Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Psych 2080 > Test Bias > Flashcards

Test Bias Flashcards

You may prefer our related Brainscape-certified flashcards:
Psychology 101 flashcards GRE® Psychology flashcards
1
Q

What is the definition of test bias?

A

gives SYSTEMATICALLY wrong results

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

What makes a test biased?

A

THE USE of the test (native vs non native english speakers)

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

Why is test bias controversial?

A

tests determine who gets to use certain resources

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

Are group differences in test scores evidence for bias?

A

NO - it is the USE that makes them biased

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

What are arguments against the argument that test are in principle biased?

A

1) scrutiny over decades/constant revision
2) consider WHOLE test
3) information vs. outcomes
4) overall prediction utility
5) criterion validity
6) results with “culture fair tests”
7) Differential Item Function analysis (DIF)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Why is scrutiny a good argument that tests aren’t in principle biased?

A

major test constantly revised

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

Why is it important to consider the WHOLE test when arguing that tests aren’t in principle biased?

A

whole test vs. individual items

- items have wide domain therefore not everyone SHOULD know them

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

Why is it important to consider the information vs. outcomes when arguing that tests aren’t in principle biased?

A

tests give info about future, not logical to blame tests (tool for testing)
e.g. blaming thermometer for temperature reading

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

Why is criterion validity a good argument that tests aren’t in principle biased?

A

test score = predictor (SAT score)
validity = success later/outcome (GPA)
*shown that this is a good predictor

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

Why is it important to consider culturally fair tests when arguing that tests aren’t in principle biased?

A

eg. Ravens is best indicator of (g) and is NON verbal therefore culturally fair

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

What is DIF analysis and why is it important in the argument that test are not in principle biased?

A

Differential Item Function analysis (DIF): term preferred to “test bias”
- method of eliminating test bias

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

What is the procedure in elimination biased questions from a test? (DIF)

A

1) from groups (e.g. 90%, 80%, 70% etc)
2) look at difference between them
3) difference = biased item

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Does removing a biased test item change the outcome for different groups?

A

NO

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

How is criterion validity evaluated between test scores and criterion scores? e.g. SAT & GPA

A

CORRELATIONS (diff correlations for diff groups)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

When is it okay to use test scores as predictors?

A

when there is a high correlation bw test scores and criterion scores e.g. SAT and GPA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is an average regression line? What is the consequence of using it?

A

when the intercept differ between two correlations, draw a line in between
RESULT: will lower one groups test score predication and raise the others

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

A steeper slope gives you a higher correlation. (T/F)

A

TRUE - up until 45 degrees

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What happens when the slopes of two correlations (bw SAT & GPA) are DIFFERENT?

A

test score good predictor for one group (steeper line)

but NOT the other therefore BIASED TEST

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are 3 tests used in testing minorities?

A

1) Chitling
2) BITCH test
3) SOMPA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is the Chitling test and what point does it try to make?

A

cannot compare intelligences b/c people raised in diff environments

21
Q

What is tested on the Chitling test as a form of intelligence?

A

street smarts

22
Q

Is the Chitling test valid?

A

NO

23
Q

Is the Chitling test a good predictor?

A

NO - does not predict college performance b/c does not test intelligence (street smarts)

24
Q

What does the BITCH test ask you to define?

A

African american slang

25
Q

Who scores the highest on the BITCH test?

A

African Americans (not surprisingly)

26
Q

What does William’s argue about the BITCH test?

A

it is analogous to the unfairness/ bias of IQ tests (for blacks)

27
Q

Why is the BITCH test not a valid predictor?

A

NO - does not test intelligence

28
Q

What is the SOMPA and what is the idea behind it?

A

System Of Multi-cultural Plurisitic Assessment

idea: knowledge is social constructed

29
Q

Is the idea that knowledge is socially constructed fact based?

A

NO - not a popular idea in science (universal facts in science)

30
Q

What else does the SOMPA suggest?

A

IQ developed to keep minorities in their place

***this is a political excerize, NOT scientific

31
Q

Why does the SOMPA think too many minorities are in EMR classes? (EMR = education able mentally retarded)

A

1) high medical problems

2) unfamiliar cultural references on tests

32
Q

Do all cultural groups have the same potential on every test?

A

NO - if one group does poorly, this is a fact about the test (not the group)

33
Q

How can you tell if someone is truly EMR if the tests used are to blame?

A

YOU CANT

34
Q

What are the 3 evals given on the SOMPA?

A

1) medical
2) social (WISCR)
3) pluralistic** (compares WISCR to communities to give ELPs)

35
Q

What are ELPs?

A

estimated learning potentials: adjusted WISCR culturally

36
Q

Does the label ‘EMR’ affect a child’s academic performance?

A

NO

37
Q

What are the 4 models of test bias?

A

1) regression
2) constant ratio
3) Cole/Darlington
4) Quota

38
Q

What is the basis of regression in modelling test bias?

A

“unqualified individualism”

- treat each personal as an individual

39
Q

Who is selected via the model of regression?

A

ppl with highest scores (regardless of group)

40
Q

Does regression lead to higher or lower avg. performance?

A

HIGHER

41
Q

What is the basis of constant ratio in modelling test bias?

A

choose so selection ratio = success ratio

42
Q

Who is selected via the model of constant ratio?

A

best candidate BUT BOOST minority so choose so selection ratio = success ratio

43
Q

How much are test scores adjusted for minority groups in the constant ratio model of test bias?

A

upwards of 50%

44
Q

Doesconstant ratio lead to higher or lower avg. performance?

A

LOWER

45
Q

What is the basis of the Cole/darlington model in test bias?

A

if there is special value in selecting a minority, then “Y” score for majority = “Y + K” is the same score for minority group

46
Q

Does Cole/Darlington lead to higher or lower avg. performance?

A

LOWER

47
Q

What is the basis of the quota model in test bias?

A

if 10% population Asian, 10% of students should be asian

48
Q

Which models use different regression equations?

A

Quota & Cole/Darlington

49
Q

Does quota lead to higher or lower avg. performance?

A

LOWER

Psych 2080 (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions