What is psychological diagnostics?
Many people know a diagnosis only as the result of a medical opinion. This primarily takes physiological characteristics into account. The relevant characteristics - the symptoms - are recorded and evaluated. In clinical-psychological diagnostics, the procedure is analogous. It is assumed that a certain combination of characteristic features can be combined to types - to syndromes - which, in the result, correspond to a diagnosis. An important difference between the two disciplines is the object of investigation. In medicine it is predominantly physical characteristics, whereas in psychological diagnostics it is predominantly - to exclusively - about psychological characteristics.
Psychological diagnostics measures characteristics of human experience and behavior. By characteristics of experience is meant internal states of consciousness and emotions. Properties of behavior refer to internal processes such as thoughts (cognitions) and externally observable actions. Their expressions are collected and utilized as information (p. 3, Schmidt-Atzert & Amelang, 2012). With the help of psychological diagnostics, questions regarding current and typical trait expressions of a person can be answered. Answering the questions follows a specific set of diagnostic rules and supports systematic and objective decision-making.
Specificity in the measurement of mental characteristics
In general, we have many standardized measurands such as the kilogram to describe a weight, as well as a meter to spatially describe a length, 38 degrees Celsius to determine the temperature of a hot summer day, or 5,000 Scoville to illustrate the spiciness of a spicy chili pepper. Many of these metrics are so familiar to us that we can no longer imagine our daily lives without the measurability of these things. We often assume that we can simply measure everything in our world. Unfortunately, this is not the case. Especially in the case of things that do not exist physically - that is, are not directly tangible. This poses a challenge to psychological diagnostics because the object itself - the properties of our experience and behavior - cannot be measured with a tape measure.
To overcome this measurement challenge, methods in psychological diagnostics have been developed to make these not directly observable characteristics measurable. Manifest traits are formulated for a latent trait of a person that exists in the world in real terms but is not directly measurable. Manifest characteristics represent the measurable properties of the latent characteristic that is not directly observable. An example of a latent trait is "empathy." In a lens model, the assignment of traits can be best represented graphically. For this, we take as an example the human ability to recognize emotions as a sub-facet of empathy (Cf. p. 25, Eid & Schmidt, 2014). It is our latent characteristic that exists in reality and cannot be observed directly.
Example: emotion recognition ability of children
Figure 1: Mapping between a child's abilities and recorded characteristics by a test. Exemplary representation as Brunswik's lens model.
In the center of the lens model, 9 stimuli are listed (Figure 1). Arrows to the left of the lenses show characteristics of the child that influence the ability to recognize emotions. To the right of the lenses represent characteristics considered by the test instrument. Green lenses represent characteristics of the child not considered by the test that influence emotion recognition competence. Assume that a child can correctly interpret the facial expressions and gestures of his or her own culture. Not, however, those of other children of cultures with which he or she has not yet had contact. For example, sticking out one's tongue has a different meaning in Tibet than in Europe. It is the same with the approving and disapproving shaking of the head in India. The red lens represents a trait that the test falsely measures but has nothing to do with the child's actual ability to recognize emotions. For example, assume that the people whose expression of emotion is to be assessed are more attractive than average. The children might be influenced by the sight of them and feel pleasure instead of recognizing the emotional expression. Because this bias is included in the test score, it unintentionally distorts the test score that is actually of interest. The gray lenses represent characteristics that affect ability and are appropriately accounted for by the test. The gray lenses therefore represent the optimal case, since these actual characteristics of the child are also captured by the test.
How well a test captures a child's true trait expression is determined by the appropriate combination of stimuli. The better the fit, the more accurate the result. Because different psychological tests capture different traits (lenses), the differences between psychological measurement and physical measurement are as follows:
- Lack of standard of comparison: Which test perfectly measures the latent construct empathy? What standard can be used to calibrate a test? There is no universally valid standard of comparison, which is why each test is at best an approximation of the true expression of the trait.
- Precision of measurement: Empathy can only be measured very roughly with the help of a single question. Significantly more questions / tests are needed to be able to perform an accurate measurement (cf. p. 20, Eid & Schmidt, 2014). The challenge is to ask only as few questions as necessary and the right questions.
Who needs it? Who benefits from psychological diagnostics?
Psychological diagnostics is not limited to the clinical field. Its scope of application is manifold.
For example, traffic psychologists prepare expert reports on the fitness of drivers to drive. They attempt to answer psychological questions by means of various diagnostic survey methods in order to make a final decision about fitness to drive. Typical instruments for this are, for example, interviews on risk behavior or a personality test that measures risk appetite.
