1. Quantitative Designs
Quantitative research tests hypotheses using numerical data and statistical analysis. It assumes a largely objective, measurable reality.
Experimental designs
The researcher manipulates an independent variable and randomly assigns participants to groups (treatment vs. control) to establish causality. Random assignment is the defining feature — it’s what allows the researcher to rule out confounding variables and make causal claims.
Quasi-experimental designs
Similar to experimental designs (there’s manipulation of a variable and comparison groups), but without random assignment. Common in field settings where randomization isn’t ethical or feasible (e.g., comparing two existing classrooms or hospital units). Causal claims are weaker than true experiments because of possible selection bias.
Nonexperimental designs
No manipulation of variables at all. These include:
Correlational studies — examining relationships between variables without implying causation
Comparative studies — comparing naturally occurring groups
Survey research — describing attitudes, behaviors, or characteristics of a population at a point in time
Treatment-control designs
A specific type of experimental/quasi-experimental structure where one group receives an intervention and another (the control) does not, allowing comparison of outcomes.
Repeated measures designs
The same participants are measured multiple times (e.g., pre-test/post-test/follow-up), which controls for individual differences and increases statistical power, but introduces concerns like practice effects and attrition.
Causal-comparative (ex post facto) designs
The researcher examines a variable that has already occurred (e.g., a naturally occurring group difference — smokers vs. non-smokers) and looks for effects, without manipulating anything. This is often confused with quasi-experimental designs, but the key distinction is that the “cause” already exists in the world before the study begins — it isn’t assigned by the researcher at all.
Single-subject designs
Intensive study of one participant (or a small number, studied individually) over time, often with repeated baseline and intervention phases (A-B-A-B designs are common). Frequently used in clinical, educational, and behavioral fields where individual-level change is the focus.
Predictive studies
Use existing variables to predict future outcomes or group membership (e.g., regression models predicting student retention from GPA and attendance). Focus is on prediction rather than explanation of causal mechanism.
2. Qualitative Designs
Qualitative research explores meaning, experience, and context, typically through non-numerical data (interviews, observations, documents). It assumes multiple, socially constructed realities.
Ethnography
Immersive, prolonged study of a culture or social group in its natural setting, typically through participant observation and fieldwork, to understand shared patterns of behavior, language, and beliefs.
Case study
In-depth exploration of a single case (an individual, organization, event, or program) or a small number of cases, bounded by time and place, using multiple sources of data (interviews, documents, observations) to build a rich, contextualized understanding.
Grounded theory
Systematic method for developing a theory that is “grounded” in data collected from the field, using iterative coding processes (open, axial, selective coding) and constant comparison, rather than testing an existing theory.
Narrative inquiry
Focuses on individuals’ stories and lived experiences told in narrative form, examining how people construct meaning through storytelling, often organized chronologically or thematically.
Phenomenological research
Explores the lived experience of a phenomenon from the perspective of those who experienced it, seeking to identify the essence or shared meaning across participants (e.g., what it’s like to live with chronic illness).
Policy analysis
Systematic examination of policies — their formation, implementation, or impact — often using document analysis, interviews with stakeholders, and contextual/historical analysis to evaluate effectiveness or inform policy decisions.
Other qualitative traditions might include content analysis, discourse analysis, action research, or historical research, depending on the field.
3. Mixed Methods, Primarily Quantitative
Combines both quantitative and qualitative data, but the study’s design, questions, and interpretation are anchored in the quantitative strand, with qualitative data playing a supporting role.
Sequential — one method follows the other (e.g., a quantitative survey followed by qualitative interviews to explain unexpected results), with quantitative data collected first or given priority in interpretation.
Concurrent — both quantitative and qualitative data collected around the same time, then merged, but quantitative results drive the primary conclusions.
Transformative — organized around a social justice or advocacy framework (e.g., examining inequities), using quantitative data as the primary lens while qualitative data adds depth or amplifies participant voice.
4. Mixed Methods, Primarily Qualitative
Same structural options (sequential, concurrent, transformative), but the qualitative strand is the main driver of the research questions and interpretation, with quantitative data playing a supporting or contextualizing role (e.g., a phenomenological study supplemented by a small survey to show prevalence of a theme).