Effects of Technology in Mathematics on Achievement, Motivation, and Attitude: A Meta-Analysis

24 articles, 4,522 subjects

Mathematics learners; grade span not recoverable from the abstract or the ERIC record.

Journal of Educational Computing Research 57(2) 283-319; first published online 28 December 2017, issue dated April 2019 - the 2019 date is the issue year and is the one recorded here. ASSESSED AT ABSTRACT LEVEL ONLY: the SAGE full text is paywalled and no OA copy exists (unpaywall reports is_oa false; ERIC has metadata only; SAGE, ResearchGate and Semantic Scholar all refused). The abstract and the ERIC record report NO effect size of any kind - only that there is "a significant overall impact of technology on student achievement, motivation, and attitudes" that "varies based on the different aspect of the intervention examined". The declared moderators are intervention type, treatment type, duration, mathematical content area and learning-environment context; conspicuously NOT study design, randomization, sample size, or outcome-measure type, which are the four moderators that this archive has established determine the answer in ed-tech. The quality_grade C is the methodology default for a meta-analysis of mixed designs and is provisional - it has not been earned by inspection.

24 articles and 4,522 subjects, reported as showing "a significant overall impact of technology on student achievement, motivation, and attitudes", with no effect size stated anywhere in the publicly available record. On the achievement question it is a small-pool restatement of ground covered by Cheung & Slavin (2013) with 74 studies and 56,886 K-12 students, which additionally reports the design-quality, sample-size, publication and implementation moderators this review does not appear to code. Its distinctive contribution - motivation and attitude outcomes - is a different outcome class (non-cognitive, largely self-report) and would not bear on the achievement verdict even if the numbers were recoverable. Recorded as an exclusion rather than dropped, so the reader can see what was considered and why it was set aside.

Unknown - the study designs in the pool could not be inspected. If the pool is mostly quasi-experimental, as is typical in this journal, selection into technology-using classrooms is uncontrolled.

We could not read the study pool or any effect size, so we cannot say whether its findings replicate. Recorded as `unclear` rather than `unreplicated`, which would assert something about the literature we have not checked. Its subject matter is fully covered, on a study pool an order of magnitude larger, by Cheung & Slavin (2013).