Research Methods Ch. 10-13

Factorial design

involve manipulating tow or more independent variables simultaneously to observe their effects on one or more dependent variables

What is the simplest factorial design?

2 × 2

Main effect

impact of the independent variable alone on the dependent variable, ignores the other independent variable

Interactions

effect of one independent variable depends on the level of another independent variable

A-B-A design

baseline phase, intervention, return to baseline

First A phase

initial observations and measurements of target behavior prior to intervention

B phase

intervention or treatment give and the behavior measured for noticeable effect

Second A phase

intervention withdrawn while still measuring behavior

Multiple Baseline Design

single-case experiment that assesses the effect of intervention across different subjects, settings, or behaviors without needing to withdraw the intervention

How are Quasi Experimental Designs different from Experimental Designs?

Quasi Experimental Designs used pre-existing or non-random groups while Experimental Designs use true random assignment

1-Shot Case Study

A single group of participants is observed only after treatment or intervention is applied with no control group or baseline measurement

1 Group Pretest Posttest Design

A single group of participants measured before and after the intervention or treatment is applied

Alternative explanations

random chance error, confounding variables, reverse causality

Descriptive Statistics

organize, simplify, and present the data

Inferential Statistics

generalize from samples to populations, hypothesis testing, relationships among variables- to make predictions

Population

all subjects available to be chosen for an experiment or study

Sample

Subjects chosen to participate in the experiment or study

Null Hypothesis

population means are equal and differences are due to random error

Research Hypothesis

population means aren’t equal

Alpha Level

probability for statistically significant results

Effect Size

measure of the magnitude of difference between groups or the strength of the relationship

Type 1 Error

we reject the null when it’s actually true- false positive

Type 2 Error

we fail to reject the null when the research hypothesis is true- false negative