Wednesday, 20 December 2017 08:40

Difference between reliability and validity Featured

Written by
Rate this item
(0 votes)

In the context of collect data in research, the terms reliability and validity are different.  Data are said to be valid when they represent what they purport to represent, and meaningful inferences can be drawn from them. In contrast, data is said to be reliable if they can be reproduced or are consistent. That is, data gathered on a particular issue will be the same when gathered by different investigators or the same investigator on separate occasions.  Therefore, validity relates to the “truthfulness” of data, and it is a property of inferences from data and exists in relation to a specific context and inferential purpose. Reliability relates to the uniformity or reproducibility of data, and it is a property of data themselves, and it is not related to the context or purpose of data collection.  Validity requires independent knowledge of the true nature or magnitude of the entity; it presupposes at least a moderate degree of reliability and is undermined by systematic error.  Reliability requires no knowledge of the true nature or magnitude of the entity; it does not presuppose validity and is undermined by random errors. Lack of validity leads to distorted inferences while lack of reliability leads to indistinct inferences.   Validity implies much more than reliability. It implies logic and well-grounded principles of evidence and, if one were to place reliability and validity on a continuum, and they would occupy opposite poles. Researchers want their measurement to be reliable, but often, reliability cannot be assured. On the other hand, validity must be assured. 

Ways of assessing reliability

Methods of testing reliability take four main forms: (1) alternate or parallel forms of the same measure administered simultaneously or (2) repeated application methods (3) the split-halves method, and (4) the internal consistency method.  The alternate or parallel method is equivalent to using same scales to weigh the same sample (multiple forms reliability).  Another popular method is called spilt-half method of testing reliability. Instead of making two distinct but supposedly identical tests or measuring instruments, the researcher constructs a single instrument containing twice as many times as he or she needs, with half of the items being redundant, or repetition of the first half.  The test-retest is a method for assessing a measure’s consistency or stability.  It is a test of consistency among different administrations. In order to determine the co-efficient for test-retest reliability, the same test is administered to a group of subjects repeatedly. If the test is reliable, scores from each administration should be similar. There should be an association between the first and the second administration. One of the major concerns of the test-retest method is that it is referred to as the memory effect. It can create an artificially high reliability coefficient. Parallel reliability test involve the use of pre-and post-test. The administration of pre-and post-test method requires that the tests are parallel or equal in what they are supposed to measure. In order to determine parallel reliability, the researcher calculates a reliability coefficient on two measures from the same group of subjects. 

Ways of assessing validity

Validity focuses on the meaningfulness of research components.  It involves the careful investigation of study to determine its competence or scientific soundness. The examination of research needs the consideration of the strengths and flaws of the methods used to produce scientific evidence.  The first method of assessing the validity of research involves content validity. Content validity is a qualitative validity where the domain of the made clear and researcher judges whether the measure is representative of the domain.  It ensures that indicators use the meaning of the concept as defined by the researcher.  The content validity can be assessed by asking a number of questions about the test or instrument and seeking the opinion of experts in the field. The second method of assessing validity is criterion-related validity. It is the measure of the degree to what the test measure corresponds to one or more referents.  Criterion validity targets the accurateness of a measure or test. Criterion validity is demonstrated by comparing a specific measure with another valid measure. It can lead to predictive validity or concurrent validity. Predictive validity predicts an identified association between a specific construct and another referent while concurrent validity exists when measures can be for the same or different constructs. The third conventional method of assessing validity applies to the concept of “construct” validity. This method to validity examines the extent to which multiple variables that encompass a theoretical construct are represented in instruments intended to measure that construct. In order to substantiate construct validity, the research needs to accumulate face, content, concurrent, predictive, convergent and discriminant validity.

Obtaining validity evidence

Validity evidence of a measure can be assessed using various methods.  In studies where it is critical to show the relationship between a test measure or score and future behaviors, the researchers use the predictive method to determine evidence of validity.  In this instance, a huge number of people take a test, and their scores are held for a predetermined time interval. The researcher that correlates these test scores with the criterion scores. When the test scores and the criterions score exhibit a strong relationship, the test is said to have shown predictive evidence of validity.  An alternate method of demonstrating validity evidence is concurrent evidence of validity. The concurrent method entails administering tow measures, the test and a subsequent measure of the quality, to the same sample at as close as the same point in time as possible.

Content validity evidence consist a demonstration of a strong relationship or linkage between the content of the selected procedure and important variables. Obtaining construct validity evidence involves operationally defining the construct, usually in terms of the underlying theoretical framework; developing hypotheses from this same theoretical basis and using logical and empirical methods to test hypotheses.

References

Sekeran U & Bougie R (2011). Research Methods for Business: A Skill Building Approach, 5th Edition. Wiley India Pvt. Limited, India.

Janet Peter is the Managing Director of MeldaResearch.Com a globally competitive custom writing service which is the premiere provider of Essay Writing Services, Research Paper Writing Services at Term Paper Writing Services at very affordable cost. For 9 years, she has helped a number of students in different academic subjects.

Read 135 times Last modified on Wednesday, 20 December 2017 08:43
janet

Dr. Janet Richard is Author of this paper and is associated with ModernGator.Com which is a global Custom Essay Writing and Term Paper Writing Company. If you would like help in Research Papers and Term Paper Help you can visit our website. 

You can place an order similar to this with us. You are assured of an authentic custom paper delivered within the given deadline besides our 24/7 customer support all through. Visit https://www.moderngator.com/place-order/ 

 

https://www.moderngator.com
2014 Difference between reliability and validity - ResearchWriting.Education .
Powered by Joomla 1.7 Templates