimensions. Responses will be retrieved on a five-point Likert scale from 1 (Strongly Disagree) to 5 (Strongly Agree).

Section C: Employee retention intentions measured by a validated retention scale based on Mobley, Griffeth, Hand & Meglino, 1979; Hom et al. (2017). Items will measure organizational commitment, job satisfaction, and likelihood of remaining in the institution in 12 months.

The Likert scale form is commonly used in organizational behavior research and permits statistical analysis of otherwise hard-to-quantify attitudes and perceptions (Bryman 2016). The questionnaire will be piloted before full deployment to identify ambiguous items and ensure respondents understand questions.

Semi-structured interview guide

20 participants will be selected for semi-structured interviews, including teachers, administrative staff, HR officers and institutional leaders. The interview guide is going to include open-ended questions about participants’ experiences with psychological safety, their impressions of supportive and non-supportive behaviors, the role of leadership in creating a safe work environment, and how these factors have influenced their decision to stay or leave their institution.

Semi-structured interviews are appropriate because they allow the researcher to follow a predefined line of inquiry but still be flexible enough to explore emergent themes raised by participants (Kvale & Brinkmann, 2015). Interviews may take place in English or Luganda as the participant prefers, to facilitate comfort and authenticity of responses. Interviews will be audio-recorded and, if consented to, transcribed verbally.

Focus Group Discussion Guide

Groups of 6-8 staff members from different functional categories will hold three focus group discussions (FGDs). The FGD guide will explore group perceptions of psychological safety, shared experiences of inclusion or exclusion, team dynamics, communication climate, and how these group factors influence retention decisions. Particularly for exploring social and organizational phenomena, FGDs produce data through interaction and group dynamics that are not apparent in individual interviews (Morgan, 1997).

Every FGD will have a researcher-moderated session and will be audio-recorded with participant consent. A research assistant will take field notes to record non-verbal cues and group dynamics. Sessions will be 45-60 minutes long and held in a neutral private environment to minimize inhibition.

Validity and Reliability

Validity

Validity is how well an instrument measures what it claims to measure (Creswell & Creswell, 2018). Three academic experts from Ugandan universities will review the instruments for content validity. Their feedback will improve item clarity, language appropriateness, & context relevance. Additionally, construct validity is ensured by using Edmondson’s (1999) psychological safety scale, which has been validated in international and African contexts (Frazier et al, 2017; Oduor & Muriithi, 2021).

For the qualitative component, validity will be enhanced through member checking, where preliminary findings will be communicated to a subset of participants to check interpretation accuracy. Longevity of engagement with the study context and detailed description of findings will build credibility (Lincoln & Guba, 1985).

Reliability

Reliability is a measure of consistency (Bryman, 2016). Internal consistency of quantitative scales will be evaluated using Cronbach’s Alpha coefficient, with a > = 0.70 acceptable threshold (Nunnally, 1978). Pilot testing of the questionnaire with 20 respondents from institutions outside of the final sample will allow reliability assessment and item fine-tuning before full data collection.

With respect to the qualitative component, reliability will be ensured by an interview protocol that is consistent across all sessions, audio-recording for verbatim transcription, and peer debriefing with a research supervisor to avoid researcher bias (Lincoln & Guba, 1985). Data coding consistency will be confirmed by inter-coder agreement for a subsample of transcripts.

Data Analysis

Quantitative Data Analysis

Quantitative data will be entered and analyzed using SPSS. Descriptive Statistics, including frequencies, means, and standard deviations regarding respondents and distribution of psychological safety and retention intention scores, will be computed.

The data will be assessed using Pearson’s correlation analysis to determine the kind of linear relationship there is between psychological safety and intentions for employee retention (Objective 1 & Research Question 1). Then, multiple linear regression analysis will determine the predictive power of individual psychological safety dimensions, i.e., leadership support, trust, communication openness, and team respect, on employee retention intentions, adjusting for demographic variables as per Objective 2 & Research Question 2. The regression model will be checked for normality/multicollinearity/homoscedasticity/linearity assumptions before interpretation. For all tests, a statistical significance level of p < 0.05 will be applied.

Qualitative Data Analysis

Qualitative data derived from interview transcripts & FGD notes will be analyzed thematically according to Braun and Clarke (2006) in a six-phase process: data familiarization, initial codes generation, theme searching, theme reviewing, theme defining and naming, and final report production. This approach identifies recurring patterns and themes regarding employee experiences of psychological safety and retention.

The qualitative data will be arranged, coded, and managed using NVivo software. First coding will be data-driven (inductive), so themes emerge from participant narratives naturally, and later coding will be informed by the study’s theoretical frame (deductive). Deviant cases and conflicting perspectives will be monitored for findings that are representative of full participant experiences rather than confirmation of assumptions.

Integration of Quantitative and Qualitative

According to the convergent parallel mixed-methods design, quantitative and qualitative results will be combined at the interpretation stage (Creswell & Plano Clark, 2018). This will include contrasting statistical results with qualitative themes to find convergence/divergence/complementarity points. Integration will occur through joint display tables or matrices displaying both sets of findings together to enable holistic interpretation and a more nuanced picture of the relation between psychological safety and retention of employees.

Ethical Considerations

To comply with ethical standards for human subjects’ research. The following ethical principles will guide the study conduct:

Informed Consent: Written informed consent forms will be provided to all participants explaining the purpose of the study, nature of participation, withdrawal without penalty, and use of data. Participation is entirely voluntary.

Confidentiality and Anonymity: Identity information will not be published about the research. Questionnaires will be filled out anonymously; pseudonyms in qualitative reports will be used to conceal interviewees/FGD participants ‘identities.

Data Security: The data collected, which will include audio recordings and transcripts, will be stored on password-protected devices only accessible by the researcher and supervisor. After five years of study completion, data will be securely deleted.

Institutional Permission: Formal permission from management of each participating institution will be sought before starting data collection. The researcher’s institution and research purpose will be communicated.

Avoidance of Harm: Interview/FGD questions will not be worded to cause psychological discomfort or anxiety. Any questions participants find uncomfortable will be declined.

Conclusion

This chapter describes the methodological framework used to investigate whether psychological safety influences employee retention in private educational establishments in Uganda. With pragmatist philosophical underpinnings, a concurrent mixed methods design allows triangulation of quantitative survey data and qualitative interview and FGD data pertaining to psychological safety and employee retention. Stringent regard for validity, reliability, and ethical principles guarantees that findings will be trustworthy, credible, and useful for administrators, HR practitioners, and policy makers in Uganda’s private education sector.

References

A power primer. (n.d.). Retrieved June 16, 2026, from https://psycnet.apa.org/doiLanding?doi=10.1037%2F0033-2909.112.1.155

Baker, D. P., Day, R., & Salas, E. (2006). Teamwork as an essential component of high-reliability organizations. Health Services Research, 41(4 II), 1576–1598. https://doi.org/10.1111/J.1475-6773.2006.00566.X;PAGE:STRING:ARTICLE/CHAPTER

Charles, B. K., Florah, O. M., Charles, B. K., & Florah, O. M. (2021). A Critical Review of Literature on Employment Selection Tests. Journal of Human Resource and Sustainability Studies, 9(3), 451–469. https://doi.org/10.4236/JHRSS.2021.93029

Determining Sample Size for Research Activities – Robert V. Krejcie, Daryle W. Morgan, 1970. (n.d.). Retrieved June 16, 2026, from https://journals.sagepub.com/doi/10.1177/001316447003000308

High‐quality relationships, psychological safety, and learning from failures in work organizations – Carmeli – 2009 – Journal of Organizational Behavior – Wiley Online Library. (n.d.). Retrieved June 16, 2026, from https://onlinelibrary.wiley.com/doi/10.1002/job.565

Liu, Y., Wang, W., & Chen, D. (2019a). Linking Ambidextrous Organizational Culture to Innovative Behavior: A Moderated Mediation Model of Psychological Empowerment and Transformational Leadership. Frontiers in Psychology, 10, 464519. https://doi.org/10.3389/FPSYG.2019.02192/TEXT

Liu, Y., Wang, W., & Chen, D. (2019b). Linking Ambidextrous Organizational Culture to Innovative Behavior: A Moderated Mediation Model of Psychological Empowerment and Transformational Leadership. Frontiers in Psychology, 10, 464519. https://doi.org/10.3389/FPSYG.2019.02192/TEXT

Making it safe: the effects of leader inclusiveness and professional status on psychological safety and improvement efforts in health care teams – Nembhard – 2006 – Journal of Organizational Behavior – Wiley Online Library. (n.d.). Retrieved June 16, 2026, from https://onlinelibrary.wiley.com/doi/10.1002/job.413

Newman, A., Donohue, R., & Eva, N. (2017). Psychological safety: A systematic review of the literature. Human Resource Management Review, 27(3), 521–535. https://doi.org/10.1016/J.HRMR.2017.01.001

One hundred years of employee turnover theory and research. (n.d.). Retrieved June 16, 2026, from https://psycnet.apa.org/doiLanding?doi=10.1037%2Fapl0000103

Paradigms Lost and Pragmatism Regained – David L. Morgan, 2007. (n.d.). Retrieved June 16, 2026, from https://journals.sagepub.com/doi/10.1177/2345678906292462

Psychological Safety: A Meta‐Analytic Review and Extension – Frazier – 2017 – Personnel Psychology – Wiley Online Library. (n.d.). Retrieved June 16, 2026, from https://onlinelibrary.wiley.com/doi/10.1111/peps.12183

Psychological Safety and Learning Behavior in Work Teams – Amy Edmondson, 1999. (n.d.). Retrieved June 16, 2026, from https://journals.sagepub.com/doi/10.2307/2666999

Psychological Safety: The History, Renaissance, and Future of an Interpersonal Construct | Annual Reviews. (n.d.). Retrieved June 16, 2026, from https://www.annualreviews.org/content/journals/10.1146/annurev-orgpsych-031413-091305

Teamwork as an Essential Component of High‐Reliability Organizations – Baker – 2006 – Health Services Research – Wiley Online Library. (n.d.). Retrieved June 16, 2026, from https://onlinelibrary.wiley.com/doi/10.1111/j.1475-6773.2006.00566.x

Using thematic analysis in psychology: Qualitative Research in Psychology: Vol 3, No 2. (n.d.). Retrieved June 16, 2026, from https://www.tandfonline.com/doi/abs/10.1191/1478088706qp063oa

Uganda Ministry of Education and Sports. (2023). Education statistical abstract 2022/2023. Ministry of Education and Sports.

Uganda National Council for Higher Education. (2022). List of accredited private universities and institutions in Uganda. UNCHE.