ERG SES H 11, ICT and Education
Currently, the testing process accompanies a person throughout his or her life. Tests are used for measuring the ability of a person to perform different types of activities, the level of qualification in the profession, knowledge of foreign languages, and many other aspects of human activity.
Rapid development of computer technologies makes computerized adaptive testing very profitable. The primary difference between Computerized Adaptive Testing (CAT) and Computerized Testing (CT) generations is that CAT tests are administered adaptively (C. Victor Bunderson, Dillon K. Inouye, James B. Olsen, 1988). This generally requires greater computer speed and computational capability and advances in psychometrics, including generally more sophisticated measurement scales. Adaptive tests provide even greater speed of administration than CT, because fewer items need to be administered for equal or greater precision.
The important advantages of CAT:
- high efficiency;
- high level of security;
- individualization of the rate of the test;
- high level of motivation for the test of the weakest students due to the exclusion of doing unnecessarily difficult tasks in the process.
Methods of testing different mode of construction of the trajectory of testing:
- With the help of the mathematical theory of educational measurement (Item Response Theory, IRT);
- With the use of neural network techniques;
- Assignment of transitions between states.
IRT is based on the theory of a latent structural analysis. To construct the path test using the method of maximizing the information in which each subsequent task selected from a bank of tests to be the most informative for estimating the level of preparedness of the test.
The purpose of this investigation is to apply the theory of latent variables based on the Rasch models to evaluate the effectiveness of Computerized Adaptive Testing (CAT) in Russia, Belarus and Bulgaria.
The following steps were taken in order to complete this investigation:
- The statistical characteristics of CAT used in Russia, Belarus, and Bulgaria were explored, and the Rasch model was selected as a measuring tool;
- Practical examples were analyzed: the range of variation of indicators and measurement objects, the distribution of the indicators, the shift of measurement results with respect to the subjects of indicator variables, the highlighted inadequacy of the data types of survey measurement models and types of violations of construct validity;
- Investigation of the accuracy of the measurement of latent variables, depending on the parameters of the tests.
1. Andrich, D. (1988). Rasch Models for Development, London, Sage Publications, Inc. 2. Anisimova, T.S., Maslak, A.A., Pozdniakov, S.A. (2005) Investigation of the accuracy of the measurement of latent variable, depending on the correlation of indicator variables. Journal of SGPI, 1(1), 126-136. 3. Bachman, L. (1990). Fundamental considerations in language testing. Oxford: Oxford University Press. 4. C. Victor Bunderson, Dillon K. Inouye, James B. Olsen (1988) The four generations of computerized educational measurement, Educational Testing service Princeton, New Jersey. 5. Dodd, B.G., De Ayala, RJ., & Koch, W.R. (1995). Computerized adaptive testing with polytomous items. Applied Psychological Measurement, 19. 5-22. 6. Freedle, R, & Kostin, 1. (1993). The prediction of TOEFL reading item difficulty: Implications for construct validity. Language Testing, 10, 133-170. 7. Lord F.M. (1980) Application of Item Response Theory to Practical Testing Problems. Hillsdale N-J. Lawrence Erlbaum Ass., Publ. 8. Maslak, A. A., & Osipov, S. A. Measurement of Latent Variables. Certificate of the state registration of the computer programs № 2013618487. Date of the state registration in the Register of the computer programs September, 10, 2013. 9. Pozdniakov, S.A. (2007) Research the accuracy of the measurement of latent variables in education. Slavyansk-on-Kuban, Publishing Center SGPI, 118. 10. Rasch, G. (1980). Probabilistic models for some intelligence and attainment tests (Expanded edition, with foreword and afterword by Benjamin D. Wright), Chicago, University of Chicago Press. 11. Wright, B.D., Masters, G.N. (1982). Rating Scale Analysis, Chicago, MESA Press.
- Search for keywords and phrases in "Text Search"
- Restrict in which part of the abstracts to search in "Where to search"
- Search for authors and in the respective field.
- For planning your conference attendance you may want to use the conference app, which will be issued some weeks before the conference
- If you are a session chair, best look up your chairing duties in the conference system (Conftool) or the app.