Facts About Psychometrics

The use of measurement can be traced back to ancient times due to its usefulness as such a branch of psychology that measures intelligence, aptitude, and personality refers to psychometrics. Aside from the measurement, psychometrics also deals with the designs to be used, modes of administration and as well as the interpretation of the different quantitative examinations. Broadly speaking, psychometrics is one discipline that quantifies and analyzes human differences. Specifically, it is regarded as the development of statistical methodology used in analyzing data in the field of psychology. (Browne, 2000). Over the centuries, psychometrics has evolved into a discipline that is mathematically oriented which aims to provide methodology in handling psychological measurement. Those methodologies arising from psychometrics have also been useful in the other fields. Particularly, the factor analysis is now being widely used in many disciplines and has gained popularity in contemporary statistical computer packages. (Winer, 1971)

The use of psychometrics is a breakthrough in psychology wherein intelligence, aptitude, or personality can now be measured using standardized questionnaires. The institution of this type of measurements has helped universities, organizations, and etc. in assessing students or potential employees although these also have associated limitations which should be considered appropriately. The analysis of the results will somehow give a clear picture of a person and this will help in the selection process.

Psychometrics and its Origin

An article published by Sir Francis Galton in 1879 describes an experiment he called "mental operations." It was later called the "free association" wherein he made an assessment of the different reactions to a roster of 75 words. He identified the rate of idea formation (50 per minute), the frequency of associations (about one half), the frequency wherein these associations might be attributed to a period of his life, and the nature of the association that occurred. The impact of this study is still of great significance today.

Three decades after, Charles Spearman started the idea of having human intelligence measured while studying. During this time Thurstone developed the concept of comparative judgment. In France, Alfred Binet applied psychometrics in creating intelligence test to evaluate children, leading to the creation of the Binet Scale. The works of these men made a vast contribution in today's intelligence testing, personality testing, and vocational testing.

Psychometric Theory

There are several distinctive areas of study involving psychometric theory. Mental tests and data analyses from these tests use a body of theory. Moreover, the classical test theory (CTT) and item response theory (IRT) are under this body theory and there is also one approach quite similar to the item response theory, but is distinct, which is the Rasch model used for measurement. Its development has been founded on requirements of measurement in physical sciences. (Rasch,1960). A study by Bechger (2003) presented a relationship between CTT and IRT. CTT assumes that measures are exchangeable. On the other hand, IRT bases on conditional independence. Yet, IRT is viewed as an extension of CTT and ideas of both theories are correlated with each other. However, there are instances wherein IRT is used to supply statistics in cases where CTT fails.

Psychological Tests

Psychological tests are given by a psychologist to formulate an assessment and evaluation to an examiner, thus is referred to as psychological assessment. Information is obtained through interview or through paper or computer in response to specific questions. Eventually, its accuracy will depend on how the respondent will seriously and honestly answer the questions.

Psychological assessment falls into different categories. The achievement and aptitude tests are frequently observed in the education and employment setting used to evaluate one's knowledge of a certain topic such as math or spelling. It is also used to assess one's capacity to master in a specific area, for instance, mechanical relationships. Intelligence tests, on the other hand, tries to measure intelligence or the ability to comprehend the world, incorporate its functioning, and apply knowledge to improve the quality of life.

According to Alfred Whitehead, "intelligence enables the individual to profit by error without being slaughtered by it." Therefore, intelligence measures potential, not what one has learned, thus supposedly not dependent on culture. For that reason, the test should be designed to be culture-free. Another category is the neuropsychological tests that attempt to gauge deficits in the ability to speak, to think, or to reason that might be a consequence of brain injury or brain damage. Occupational tests endeavors to counterpart the test-taker's interest with that of individuals with known careers. The basis in this type of test is that if the person's interests match with, say, things that interest most doctors, then this person will probably make a good doctor himself. Personality tests appraise one's basic personality and are usually used in researches and forensics to establish clinical diagnoses. Two famous personality tests are the Minnesota Multiphasic Personality Inventory (MMPI) and the Rorschach or the "Inkblot Test." There is also a category called specific clinical tests which attempts to determine specific clinical matters like anxiety or level of depression.

Justifications for Using Tests

The main reasons why psychological tests are given are the fact that they interconnected with each other. First, information are easily accessed using tests rather than by clinical interview. Interviews can be subjective although objectively should be employed thus psychological tests could be helpful to avoid bias Mostly, psychologists are inept in dealing with people so they rather administer test than conduct an interview. With is, interviews can then be used as the final step in the selection process. Second, the information obtained form tests are proven to be scientifically consistent than from an interview especially when it comes to legal matters wherein decisions should be based on standardized information. Furthermore, since these tests are standardized the intelligence or personality of the subjects can be gauged comprehensively. Lastly, it has been shown that it's harder to lie on a test than with a clinical interview since the results of the tests are rather subjective and interviews can be affected and influenced by a number of factors.

Validity and Reliability

Generally, problem with psychological tests lie in their ability to measure what they supposedly have to measure. Validity is otherwise defined as the accuracy or the usefulness of a test. To illustrate, suppose a bank owner would want to know which among the job applicants would work well in his bank, an arithmetic test would not be able to solely determine job success especially if other skills are required for the job.

