Data Analytic Techniques

Question 1

ANOVA: which way?

Answer 1

(Mertler & Vannatta 2013)
One way: 1 IV
Two-way: 2IV
Factorial: 2+ IVs

Question 2

Mixed-design ANOVA

Answer 2

(Mertler & Vannatta 2013)

has at least one between-subjects and one within-subjects variable

Question 3

Main Effects

Answer 3

(Mertler & Vannatta 2013)

The difference in the averages between groups, ignoring the other IVs

Question 4

Interaction Effects

Answer 4

(Mertler & Vannatta 2013)
Requires Factorial design
The effect of one IV depends on the level of another

Question 5

Mediator

Answer 5

a variable that explains the relationship between other variables

Question 6

Moderator

Answer 6

A variable that influences the strength of a relationship between other variables

Question 7

Bivariate Regression

Answer 7

(Mertler & Vannatta 2013)

To what degree can one quantitative variable predict the other?

Question 8

Linear Regression

Answer 8

(Mertler & Vannatta 2013)

When both predictors and response variables are continuous and linear

Question 9

Construct Validity

Answer 9

(Kazdin 2017)

the operationalization of the intervention/measure that is responsible for the effect found in the experiment

Question 10

CASE

Answer 10

(Kazdin 2017) Threats to construct validity
Cues of experimental situation
Attention and Contact with clients
Single operations and narrow stimulus sampling
Experimenter Expectancies

Question 11

Internal Validity

Answer 11

(Kazdin 2017)

The extent to which the experimental manipulation, rather than extraneous factors, accounts for results

Question 12

DIRT CHASMS

Answer 12

(Kazdin 2017) Threats to internal validity
Diffusion of treatment (similarity between interventions)
Instrumentation
Regression, statistical
Testing exposure
Combination of selection + other threats (e.g. groups are different in an important way)
History
Attrition
Selection bias
Maturation
Special Treatment/Reaction of Controls

Question 13

External Validity

Answer 13

(Kazdin 2017) the extent to which results can be generalized

Question 14

MRS TReNTS

Answer 14

(Kazdin 2017) Threats to external validity
Multiple-Treatment effect
Reactivity to experimental status
Sample Characteristics
Test Sensitization
Reactivity to assessment
Novelty Effects
Timing of Measurement
Stimulus Characteristics

Question 15

Standard Multiple Regression

Answer 15

(Grimm & Yarnold 1995)
2 or more continuous IVs
1 continuous DV
All variables entered simultaneously

Question 16

Hierarchical Multiple Regression

Answer 16

(Grimm & Yarnold 1995)
2 or more continuous IVs
1 continuous DV
Variable order is assigned

Question 17

Logistic Regression

Answer 17

(Grimm & Yarnold 1995)
2 or more continuous IVs
1 categorical DV

Question 18

Path Analysis

Answer 18

(Grimm & Yarnold 1995)

• Similar to Multiple Regression
• Tests models of causal relationships among variables
• Does not prove causation
• Includes only observed/measured variables

Question 19

Structural Equation Modeling

Answer 19

(Grimm & Yarnold 1995)

Model testing including latent constructs and manifest variables

Question 20

Factor Analysis

Answer 20

(Costa & McCrae 1991)

Analyzes covariance and reduces to factors. Factors predict variables.

Question 21

Principle Components Analysis

Answer 21

(Grimm & Yarnold 1995)

Mathematical reduction of variance - aggregates variables

Question 22

Statistical Conclusion Validity

Answer 22

(Shadish, Cook & Campbell 2002) Threats: Monday Law and order SVU

• Multiple comparisons and error rate
• Low statistical power
• Subject heterogeneity
• Variability in procedures
• Unreliability of the measures