Computer-aided psychodiagnostics

from Wikipedia, the free encyclopedia

Under computerized Psychodiagnostics refers to the use of information technology as a means of psychological tests and other psycho diagnostic methods of gathering information to carry out, evaluate or interpret and automatically create findings and to support diagnostic decisions. In psychological diagnostics , this competes with " paper-and-pencil diagnostics ", where the procedures are carried out using paper.

Definition

“Computer-assisted” emphasizes that it is an aid for psychological diagnosis, which proceeds according to the concept of the psychodiagnostic process . The (human) diagnostician is responsible for finding a diagnosis or making a diagnostic decision, who questions all results and, if necessary, supplements them with further examination methods (cf. the role definitions of DIN 33430 for suitability diagnostics ).

  • Synonyms are internet-based psychodiagnostics, computer-based psychodiagnostics, computer-assisted psychodiagnostics. Computer diagnostics is also common, but ambiguous in the sense of “diagnostics of computers”.
  • Instead of psychodiagnostics, psychological diagnostics can also be used.
  • Since the technology used is no longer limited to computers in the classic sense, there is increasing talk of IT-based, IT-supported (IT stands for information technology).

A "fully automated" test application such as B. With online self-assessments it is also possible

The abbreviation CBT computer based testing or in German-speaking computer-based testing is used for this purpose, the English-language abbreviation CAT for computer assisted testing is also used in the German language, but also for computerized adaptive testing ( Computerized adaptive testing use) as a sub-form.

Influence on the diagnostic process

The interconnection of conventional and IT-based elements can be very diverse. Different parts of the diagnostic process can be supported by the use of computers:

  • the question-specific selection of suitable tests through the use of information and research tools and the preparation or administration (management of persons and results, compilation of test batteries , generation of login information for the tested).
  • the implementation as a presentation and processing of psychological test procedures
  • the evaluation and interpretation as well as automatic generation of findings (so-called reports that explain the procedure and result in detail for the diagnosed person, mostly using result-dependent text modules).
  • support for diagnosis and decision-making ( expert systems or knowledge-based systems ).

There are several possible ways to use information technology to conduct tests.

  • Conventional implementation as a paper and pencil test, using machine-readable forms that are automatically evaluated. A variant of this are evaluation services, where the completed test is sent to the evaluation office by fax, and the result is returned by email.
  • Program-based methods that require the prior installation of a program in order to carry out tests on a local computer.
  • Web testing or Internet testing (also online testing), which enable server-based implementation directly on the Internet . This also enables a spatial separation between the diagnostician and the diagnosed person; preparation, implementation and result interpretation can be separated in time.
    • One variant is the use of so-called players, which are loaded from the Internet onto the local computer of the diagnosed person and control the execution of the test there. No permanent internet contact is necessary during the session, the bandwidth is not overloaded during group tests and precise tests can be carried out in this way.
    • Outpatient assessment is a variant of data collection directly in a situation to be assessed, earlier using a palmtop and local program, today also with smartphones and tablet computers using the Internet.

The introduction of the Internet has led to a great variety of methods, some of which still have to be worked up theoretically. It is characteristic that the administration by the specialist (initiating the testing, evaluation) and the actual test execution are locally separated, the specialist no longer has to be in the same place as the person being tested. Another variety are online assessments , where those tested often receive the results immediately after the test and usually no specialist is required to interpret and explain them (especially for online self-assessments , which are used for student advice). Here, too, the specialist continues to be responsible for ensuring that the results are understandable and that no damage can occur as a result of any uncertainty or misinterpretation.

Implementation forms

With reference to Dave Bartram , a distinction is made today between four modes of computer-aided diagnostics. All four modes apply to Internet testing; locally permanently installed programs can also be used for supervised testing.

