Lecture 8 Flashcards

1
Q

To interpret test scores, scores have to be compared to one or more reference points, Flanagan distinguished four different types of reference points…

A
  1. norm-referenced test score interpretation: test takers score is compared to other test takers
  2. test-referenced test score interpretation: test takers score is compared to his scores on other tests or subtests
  3. occasion-referenced test score interpretation: test takers test score is compared to his score from the same test from another occasion
  4. criterion-referenced test score interpretation: test taker’s test score is compared to an external criterion or standard
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What means norm-referenced test score interpretation?

A

that the position of the test taker’s score is assessed with respect to the scores of the other test taker’s.

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

Name 4 sampling methods

A
  1. simple random
  2. stratified random
  3. cluster
  4. judgemental sampling
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is probability sampling?

A

simple random, stratified random and cluster sampling are probability sampling methods, and the generalization of sample characteristics to population characteristics is justified by probability theory

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

What is non-probability sampling?

A

judgemental sampling is a form of non-probability sampling. generalization of the sample to the population can only be made by judgement, and not by probability theory

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

What is stratified random sampling?

A

the population is split into non-overlapping strata. Then, randomly sample persons per stratum, with sample sizes according to proportion of persons in each stratum (=proportional allocation)

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

What is cluster sampling?

A

simple random sampling of groups (clusters) of persons from a population of groups of persons. This can be done in two stages: one-stage cluster sampling-> clusters are randomly selected, and all persons of the selected clusters are tested.
two-stage cluster sampling -> clusters are randomly selected from a population of clusters, and only a simple random sample of persons from each cluster is tested, and so on.

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

What is judgemental sampling?

A

persons are selected purposely and not by random selection.

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

What is the difference between distribution shape-invariant transformations and distribution shape-enforcing transformations?

A

distribution shape-invariant transformations -> scale transformation that do not change the shape of the test score distribution
distribution shape-enforcing transformations -> transformations that enforce a shape on the distribution

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

What is done by an distribution shape-invariant transformation?

A

these transformations transforms the original origin and unit of a scale to a new scale that has another origin and unit, but leaves the shape.

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

How is the z-score transformation computed? (distribution shape-invariant transformation)

A

Zj = (TS - gemTS) / (s)

–> Standardized Z score = test score minus sample mean divided by the standard deviation of the test

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

name to characteristics of Z-scores

A
  1. mean = 0

2. SD = 1

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

What is the general format of a linear transformation?

A

Yj = bXj + a

Yj = normed score of person j
Xj = raw score of person j
b = slope (sd)
a = intercept (mean)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Name 3 common shape invariant transformations, along with their mean and SD

A

Type of transformation Mean SD
Z-score 0 1
T-score 50 10
IQ-score 100 15

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

What is done by a distribution shape-enforcing scale transformation?

A

this transformation not only changes the mean and SD, but also changes the shape of the test score distribution.

The normed scores and raw scores have different distributions

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

Name 4 common shape-enforcing scale transformations

A
  1. percentile scores
  2. decile scores
  3. stanine scores
  4. normalized scores
17
Q

What does a percentile score indicate?

A

the percentage of test takers who have a smaller test score

18
Q

What is the distribution of percentiles?

A

the percentile transformation enforce a uniform distribution of the test scores. A distribution that has an equal percentage of test takers per score.

19
Q

How is the decile score obtained?

A

by splitting the empirical frequency distribution into groups of 10%

20
Q

What is the distribution of the stanine scores?

A

the stanine scale transforms the test score distribution to a normal distribution

21
Q

What is test-referenced test score interpretation?

A

a test taker’s test or subtest score is compared to his scores on other tests and subtests. the score is interpreted within a test score profile

22
Q

What is occasion referenced test score interpretation?

A

a test taker’s score is compared with administrations of the same test on different occasions. The same test is administered to the same test taker on different occasions, and the test scores are compared.

23
Q

What is a response shift?

A

when the test taker recalibrates the item response scale from pretest to posttest

24
Q

Test scores at two measurement occasions are typically compared using …

A

a difference score:

Dj = S2j - S1j