Study guide
Interpreting Quantitative and Qualitative Results in Public Health Studies
Certified in Public Health CPH
Track your progress as you read
Finished sections get a check mark.
1 Foundations of Public Health Data and Analytics
Primary vs Secondary Data: Sources, Applications, and Selection Data Collection Instruments, Validated Methods, and Survey Design Data Cleaning, Management, and Descriptive Statistics for Validity
2 Disease Frequency, Interpretation, and Informatics
Mortality, Morbidity, Prevalence, Incidence, and Risk Data Interpretation Interpreting Quantitative and Qualitative Results in Public Health Studies Surveillance Systems, Registries, Data Synthesis, Visualization, and Informatics
3 Communication Foundations and Planning
Professional Communication, Needs Assessment, and Communication Plans Health Literacy and Culturally Appropriate Communication Approaches Risk Communication and Crisis Communication Models and Principles
4 Media, Ethics, and Data Translation in Communication
Social, Digital, and Traditional Media in Public Health Communication Communication Evaluation, Ethics, Facilitation Skills, and Data Translation for Diverse Audiences
5 Infectious and Chronic Disease Prevention
Health Hazard Identification: Frequency and Impact Assessment Infectious Disease Prevention: Transmission Modes, Risk, and Protective Factors Chronic Disease Prevention: Risk and Protective Factors
6 Environmental Health, Injury, and Emergency Management
Environmental Hazards and Mental Health/Substance Use Disorder Prevention Unintentional and Intentional Injury Prevention Public Health Emergency Management Principles and Response
7 Program Planning and Intervention Design
Needs Assessment, Context-Specific Intervention Design, and Health Equity Lens Objectives, Evaluation Metrics, and Project Management for Program Planning Formative, Process, and Outcome Evaluation Types
8 Evaluation Methods, Implementation, and Resources
Quantitative, Qualitative, and Mixed Evaluation Methods and Report Quality Using Evaluation Results, Implementation Fidelity, and Resource Evaluation
9 Collaboration, Partnership, and Community Engagement
Cross-Sector Partnerships and Stakeholder Engagement The Broader Public Health System and Governmental Agency Partnerships Collaboration Strategies and Community Collective Action Evaluating Coalition and Partnership Effectiveness
10 Health Equity and Social Justice
Culturally Appropriate Assessment and Engagement with Diverse Communities Values, Perspectives, and Health Behaviors in Diverse Populations Health Disparities, Systemic Racism, and Discrimination Access to Services, Environmental Justice, and Human Rights Principles
11 Leadership: Strategy, Change, and Teams
Strategic Planning, Resource Allocation, and Organizational Performance Evaluation Change Management, Conflict Management, and Negotiation Skills Team Building, Performance Standards, Motivation, and Innovation
12 Leadership: Decision Making, Capacity, and CQI
Data-Driven Decision Making, Collaborative Problem Solving, and Capacity Building Vision Communication and Continuous Quality Improvement
13 Program and Resource Management
Budgeting, Financial Proposals, and Contract Management Operations Management, Resource Forecasting, and Program Sustainability CHA/CHIP Implementation and Project Team Coordination Personnel Management, Constructive Feedback, and Retention
14 Law and Ethics
Privacy, Confidentiality, and Regulatory Compliance Ethical Principles, Data Ethics, and Conflicts of Interest International Health Regulations and Health Impact Assessment of Laws
15 Policy in Public Health
The Governmental Policy-Making Process and Policy Implementation Policy Analysis: Evidence, Feasibility, and Impact Assessment Advocacy, Defending Policies, Educating Decision Makers, and Equitable Policy Design
Interpreting Quantitative and Qualitative Results in Public Health Studies
Key Takeaways Explain This Topic AI
Quantitative results answer how much and how many through numeric estimates, confidence intervals, and p-values, while qualitative results answer why and how through thematic, narrative, and contextual findings.
A 95% confidence interval gives a range of plausible values for the population parameter; a wide interval signals low precision and a small sample or high variability.
Statistical significance (p < 0.05) does not imply clinical or public health significance; effect size and confidence interval width must accompany p-values.
Qualitative trustworthiness is judged by credibility, transferability, dependability, and confirmability, not by p-values or generalizability to a sampling frame.
Mixed-methods interpretation triangulates numeric effect estimates with participant narratives to produce convergent, complementary, or divergent evidence.
Last updated: July 2026
Quick Answer: Quantitative interpretation focuses on point estimates, confidence intervals, p-values, and effect sizes; qualitative interpretation focuses on themes, context, and trustworthiness criteria (credibility, transferability, dependability, confirmability). Mixed-methods studies triangulate both to converge on a defensible conclusion.
