Reliability
Reliability refers to the attribute of consistency in measurement (Greogory 2011).
Test reliability is the consistency of scores obtained by the same persons when retested with the identical test or with an equivalent from of the test (Anastasi & Urbina, 2005).
The concept of reliability underlies the computation of the error of measurement of a single score, whereby we can predict the range of fluctuation likely to occur in a single individual’s score as a result of irrelevant or unknown chance factors.
It has been used to cover several aspects of score consistency.
Test reliability indicates the extent to which individual differences in test score are attributable to true differences in the characteristics under consideration and the extent to which they are attribute to chance errors.
Measure of test reliability make it possible to estimate what proportion of the total variances of test score is error variance.
Since all types of reliability are concerned with the degree of consistency or agreement between two independently derived sets of scores, they can all be expressed in terms of a correlation coefficient.
Types of Reliability
There are mainly two types such as-
1. Reliability as temporal stability
* Test-retest reliability
The most obvious method for finding the reliability of test score is by repeating the identical test on a second occasion. A test retest reliability coefficient is compute to determine whether a test measures consistency from one time to another.
This coefficient is found by correlating the scores obtained by a group of people on one administration with their scores on a second administration of a test. This coefficient is also known as a coefficient of stability.
This type of analysis is of value only when we measure traits or characteristics that do not change over time.
* Alternate from reliability
One way of avoiding the difficulties encountered in the test retest reliability is through the use of alternate from of the test, one consisting of similar items (with some difficulty level) but not the same items.
Alternate forms reliability compares two equivalent forms of a test that measure the same attributes. It is also known as parallel from or equivalent forms reliability.
2. Reliability as internal consistency
For avoiding the difficulties of test retest and alternate forms, a less direct method of taking account the effects or reliability of different sample of test items was developed. This is the method of internal consistency including-
• Split-half reliability;
Split half reliability is estimated by correlating the pairs of scores obtained from equivalent halves of test administered only once. In this simplest of internal consistency approach, a single test is viewed as composed of two parts measuring in some thing.
The split half approach to reliability is useful in cases where the test retest method is forbiddingly expensive or where retesting is potentially misleading.
• spearman-brown formula
The general formula can be used to assess a variety of different question about test length and reliability. It is used extensively in calculating the corrected split-half reliability.
• Cronbach’s Alpha
Using the spearman-brown correction is not always advisable. When the two halves of a test have unequal variances, Cronbach’s alpha can be used.
Coefficient alpha may be thought of as the mean of all possible split-half coefficients, corrected by the spearman-brown formula.
• Coefficient Theta
One little used index of internal consistency is that of coefficient theta. Theta is based on principle component analysis of the test items. It differentially weights items that correlate more with each other that does theta of IRT.
• Kunder-Richardson 20
Kunder and Richardson 1937 greatly advanced reliability assessment by developing methods for evaluating reliability within a single test administration.
There approach does not depend on some arbitrary splitting of the test into halves. Decisions about how to split test into halves cause many potential problems for split-half reliability.
• Kunder-Richardson 21
In addition to KR 20, Kunder and Richardson presented another formula KR 21, a special case of the reliability formula that does not require the calculation of the p’s and q’s for every item.
There is another type of Reliability
Inter-scorer reliability
Reliability that describe consistency of judgments, or rating across individuals or raters/judges are called inter-score or inter-rater reliability.
Which types of reliability is Appropriate:
To answer this question we need to know the nature and purpose of the individual test in question.
• For test designed to be administered more than once, it would be reasonable to expect that the test demonstrate reliability across time case test retest reliability is appropriate.
• For tests which purport to possess factorial purity, coefficient alpha would be essential.
• Split half methods work well for instruments which have items carefully ordered according to difficulty level.
• Inter-scorer reliability is appropriate for any test which involves subjectivity of scouring.
এতে সদস্যতা:
মন্তব্যগুলি পোস্ট করুন (Atom)
কোন মন্তব্য নেই:
একটি মন্তব্য পোস্ট করুন