CHAPTER 4: RELIABILITY Flashcards

(36 cards)

1
Q

Who is the founder of the advancement of reliability?

A

Charles Spearman

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

He wrote book “ The proof of measurement of associatuon between 2 things”

A

Charles Spearman

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

Who has this basic notion of sampling error?

A

Abraham De Moivre

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

Who created the Product Momentum Correlation?

A

Karl Pearson

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

Who is the founder of the foundation of reliability?

A

Edward Thorndike

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

He wrote the book “Into the theory of mental and social measurement”

A

Edward Thorndike

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

What are the three basic princples of Reliability?

A

SEM, Domain Sampling Method and IRT

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

A principle which the deviation of error of the true score and observed score is important to know the reliability of the test.

A

SEM

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

It is the reliability of variance of observed score in a short test and variance of long run to know the true score.

A

Domain Sampling Method

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

It uses computer that focus on the range of item difficulty to assess individual’s ability

A

IRT

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

What are the four sources of error?

A

test- retest method, parellel forms method, Internal consistency method and difference score

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

A method which to test the reliability of the test, a test used in different occassion. It measures “traits” over time

A

test -retest method

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

A method which we evaluate the test accross different forms of the test.
There are two forms of test but has the same item of difficulty

A

parallel forms method

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

A method which people are examine in the same subset of items selected from a same form of measurement

A

Internal consistency

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

It is the difference of the score of two groups.

A

difference score

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

What are the two effects of error?

A

Carry over and Practice effect

17
Q

The effect which the first administration of the test is greater than the second one that leads to inconsistency or gap of the test results

18
Q

The effect which the second test result is high due to practice

A

Practice effect

19
Q

A method of parallel forms

A

Alternate forms of parallel

20
Q

What are the three methods of internal consistency?

A

Split half, KR20 Formula and Coefficient Alpha

21
Q

It is way which a long test is split into half which the first half of the result is compare to the second one.

22
Q

How to divide the items of the test?

A

Odd-even system

23
Q

It is a formula which used to have a correct system of splitting the long test which has the correlation of two halves.

A

Spearman-Brown Formula

24
Q

What formula is this?

r = 2r/1-r

A

Spearman-Brown Formula

25
A way of splitting which is used when has 2 unequal variance for non-dichotomous
Cronbach's Coefficient Alpha
26
It is a method of test which calculates the reliability are dichotomous 0 or 1 or (right or wrong)
KR20 Formula
27
It is a method of test which describes the variance of items whether or not they are in a wrong or right format.
Coefficient Alpha
28
A method used for Difference of score
Kappa Statistics
29
He established the Kappa Statistics
J. Cohen
30
It is the best method of the level of agreement of 2 raters
Kappa Statistics
31
+1. -1 0. 75 0. 40 - .74 0. 39
Perfect agreement Excellent agreement Fair / good agreement Poor agreement
32
What formula is this? | S square root of 1 - r
SEM
33
How reliable is reliable?
1. Depends on the use of tests 2. Usually needed is a test with .70 - ,80 3. If clinics, to the patient's future greater than .95 is needed
34
How to improve reliability?
1. Increase the no. of items based on Sampling Domain Effect and the Pearson - Brown method 2. Perform Factor and Item analysis 3. Correction of Attenuation
35
It examines the correlation between each item and total score for the test
Discriminability analysis
36
It examines the test if there have been no error
Correction of Attenuation