Abstract: Original method to study students’ conceptual knowledge was reported at previous ITHEA
conferences. According to this method, to show their educational achievements students had to pass through
some testing procedure, which requires combining of basic terms from a learnt course into interrelated pairs.
Then students’ answers, saved in text files, were analyzed by computer software that joined all interrelated
concepts into groups, independent from each other. The larger these groups were the better students’ success in
digestion of learnt course was supposed.
As human cognitive processes are very complex, it is important to make sure that experimental method is really
valid and steady. The present paper examines stability of the experimental results; new implemented experiments
were also aimed at estimation of method’s inaccuracy. Several diversified tests were made to assure that method
error do not deface observed picture: the difference between repeated tests was evaluated; the stability of
student’s pair selection and of knowledge fragmentation were analyzed to make sure that we can trust the results
of experiments; regression towards the mean effect, which can skew the outcome of low-performed students, was
estimated – the situation was found to be secure; another discipline was checked using the same method and
there was no noticeable dependence of numeric scaled gauges from a learning content.
Detailed study shows that the generality of students demonstrate a growth of the proposed measure of learning
success and its magnitude distinctly exceeds inaccuracy of method. The value of this inaccuracy was also
estimated.
Keywords: concept, relation, semantic network, knowledge representation, learning
ACM Classification Keywords: H.1.2 User/Machine Systems – Human information processing, I.2.4 Knowledge
Representation Formalisms and Methods, Experimentation – Semantic networks, I.2.6 Learning, Human factors –
Concept learning, Knowledge acquisition.
Link:
SEMANTIC NET FROM CONCEPTS AS A MODEL OF STUDENT’S KNOWLEDGE:
HOW STABLE ARE THE RESULTS OF EXPERIMENTAL STUDY?
Evgeny A. Eremin
http://www.foibg.com/ijima/vol03/ijima03-01-p04.pdf
Wiki
(1.00/10)