  • Open mode : testing without supervision (unproctored), those tested remain anonymous and cannot be identified. There is no registration option for users, logins must be avoided or verifiably be the same for all persons. Results are offered for download via the Internet portal immediately after the test. Example: Online self-assessments for the choice of degree, where you only admit your own mistakes and weaknesses in the case of personality traits if you are sure that the result will not be used for admission regarding the anonymity of the test.
  • Controlled mode (Controlled mode): Test also without supervision (unproctored). However, access authorization is required (login with user name and password or a personalized link to start the test). This case also applies if an e-mail address is to be provided to send the results. Example: all commercially offered tests where individualized payment creates individual access or a user administration is used for the tested.
  • Supervised Mode (Supervised mode): Test under supervision (proctored). The identity of the person being tested is checked; a supervisor is at least available during the test (who helps with technical problems, for example). The necessary technology is made available to the tested person. Login is made by the specialist who starts the test and checks that it has ended properly. As a rule, the results are then also explained by the specialist. Example: if tests are carried out in a psychological practice in a separate room / on a separate PC.
  • Controlled mode (managed mode): Also testing under supervision (proctored). Exact control of the identity of the people and supervision of the execution of the test (for example with regard to the avoidance of the use of unauthorized aids, the use of recording technology - to "spy on" the test - or unauthorized agreements in group tests). Example: if admission or aptitude tests are relevant for the future of the person and these tests e.g. B. be carried out in specially certified test centers.

Technical modes

When testing, four modes can be distinguished, which are to be assessed differently in terms of data protection and convenience: Those tested must be informed of these modes and possible risks in corresponding declarations of consent. A distinction must be made between the preparation and selection of the tests to be carried out (diagnostician), the actual test execution, the evaluation and the data storage / archiving.

  • Local testing: preparation, testing, evaluation on a local PC or device, data protection can be implemented in the strictest way, since no personal data or results leave the device. They are also archived there. This classic mode corresponds to the previously necessary permanent installation of a program on a local device.
  • Portable testing: preparation and evaluation on the diagnostician's device, testing on another device possible at any other time (e.g. when diagnosed), the complete test program is e.g. B. provided on a USB stick, where the results are saved and then read in again on the PC of the diagnostician. Data protection is also easy to implement, especially if the data is also encrypted on the USB stick and archived on a device.
  • Intranet testing : Preparation, testing and evaluation possible on different PCs, the connecting server is in the corresponding facility in the intranet and data does not get outside. Data protection: as far as possible, how the intranet is secured against external access. This is usually a prerequisite and the collection of sensitive personal data and archiving takes place in patient or client administrations in appropriate institutions.
  • Internet testing: Use of a central server at the test provider, preparation, testing and evaluation is possible on various devices that are connected via the Internet . The diagnostician and the diagnosed person can be located in different places (so-called temporally and spatially decoupled diagnostic process). Most providers guarantee corresponding data protection conditions and encryption. Nevertheless, sensitive personal data leave the facility and there is the option of making external communication anonymous with the appropriate person codes and then relinking the test result to the person at the facility. Archiving can take place at the test provider or locally ( download and delete on the Internet server). Disadvantages can be the susceptibility to errors in the correct assignment and the impossibility of personalized findings (with the longer evaluation reports that are customary today). Age, gender and possibly other data such as occupation or level of education must then be transferred to the Internet server if they are necessary for the selection of the correct standard . This is the standard configuration for online assessments .

Advantages and disadvantages

Advantages and disadvantages are often discussed compared to paper-and-pencil diagnostics . The advantages of computer-aided psychological diagnostics include, for example:

  • Time economy, many test subjects can be reached in less time
  • a quick and less error-prone evaluation and output of the test results
  • Interpretation aids through expert systems, automatic classification and other methods
  • more objectivity in the test execution
  • the ability to specify adaptive test procedures (response-dependent process design)
  • the use of multimedia components is made possible
  • By networking several test stations (locally or via the Internet) under a central evaluation, diagnostics is made possible even with local separation of diagnosed and diagnosed
  • behavior-based evaluation through the use of log files , also in connection with data mining

Disadvantages can be:

  • Reduction of human communication, direct interaction and behavioral observation - especially in clinical use as a shortening of the anamnesis
  • Errors due to uncritical adoption of non-transparent automated findings, especially when non-specialists develop and evaluate the tests (see training standards of DIN 33430 )
  • Influence of competence in handling computers on the part of the test participants, which is partly age-dependent (see ICT literacy )
  • Dependency on software and hardware (e.g. compatibility of end devices)
  • major hurdles in test development and evaluation (e.g. programming, financing)

Test systems

In addition to numerous individual programs, there are test systems that combine subject management, testing, evaluation and other auxiliary programs for psychological diagnostics under a uniform interface. Examples in the German-speaking area are the Hogrefe test system , the Vienna test system or the CAT test system .

