110

Introduction of Reliability analysis in SPSS

In this section, we are going to learn about Reliability analysis. Reliability analysis is one of the fundamental types of analysis and the stats that we need to report if we are doing a study. Suppose we are doing a study for PhD basis or publication purpose or ever general understanding purpose. In that case, it’s often essential to communicate the reliability of instruments that we have used in our study. By instruments, we mean the kind of test or the scales that we are using in our study.

• Whenever we plan a study, we use many scales or tests constructed by other scholars in our study. Scholars are those who construct scales or tests. They establish the reliability and validity of it. Only a scale or test which has adequate reliability or validity is accepted for the publication purpose.
• Suppose we have seen some corporate reports or even the World Bank report in which scholars sometimes use some measures or the test or scales, but they do not report its reliability or validity. It’s crucial that if we are using any scales in our research and the scales are given by others or constructed by ourselves, always communicate the reliability of the scale on our sample.
• It may be that the reliability of the scale is good enough on the sample or the kind of situation used by the original author. But in our case or sample, there is no adequate reliability instrument that we are using. So, in that case, all our conclusions are meaningless because the kind of tool that we are used or the fundamental test that we are using is not a reliable scale. So it’s essential to establish the reliability of the scale that we have used in our research.

Reliability

• The word reliability has been derived from the word reliable. The reliable word we often used in our daily life. For example, we say that person A is more reliable than person B or brand A is more reliable than brand B or brand C or some other brand. If we speak in such a manner, we are communicating that there are certain expectations that we are having with the Whenever we use that person or brand, those expectations are going to be fulfilled. Expectations can be a quality expectation, the expectation of discipline, consistency, honesty or any other criteria. If a person repeatedly shows stable trades or characteristics over a period of time, we say that the person is reliable.
• If a person is honest at one moment but dishonest at another moment, we cannot say that a person is a reliable person or a reliably honest person. The same happens in the case of a test. Reliability refers to the consistency of a test; It is an important aspect of the quality of a test, along with the validity of the test. Now, whenever we use a test or scale, our primary purpose is to measure the true score on any constructs that we are targeting.
• Suppose we measure the happiness of the subjects. So, in that case, a reliable instrument is one instrument that gives us the true scores of the subject’s happiness. For example, if a subject is happier, our scale must capture a higher amount of happiness in the subjects. If the subjects are less happy, in that case, our scale must capture the lower amount of happiness among the subjects. If our test consistently gives a stable result over a period of time, we say that our test is reliable. It’s the same as a diagnostic machine.
• Suppose somebody gets diagnosed by a machine, and the machine gives a specific report that he has a low cholesterol level. Suppose he again gets diagnosed by the same machine, and tomorrow or after some time, if everything remains the same, he is being diagnosed with again the same cholesterol level. Now we can say that this machine gives a reliable result and accurately reflects the cholesterol level of the subject. So the reliability of a scale means to what extent a scale or a test or an instrument offers consistent estimates of the desired characteristics.
• Reliability is often measured through the internal consistency of a scale. Internal consistency means to what extent various parts of the scale truly capture the desired characteristics in a similar direction. Suppose we develop a scale, and that scale has four If there is a good amount of correlation among various dimensions of the scale, the scale will have an excellent internal consistency, and the same is true for the items of the scale.

---