Eppp Research Design Review

Front 1

Independant variable

Back 1

Causes variablity in the Dependent Variable; "treatment" or "intervention"

Front 2

Dependent variable

Back 2

Presummed to be affected by Independent Variable; observed & measured by researcher

Front 3

If A, then B....

Back 3

A = Independent variable
B = Dependent variable

Front 4

Experimental vs quasiexperiemental experiemnets

Back 4

Experimental = randomized
Quasi-experiemental = can't randomize

Front 5

Internal validity

Back 5

Allows researcher to determine:
1) if there is a statistically significant relationship bt variables
2) If observed behavior is a causal one

Front 6

External validity

Back 6

if relationship bt variables generalizes

Front 7

Threats to Internal validity

Back 7

1) History
2) Maturation
3) Statistical regression
4) Selection = when subjects from one group initially differ from subjects in other groups in important ways

Front 8

Threats to External validity

Back 8

REACTIVITY:
1) Demand caharacteristics = cues that inform participants how they are expected to act
2) Hawthorne effect = tendency to perform better b/c of the attention they are receiving because of being a participant
3) Experiemntal expectancy = experimenter unintentionally providing cues about the purpose of the experiment

ORDER EFFECTS (overcome with counterbalancing)

Front 9

Between group designs

Back 9

Comparisons of levels of IV are made across different groups

Front 10

Within subject design

Back 10

each subject recieves all levels of IV at different times

Front 11

Mixed designs

Back 11

combines BGD and WSD

Front 12

Factorial designs

Back 12

- Used whenever study includes 2 or more IV
- Provides info on main effects and interaction effects
- Interaction effect is when the effects of one IV are different at different levels of another IV

Front 13

Single subject designs

Back 13

- includes at least one baseline phase and one treatment phase
-DV is measured repeatedly during each phase of study
- multiple baseline designused when removal of tx would be undesirable