| Abstract [eng] |
The work analyses computer literacy test design methods. Following the results of the analysis, the need for developing an intelligent computer literacy test design method was identified. In order to create a compact test an adaptive selection of questions according to the professions of the test takers and area of their activities was used. Prior to the test, computer literacy skills of a test taker are defined and the first test question is selected with regard of this information. It was found out that the way of the work with the computer depends on the occupation; therefore, each question can be given occupational weights. These weights help to select questions for a test taker based on characteristic computer skills of people of different occupations. Clustering of statistical data using self-organizing neural networks helped to estimate test taker‘s initial computer literacy skills level. Shortening of an adaptively constructed test up to certain limit does not have a significant influence on the final test results. The obtained results of the research can be used for testing computer literacy skills at educational institutions, various training and testing centres or for the purpose of qualification improvement. |
