No abstract

Non-cognitive factors have been considered as particularly important aspects in shaping students' academic achievement. The current study aimed to examine a number of these factors in relation to the prediction of mathematics achievement among a representative sample of students in Greece. The sample consisted of 5,125 15-year-old students who participated in the Programme for International Student Assessment (PISA) 2012. The study considered the extent to which students' mathematics self-beliefs, motivation towards mathematics and attitudes towards school contributed to the prediction of their mathematics achievement. Multilevel modelling assessed both individual and school level variation, revealing that students' mathematics self-efficacy, anxiety, self-concept, instrumental motivation and attitudes towards school were statistically significant predictors of their mathematics achievement, even after controlling for their gender and school socioeconomic status (SES). Moreover, students were found to differ in the levels of their self-beliefs, motivation and attitudes based on their gender.

The linguistic complexity of many text-based tests can be a source of construct-irrelevant variance, as test-takers' performance may be affected by factors that are beyond the focus of the assessment itself, such as reading comprehension skills. This experimental study examined the extent to which the use of animated videos, as opposed to written text, could (i) reduce construct-irrelevant variance attributed to language and reading skills and (ii) impact test-takers' reactions to a situational judgment test. The results indicated that the variance attributed to construct-irrelevant factors was lower by 9.5% in the animated version of the test. In addition, those who took the animated test perceived it to be more valid, fair, and enjoyable, than those who took the text-based test. They also rated the language used as less difficult to understand. The implications of these findings are discussed.

The role of digital technology in assessment has received a great deal of attention in recent years. Naturally, technology offers many practical benefits, such as increased efficiency with regard to the design, implementation and scoring of existing assessments. More importantly, it also has the potential to have profound, transformative effects on the field of assessment by facilitating the integration of formative activities with accountability requirements and broadening the range of abilities and the scope of constructs that can be assessed. This article provides an overview of the current state-ofthe-art in digital technology-based assessment, with particular reference to advances in the automated scoring of constructed responses, the assessment of complex 21 st century skills in large-scale assessments and innovations involving high fidelity virtual reality simulations. Key challenges with respect to each are highlighted before the extent to which digital technology is truly transforming assessment is considered. | I NTR OD U CTI ONThe interconnectedness of technology and assessment 1 has a long history. Madaus (2001) refers to a technology as any body of special knowledge, skills and procedures that people use to satisfy a need, solve a problem or attain a societal, economic, or educational goal. Under this definition, it is clear that educational assessment itself is a technology and, as Kellaghan and Madaus (2003) point out, one that has existed since an external civil service examination system was invented in China around 1100 BC. While the system changed over the centuries until its demise in 1905, the general approach involved candidates writing up to eight essays within an allocated time period in which they explained ideas from the Four Books and Five Classics of Confucianism (www.sacu.org/examinations.html). In many respects, the artefacts and processes of assessment that make it a technology have been visible from the very beginning in the form of test booklets and answer sheets, as well as in the specialist knowledge, skills and language of its community of practitioners. What has changed in the recent past is the speed with which one technology (digital technology 2 ) is transforming another (assessment) in terms of the constructs that can be measured and the types of 160 | How to cite this article: O'Leary M, Scully D, Karakolidis A, Pitsia V. The state-of-the-art in digital technologybased assessment. Eur J Educ. 2018;53:160-175. https://doi.