Method for Forming Linear Scales for Assessing Learning Outcomes with Control of the Adequacy of Indicator Variables

Abstract

The purpose of the research is to construct a linear scale for assessing learning outcomes, and to assess the relevance of the indicator variables involved in the formation of the scale.Methods. An analysis of existing approaches to assessing competencies has shown that the most effective approach is the activity-based approach, which involves assessing competencies as a result of current, milestone and/or final control in a discipline or group of disciplines. For this, a scale is needed ‒ a measuring tool that allows you to assess the level of knowledge in a given competency. The quality of the measurement instrument was assessed within the framework of latent variable theory based on the Rasch model. We used the interactive environment “Measurement of Latent Variables, developed in the laboratory of objective measurements of Kuban State University, and the RUMM 2020 software, which is widely used in our country for processing student testing results.Results. For the universal competence Able to manage one’s time, build and implement a trajectory of self-development based on the principles of lifelong education, software and a pool of test tasks have been developed. Using latent variable theory, the validity of a linear scale for this competency was tested. Linear scale indicators have been identified that most adequately characterize the quality of knowledge level assessment, and indicators with the greatest and least sensitivity to the level of knowledge in this competency have been identified.Conclusion. The developed linear scale for assessing learning outcomes was used to measure the level of development of the universal competence UC-6 among students of Kuban State University. The results obtained are necessary for monitoring the formation of the UC-6 competence and for determining the factors influencing the formation of this competence.