Validity and Reliability: The close relationship between them
Reliability and validity are interrelated, and there is a requirement that a measurement tool should be both reliable and valid, as the exercise shows in the third dial. The first dial shows that a test can yield a given measurement over and over, and that is reliability, but not the measure that it intendeds to measure (i.e., it is not valid). The second dial shows that a test might do what it is supposed to do, but not in a consistent manner (e.e., it is valid, but not reliable). Keep in mind that you cannot have a valid test without it first being reliable, because in order for an instrument to do what it is supposed to do, it must first do it consistently.
EXERCISE 1
There is a quantified link between reliability and validity. This relationship expresses that the maximum level of validity is equal to the square root of the reliability. For example, if the reliability coefficient for a test of communication skills is 0.79, the validity coefficient cannot be larger than 0.88, which is the square root of 0.79.
Now that you got some insights on the reliability and validity of the instruments, you can use your knowledge in a research or experiment. By now you should understood that if these instruments are not reliable and valid, then the results of your investigation will be uncertain.
As undergraduate student, you might choose to conduct your own a research based on a quantitative design as part of a semester project or the graduation thesis. Before creating a new instrument in order to test a new hypothesis, first you must know that establishing the reliability and validity of an instrument can even take years of intensive work. This warning is not to discourage you, but is for raising your awareness that a new hypothesis cannot be tested until you are sure that the measurement instrument works as it should.
When you are at the beginning of the road as researcher and lack sufficient experience with determining reliability and validity of a measurement tool, the best option for you is to find an instrument with established reliability and validity. You can find such instruments in the scientific articles and in books dedicated to instruments already tested for reliability and validity. For example, in the article “Brief report: Intimacy, passion, and commitment in romantic relationships – Validation of a ‘triangular love scale’ for adolescents” (Overbeek et al., 2007), you can find in the appendix the integral Triangular love scale, with its 19 items, grouped into the three investigated dimensions: intimacy, passion and commitment. Another example is the work of Corcoran and Fischer (2000), who published an extensive number of standardized instruments in two volumes, appropriate to measure the majority of the problems encountered in the clinic and therapeutic work of professionals in all the human services (psychology, social work, counselling, psychiatry, nursing) with individuals, couples, groups, or families. Index of Marital Satisfaction, Life Distress Inventory, Positive Feelings Questionnaire, Adolescent-Family Inventory of Life Events and Changes, Attitude toward the Provision of Long-Term Care, Social Support Index, Children’s Beliefs about Parental Divorce Scale, Compulsive Eating Scale, Rosenberg Self-Esteem Scale, Aggression Inventory, Problem-solving Inventory are just a few examples of the standardized assessment instruments comprised in these two volumes.