There are some other test systems whose implementation problem on the market is primarily that suitable and requested tests must be offered in the system with sufficient bandwidth and where the authors are more oriented towards the larger test systems. Test development is traditionally very complex, and test systems are only successful in the long term if sufficient resources are available for test development.

The traditional test systems are more oriented towards the test application (test execution, evaluation) and leave the interpretation and decision-making to the diagnosticians. In addition, so-called process-oriented systems have developed that also provide greater support for decision-making (e.g. diagnostics in application portals) and offer ready-made evaluation options. There is also a difference in whether the test system is designed to be universal (for all questions) or whether it specializes in certain questions. Such systems can be more strongly oriented towards the diagnostic process of this question and deliver process selection or cross-process evaluation, for example based on a requirements analysis.

In the context of empirical educational sciences or international comparative educational studies , various test systems are used, which are particularly designed to be able to test a large number of study participants at the same time. For research there are also test systems for specific tasks (e.g. complex problems ), which are intended to combine and simplify the creation of computerized tasks and test execution.

literature

  • JF Booth: Computer Diagnostics. In: RS Jäger, F. Petermann (ed.): Psychological diagnostics - a textbook. 2nd Edition. Psychologie Verlags Union, Weinheim 1992.
  • J. Hageböck: Computer-aided diagnostics in psychology. Hogrefe, Göttingen 1994.
  • D. Klinck: Computer-assisted diagnostics. Hogrefe, Göttingen 2002.
  • KD Kubinger: Introduction to Psychological Diagnostics. Beltz, Psychologie Verlags Union, Weinheim 1995.
  • KD Kubinger: Advantages and disadvantages of computer diagnostics. In: Psychology in Austria. Volume 13, No. 1-2, 1993, pp. 25-29.
  • T. Walter, G. Schuhfried: Computer-aided psychological diagnostics. In: G. Mehta (Ed.): The practice of psychology. Springer, Vienna / New York 2004.
  • FH Wilhelm, MC Pfaltz: Computer-aided diagnostics and new media. In: J. Magraf, S. Schneider (Ed.): Textbook of behavior therapy. Springer, Heidelberg 2009, pp. 409-432.
  • E. Volz-Sidiropoulou: Computer-based psychodiagnostics. In: H.-J. Fisseni (ed.): Textbook of psychological diagnostics. Hogrefe, Göttingen 2004.

Individual evidence

  1. a b Computer-based diagnostics in: DORSCH Lexikon der Psychologie
  2. Dave Bartram, Ron Hambleton: Computer-Based Testing and the Internet. John Wiley & Sons, 2005, p. 168. (books.google.ch)
  3. Attention, in the English-speaking world, test administration means the execution of tests. In the German-speaking world, the "administrative work" of the diagnostician is meant.
  4. see also K.-D. Hänsgen: Help with the Hogrefe TestSystem.
  5. ^ D. Bartram: The Changing Face of Testing. In: The Psychologist. Vol 18, No 11, pp. 666-668.
  6. Example video for the processes in a test center, here Pearson VUE at the GMAT test (English)
  7. Klaus-Dieter Hänsgen: Handbook for the Hogrefe test system. Hogrefe Verlag, Göttingen 2011. See also the help system for the Hogrefe test system version 4
  8. K.-D. Hänsgen, M. Perrez: Computer-aided diagnostics in family and upbringing: Approaches and perspective. In: Psychology in Education and Teaching. Issue 3, 2001.
  9. z. B. eligo or PERLS as requirement-related suitability diagnostics, cf. International personnel selection: search, selection, integration Anja Peitz 2002 Symposium Publishing
  10. CBA ItemBuilder . Website of the German Institute for International Educational Research. Retrieved May 11, 2016.