Enhancing statistics proficiency through computer-assisted feedback

  • Mehmet Filiz

Student thesis: Doctoral ThesisDoctor of Philosophy


Statistics courses are compulsory in numerous college and undergraduate programs, yet students enrolling in these courses can encounter some difficulties including insufficient statistical knowledge, unsuccessful execution of advanced learning strategies, negative feelings about statistics and high statistics anxiety. These present colleges and universities with challenges that need addressing to ensure student success.

This research explored the effectiveness of pedagogies of statistics learning that included computer-assisted feedback when studying independently, computer- assisted feedback combined with teacher feedback and/or computer-assisted feedback combined with peer feedback. The differential effects of independent study, teacher feedback and peer feedback when combined with computer assistant feedback were determined using a randomised controlled trial. The trial was conducted with 53 undergraduate/graduate students at Queen’s University Belfast who had previously taken at least one statistics related course. Participants completed pre-tests and were then randomly allocated to three conditions/groups using learning supported by teacher feedback, by peer feedback or undertaken as independent study-based pedagogies. In each condition, participants were asked to create initial concept maps, undertake a series of learning units utilising the allocated pedagogy and generate revised concept maps (note that they completed concept maps before and after each learning unit). In the teacher feedback pedagogy condition, students were given the teacher-based feedback on their concept maps, highlighting discrepancies between their concept maps and the teacher’s ‘model’ concept map. In the ‘model’ feedback maps, students could access to linked explanations, examples, problems and reflective activities related to their areas of weaker conceptual understanding. Students studying with peer feedback pedagogy were provided with dyad-based feedback on their concept maps, demonstrating the similarities and differences between their concept maps, their peer’s concept map and the teacher’s ‘model’ concept map. Then, students were prompted to ask questions of each other concerning the identified similarities and differences in the peer feedback. Finally, those undertaking the learning units with independent study received a collection of instructional materials and were requested to study these materials. After completing all the appropriate learning units, participants completed a series of post-tests (these were identical to pre-tests). A process evaluation was conducted that included data gathering using semi-structured interviews with a randomly selected group of five participants concerning their learning experiences in using the Concept Maps for Learning website (CMfL).

Data analysis indicated that providing teacher feedback, peer feedback and independent study pedagogies did not lead to statistically significant gains for students regarding their superficial knowledge of ANOVA concepts. Conversely, students studying using the teacher feedback pedagogy significantly improved deep knowledge of ANOVA concepts when compared to those who undertook the independent study pedagogy. Data indicated that student planned to use only selfmonitoring learning strategies in studying statistics, but that while studying the learning units, students in the receipt of the teacher feedback and peer feedback pedagogies executed elaboration and monitoring learning strategies. In addition to this, students undertaking work using the peer feedback pedagogy used more selfregulation learning strategies when compared to those studying the other pedagogies. Students in the receipt of peer feedback reported that they found statistics useful, relevant and worthy in their personal and professional life compare to those studying using the independent study pedagogy. Students found the CMfL website effective, easy to learn from, efficient to use, and satisfactory for purpose. They highlighted that the design of the CMfL website might lighten the workload of teachers. Although creating initial concept maps was challenging for many, this task activated their prior knowledge.

For future studies, repeating this study with sample from different contexts and/or a larger sample size and developing more learning units are recommended. It was also argued that there is a need for further studies aiming to explore the effects of different teaching/pedagogic approaches on students’ use of cognitive and metacognitive learning strategies and their attitudes to learning statistics with technology.
Date of AwardDec 2019
Original languageEnglish
Awarding Institution
  • Queen's University Belfast
SponsorsTurkish Ministry of National Education
SupervisorAllen Thurston (Supervisor) & Sarah Miller (Supervisor)

Cite this