Test-retest Reliability


Lesson 5


Short Answer


1. Compare and contrast test-retest reliability with parallel forms reliability.


Test-retest reliability is a measure of the correlation between the scores of the same people on the same test given on two different occasions. Parallel forms are the different versions of a test used to assess test reliability; the change of forms reduces effects of direct practice, memory, or the desire of an individual to appear consistent on the same items. A perfectly reliable test-retest experiment will give a correlation coefficient of +1.00; in other words, the identical pattern of scores emerges both times and the most ideal situation is perfectly correlated. The only draw back is that the subjects to study may know the answers of the tests because they have finished the same test before. The parallel forms prevent it from happening – it reduces the effects of direct practice of the test questions, memory of the test questions, and the desire of an individual to appear consistent from one test to the next.


2. In psychological testing, what does standardization mean? Why do we need it?


Standardization means that a set of uniform procedures is set for treating each participant in a test, interview, or experiment, or for recording data. It is important because when a research is carried out, there will be usually more than one subject to study; fairness is needed so that all subjects to study can act and participate in the same status, minimizing the errors as much as possible. That is, the subjects to study should be treated equally – same questions, same manners of the researchers, same environment etc.


3. Explain the difference between criterion validity and construct validity?


Criterion validity is the degree to which test scores indicate a result on a specific measure that is consistent with some other criterion of the characteristic being assessed. Construct validity is relevant only for tests that measure hypothesis constructs; it is not directly observable but is inferred on the basis of observations of behavior. The main difference is that criterion validity tries to predict the behavior, while construct validity observes the memory-related behaviors.


4. What are the drawbacks of using judge’s rating to make an assessment? Under what conditions are these ratings most reliable?


The primary goal of psychological assessment is to make accurate assessments with minimum bias. Since judge’s rating is always drawn by human beings, the assessment made by an assessor’s judgment is always affected by human emotions such as the arbitrary criteria of sex, race, nationality, privilege, or physical appearance. It is most reliable when there is a wide variety of ethnic groups to study, since it can help justify discriminatory; and other information concerning individuals’ adaptive skills should be absent, because the different cultural background may give different results on a specific-cultural-based research.


5. Describe the three components of intelligence that have been identified by Sternberg. Provide your own examples of each.


Sternberg has identified three triarchic theory of intelligence, and the three components are componential intelligence, experiential intelligence, and contextual intelligence.


Componential intelligence is defined by the components that underlie thinking and problem solving; it is for selecting strategies and monitoring progress toward success. For example, when a person is given a mathematical question such as “Find A, B and C when AAB + BBC = 44055”; one’s componential intelligence will acquire knowledge to learn new facts, then perform the components to find the strategies and techniques to solve the problem. Experiential intelligence captures people’s ability to deal with two extremes: novel and very routine problems. For example, a bus driver may have high experiential intelligence because the route that he drives everyday is the same, which he can perform his actions with the least amount of “new” thought; but when the road is currently blocked due to the flooding, he can use his experiential intelligence to change his initial route to a new one with no error. Contextual intelligence is reflected in the practical management of day-to-day affairs. It involves one’s ability to adapt to new and different contexts, select appropriate contexts, and effectively shapes your environment to suit one’s needs. For example, a factory “newbie” doesn’t know how to operate the machines efficiently, but he will operate efficiently soon because he “get used