DIGITAL LIBRARY
THE ANALYSIS OF STUDENT BEHAVIOUR IN THE STATE ONLINE DIAGNOSTIC TESTS: EXPERIENCE OF LATVIA
1 Riga Technical University (RTU) (LATVIA)
2 Vilnius Gediminas Technical University, Faculty of Electronics (LITHUANIA)
3 Vilnius Gediminas Technical University (VGTU) (LITHUANIA)
About this paper:
Appears in: ICERI2019 Proceedings
Publication year: 2019
Pages: 2907-2913
ISBN: 978-84-09-14755-7
ISSN: 2340-1095
doi: 10.21125/iceri.2019.0741
Conference name: 12th annual International Conference of Education, Research and Innovation
Dates: 11-13 November, 2019
Location: Seville, Spain
Abstract:
For the past four years, Latvian schools were offered the opportunity to choose how to organize the state diagnostic tests – paper-based or online. Online tests modernize and facilitate the assessment and analysis process, as well as ensure sustainable elaboration of diagnostic tests and their customization to the new science subject model and requirements of modern times. The analysis of the results allows to look for the new approaches and methods, innovative information systems and technologies in order to enhance certain skills, abilities, and attitudes. This article analyses students’ behaviour during the state online diagnostic test. Online diagnostic test was prepared by uzdevumi.lv and the tasks were created by the National Examination Centre. The system of the online diagnostic test has the functionality to observe every detail of the student behaviour while taking the test (e.g. how many times students returned to the same task, did they change the answers, if yes, which questions and how many times were corrected, etc.).

State diagnostic test is one of the diagnostic tools used to ascertain the acquisition of students’ basic skills on finishing the certain level of education according to the national education standards and the subject curriculum requirements. In order to get valid and reliable data, the participation of all types of educational institutions from all regions of the country is necessary. The obtained results of the state diagnostic tests are used to improve the learning process. In order to assess students’ achievements, the test contains tasks with different difficulty and cognitive levels, the content of which relatively covers all aspects contained in the education curriculum. Task content is selected according to real life situations which they might face in everyday life and would be able to deal with them successfully. The diagnostic test contains tasks to examine knowledge and skills which are required to successfully continue the education. It consists of 30 tasks with different levels of difficulty and response types (multiple choice, fill in the missing word, explain the choice in written form). 22 tasks of the test were evaluated automatically by the computer program and the other 8 were reviewed by the teacher online.

The research was conducted among sixth grade (11-13 years old) students in Latvian schools. (This year online diagnostic tests were completed by 8564 students from 249 schools).

The purpose of the research – to analyse the peculiarity of students’ behaviour during the online diagnostic test.

Research tasks:
1) To analyse in which tasks and how many times the answers were changed by students;
2) To ascertain how many students reviewed the answers before submitting;
3) To analyse the correlation between reviewed tasks and correct or incorrect answers;
4) To compare if the diagnostic tests were solved by clever and unskilled students equally (the correlation between skills of the student and behaviour in the test).

The recommendation model for improving the learning process was created while analysing the obtained results and using forecasting tools.

Research methods:
1) Analysis of scientific literature;
2) Descriptive statistics and dependency analysis were used to process the data.

The analysis of the research data was carried out using Classical Test Theory (CTT) and Test Analysis Program ITEMAN™ for Windows.
The charts and tables were created using MatLab & Tableau Public
Keywords:
Student behaviour, online tests, skills, knowledge.