HMPYC80 Research Methodology Exam Notes for UNISA: Honours in Psychology

Research methodology is the foundation of credible psychological science. For HMPYC80, the central goal is not simply to memorise definitions, but to understand how research questions become designs, designs become data, and data become evidence. These exam notes bring together the main concepts, logic, and application points that commonly appear in honours-level research methodology assessments in psychology, with a strong UNISA and South African academic context.

1. The Purpose and Logic of Research Methodology in Psychology

Research methodology is the study of how knowledge is produced, evaluated, and justified. In psychology, it answers a practical and philosophical question: how do we know that a claim about human thoughts, emotions, behaviour, or social interaction is trustworthy? HMPYC80 places this question at the centre because honours students are expected to read research critically, design feasible studies, and interpret findings with intellectual discipline.

What research methodology is really about

At its core, methodology is broader than “methods.” Methods are the techniques used to collect and analyse data, such as interviews, questionnaires, experiments, or statistical tests. Methodology includes the logic behind choosing those methods. It asks why a particular approach is suitable for a particular research question, what assumptions are being made, and what limitations follow from those assumptions. A student who understands methodology can justify why a qualitative interview study may be more appropriate than a survey, or why a cross-sectional design cannot establish causality.

In psychology, this matters because human behaviour is complex, contextual, and often influenced by factors that are difficult to observe directly. Emotions, attitudes, motivation, identity, coping, trauma, and social support are not always visible in the same way as physical phenomena. Methodology therefore provides the framework for deciding whether a construct should be measured quantitatively, explored qualitatively, or approached through a mixed-methods design.

The scientific logic of psychological inquiry

Psychology relies on systematic inquiry rather than intuition. This means that claims should be based on observable evidence that can be examined by others. The logic of science in psychology typically involves the following sequence:

Observe a phenomenon
For example, students may appear to experience high anxiety during examinations.
Formulate a question
Why does exam anxiety increase for some students more than others?
Develop a theory or explanation
A theory may suggest that low self-efficacy and poor coping strategies contribute to anxiety.
Derive a hypothesis or qualitative focus
One might predict that students with lower perceived self-efficacy report higher anxiety.
Collect data systematically
Use a validated scale, interview protocol, or observation schedule.
Analyse the data
Apply statistical tests or interpret themes.
Draw conclusions and refine theory
Evidence may support, weaken, or complicate the initial idea.

This sequence is important because it demonstrates that research is cumulative. Each study adds to a broader body of knowledge, even when the result is negative or mixed. In psychology, a “negative” result is not a failure; it may show that a theory needs revision, that a measure is inadequate, or that a relationship is more context-dependent than expected.

Research problems in the South African context

HMPYC80 should be understood within the realities of South African higher education and psychology practice. Research questions often arise from issues such as student mental health, inequality, family dynamics, violence, unemployment, language diversity, access to services, and community resilience. A methodology that works in one context may not work equally well in another. For example, a questionnaire developed in an urban, English-speaking, middle-income sample may need adaptation before it is used in a multilingual, rural, or economically diverse population.

This is why methodology is also about fit. Research must fit the population, the setting, the ethical environment, and the practical constraints. In the UNISA context, many students study while working or caring for families, so research design must often be efficient, ethical, and realistic. Methodology helps researchers balance ambition with feasibility.

Key terms and conceptual distinctions

A strong exam answer usually distinguishes between closely related terms:

Methodology: the overall logic and justification of research choices
Methods: the specific tools and procedures used to gather and analyse data
Theory: a system of ideas explaining phenomena
Hypothesis: a testable prediction derived from a theory
Concept: an abstract idea, such as stress or resilience
Variable: a measurable characteristic that can differ across cases
Construct: a theoretical concept that must be operationalised
Operationalisation: defining how a construct will be measured or observed

These distinctions matter because exam questions often require precision. A student might be asked to explain why “stress” is a construct, “stress score” is a variable, and “number of missed classes” could be one operational indicator of stress-related functioning. Methodological sophistication begins with careful language.

