The impact of task difficulty and performance scores on student engagement and progression

Abstract

Background: This article considers the impact of differential task difficulty on student engagement and progression within an Irish primary school context. Gaining and maintaining student engagement during learning tasks such as homework is a significant and understandable on-going challenge for teachers. The findings of this study hold the potential to support teachers’ decision-making processes regarding the development of student tasks.

Purpose: The research study aimed to explore the impact of task difficulty on student engagement and subsequent progression in the computerised navigation task Pac-Man. The central research questions addressed in this article were; do subtle variances in task difficulty impact on student volition and consequently, will this result in a significant variance in students’ levels of improvement?

Sample: Sixty students from a large urban, coeducational primary school in the south of Ireland were identified as a suitable sample cohort. All students were in their final year of primary school within the Irish education system and were between 11 and 12 years of age.

Design and methods: The study employed the use of the popular arcade game Pac-Man. In a test–retest approach, 60 primary school students completed the standard computerised navigation task with a seven-day interval. Between testing, participants were randomly subdivided into three cohorts. Each cohort of 20 participants received a different version of the Pac-Man game to practise with for one week. Cohort A received a version of the Pac-Man game of lesser difficulty, Cohort B received the standard Pac-Man game and Cohort C received a version of greater difficulty. A paired-samples t-test (repeated measures) was employed to compare the scores achieved by each of the three cohorts both pre- and post-practice. As an indication of the resulting effect size for each cohort the eta-squared statistic was subsequently calculated. In order to support any future meta-analysis, Cohen’s d statistic is also provided in this paper. Analysis of variance (ANOVA) was employed to explore differences between groups with regard to progression scores and number of games played when practising.

Results: The results of this small scale study found the cohort who received the easier version of the task presented the greatest overall improvement in performance between the pre- and post-tests. No statistically significant difference was found in the change in scores of the three cohorts – potentially due to the small sample size. However, paying attention to the size of the effect indicated that, over seven days, there was an 80% improvement in performance for Cohort A, 63% improvement for Cohort B and 26% improvement for Cohort C. The results highlight the negative impact of increased task difficulty on students’ volition and consequently, on overall progress in the task.

Conclusions: Further research with larger student populations would be needed to assess the generalisability of the results. However, the findings suggest that when designing tasks to promote student learning, particularly self-directed tasks such as homework, it is important that teachers afford ample opportunity for student success.

Keywords:

• task difficulty
• student performance
• engagement
• self-efficacy
• homework

Notes

1. A non-deterministic version of Pac-Man, often referred to as Ms. Pac-Man, was employed and adapted for this study, as it ensured that participants could not learn or predict the movement of the game’s characters. The ability to predict the characters’ movements would have reduced the skill level required to complete the task and distorted the findings. It was developed by extending the C# implementation of the game provided by Flensbank andYannakakis (available at: http://mspacmanai.codeplex.com/).