Research & Stats Exam II

How do you determine how many groups there are in a 2×3×2 experiment?

You multiply the numbers together

2×3×2 = 12

2×2 Experiment

An experimental design with 2 independent variables, each having 2 levels.

3×3 Experiment

An experimental design with 2 independent variables, each having 3 levels.

Main effect

Does this one IV effect the outcome overall

Interactions

When the effect of one IV on the DV depends on the level of another IV

Quasi- experimental designs

Only partial control over the IV

Conditions assigned without randomization

What makes something a quasi experiment vs a true experiment?

Random assignment

Causation and quasi experimental designs

Causation means that one variable directly causes a change in another variable.

Subject variables

Variables we can't mainpulate only select

• Gender, age, personality, and group membership

Person by treatment design

The effect of a treatment depends on the person receiving it.

• Measure one variable, manipuate another

Nonequivalent control groups design

Researchers compare two or more groups, but participants are not randomly assigned to those groups.

• Treatment group and comparison group

• Pretest and posttest for both groups

Pros of quasi experimental designs

Ethics

External validity

Construct validity

Cons of quasi experimental designs

More “messy”

Less internal validity

Small N Design

A study that examines a small number of participants, to track conditions, behaviors and changes over time

Small N designs - pros

Provides detailed information about individual behavior.

Tracks changes over time.

Can help researchers determine whether a treatment affects behavior.

Small N designs - cons

Findings may not generalize to everyone.

Results can be affected by individual circumstances.

Some designs have difficulty ruling out alternative explanations.

Qualitative data

Information that describes qualities, characteristics, experiences, or behaviors rather than measuring them with numbers

Observational Study

Research methods in which researchers observe and record people's behavior without manipulating variables or assigning participants to treatments.

Naturalistic observation

Researchers observe behavior in a person's normal environment.

Subject Reactivity

Occurs when participants change their behavior because they know they are being observed or studied.

Systematic Observation

Research method in which researchers observe and record specific behaviors, and what behavivors to observe, how to measure them, and when to record them before the study begins

Pros of Observational Research

Researchers can observe people in real-life settings.

Useful when manipulating a variable would be unethical or impractical.

Studies real world behavior

Can help researchers develop hypotheses for future experiments.

Cons of observational research

Cannot establish causation

Subject reactivity

Time-consuming

Limited control

Sampling problems

Archival Data

Researchers examine existing records (public or private) in order to test a hypothesis

Why do archival data

It can be less expensive, data may already exist, may be impossible to study in other ways

Pros of Archival Data

Saves time

Large amounts of data

Studies past behavior

Cons of archival data

Internal validity issues - you have little control

Biases in collecting data will be built into your study

Interrater reliability

The degree on which two or more researchers or observers agree when measuring Or rating the same thing

Importance of Operational Definitions in Observational Research

Explains exactly how a researcher defines and measures a behavior.

Makes behaviors clear and specific.

Helps researchers measure the same behavior consistently.

Reduces confusion and subjective interpretation.

Allows other researchers to repeat the study.

Importance of Interrater Reliability in Observational Research

Shows whether observations are consistent across researchers.

Increases confidence in the research results.

Helps ensure that behaviors are being recorded consistently.

Importance of operational definitions in archival research

Ensures researchers know exactly what information to look for in existing records.

Makes data collection consistent.

Reduces confusion and subjective interpretation.

Helps researchers measure variables accurately.

Importance of Interrater Reliability in Archival Research

Ensures different researchers interpret records similarly

Increases the consistency and trustworthiness of the findings.

Direct Replication

As close as possible to the original, with a similar sample

Conceptual Replication

Explore same research question using different procedures

Replication + Extension

Replicate original experiment and add variables

Participant Replication

Same procedure, different population

Why might studies replicate and when should we worry?

What do we learn from failed replications?

Evidence of moderators - variables that changes relationship between 2 variables

Evidence of boundary conditions - The Who, where, when of a theory

Meta analysis

Combines data from several studies to examine patterns and treads

• Treats each study as a data point, ideally includes unpublished studies

What do we learn from meta analysis

Mathematical summary of a literature

What conclusions can we draw from the entire body of literature

Factorial ANOVA

Used to examine the effects of two or more independent variables (factors) on one dependent variable

It helps us determine whether each independent variable affects the outcome and whether the variables interact with each other.

One-Way Chi Square

How well does the data from our sample “fit” with the pattern of results that we would expect in the population

• We have one variable that is categorical

Two way chi-square

Looking at the relationship between two or move variables

Looks for independence between variables - are the observed frequencies in one group different from the observed frequencies in another group

Use when we have two variables and they are both categorical

What are we calculating when we do an ANOVA?

Whether there is a statistically significant difference between the means of two or more groups.

The f value, The p value And the degrees of freedom?

What are we comparing in a Chi-Square?

the X² to The critical value