Mod 4: Quantitative Research

What is a population?

Any group of interest that can be generalized to (i.e. grade school children in Atlanta, GA)

What is a sample?

A subset of the population selected to represent the larger group

What is a variable?

Anything that can change or vary between people or situations and can be measured

What is a construct?

An abstract concept that cannot be directly measured (e.g., happiness, anxiety)

What is an operational definition?

The exact way a construct is measured (e.g., score on happiness questionnaire)

Example of operational definition for happiness?

Participant's score on the Measure of Happiness scale, where higher scores mean greater happiness

Example of operational definition for anxiety?

Participant's score on the Patient Health Questionnaire, where lower scores mean lower anxiety

What is an independent variable (IV)?

The variable that influences the outcome or is being tested

What is a dependent variable (DV)?

The outcome variable that is influenced by the IV

Example of IV in OT study?

OT social support group

Example of DV in OT study?

Self-efficacy

What is a discrete variable?

A variable measured in whole units (categorical—ordinal or nominal, e.g., gender, letter grade)

What is a continuous variable?

A variable measured along a continuum and can include fractions (e.g., weight, score, duration)

What are parametric tests used for?

Tests with continuous data, normal distribution, large sample size, comparing means

Examples of parametric tests

T-tests, ANOVA

What are nonparametric tests used for?

Tests with small samples, any distribution, ordinal/nominal data, comparing medians

Examples of nonparametric tests

Mann-Whitney, Wilcoxon signed-rank

What is the purpose of experimental quantitative design?

To test the effect or efficacy of an intervention

What are the three main types of experimental designs?

True, quasi, single-subject

What are the exploratory quantitative designs?

Correlational, predictive, survey

Why use experimental designs?

To determine if an intervention is effective or replicable for other populations

What is a true experimental design?

Participants are randomly assigned to groups, includes control group(s)

What is a quasi-experimental design?

No random assignment

groups may not be equivalent

What is a single-subject design?

Focuses on one individual to observe changes across conditions

In a true experiment, what does "between-subjects" mean?

Different participants in each condition

In a true experiment, what does "within-subjects" mean?

Same participants experience all conditions

What is a randomized block design?

Participants are grouped by characteristics, then randomized to treatments

What is a pretest-posttest control design?

Measures before and after an intervention compared to control group

What is a posttest-only control design?

Only measures outcomes after the intervention

What is a multifactor (independent) design?

Examines effects of two or more IVs on one DV

What is a repeated measures design?

Participants serve as their own control and are tested under all conditions

What is a crossover design?

Participants receive multiple treatments in different orders to compare effects

What is an advantage of repeated measures design?

Controls for individual differences

What is a disadvantage of repeated measures design?

Practice or carryover effects

When should crossover designs be avoided?

When treatment effects are long-lasting or slow to appear

What are the characteristics of a quasi-experimental design?

No random assignment, uses existing groups, still tests intervention effects

What is a one-group pretest-posttest design?

Measures same group before and after intervention without control

What is a time series design?

Multiple measurements over time to observe treatment effects

What is a non-equivalent pretest-posttest control design?

Uses two groups without randomization, both measured before and after

Why do OTs like quasi-experimental designs?

They're practical in clinical and community settings

Who else commonly uses quasi-experimental designs?

Educators and healthcare researchers

In a two-way factorial design, what is being tested?

Effects of two IVs and their interaction on one DV

Example of two IVs in a study

Exercise intensity (moderate/vigorous) and exercise location (home/community)

What is the DV in a two-way factorial design example?

Self-efficacy

What does "parametric" generally refer to?

Tests using continuous data and normal distributions

What does "nonparametric" generally refer to?

Tests using categorical data and non-normal distributions

What is a correlational design?

Examines relationships between variables

What is a predictive design?

Uses one variable to predict another

What is a survey design?

Collects self-report data from a large sample