Research Designs & Statistics Vocabulary

A comprehensive set of vocabulary flashcards covering key research design and statistical concepts drawn from the lecture notes.

Descriptive Study

Research that describes characteristics of a population or phenomenon without investigating cause-and-effect relationships (e.g., surveys, qualitative studies).

Survey

A descriptive research method that collects self-reported data from participants through questionnaires or interviews.

Qualitative Research

Investigates participants’ experiences or perceptions using non-numeric data such as words, images, or observations.

Observational Study

Research in which the investigator observes outcomes without manipulating variables (e.g., cohort studies).

Cohort Study

An observational design that follows a group with shared characteristics over time to assess outcomes.

Correlational Study

Examines the statistical relationship between two or more variables without inferring causation.

Comparative Study

Compares two or more groups or conditions to evaluate differences in outcomes.

Prospective Study

Collects data forward in time from the point of study initiation.

Retrospective Study

Analyzes existing data or past records to investigate outcomes that have already occurred.

Experimental Study

Research in which the investigator manipulates an independent variable and uses random assignment to establish cause-and-effect.

True Experimental Design

Includes random assignment, control groups, and manipulation of variables.

Quasi-Experimental Design

Lacks random assignment but still involves an intervention or manipulation.

Single-Subject Design

Research in which one participant (or small number) serves as both control and experimental unit; often includes blinding of participant and assessor.

Sample

The group of individuals actually studied, intended to represent the target population.

Randomization

Process of assigning participants to groups by chance to reduce bias.

Blinding

Concealing group assignment from participants, investigators, or assessors to minimize bias.

Single-Blinding

Only the participant (or assessor) is unaware of group assignment.

Double-Blinding

Both participants and investigators/assessors are unaware of group assignment.

Intention-to-Treat Analysis

Statistical approach that analyzes participants in the groups to which they were originally assigned regardless of adherence.

Convenience Sample

Participants are selected based on ease of access rather than random procedures.

Cross-Sectional Study

Collects data from a population at one specific point in time.

Longitudinal Study

Follows the same participants over an extended period to observe changes.

Random Sample

Every member of the population has an equal chance of being selected for the study

Random Assignment

Randomly allocates study participants to different interventions or groups. Ensures that all groups will be similar in the characteristics that are thought to affect outcome before the treatment begins

Variable

Any characteristic or factor that can vary and be measured or manipulated.

Equivalent Samples

Groups that are similar on key characteristics at baseline, reducing confounding.

Statistical Power

Probability that a test will detect a true difference when one exists.

Patient-Reported Outcome Measure (PROM)

Instrument in which patients provide information about their own health status.

Objective Measure

Outcome assessed through direct observation or devices, minimizing subjectivity.

Subjective Measure

Outcome based on personal judgment or patient self-report.

Validity

Extent to which a measure accurately reflects the concept it intends to assess.

Content Validity

Degree to which a test covers all relevant facets of the construct.

Construct Validity

Degree to which a test truly measures the theoretical construct it claims to assess.

Reliability

Consistency or repeatability of a measurement.

Index Test

The diagnostic test under investigation.

Gold Standard

The best available method against which new tests are compared.

Nominal Data

Categorical data with no inherent order (e.g., yes/no, blood type).

Ordinal Data

Categorical data with a ranked order but unequal intervals (e.g., pain scale 0-10).

Interval Data

Numeric data with equal intervals and no true zero (e.g., temperature in °C).

Ratio Data

Numeric data with equal intervals and an absolute zero (e.g., weight, time).

Mean (x̄)

Arithmetic average of a set of values.

Median

Middle score in an ordered data set.

Mode

Most frequently occurring value in a data set.

Range

Difference between highest and lowest values in a data set.

Skew

Asymmetry in a distribution; can be positive (right) or negative (left).

Normal Distribution

Symmetrical, bell-shaped distribution where mean = median = mode.

Non-Parametric Data

Data that do not meet parametric test assumptions (normality, equal variance). abnormal distribution of data

Hypothesis Testing

Statistical procedure to decide whether to reject a null hypothesis.

Null Hypothesis (H0)

Statement that there is no effect or difference between groups.

Directional/Research Hypothesis

Predicts the existence and direction of an effect (e.g., treatment A > treatment B).

Standard Error of Measurement (SEM)

Estimate of the precision of individual scores; reflects measurement error.

Standard Deviation (SD)

Measure of variability indicating how much scores differ from the mean. one representation of the variability in a sample

Confidence Interval (CI)

Range of values that include the real or true mean for a population of people.

A Priori Power Calculation

Sample size estimation performed before data collection to ensure adequate power.

Alpha Level (α)

Threshold probability for rejecting null hypothesis, commonly set at 0.05.

Pearson Product-Moment Correlation

Measures linear relationship between two continuous variables (r).

Intraclass Correlation Coefficient (ICC)

Reliability statistic that accounts for agreement and absolute values for continuous data.

Kappa (κ)

Statistic measuring agreement for categorical data beyond chance.

Spearman’s Rho (ρ)

Non-parametric correlation for ranked/ordinal data.

Regression

Statistical method for predicting an outcome based on one or more predictors.

Chi-Square Test (χ²)

Analyzes differences in categorical (nominal) data frequencies.

Mann-Whitney U Test

Non-parametric test comparing two independent groups on ordinal/continuous data.

Wilcoxon Signed-Rank Test

Non-parametric test comparing two related samples on ordinal/continuous data.

Kruskal-Wallis Test

Non-parametric alternative to one-way ANOVA for three or more independent groups.

Friedman Test

Non-parametric alternative to repeated-measures ANOVA for ordinal/continuous data.

T-Test

Parametric test comparing means of two groups (independent or paired).

ANOVA (Analysis of Variance)

Parametric test comparing means of three or more groups.

Repeated-Measures ANOVA

Compares multiple measurements taken from the same participants across time or conditions.

Sensitivity (SnNOut)

The proportion of people who actually have the condition who test POSITIVE. Ability of a test to correctly identify those with the condition;

False Negative

Test result that erroneously indicates absence of a condition when it is present.

Specificity (SpPIn)

The proportion of people who are free of the condition who test NEGATIVE Ability of a test to correctly identify those without the condition;

False Positive

Test result that erroneously indicates presence of a condition when it is absent.

Positive Predictive Value (PPV)

The proportion of people with a positive index test who actually have the condition of interest. Probability that a person with a positive test truly has the condition.

Negative Predictive Value (NPV)

The proportion of people with a negative index test who are actually free of the condition of index. Probability that a person with a negative test truly does not have the condition.

Receiver Operating Characteristic (ROC) Curve

Graph plotting sensitivity vs (1-specificity) across test thresholds.

Likelihood Ratio

Ratio expressing how much a test result will change the odds of having a disease.

Nomogram

Graphical tool that uses likelihood ratios to convert pre-test to post-test probabilities.

Risk Ratio (Relative Risk)

Ratio of event probability in exposed vs unexposed groups.

Odds Ratio

Ratio of odds of an event occurring in one group to the odds in another group.

Celeration Line

Trend line used in single-subject designs to visualize rate of change over time.

P-Value

Probability of observing data as extreme as those collected if the null hypothesis is true.

Effect Size

Quantitative measure of the magnitude of a difference or relationship (e.g., Cohen’s d).

Risk of Bias

Likelihood that study design, conduct, or analysis systematically alters results.

Type I Error (α Error)

False positive; concluding a significant difference exists when it does not.

Type II Error (β Error)

False negative; failing to detect a real difference due to low power.

Clinical Meaningfulness

Extent to which study results have practical importance for patient care.

Minimal Detectable Change (MDC)

Smallest amount of change that exceeds measurement error.

Minimal Clinically Important Difference (MCID)

Smallest change in an outcome perceived as beneficial by patients.

Peer-Reviewed Publication

Article evaluated by independent experts before publication.

Primary Study

Original research collecting and analyzing new data.

Psychometric Study

Research evaluating measurement properties such as reliability and validity.

Case Report

Detailed description of a single patient’s clinical presentation and outcomes.

Case-Control Study

Observational design comparing individuals with a condition to those without to identify risk factors.

Observational Study

Research in which investigators assess outcomes without intervention.

Survey Study

Research collecting data through questionnaires to describe characteristics or opinions.

Qualitative Method – Photovoice

Participants use photographs to document and discuss their experiences.

Mixed Methods Research

Combines quantitative and qualitative approaches in one study.

Randomized Controlled Trial (RCT)

Participants are randomly allocated to intervention or control groups to test efficacy.

Quasi-Experimental Study

Intervention study without full randomization or control conditions.

Literature Review

Narrative summary of existing research on a topic.