Fundamentals
Empirical Testing
how scientific knowledge is primarily acquired—through direct observation and experimentation, rather than just theory.
Independent Variable
The variable that an experimenter manipulates or changes to test its effect on an outcome.
Nominal Data
Categorical data where items are in different categories without any rank order or quantitative difference between them. An example would be political party affiliation
ABAB Design (Withdrawal Design)
An experimental single-subject design involving repeated alternation between baseline (A) and treatment (B) phases. This design aims to demonstrate a causal link by observing changes in the dependent variable with the introduction and removal of treatment.
Case Study
a qualitative research approach involving a detailed description of a single entity, which could be an individual, a clinic, or a program. It typically moves beyond mere description to interpret what is observed within that single unit.
Replication
If a study's results cannot be replicated, their validity is significantly questioned.
Dependent Variable
The variable that is measured or observed, where changes are expected as a result of the independent variable's manipulation.
Ordinal Data
Data that has an order or rank, but the difference between each level is not necessarily equal. An example is finishing first, second, or third in a race, where the time difference between positions can vary.
ANCOVA (Analysis of Covariance)
A statistical analysis extending ANOVA by including one or more covariates to statistically adjust for initial group differences or to remove the influence of an uncontrolled variable. It aims to equate groups on a specific variable when randomization is imperfect.
Grounded Theory
a qualitative research approach that aims to develop a theory directly from the data collected, rather than testing a pre-existing hypothesis. The theory emerges incrementally through the "constant comparative method" as the study progresses.
Probabilistic Knowledge
Scientific findings are rarely absolute certainties, but rather understandings expressed in terms of likelihoods and probabilities. It shapes how we interpret and apply research outcomes.
Confounding Variable
Factors that might affect the outcome (dependent variable) but are not the independent variable and whose influence needs to be controlled.
Gain Scores
Scores calculated by subtracting a participant's pre-treatment score from their post-treatment score, used to quantify individual improvement. These scores can be analyzed in some between-subject designs.
Baseline (Phase A)
The initial phase in single-subject designs where the dependent variable is repeatedly measured before any intervention is introduced. It serves as a control condition, establishing the natural behavior or trend of the dependent variable without treatment.
Member (Participant) Checking
a trustworthiness strategy in qualitative research where researchers take their interpretations, summaries, or developed themes back to the participants to verify their accuracy. This ensures the findings reflect the participants' lived experiences.
Testing Effects
The more you test the person the more they change their behaviour while being tested
Internal Validity
The degree of confidence in a study's findings, indicating how well the independent variable is demonstrated to be responsible for the observed changes in the dependent variable.
Critical Value
A value from a statistical distribution that a test statistic must exceed (or fall below) to be considered statistically significant at a chosen alpha level and given degrees of freedom.
Interaction Effect
Occurs in factorial designs when the effect of one independent variable on the dependent variable changes depending on the level of another independent variable. The combined effect is different from the sum of their individual effects.
Phenomenology
A qualitative research approach typically used when little is known about a phenomenon, aiming to give voice to the people being studied and draw meaning from their complex realities and lived experiences. Unstructured interviews are its primary tool.
Quasi-experimental Designs
where full randomization isn't possible, offering practical methods to investigate causal relationships while acknowledging inherent limitations compared to true experiments.
External Validity
How well the findings from a study can be generalized or extended to the real world and the population of interest.
Hawthorne Effect
A phenomenon where participants' behavior changes simply because they are aware of being part of a study or being observed. Control groups are often used to account for this effect.
Multiple Baseline Design
An experimental single-subject design that demonstrates treatment effectiveness by staggering the introduction of an intervention across different baselines, such as different behaviors, participants, or settings.
Saturation
A trustworthiness strategy in qualitative research where data collection continues until no new information, themes, or insights emerge from additional participants or data sources. It indicates that a comprehensive understanding of the phenomenon has been achieved.