Why methodology matters for psychological credibility

Psychology has sometimes been criticised for weak replication, sampling bias, and overreliance on convenience samples, particularly student samples. Methodology addresses these concerns by strengthening transparency, reproducibility, and validity. A study can only be trusted if its procedures are clear enough that another researcher could reasonably repeat them and obtain comparable evidence. Even if the exact same result does not recur, the logic of the inquiry should be traceable.

For honours students, this means learning to ask not only “What did the study find?” but also:

How was the question framed?
Were the methods appropriate to the question?
Was the sample suitable?
Were the measures valid and reliable?
Could the findings be explained by alternative factors?
What ethical issues were present?
How far can the conclusions be generalised?

A strong methodological mindset is therefore critical, not only for passing exams, but for becoming a responsible consumer and producer of psychological knowledge.

2. Research Paradigms, Approaches, and Designs

Research methodology begins with a philosophical choice about what counts as knowledge. Different paradigms lead to different questions, designs, and forms of evidence. In HMPYC80, students are expected to understand not only what a design is, but why a design is aligned with a particular way of thinking about reality and knowledge.

Positivist, interpretivist, and pragmatic orientations

A positivist or post-positivist approach assumes that reality exists independently of the researcher and that it can be measured, at least imperfectly, through systematic observation. This approach usually supports quantitative methods, hypothesis testing, and statistical analysis. It is especially useful when the goal is to test relationships between variables, estimate prevalence, compare groups, or examine causal patterns.

An interpretivist approach assumes that reality is socially constructed and that meaning is produced through interaction, language, and context. Rather than seeking universal laws, interpretivist research explores lived experience, perspective, and social meaning. This approach often uses qualitative methods such as interviews, focus groups, and thematic analysis.

A pragmatic orientation is driven by the research problem rather than by a strict philosophical commitment. Pragmatism allows researchers to combine methods when necessary. If the problem requires both numerical trends and rich explanation, mixed methods may be the best option. In psychological research, pragmatism is often valuable because human problems rarely fit neatly into one paradigm.

Quantitative research

Quantitative research is designed to measure variables numerically and analyse patterns statistically. It is suitable for questions such as:

Does perceived social support predict lower depression scores?
Is there a difference in academic burnout between first-year and final-year students?
What proportion of respondents report moderate-to-high anxiety?

The strength of quantitative research lies in standardisation, comparability, and statistical inference. Common quantitative designs include:

Descriptive surveys
Correlational studies
Experimental designs
Quasi-experimental designs
Longitudinal studies
Cross-sectional studies

Quantitative research is often associated with structured questionnaires, scales, behavioural counts, and controlled procedures. Because results are numerical, they can be summarised efficiently and used to identify trends across larger samples. However, quantitative research can oversimplify complex experiences if the measures are poorly chosen or if the context is ignored.

Qualitative research

Qualitative research seeks deep understanding of meaning, experience, and process. It is appropriate when the researcher wants to know how people experience something, why they interpret it a certain way, or what social meanings shape behaviour. Typical qualitative methods include:

In-depth interviews
Focus groups
Observations
Document analysis
Narrative inquiry
Case studies
Phenomenological approaches
Grounded theory approaches

Qualitative research is especially helpful for exploring sensitive or under-researched issues, such as stigma, identity, grief, trauma, or the lived experience of disability. It can reveal nuances that surveys may miss. For example, a study of student mental health may discover that the main stressor is not simply workload, but the cumulative pressure of financial strain, family expectations, transport difficulties, and digital access problems.

The limitation of qualitative research is that it typically involves smaller samples and does not aim for statistical generalisation. Instead, it aims for depth, credibility, and analytical insight.

Mixed methods research

Mixed methods research combines quantitative and qualitative approaches within one study or programme of research. The purpose is to gain both breadth and depth. For example, a researcher might first use a survey to identify levels of anxiety among students and then conduct interviews to understand what those anxiety scores mean in everyday life.

Mixed methods can be designed in several ways:

Sequential explanatory design
Quantitative data are collected first, followed by qualitative data to explain the results.
Sequential exploratory design
Qualitative data are collected first to explore a phenomenon, followed by quantitative data to test or generalise findings.
Concurrent triangulation design
Both types of data are collected at roughly the same time and compared.

