1.2 Research Methods in Educational Psychology

Research Approaches

Quantitative and Qualitative Research Methods

Research in educational psychology falls into three broad categories, each suited to different kinds of questions.

Quantitative research collects and analyzes numerical data to test hypotheses, identify trends, and make generalizations. Think large-scale surveys or standardized test scores. Because the data are numerical, you can run statistical analyses and compare groups with precision.

Qualitative research gathers non-numerical data to build an in-depth understanding of experiences, perspectives, and processes. The tools here are interviews, observations, and focus groups. Instead of producing a number, qualitative work produces rich descriptions that capture context and meaning.

Mixed methods research combines both approaches in a single study. The goal is to leverage the strengths of each: quantitative data can show what is happening across a large sample, while qualitative data can explain why it's happening.

When Each Approach Fits

• Quantitative research works well when you need to establish cause-and-effect relationships, compare groups, or predict outcomes. For example, you might measure whether a new teaching method improves test scores across 500 students.
• Qualitative research is a better fit for exploring complex issues or understanding individual experiences. For example, interviewing students about how they perceive classroom climate reveals nuances a survey might miss.
• Mixed methods research is useful when you want to triangulate findings or use one type of data to explain the other. A researcher might conduct interviews first to identify key themes, then build a survey based on those themes to test the findings at scale.

Research Designs

Experimental and Quasi-Experimental Designs

Experimental design is the gold standard for establishing causation. The researcher manipulates an independent variable, measures its effect on a dependent variable, and controls for other factors. The defining feature is random assignment: participants are randomly placed into treatment or control groups, which helps rule out pre-existing differences between groups.

Quasi-experimental design follows the same logic but without random assignment. This is far more common in educational settings because you usually can't randomly shuffle students between classrooms. Instead, you compare intact groups (like two existing class sections taught with different methods). The tradeoff is that you have less confidence that differences in outcomes are caused by the treatment rather than by pre-existing group differences.

Non-Experimental Designs

Non-experimental designs don't involve manipulating variables at all. They describe, explore, or identify relationships.

• Correlational studies examine the relationship between two or more variables without manipulating them. They can tell you that parental involvement and student motivation tend to go together, but they cannot tell you that one causes the other. This is the classic "correlation does not equal causation" situation.
• Case studies involve in-depth exploration of a single case or a small number of cases within their real-life context. A researcher might spend months documenting how one school implements a new curriculum, capturing details that broader studies would miss.
• Surveys collect data from a sample of individuals using questionnaires or interviews. They're efficient for gathering information about characteristics, opinions, or behaviors across many people, such as assessing teachers' attitudes toward professional development.

Longitudinal and Cross-Sectional Designs

These two designs differ in how they handle time.

Longitudinal research follows the same participants over an extended period, collecting data at multiple points. This lets you track change and development directly. For example, following a cohort of students from kindergarten through high school reveals how academic growth unfolds over time. The downside: longitudinal studies are expensive and participants may drop out along the way (a problem called attrition).

Cross-sectional research collects data from different groups at a single point in time. Comparing the reading skills of students in grades 3, 5, and 7 all at once is faster and cheaper than waiting years. However, the groups may differ in ways beyond just age or grade level, so you can't be as confident about developmental conclusions.

Research Quality and Standards

Validity and Reliability

Two concepts determine whether a study's findings are worth trusting.

Validity is the extent to which a study actually measures what it claims to measure and the accuracy of the conclusions drawn. If a test is supposed to assess reading comprehension but mostly tests vocabulary, it has a validity problem. There are several types (content validity, construct validity, internal validity, external validity), but the core question is always: Are we measuring the right thing, and are our conclusions justified?

Reliability is about consistency. A reliable measure produces stable, repeatable results. If two different observers use the same classroom observation protocol and get similar ratings, that protocol has good inter-rater reliability. If a student takes the same test twice under similar conditions and scores similarly, the test has good test-retest reliability.

Both matter. A measure can be reliable without being valid (it consistently measures the wrong thing), but it cannot be valid without being reliable. Researchers need to establish both for their instruments and procedures before their findings can be considered credible.

Ethical Considerations in Educational Research

Educational research involves real people, often children, so ethical standards are non-negotiable. The core principles include:

• Informed consent: Participants (or their parents/guardians for minors) must understand the study's purpose, procedures, and any potential risks before agreeing to participate.
• Confidentiality: Researchers must protect participants' identities and personal data.
• Minimizing harm: Studies should not expose participants to unnecessary physical or psychological risk.
• Transparency: Researchers should be honest about their methods and report findings accurately, including results that don't support their hypothesis.

Institutional Review Boards (IRBs) review research proposals before data collection begins. Their job is to ensure the study meets ethical standards and that participant protections are adequate. Any study involving human participants at a university or research institution typically requires IRB approval.