In a school psychological counseling center, a child's conspicuous behavior is examined by means of an intelligence test to determine whether the child is over- or underchallenged in class (Cf. p. 20, Eid & Schmidt., 2014).
In forensic decision-making, such as the culpability of offenders, intelligence tests can be used to answer the question of the presence of below-average cognitive performance (cf. p. 7, Schmidt-Atzert & Amelang, 2012).
For personnel selection procedures for trainees, performance tests are often used in which, for example, general knowledge, skills in scientific subjects or intelligence are measured. For candidates with professional experience, structured interviews are usually conducted or work samples are collected. Whether a software developer can solve complex technical problems can be tested appropriately using a concrete situation.
Psychological diagnostics is generally useful to anyone who wants to make an accurate, objective decision about a person. It is supported by existing knowledge that can be drawn upon. For example, well researched existing instruments such as intelligence tests. In addition, it also supports scientific quality standards that are required of psychological tests in order to design a useful instrument in the first place. Finally, psychodiagnostics benefit not only those who use them, but also those who are assessed. Assessed candidates benefit from a comprehensible, transparent, objective test result.
How do I choose a psychological test?
First and foremost, a concept must be designed in order to be able to answer psychological questions. This concept defines which criteria (latent characteristics) are relevant and on the basis of which characteristic expressions (manifest characteristics ) the psychological questions are answered. Once the significant constructs have been defined, appropriate diagnostic tests can be sought. For each psychological test instrument there is a manual in which it can be checked whether the construct sought can be captured by the test. If there are several tests to choose from, they can be compared on the basis of the quality criteria. It is recommended to choose the test that has the higher characteristic values for the construct of interest in comparison.
Quality criteria are quality characteristics that can be applied to a psychological test. They can be used in different ways. They are used to select a suitable test for the psychological question of interest, to compare tests with each other, or to develop tests further. There are three main quality criteria. They are listed below in ascending order of priority:
- Objectivity (independence),
- Reliability (dependability) and
- Validity (validity).
A test is considered objective if the test values were obtained independently of the person performing the test. Conversely, the result must not have been subjectively influenced by the test user. Objectivity includes the performance, evaluation and interpretation of a test.
A test is considered reliable if the same result is obtained after a repeated measurement. It must be taken into account here that there are characteristics that are allowed to change. For example, characteristics that are unstable over time, such as well-being, or characteristics that change over a lifetime, such as emotional stability. The important point here is that the variation in the test result may only be explained by changes in the trait carrier - the person tested. Not by a measurement error of the test.
A test is considered valid if it really measures what it claims to measure (cf. p. 49 Eid & Schmidt, 2014). This is the case when the construct of interest is the real cause of the test result. In other words, a person with high mental abilities will achieve a high test score as well as a person with low abilities will achieve a lower test score. This described view of validity refers to the goodness of the test itself. Another view of validity describes whether the test is used for the right purpose. Here, validity describes the meaningfulness with which a test should be used to answer a question (Cf. p. 49 Eid & Schmidt, 2014). An attention test, for example, is more useful in a recruitment process for air traffic controllers than for competitive athletes. Again, the use of a compatibility test (personality) in team sports makes more sense than for air traffic controllers.
Test review - who tests the tests?
The Federation of German Psychological Associations has established a board with the mandate to evaluate psychological diagnostic procedures. The test assessment system of the Test Curatorium publishes reviews on an irregular basis. In it are ratings in the main quality criteria: Objectivity, reliability, validity to general information, description as well as diagnostic objective. Based on a tabular presentation, one can quickly see whether the requirements are fully, largely, partially or not met. The reviews can be viewed free of charge and are a valuable source for test users. An overview can be found here: https://www.bdp-verband.de/publikationen/testrezensionen.
How good does a test has to be?
Insofar as one is concerned with the metrics of the main quality criteria, one might get the impression that there can be a test that could deliver a perfect measurement. The probability is extremely low that there will ever be a test that delivers undoubtedly causal results (p. 470 Döring & Bortz, 2016). In truth, there need not be a perfect measurement either. What is important is that the test provides added value and contributes positively to correct decision making.
Diagnostic and testing board of trustees of the Federation of German Psychological Associations. (2017). Quality standards for psychological assessments.
Döring, N., & Bortz, J. (2016). Research methods and evaluation. Wiesbaden: Springerverlag.
Eid, M., & Schmidt, K. (2014). Test theory and test construction. Hogrefe Verlag.
Fort Walton Beach, F. L. (2001). Psychological Testing and Psychological Assessment.
Schmidt-Atzert, L., & Amelang, M. (2012). Psychological diagnostics (Textbook with online materials). Springer Science & Business Media.