The strength of mixed methods is that it can offset the weaknesses of each approach. The challenge is complexity: it requires careful planning, sufficient time, and a clear rationale for integration. In an honours-level exam, it is important to show that mixed methods are not simply “doing everything,” but using complementary forms of evidence strategically.

Research designs and their purposes

A design is the blueprint for how a study will be conducted. It connects the research question to the type of evidence needed. The most common designs in psychology include:

Descriptive design

Used to describe characteristics, frequencies, or patterns. It answers “what is happening?” rather than “why is it happening?”

Correlational design

Used to examine the association between variables. It can show whether variables move together, but not whether one causes the other.

Experimental design

Used to test causal relationships by manipulating an independent variable and measuring its effect on a dependent variable, ideally with random assignment.

Quasi-experimental design

Similar to experimental design, but without random assignment. Often used when randomisation is impractical or unethical.

Longitudinal design

Collects data from the same participants over time to study change, development, or temporal ordering.

Cross-sectional design

Collects data at one point in time from different participants or groups.

Comparing designs in exam-friendly form

Design	Main purpose	Strength	Limitation	Example in psychology
Descriptive	Describe a phenomenon	Simple and informative	Cannot explain cause	Prevalence of test anxiety among students
Correlational	Assess association	Useful for prediction	No causal inference	Relationship between sleep quality and mood
Experimental	Test cause and effect	Strongest causal evidence	May be artificial or unethical	Effect of mindfulness on stress
Quasi-experimental	Compare naturally existing groups	Practical in real settings	Selection bias risk	Comparing stress across faculties
Longitudinal	Study change over time	Captures developmental patterns	Time-consuming	Coping skills across a semester
Cross-sectional	Compare groups at one time	Fast and economical	Cannot track change	Burnout across year levels

Understanding these designs is essential because exam questions often ask students to identify a design from a scenario, justify why it was used, or critique its limitations.

3. Sampling, Measurement, and Operationalisation

A well-designed study can still fail if the sample is inappropriate or the measurement strategy is weak. Research methodology therefore requires attention to who is studied, how they are selected, and how concepts are turned into data. These issues are central to both validity and ethics.

Population, sample, and sampling frame

The population is the larger group to which the researcher wants to generalise. For example, a study may aim to understand undergraduate psychology students at a university, all honours students in South Africa, or adults in a particular community.

A sample is the subset of the population that actually participates in the study. Because it is usually impossible to study every member of a population, researchers rely on samples.

The sampling frame is the list or source from which the sample is drawn. If the frame is incomplete, the sample may be biased. For example, an online survey distributed only through email may exclude students with limited internet access or those who check email infrequently.

Probability and non-probability sampling

Sampling strategies are broadly grouped into probability and non-probability methods.

Probability sampling

In probability sampling, each member of the population has a known chance of selection. This supports stronger generalisation. Common forms include:

Simple random sampling
Systematic sampling
Stratified sampling
Cluster sampling

Stratified sampling is particularly useful when the researcher wants representation from important subgroups, such as gender, study level, language group, or campus location. Cluster sampling is practical when the population is geographically dispersed.

Non-probability sampling

In non-probability sampling, not every member has a known chance of selection. It is often more feasible in educational and psychological research. Common forms include:

Convenience sampling
Purposive sampling
Snowball sampling
Quota sampling

Convenience sampling is common in student research because it is easy and inexpensive, but it may produce biased samples. Purposive sampling is useful when participants must meet specific criteria, such as having experienced a particular event. Snowball sampling is useful for hard-to-reach populations.

Sample size and representativeness

Sample size affects the stability and interpretability of findings. Larger samples generally improve precision, but size alone does not guarantee quality. A large biased sample can still be misleading. The goal is not only size but fit between sample and research question.

Representativeness means that the sample resembles the population on key characteristics relevant to the study. For example, if a study on student stress includes only first-year students, it may not represent the broader undergraduate experience. Similarly, if a study on workplace burnout includes only office workers, it may not capture the realities of shift workers or informal-sector employees.

