Psych Stats Quiz 5/8/23 Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

rank sum test

A

nonparametric equivalent of the independent t

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

why use rank sum

A

when you want to compare two samples of ordinal data because the assumption of interval/ratio is not met

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

disadvantages of rank sum

A

less powerful than parametric tests - type II error is more likely
null hypothesis is more general

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

t-test for dependent samples

A

looking for differences within subjects
structurally similar to other t statistics but based on difference score (D)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

t-test for dependent samples: advantages

A
  • use same subjects in all treatment conditions - no risk that subjects in one condition are substantially different from subjects in another
  • only need half the subjects for the same n
  • more powerful than independent sample t-test because its influenced by variability that is reduced by using the same subjects
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

t-test for dependent samples: within-subjects design

A

each subject contributes a score to each sample/condition in an experiment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

t-test for dependent samples: matched-group design

A

pairs of subjects who are similar - one member of the pair is assigned to one sample and the other is assigned to the second sample
- each subject contributes one score

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

steps for dependent samples

A

1) state hypotheses (always = 0 and not = 0)
2) Find critical region: df = n-1 and then use t table to find value
3) Difference scores: after score - before score
4) sample mean: average D values
5) sum of squares = sum of each score square - (sum of scores)squared divided by n
6) standard deviation: square root of SS/n-1
7) standard error: sd/square root of n
8) t = average of D/standard error
9) fail to reject the null if t falls within critical region

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

steps for dependent samples

A

1) state hypotheses (always = 0 and not = 0)
2) Find critical region: df = n-1 and then use t table to find value
3) Difference scores: after score - before score
4) sample mean: average D values
5) sum of squares = sum of each score square - (sum of scores)squared divided by n
6) standard deviation: square root of SS/n-1
7) standard error: sd/square root of n
8) t = average of D/standard error
9) fail to reject the null if t falls within critical region

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

naturally occurring population

A

present without any intervention by the investigator

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

hypothetical population

A

does not exist until they are actually measured

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

confound

A

systematic differences between two groups OTHER than the independent variable
arises due to differences between the groups that existed before the independent variable was applied

How well did you know this?
1
Not at all
2
3
4
5
Perfectly