Distinctive Ability of Concept Maps for Assessing Levels of Competence. Pilot study

Yulia Tyumeneva - Candidate of Sciences in Psychology, Associate Professor, Institute of Education, National Research University Higher School of Economics. E-mail: jutu@yandex.ruAnastasiya Kapuza - Postgraduate Student, National Research University Higher School of Economics. E-mail: a. v.kapuza@gmail.comKseniya Vergeles - Master Student, National Research University Higher School of Economics. E-mail: vergeles.k.soc@gmail.comAddress: 20 Myasnitskaya Str., 101000 Moscow, Russian Federation.Previous research has proved the concept mapping is an effective tool to evaluate knowledge structure, but usually the concept mapping served to foster and trace individual progress in specific field of knowledge. No attention was paid to identifying or verifying the formal indicators of concept maps or their sensitiveness to the level of competence in a specific field of knowledge. However it will make possible to use concept mapping as a standardized tool. In the current study some possible indicators are suggested based on concept maps of experts (n = 4) and novices (n = 9) in the field of data analysis. Experts and novices constructed their concept maps individually after receiving standardized instructions and brief training. Formal indicators were based on interpreting concept map as a graph. Specifically, indicators such as generality of concepts used, structure coherence, proportions of singular, complex concepts, etc. were expected to be discriminative for different levels of competence. We found that nearly all indicators actually discriminate between experts and novices. In addition, a few qualitative parameters of concept maps were identified (availability of key concepts, existence of erroneous relationships, procedural/conceptual nature of knowledge) which also differed across groups. As a result, concept mapping look potentially helpful for standardized evaluation of competence levels if we use the formal indicators. Although further research on extended and heterogeneous samples is required to test stability and generalizability of this formal approach to the concept mapping.