In exam answers, it is often valuable to explain that representativeness depends on the research aim. In some qualitative studies, representativeness is not the main goal; rather, the focus is depth and richness of information. In such cases, the researcher seeks information-rich cases rather than statistical representation.

Operationalisation

Operationalisation is the process of translating abstract ideas into observable indicators. This is one of the most important concepts in HMPYC80 because many psychological constructs cannot be measured directly.

For example:

Anxiety may be operationalised as a score on a standard anxiety inventory
Academic performance may be operationalised as grade average
Social support may be operationalised as a perceived support scale or number of supportive contacts
Aggression may be operationalised as self-reported frequency, behavioural observation, or disciplinary records

A strong operational definition must be:

clear,
measurable,
relevant to the construct,
and appropriate to the context.

Poor operationalisation creates invalid research. If “resilience” is measured only by whether someone reports smiling often, the measure may miss endurance, coping, and recovery. Good methodology requires distinguishing the indicator from the full construct.

Reliability and validity

Reliability refers to the consistency of a measure. If a scale is reliable, it produces stable results under consistent conditions. Types of reliability include:

Test-retest reliability: consistency over time
Internal consistency: whether items measure the same underlying construct
Inter-rater reliability: agreement between observers or coders
Parallel forms reliability: consistency across equivalent versions of a test

Validity refers to whether a measure actually measures what it claims to measure. Types include:

Content validity: coverage of the full construct domain
Construct validity: alignment with the theoretical concept
Criterion validity: relation to a relevant external criterion
Face validity: whether the measure appears appropriate on the surface

Reliability is necessary but not sufficient for validity. A measure can be consistently wrong. For instance, a scale that repeatedly captures only one narrow aspect of depression may be reliable but not valid for the broader construct.

Measurement error and bias

Measurement error is the difference between the true value and the observed value. It may come from participant misunderstanding, poor wording, social desirability, fatigue, interviewer influence, or faulty instruments. Bias is a systematic error that pushes results in one direction.

Common forms of bias include:

Sampling bias
Response bias
Social desirability bias
Recall bias
Observer bias
Confirmation bias

These sources of error matter because they can distort findings without being obvious. For example, if students fear that admitting distress will make them seem weak, they may underreport anxiety. If the researcher strongly expects a certain outcome, that expectation may subtly shape coding or interpretation. Methodology teaches students to identify these threats and reduce them through careful design.

4. Data Collection, Ethics, and Quality in Psychological Research

Data collection is where methodological planning becomes practical reality. At this stage, the researcher interacts with participants, records observations, or gathers documents. Because psychology deals with people, ethics is inseparable from method. No study is methodologically sound if it ignores participant rights, dignity, and welfare.

Common data collection methods

Questionnaires and surveys

Questionnaires are widely used in psychology because they can reach many participants efficiently and produce standardised data. They may include closed-ended items, rating scales, and demographic questions. The major advantage is comparability. The main risk is shallow or inaccurate responses if the items are poorly written or the topic is sensitive.

A good questionnaire should have:

clear wording,
one idea per item,
appropriate response options,
balanced scale anchors,
and instructions that minimise confusion.

Interviews

Interviews allow the researcher to explore experiences in depth. They may be structured, semi-structured, or unstructured. In honours research, semi-structured interviews are often especially useful because they balance consistency with flexibility. The interviewer follows a guide but can probe interesting responses.

Interviews are valuable when the researcher wants nuance, meaning, or narrative. However, they require skill, time, and careful reflexivity. The interviewer’s presence can shape what participants disclose.

Focus groups

Focus groups gather small groups of participants for guided discussion. They are useful for exploring social norms, shared perceptions, and group interaction. One advantage is that participants may generate ideas that would not emerge in individual interviews. One challenge is that dominant voices can silence quieter members, so facilitation matters.

Observation

Observation involves systematically watching behaviour in natural or controlled settings. It can be overt or covert, participant or non-participant. Observation is useful when self-report may be unreliable or when behaviour itself is the focus. However, observers may influence behaviour, and interpretation can be subjective without clear coding schemes.

