RELIABILITY AND VALIDITY

Reliability and Validity

Reliability and Validity

Introduction

The results of any research depend on the method selected by the researcher to carry out the study. However, any method cannot be regarded as appropriate if it is not reliable or valid. Though, there are different approaches and techniques used by researchers to measure the reliability and validity of any research, however, for researches related to human service the techniques that are used to test the reliability and validity will come under discussion in this study.

Validity

The term validity refers to the quality of being valid and rigorous (Saunders, Lewis and Thornhill, 2007). In human services researches this technique is used to validate the results or outcomes of the study.

Types of Validity

In human services research, validity can be divided into four different types (Saunders, Lewis and Thornhill, 2007).

Conclusion validity

This type of validity inquires the relationship between the research, it methods and its outcomes. For instance, if a research attempts to find out the relationship between training and job satisfaction among employees of any organization, than conclusion validity would assess what research design was followed and what are the outcomes of the research.

Internal validity

Internal validity is largely concern about cause and effect relationship. For instance, is there any improvement in job satisfaction level after training?

Construct validity

Construct validity attempts to find out that what study was about to measure and what is the actual result. This can be understand in our example like, if the study proposed a hypothesis that training increases the level of job satisfaction, and the outcomes or results explain that training and job satisfaction level shares a positive relationship, that is by training employees the increased level of job satisfaction observed. This means that construct validity is positive.

External validity

The term external validity means that to what extent the results of a study can be generalized. For instance, the study of training and job satisfaction can be considered valid externally, if the results can be applied to every organization working in the same industry (Saunders, Lewis and Thornhill, 2007).

Reliability

Reliability refers to the consistency of a measure, score, or test. Reliability occurs more often in statistical quantitative studies and less frequently in qualitative studies. Since the world of research with human subjects is not perfect, researchers developed a number of techniques for estimating reliability or the degree of error in measurement. One such ...