302 Exam 3

Question 1

Correlations (def)

Answer 1

provide the strength and direction of relationship between 2 variables

Question 2

Bivariate Correlations

Answer 2

determines relationship between 2 measured variables
-describes the degree of the relationship

Question 3

What should you use if two variables are continuous?

Answer 3

Correlations and/or Regressions

Question 4

What does a correlation test? (3)

Answer 4

1) the null hypothesis that 2 variables are not related
2) type of relationship (+/-)
3) the strength of a relationship

Question 5

What are the 4 types of associations?

Answer 5

1) positive
2) negative
3) zero
4) curvilinear

Question 6

What does a t-test measure?

Answer 6

(use for 1 or more categorical variables)
-how much variables defer from each other
-whether the means between 2 groups are statistically significant

Question 7

what are the 4 types of validities?

Answer 7

1) construct: of each variable
2) external: generalizable to defined population
3) statistical: how well the data supports the conclusion
4) internal: does it establish cause/effect

Question 8

Effect Size (statistical validity definition)

Answer 8

-describes the strength of the association
-p<0.05 = statistically significant

Question 9

Confidence Intervals (w/ vs w/out zero)

Answer 9

-CIs containing zero are not statistically significant
-CIs not containing zero are statistically significant

Question 10

What is an outlier?

Answer 10

an extreme score than can have an effect on the overall relationship

Question 11

What is internal validity?

Answer 11

the extent to which a study accurately establishes a cause-and-effect relationship between variables

Question 12

What is external validity?

Answer 12

the extent to which the findings of a study can be generalized

Question 13

What types of questions can regression answer? (3)

(think IV and DV)

Answer 13

1) how well the IV predicts the DV.
2) which IV (if multiple) was the best predictor.
3) whether an IV still predicts an outcome when the effect of another variable is controlled for.

Question 14

Assumptions of Regressions (4)

Answer 14

1) Linearity: relationship between x/y is linear
2) Homoscedasticity: the variance of residual is the same for any value of X
3) Independence: observations are independent of each other
4) Normality: for any fixed value of x, y is normally distributed

Question 15

Sample Linear Regression

Answer 15

similarities to correlation, direction and significance of the association between variables
-allows us to make predictions

Question 16

Moderating variable (def)

Answer 16

when the relationship between 2 variables changes depending on a third variable

Question 17

what 3 things do you need to determine causality?

Answer 17

1) covariance
2) temporal precedence
3) internal validity

Question 18

Multivariate designs can…

Answer 18

1) suggest causality
2) give individual correlations
(cross sectional, autocorrelations, cross log correlations)

Question 19

what does a cross sectional correlation determine?

Answer 19

if 2 variables measured at the same time are associated

Question 20

what do autocorrelations determine?

Answer 20

association of each variable with itself across time

Question 21

what do cross lag correlations determine?

Answer 21

if an earlier measure of one variable is associated with the later measure of another

Question 22

what are the 4 possible cross lag patterns

Answer 22

1) one
2) other
3) both
4) neither

Question 23

what is the purpose of multiple regressions?

Answer 23

ruling out third variables

Question 24

Criterion Variable (multiple regression def)

Answer 24

variable we are most interested in understanding or predicting (similar to DV)

Question 25

Predictor Variable (multiple regression def)

Answer 25

any variable that is not the criterion

Question 26

Beta Basics (what is it?)

Answer 26

-similar to r but reveals more -shows direction/strength of relationship

Question 27

Beta statistical significance

Answer 27

if p/sig is less than 0.05 than it is significant, true, and replicable (greater then/equal to 0.05 is not significant)

Question 28

mediators

Answer 28

internal to the causal variable in the bivariate relationship

Question 29

moderators

Answer 29

-third variable -external to the 2 variables in the bivariate relationship

Question 30

Experiment (def)

Answer 30

manipulate one variable (independent), measure another (dependent)

Question 31

experimental variables (7)

Answer 31

1) manipulated 2) measured 3) independent 4) dependent 5) control variables (constants) 6) covariance

Question 32

Covariance (def)

Answer 32

determines the relationship between the movements of 2 random variables

Question 33

temporal precedence

Answer 33

establishes cause and effect relationships

Question 34

internal validity

Answer 34

refers to extent to which a study accurately establishes a cause and effect relationship

Question 35

what are the three threats to internal validity?

Answer 35

1) design confounds 2) systematic variability 3) selection effects

Question 36

Systematic Variability (def)

Answer 36

a second variable varies systematically with the IV -cannot establish cause

Question 37

Unsystematic Variability (def)

Answer 37

random variability (occurs in both groups) -makes it more difficult to detect changes in DV

Question 38

Selection Effects (def)

Answer 38

kinds of participants at one level of the IV are systematically different than another level

Question 39

Group design (def)

Answer 39

uses different groups of participants for several levels of measurement

Question 40

Independent group design (posstest only)

Answer 40

-random assignment -test on DV once

Question 41

Independent group design (pretest/posttest)

Answer 41

-random assignment to IV -test on DV twice

Question 42

what is a disadvantage or pre/post test design?

Answer 42

participants may change their behavior due to exposure to the pretest

Question 43

Concurrent measures design (process)

Answer 43

participants are exposed to all levels of the same IV at the same time, then given a single preference dependent measure (ex. one group, tries 2 flavors, picks favorite)

Question 44

repeated measure design (process) (within group design)

Answer 44

participants are measured on the DV more than once (after each exposure to IV) (within group design)

Question 45

what are advantages of within group designs? (2)

Answer 45

1) participants are ensured to be equivalent 2) fewer participants are required

Question 46

What 2 things can determine cause?

Answer 46

1) covariance 2) temporal precedence **internal validity can NOT

Question 47

what is a threat to internal validity?

Answer 47

exposure to one condition can change your reaction to another (carryover, practice, and order effects) -fix this with counterbalancing

Question 48

Partial Counterbalancing

Answer 48

conditions presented randomly for each participant