Document and archival analysis

Researchers may analyse existing documents, policies, diaries, media texts, reports, or administrative records. This is useful when studying historical or institutional patterns. The challenge is that the material was often created for a purpose other than research, which may affect completeness or accuracy.

Ethical principles in psychological research

Ethics is not an optional extra; it is integral to responsible methodology. The standard ethical principles include:

Respect for persons: participants must be treated as autonomous agents
Beneficence: maximise benefits and minimise harm
Justice: distribute burdens and benefits fairly
Confidentiality and privacy: protect participant information
Informed consent: ensure participants understand the study
Right to withdraw: participation must be voluntary
Debriefing: explain the study after participation when needed

In psychology, ethical issues often arise because research may involve sensitive experiences such as trauma, substance use, abuse, mental health, sexuality, family conflict, or discrimination. A researcher must consider whether asking a question might cause distress, whether support resources are available, and whether the participant’s privacy could be compromised.

Informed consent in practice

Informed consent means more than signing a form. Participants should understand:

the purpose of the study,
what they will be asked to do,
possible risks and benefits,
how data will be used,
whether anonymity or confidentiality is guaranteed,
and that participation is voluntary.

For online research, consent may be embedded in a digital information sheet, but the principle is the same. In South African educational settings, language accessibility is important. Consent documents should be understandable to the target participants, not merely technically correct.

Confidentiality, anonymity, and data protection

Confidentiality means that participant information is not disclosed improperly. Anonymity means that the researcher cannot link data to a participant’s identity. These are related but not identical. A study may be confidential without being fully anonymous, especially if identifiers are stored separately from data.

With digital data, protection becomes even more important. Files should be stored securely, access should be limited, and identifiers should be removed where possible. In academic settings, researchers must be especially careful when reporting small samples or distinctive cases, because descriptions may inadvertently reveal identities.

Quality criteria in quantitative and qualitative research

Quantitative studies are judged using criteria such as:

reliability,
validity,
internal validity,
external validity,
and statistical conclusion validity.

Qualitative studies are often evaluated using:

credibility
transferability
dependability
confirmability

Credibility asks whether the findings are believable. Transferability concerns whether insights may apply to similar contexts. Dependability asks whether the process is logical and documented. Confirmability focuses on whether findings are shaped by the data rather than by researcher bias.

Managing quality through procedure

Quality is strengthened through practical steps such as:

Piloting instruments before full data collection
Using clear instructions
Training interviewers or observers
Standardising procedures
Keeping detailed records
Reflecting on positionality and bias
Triangulating data where appropriate

For example, if a study measures student stress using a survey and interviews, the researcher may compare the numerical trends with the qualitative explanations. If the two forms of data point in different directions, that divergence becomes analytically important rather than a problem to hide.

5. Data Analysis, Interpretation, and Exam Strategy for HMPYC80

The final stage of research methodology is making sense of data. Analysis is not simply a technical afterthought; it is the process through which evidence becomes knowledge. HMPYC80 students are often assessed on whether they can identify the correct analytical logic for a study, interpret results carefully, and avoid overclaiming.

Quantitative data analysis

Quantitative analysis begins by organising and cleaning data. This may involve checking for missing values, coding responses, identifying outliers, and ensuring that variables are labelled correctly. Once the dataset is prepared, the researcher uses descriptive and inferential statistics.

Descriptive statistics

These summarise the data.

Mean: average score
Median: middle score
Mode: most frequent score
Standard deviation: spread around the mean
Range: difference between highest and lowest scores
Frequencies and percentages: useful for categorical data

For example, if a student stress survey shows an average score of 32 with a standard deviation of 8, that tells us the typical level and the spread of responses. But the mean alone does not show whether the distribution is skewed or whether there are subgroups with different patterns.

Inferential statistics

Inferential statistics help researchers draw conclusions from a sample about a population or test whether observed differences are likely due to chance. Common tests include:

t-tests for comparing two groups
ANOVA for comparing more than two groups
Chi-square tests for categorical variables
Correlation coefficients for association
Regression analysis for prediction
Non-parametric tests when assumptions are not met

The choice of test depends on the research question, variable types, and assumptions. A frequent exam error is to memorise test names without understanding why they are used. The better answer explains the logic: if the question compares average anxiety scores between men and women, a t-test may be appropriate; if it examines the relationship between hours of sleep and anxiety scores, correlation or regression may be more suitable.

