Investigating the Psychometric Structure of Bebras Contest: Towards Mesuring Computational Thinking Skills

• Published in: 2015 International Conference on Learning and Teaching in Computing and Engineering pp. 62 - 69
• Main Authors: Hubwieser, Peter, Muhling, Andreas
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-04-2015
• Subjects: Algorithm design and analysis; Computer science; contest; Fitting; Joints; latent trait analysis; Mathematical model; Numerical models; psychometricconstruct
• DOI: 10.1109/LaTiCE.2015.19

The "Bebras" is an international large scale contest that aims to motivate children for engaging in Computer Science. In this paper we present an investigation of the psychometric factors that determine the success in this contest, e.g. Abilities in Computational Thinking or intelligence factors. For this purpose, we looked for subsets of Bebras questions that are requiring joint psychometric constructs, according to the responses of the test persons, applying latent trait analysis on all combinations of questions. After identifying suitable sets of questions, we tested the fit of the mono-parametric Rasch Model and evaluated the distribution of person parameters. Additionally, the identified sets of questions were analyzed by qualitative methods to find out the nature of these common psychometric constructs. As a test bed for feasibility studies, we have chosen the Bebras Contests of Germany 2009. We have found and analyzed several sets of questions that met our requirements, analyzed the questions, and identified differences in the performance of boys and girls in these sets.