Interpreting Quantitative Results
A point estimate (a mean, proportion, RR, OR) is the best single guess for a population parameter, but it is always flanked by uncertainty. A 95% confidence interval (CI) is interpreted as: if the study were repeated many times, 95% of the resulting intervals would contain the true parameter. A common misreading is that there is a 95% probability the true value lies in this specific interval — once computed, the parameter is either inside or outside. The width of the CI reflects precision: wide intervals come from small samples, high variability, or both. A CI that crosses the null (RR = 1, OR = 1, difference = 0) means the result is not statistically significant at that alpha level.
A p-value is the probability of observing data as extreme as or more extreme than what was observed, assuming the null hypothesis is true. A p-value below the pre-specified alpha (commonly 0.05) leads to rejecting the null; it does not measure the probability that the null is true, nor does it measure the size or importance of an effect. A very small p-value from a huge sample can correspond to a trivial effect size. Always report and interpret the effect size (the magnitude of the difference or association) alongside the p-value and CI.
- Statistical concept What it tells you Common misinterpretation
- Point estimate Best single guess of parameter Treats it as the exact true value
95% confidence interval Range of plausible values for parameter Treats it as a 95% probability the parameter is in this interval
- p-value Compatibility of data with null hypothesis Treats it as probability the null is true
- Effect size Magnitude of difference or association Assumes statistical significance means large effect
- Interpreting Qualitative Results
Qualitative findings are not summarized by means and p-values; they are summarized by themes , patterns , narratives , and contextual explanations . Common analysis approaches include thematic analysis (identifying recurring themes across transcripts), grounded theory (building theory inductively from data), phenomenology (lived experience descriptions), and content analysis (systematic coding of text). Findings are typically supported by direct participant quotes that illustrate each theme.
Qualitative trustworthiness replaces the quantitative notions of validity and reliability with four criteria from Lincoln and Guba:
Credibility (parallel to internal validity): do the findings represent the participants' realities? Established through prolonged engagement, triangulation of sources, member checking, and peer debriefing.
Transferability (parallel to external validity): can findings transfer to similar contexts? Established through thick description of the setting, participants, and context so readers can judge applicability.
Dependability (parallel to reliability): are the findings consistent and the process auditable? Established through an audit trail and external review.
Confirmability (parallel to objectivity): do the findings reflect the data, not researcher bias? Established through reflexivity, audit trails, and confirmability audit.
Mixed-Methods Interpretation and Triangulation
Mixed-methods designs combine quantitative and qualitative components. Convergent mixed methods collect both forms simultaneously and compare findings: do the numbers and the narratives agree? Explanatory sequential designs collect quantitative first, then use qualitative follow-up to explain unexpected quantitative results. Exploratory sequential designs start with qualitative exploration, then build a quantitative instrument or test hypotheses. Triangulation is the process of comparing evidence from different sources or methods; results can be convergent (both methods agree), complementary (each method adds a distinct facet), or divergent (methods disagree and the discrepancy itself becomes the finding to investigate).
A common interpretation error is treating qualitative findings as hypothesis-generating only and quantitative findings as the real truth. Both carry inferential weight; a well-conducted qualitative study with thick description and member-checked themes can support strong conclusions about context, acceptability, and mechanisms that numeric data cannot capture. Conversely, a quantitative study with a biased sampling frame can produce precise but misleading estimates. Public health reports should explicitly state the type of evidence, its strengths and limitations, and how convergent or divergent findings were reconciled.
Effect modification and confounding also shape interpretation. Effect modification (interaction) means the association differs across strata of a third variable (e.g., a vaccine works well in adults but not elderly); it should be reported as stratum-specific estimates, not a single pooled estimate. Confounding means a third variable distorts the exposure-outcome association; it is addressed through study design (restriction, matching, randomization) or analysis (stratification, regression adjustment). A result that flips direction after adjusting for a confounder (Simpson's paradox) demands explicit discussion in any report.
Worked example of confounding: A cross-sectional study finds coffee drinkers have higher pancreatic cancer rates (RR = 2.0). After adjusting for smoking — a confounder because smoking causes pancreatic cancer and coffee drinkers are more likely to smoke — the adjusted RR drops to 1.1. The apparent association was confounded by smoking. Worked example of effect modification: A vaccine trial finds overall efficacy of 70%, but stratified analysis shows 90% efficacy in adults under 65 and only 20% efficacy in adults over 65. Reporting only the pooled 70% would mask the important finding that the vaccine's effect differs by age group. The correct approach is to report stratum-specific estimates and test for interaction, not to pool across strata. This distinction — confounding distorts and should be removed, effect modification is real and should be reported — is a frequent CPH exam theme.