Interpreting statistical significance

Statistical significance indicates whether an observed result is unlikely to have occurred by chance under a null hypothesis. However, significance does not automatically mean practical importance. A very small effect can be statistically significant in a large sample, while a meaningful effect may not reach significance in a small sample.

Students should therefore distinguish between:

statistical significance
effect size
practical significance

An effect size tells us how large or meaningful the relationship or difference is. In psychology, this is crucial because trivial effects can be overstated when only p-values are considered.

Qualitative data analysis

Qualitative analysis is typically iterative and interpretive. The researcher reads the data repeatedly, codes meaningful segments, and develops themes or conceptual categories. Common approaches include:

Thematic analysis
Content analysis
Narrative analysis
Phenomenological analysis
Grounded theory coding

A basic thematic analysis process can be summarised as follows:

Familiarise yourself with the data
Generate initial codes
Search for patterns across codes
Develop themes
Review and refine themes
Define and name themes
Write up findings with evidence from the data

Qualitative findings are strengthened by rich excerpts, careful explanation, and transparency about how themes were developed. The analyst should not force the data into preconceived categories without justification.

Mixed-methods integration

In mixed methods, the analytical challenge is not just analysing two sets of data separately, but integrating them. Integration may happen by:

merging datasets,
connecting one phase to the next,
or embedding one type of data inside another.

For example, a student mental health study might find high survey scores for stress among final-year students. Interviews may then reveal that the stress is driven by financial pressure, uncertainty about employment, and family responsibilities. The integration of numbers and narratives produces a fuller picture than either alone.

How to answer HMPYC80 exam questions well

Exams in research methodology often test application rather than memorisation. Strong answers usually do the following:

define the key concept accurately,
apply it to the scenario,
explain why it matters,
identify limitations or alternatives,
and use correct methodological vocabulary.

A useful exam structure is:

State the concept or design
Explain it clearly
Apply it to the example
Discuss strengths and limitations
Conclude with methodological judgement

For instance, if asked whether a cross-sectional design can establish causality, a good answer would explain that it cannot determine temporal precedence, because variables are measured at one point in time. It may identify association, but not cause and effect. The answer should also mention that longitudinal or experimental designs are better suited to causal questions, though they have their own practical and ethical constraints.

Common mistakes to avoid

Students often lose marks because they:

confuse methodology with methods,
claim causation from correlational data,
ignore sampling limitations,
misuse statistical terminology,
treat validity and reliability as the same thing,
oversimplify qualitative analysis,
or make ethical claims without explaining the issue.

Avoiding these mistakes requires disciplined reading and careful terminology. In psychology, precision is not optional. A well-chosen word can completely change the interpretation of a finding.

Final revision themes for HMPYC80

The most important revision clusters for this module are:

the difference between paradigms, methods, and methodology
the strengths and limits of quantitative, qualitative, and mixed methods
the logic of research design and causal inference
the role of sampling in generalisation
the meaning of operationalisation, reliability, and validity
the ethical principles governing psychological research
the interpretation of statistical and thematic findings
the ability to critique a study’s fit, credibility, and limitations

A well-prepared student should be able to read a scenario and identify what approach is being used, what the design can legitimately conclude, what threats to quality exist, and how the study could be improved. That ability is the real purpose of research methodology in honours-level psychology: not merely to pass an exam, but to think like a careful psychological researcher.

6. High-Yield Revision Tables, Case Examples, and Exam Phrasing

Methodology becomes easier to remember when it is tied to concrete examples. In HMPYC80, exam questions often present a short research scenario and ask students to identify the design, evaluate the method, or propose improvements. The following examples help anchor the theory in practical psychological research.

Example 1: Student stress and academic workload

Suppose a researcher wants to investigate whether workload is related to stress among honours psychology students at a South African university. The researcher distributes an online questionnaire to 120 students and measures self-reported stress, hours spent studying, employment status, and perceived support.

