Validity Within Psychological Measurement
The goal of psychological measurement is to generate findings that are both reliable and valid (Gerrig & Zimbardo, 2010). Reliabilty is the consistency or dependability of behavioral from psychological and or experimental research. A result is considered reliable when the outcome is the same after testing has been done under similar conditions at different times. An employee arriving to work fifteen minutes early every Monday morning for two years demonstrates reliability. Validity is quantifying the results of a test. For example, being able to tell to what degree someone would be happy or sad in a situation. Without a clear and precise definition of the property in question, a test free of ambiguity is difficult to design. Intelligence tests are a well-known example of this difficulty: as psychologists do not agree on the definition of intelligence, the controversies on intelligence test scores are permanent.
Self-report measures are the most unreliable method of data collection. It requires the subject to answer the researcher’s question(s) verbally or written. Most people cannot/will not completely disclose information that may be relevant to the study being performed. Data collection could be standardized via written questionnaires or by interview. Questionnaires may be fact, behavior, or open-ended questions. They allow data collection to occur without the presence of the researcher. The interview process requires the researcher to communicate with the person(s) being observed. Being an interactive method, to get the most factual, truthful responses, the researcher needs to build a rapport with the subject.
Behavioral measures are ways to study overt actions and observable and recordable reactions (Gerrig & Zimbardo, 2010). Observation is one of the primary ways researchers study what people do. Normally, they focus on the process or product of behavior and can be performed in two ways: direct and naturalistic observation.