If you ask a classmate questions about course content and he is correct sometimes but at other times he gives you the wrong information, you might say that he is not reliable. You may want to request help from someone who knows the correct answer every time—perhaps your Instructor! The concept is the same in research. It is important to be confident that when you measure a variable repeatedly, it will have the same result. That is what reliability means in research.
Another important concept in research is validity. You might have a friend who laughs at all of your jokes. You determine that you must be very funny because your friend laughs. You are measuring your ability to be funny based on laughing behavior. However, it is possible that your friend is very polite and does not want to hurt your feelings by not laughing. In that case, the laugh is really measuring your friend’s politeness and not your expertise in telling jokes. The “laugh” measure, in this case, is not valid. It does not measure what you think it is measuring.
The concept of validity is critical in research, too. If you develop a scale to measure the conceptual variable of anxiety, but it really measures fear, the scale is not valid. Conceptual variables are abstract ideas that form the basis of research designs. What does the concept of anxiety mean to you? How would you define anxiety to another person? Only after conceptual variables are precisely defined can they be turned into measureable variables.
In this Discussion, you will turn a conceptual variable into a measureable variable by creating and administering your own scale, and by considering the scale’s reliability and validity.
With these thoughts in mind:
Stangor, C. (2015). Research methods for the behavioral sciences (5th ed.). Stamford, CT: Cengage Learning.