Test #2: Chapter 8

Front 1

Between-subjects design

Back 1

Different groups of scores are obtained from different groups of participants

- Charateristics:

-- different participants/groups compared over each treatment/experimental condition (levels of the IV)

-- DV measured for each participant

-- only one score per participant

-- examine if differences exist between two or more treatment conditions/groups

Front 2

Independent measures design

Back 2

independent scores

- only one score for each participant

- each participant exposed to only one level of the independent variable

- determines whether differences exists between two or more treatment conditions/groups

Front 3

B/W subjects design: Advantages

Back 3

can be used for variety of research questions

- time-related factors (maturation, history, regression to the mean) and confounds not an issue

- Results in measurements relatively uncontaminated from other treatment factors

Front 4

B/W subjects design: Disadvantages

Back 4

Requires relatively large number of participants

- (pre-existing) individual differences (gender, age, level of education)

-- extraneous variables

-- may become confounding variables (groups not equivalent at the beginning)

-- can produce wide variance in scores making it difficult to determine if treatment had an effect

Front 5

Individual differences (as confounding variable)

Back 5

any extraneous variable that systematically differentiates the groups is a confounding variable (age, intelligence, speed)

- difference in DV between groups (what caused it? IV or confound?)

- assignment bias

- threats due to environmental variables that vary systematically between groups (change of room, temperature, time of day)

Front 6

Limiting confounding: Random assignment

Back 6

random process used to assign participants to groups (treatment conditions)

-ensure same chance of assignment to group

- restricted random assignment

-- ensures equal group size

- block randomization

-- ensures each condition has a participant randomly assigned to it before any condition is repeated a second time

Front 7

limiting confounding: Matching groups

Back 7

ensures groups are matched on particular variable(s) (those likely to influence DV)

- Three steps:

-- identification of variable(s)

-- measurement of matching variable(s)

-- assignment to groups by means of the restricted random assignment to balance groups

Front 8

Limiting confounding: Holding a variable constant

Back 8

- identify an EV that influences the DV

- keep at same level for all groups

-- ex. same age, occupation

- Restricting range of variation (to avoid confounds)

-- ex. age (18-21), IQ (100-110)

Front 9

Systematic between-groups variance

Back 9

-experimental variance (due to IV)

- extraneous variance (due to confounding variables)

Front 10

Non-systematic within-groups error variance

Back 10

- due to chance factors and individual differences

- ideal:

-- little variance within treatments

-- big differences between treatments (mean values)

Front 11

Attrition

Back 11

withdrawal from a study before it is completed

- Differential attrition:

-- significantly more participants drop out of one treatment condition than the others

-- threat to internal validity (if there is a negative experience among subjects)

Front 12

Diffusion

Back 12

spread of treatment effects from the experimental group to the control group (participants talk to each other)

- comparing conditions that are more similar than intended

- threat to internal validity (masks or wipes out true effect of treatment)

Front 13

Compensatory equalization

Back 13

- when untreated groups learn about treatment being received by another group and demand treatment

- want equal compensation

Front 14

Compensatory rivalry

Back 14

- when untreated groups learn about treatment received by another group and work harder to show they can perform just as well as special treatment condition

- level of motivation is confounding

Front 15

Resentful demoralization

Back 15

- opposite of compensatory rivalry

- when untreated group learns about treatment received by other group and become less productive and motivated because they resent the expected superiority of treated group

- treatment appears to make a difference between groups

- untreated group internalize inferiority, creating false positive

Front 16

Two-group mean difference

Back 16

- single factor two group design

- manipulates one IV with only 2 levels (usually control vs. experimental/treatment group)

- means computed for each group

- independent measures T-Test

-- evaluates size of difference between means of 2 groups

Front 17

Single factor multiple-group design

Back 17

- dosage levels (no drug, placebo, low dose, high dose)

- single factor Analysis of Variance (ANOVA)

- if significant, a post-hoc test is used to determine exactly which groups are significantly different from each other

Front 18

Comparing proportions

Back 18

nominal or ordinal scale

- each participant is classified into category

- data consists of frequency count of participants in each category

- Chi-Square test for Independence

-- compares proportions in one treatment with proportions in other conditions