Groth Chapter 1

Question 1

What are the 5 primary methods of obtaining reliability?

Answer 1

1. Test-Retest: the extent to which the test produces consistent results on retesting.
2. Alternate Form: the relative accuracy of a test at a given time
3. Split Half and Coefficient Alpha: the internal consistency of the items
4. Interscorer: the degree of agreement between two examiners.

Question 2

Another way of saying this

Answer 2

Reliability can be:

1. Time to time - Test Retest
2. Form to form - Alternate Forms
3. Item to Item (split half/coefficient alpha
4. Scorer to Scorer - Inter-scorer

Question 3

Reliability addresses:

Validity:

Answer 3

consistency

measures what the test is intended to measure

Question 4

Name the three main methods of establishing Validity

Answer 4

1. Content Related
2. Criterion Related
3. Construct Related

Question 5

Content Related Validity:

Answer 5

Refers to the representativeness and relevance of the assessment instrument to the construct being measured. A test high in content validity must cover all major aspects of the content area and must do so in the correct proportion. (IQ tests should have questions that measure your cognitive ability)

Question 6

Criterion Validity:

Answer 6

Criterion validity is determined by comparing test scores with some sort of performance or outside measure.

1. Concurrent Validity: refers to measurements taken at the same time as the test. ie) an IQ test and academic achievement test at the same time. Is Mr Jones maladjusted?
2. Predictive Validity: refers to outside measurements that were taken some time after the test scores. ie) done by correlating the IQ test scores with academic achievement a year after the initial testing. Is Mr Jones likely to be come maladjusted?

Question 7

Construct Validity:

Answer 7

Is to assess the extent to which the test measures a theoretical construct or trait.

1. careful analysis of the trait
2. consideration of the way sin which the trait should relate to other variables
3. test designer needs to test whether these hypothesized relationships actually exist.

Question 8

Another definition of Construct Validity:

Answer 8

The extent to which a test truly measures a proposed psychological ability or skill and is related to an underlying theory or model of behaviour.