Validity and Reability

METHODS OF DETERMINING VALIDITY

In general sense, validity is concerned with the extent to which tests results serve their intended use. Ex: test resultst might be used to describe the types of performance an individualcan demonstrate, or to predict an individual’s probable success in some future activity. Questions of validity are questionsof what may properly be inferred from a test score(American Psychological Association,1974).

Concept of validity:Validity refers to interpretation of test results (not to the test it self).
Validity is inferred from available evidence (not measured).
Validity is specific to a particular use (selection,placement,evaluation of learning,so forth.
Validity is expressed by degree (for example,high, moderate,low).

Basic Type of Validity
Type Question to be answer
Content validity How adequately does the test content sample the larger domain of situations it represents?
Criterion-related validities How well does test performance predict future performance (predictive validity) or estimate present standing (concurrent validity)on some other valued masure called a criterion?
Construct validity How well can test performance be explained in terms of psychogical attributes?

Content validity: is matter of determining whether the sample is representative of the large domain it is supposed to represent. It is especially important in achievement testing. We can built a test that has high content validity: (1)identifying the subject-matter topic and the learning outcomes to be measured, (2) preparing a set of specification,(3) constructing a set test that closely fits the set os specifications.

Criterion-Related Validities
There are 2 types: the first concerned with the use of test performance to predict future performance on some other valued measure called a criterion. The second concerned with the use of test performance to estimate current performance in some criterion. The key of both: the test scores, the criterion to be predicted.
Expectancy table is a simple and practical means of expresssing criterion-related validity, and especially useful for making predictions from test scores. It can use with criterion-referenced as well as norm-referenced tests.

Construct Validity
The aim is to identify all the factors that influence test performance and to determine the degree of influence of each. The process: identifying the construct that might account for test performance, formulating testable hypotheses from the theory surrounding each construct, gathering data to test these hypotheses.
These hypotheses might be stated in harmony with any general type of evidence:
1. Differences between known groups.
2. Changes in perfomance with training.
3. Correlations with other test.
4. Internal consistency.
5. Analysis of the test-taking process.
METHODS OF DETERMINING RELIABILITY
Realibility refers to the consistency of the test scores-that is, to how consistent they are from one measurement to another. Because of the ever present errors of measurement ,we can expect a certain amount of variation in test performance from one time to another,from one sample of items to, and from one part of the test to another.
Test-Retest Method
It requires administering the same form of the test to the same group after some time interval. The time between the two administrations may be just a few days/several years. The length of the time interval should fit the type of interpretation to be made from the results. Test-retest reliabiity coefficients are influenced both by errors another within the measurement procedure and by the day-to-day stability of the students’ responses.
Methods of Estimating Realibility
Method Type of Information Provided
1.Test-retest method 1.The stability of test scores over some given period of time.
2.Equivalent-forms method 2.The consistency of the test scores over different forms of the test (that is, different samples of items).
3.Test-retest with equivalent forms 3.the consistency of test scores over both a time interval and different forms of the test.
4.Internal-consistency methods 4.The consisncy of test scores over different parts of test.
Equivalent-Forms Method
Two equivalent forms of test(alternate/parallel) are administrated to the same group during the same testing session to measure the same abilities.
Test-Retest Method with Equivalent Forms
Correlation coefficient based on the odd and even items, for total test applying the Spearman-Brown:
Reliability of total test = 2 X reliability for ½ test
1 X reliability for ½ test Kuder-Richardson,3 type of info: number of items(K), means(M), standard deviation(s).
Reliability estimate(KR21)= 1 M (K-M)
K(S2)

Standard Errors of Measurement
= s √1-rn s: standard deviation, rn:reliability coefficient

Reliability of Criterion-Referenced Mastery Test
Use with criterion-referenced tests, they are likely to provide misleading results. When a criterion-referenced test is used to determine mastery(which is its major use), our primary concern is with how consistenly our test classifies masters and nonmasters.