This is most likely a correlational, cross-sectional, quantitative study. It can show whether workload and stress are associated, but it cannot prove that workload causes stress. The sample may be limited if only students who regularly check email respond. If the questionnaire is not validated, reliability and validity may be weak. If the sample includes mostly women or mostly urban students, generalisation may also be limited.

A stronger version of the study might use stratified sampling, a validated stress scale, and perhaps follow-up interviews with a smaller subset of participants. That would provide both statistical patterns and contextual detail.

Example 2: Coping after family loss

A researcher wants to understand how young adults cope after losing a parent. Because the issue is deeply personal and underexplored, the researcher conducts semi-structured interviews with 15 participants.

This is a qualitative study, probably using phenomenological or thematic analysis. The goal is depth, not statistical generalisation. The sample is small but appropriate if the participants are chosen purposively because they have direct experience of the phenomenon. Ethical sensitivity is essential because the topic is emotionally difficult. The interviewer should be trained to handle distress, and referral information should be available if needed.

Example 3: Mindfulness intervention

A lecturer wants to test whether a mindfulness programme reduces examination anxiety. One group of students participates in the programme, while another group does not. Anxiety is measured before and after the intervention.

This is an experimental or quasi-experimental design, depending on whether participants are randomly assigned. If random assignment is used, the researcher has stronger grounds for causal inference. If groups are pre-existing, selection bias becomes a concern. Pre-test and post-test measures help show change over time, but the study must also consider confounding variables such as previous mindfulness experience or differences in workload.

Example 4: Community attitudes toward mental health

A local NGO wants to understand how community members perceive depression and help-seeking. A focus group approach may be suitable because it captures shared beliefs, social stigma, and group norms. The analysis may reveal that people understand depression through moral, spiritual, or biomedical lenses. This would be difficult to capture through a survey alone. However, focus groups also risk conformity pressure; participants may avoid expressing unpopular views in front of others.

Revision table: concept, meaning, and exam clue

Concept	Meaning	Exam clue
Methodology	Logic and justification of research choices	“Why this approach?”
Method	Specific data collection/analysis tool	“How was data collected?”
Validity	Accuracy of measurement or inference	“Does it measure what it claims?”
Reliability	Consistency of measurement	“Would it give stable results?”
Population	Entire group of interest	“To whom do the findings apply?”
Sample	Actual participants studied	“Who took part?”
Operationalisation	Turning abstract ideas into measures	“How was stress measured?”
Correlation	Relationship between variables	“Are the variables linked?”
Causation	One variable produces change in another	“Did X cause Y?”
Thematic analysis	Identifying patterns in qualitative data	“What themes emerged?”

Strong exam phrasing

Using disciplined language improves marks. Compare the weak and strong versions below.

Weak:
“The study proves that stress causes poor marks.”

Stronger:
“The study suggests an association between stress and academic performance, but because the design is cross-sectional, causal conclusions cannot be drawn.”

Weak:
“The sample is good because it is many students.”

Stronger:
“The sample size is acceptable, but representativeness depends on whether the selected students reflect the broader target population.”

Weak:
“Qualitative research is subjective.”

Stronger:
“Qualitative research is interpretive, and its trustworthiness depends on systematic analysis, transparency, and reflexive attention to the researcher’s role.”

Final condensed checklist for revision

Before an HMPYC80 exam, check that you can do the following:

Distinguish methodology, method, and design
Identify when a study is quantitative, qualitative, or mixed methods
Explain why certain sampling strategies are used
Define and apply operationalisation
Differentiate reliability from validity
Recognise ethical concerns in psychological research
Interpret the limits of correlation, cross-sectional, and quasi-experimental designs
Describe the basic logic of thematic and statistical analysis
Critically evaluate a study’s strengths and weaknesses
Write answers with methodological precision and confidence

Research methodology is one of the most important pillars of psychology because it determines whether evidence is meaningful, defensible, and useful. For HMPYC80, success depends on thinking carefully about the connection between question, design, data, ethics, and interpretation. A student who masters those connections is not only prepared for the exam, but also prepared for advanced psychological research in a demanding and diverse South African context.