On the other hand, reliability refers to the extent and ability of the test and its measurement to remain consistent despite repeated tests of same subject in identical conditions. Therefore, a test is said to be unreliable once repeated measurements yield different results. (Rudner 2001). Reliability does not necessarily imply validity. A reliable measurement is a measurement taken consistently, without taking into account that if it's measuring what it is supposed to measure. In terms of precision and accuracy, reliability is precision and validity is accuracy. It is also important to note that without reliability, there is also no validity. A thermometer as such, may be one way to do a valid measurement of temperature but if an electronic thermometer has bad batteries, an unreliable result might be taken thus making the reading invalid until such time that the batteries are changed. Another important thing to take note is that no psychological test is completely valid or reliable since the human psyche is too complex to know anything with full confidence. As a result, even after some extensive research, there can always be such uncertainty.

Stretching Validity

Problems with validity occur once a test is administered for a purpose it wasn't intended for (Richmond, 1997). For example, MMPI which is designed to measure pathological personality traits is oftentimes used in law enforcement as a screening tool. In fact, seminary students, paramedics, pilots, and nuclear facility workers also utilize this test in screening.

Logically, people know what makes a good police like they should have good self-esteem, energetic, has impulse control, or should be fair and objective. But these qualities are complex and it is quite difficult to measure such. For instance, asking a person if he is honest, most probably this person would answer "yes." But genuine honesty could be revealed under temptations of dishonesty and there is no such test that could measure this. So in selecting good candidates as police offices, it is not reliable to select the good traits , then the nest option is to choose the obvious "negative qualities." This is where the measurement of the level of psychopathology comes in. it tells if someone is habitually depressed or is thinking bizarrely, or is socially inhibited, and so on.. It would now be easy to determine if one would make as a good police officer if he is free of those "negative traits" but there are still a lot more in a career rather than just being void from psychopathology. Still MMPI is used as a rough screening instrument to get rid of obvious problems. Nonetheless if MMPI is used solely for selection, it is said that its validity is stretched way beyond reason, and that a serious blunder would be committed since those who don't have the negative qualities but also don't have the right qualities would get accepted anyway. Those who showed some scale elevations can be eliminated even if they are doing well in life. This has been the reason why a psychologist should base his interpretation on the overall profile and has to make considerations in the light of the historical information of the person such as academic and job performance, recommendations, and so on.

Psychological tests are not considered magic. Mistakes will be made even if a decision is made by a qualified psychologist basing on all evidences and there will still be doubts. Hence, a false negative result happens when, for instance a qualified person is rejected for a job. Meanwhile, a false positive error occurs when a person who is not truly qualified for the job gets accepted and this can cause issues on the validity and reliability of the tests.
Furthermore, it is important to consider that the tests are not biased and proper assessment should be employed and this could be achieved by using objective measurements as much as possible in the first steps of the process and subjective measurement can then be used as a concluding mark. What is important is that in using these tests , qualified applicants will be accepted in the job through the utilization of psychometrics.

Conclusion

Mostly, psychological tests are developed in the course of sound scientific principles. A psychologist needs to learn the scientific principles of test construction even if he has no desire to create a test. Psychologists should have the competence to evaluate the scientific value of any test before it will be used clinically. It is unfortunate though that some tests are accepted as scientific. For instance the Myers-Briggs Type Indicator and Enneagram which are used as instruments in assessing personality types are based only in pseudoscience. Then there is also the Rorschach test to assess the inner psychological experience of a person.

There are already several methods to administer and score this test but it would be of great surprise to obtain identical findings if this test is given by two psychologists to one person. It has also a similar case with the Thematic Apperception Test (TAT), asking a person to interpret pictures through stories and the Draw-A-Person and House-Tree-Person asking a person to draw pictures. These tests are usually not objectively scored and the results' validity is held questionable. Psychological testing, in some ways, can both be valid as well as reliable but in other ways it does not realize more than a hazy evaluation of a person.

In the final analysis, psychometric measurements should be used properly to ensure appropriated results. It is important therefore that everything uses standardized testing materials and scientifically proven methods otherwise everything will be forfeited and lose the primary objective in using psychometrics. Measurement has indeed made everything more convenient but it also entails careful assessment to avoid complications.
References

Alfred N. Whitehead (1929). Process and Reality. New York: Macmillan.

Browne, Michael (2000). Psychometrics. Journal of the American Statistical Association, Vol. 95 (450): 351.

Rasch, G. (1980). Probabilistic models for some intelligence and attainment tests. Chicago: The University of Chicago Press.

Richmond, Raymond Lloyd. (2008). A Guide to Psychology and its practice. San Francisco, California USA.

Rudner, L.M., & Shafer, W.D. (2001). Reliability. ERIC Digest. College Park, MD: ERIC Clearinghouse on Assessment and Evaluation.

Winer, B. J. (1971). Statistical Principles in Experimental Design. New York: McGraw-Hill.