Impact of dynamic soundfield on delivering improvements in educational attainment and closing the attainment gap with young learners in mainstream primary school

Abstract

INTRODUCTION: Young people in classrooms are generally listening to speech in an acoustically complex environment composed of long reverberation times and multiple competing voices generated by the room occupants. The challenges of listening to conversation when there are several competing voices is known as the ‘cocktail party’ or ‘multi-talker listening’ effect. Many of the cues that help to filter speech in noise are sometimes inaccessible to young people due to an auditory system that is going through a process of maturation; young people have limited life experience and an underdeveloped knowledge of linguistic rules. Education systems throughout the UK and beyond have gone through a period of reform, with the traditional didactic teaching methods supplemented by more interactive and student-focused learning which are commonly associated with higher noise levels. Several studies have shown that noise levels have a detrimental effect on learning and cognitive function. At the same time, concerns over the general attainment levels in schools and the gap between learners from the most and least deprived areas have come to the forefront of government policy.

AIMS: The primary purpose of this research was to evaluate the effects of dynamic soundfield technology on educational attainment with young primary three learners and to determine if its efficacy is moderated by the speech weighted C50 properties of the classrooms. Dynamic soundfield is a technology that monitors the level of background noise in a classroom and adapts the level of gain provided to the teachers’ voice to enhance speech intelligibility in noise. Another primary aim was to explore and evaluate the effects of dynamic soundfield on learners from Scottish Index of Multiple Deprivation (SIMD) 1 and 5 to discover if it contributed to closing the attainment gap between learners from the most and least deprived areas of Scotland. One of the secondary aims of this study was to measure the acoustic properties of the occupied and unoccupied research classrooms and to establish quantitative data on the overall use of the dynamic soundfield systems in each intervention classroom. In addition, the young learners’ and teachers’ experience of noise when exposed to the intervention was compared to the control. The findings from the research are discussed with reference to the literature and in relation to implications for practice.

METHODS: This research was a longitudinal repeat measure study design. The 495 primary three learners from 13 schools were allocated to control and intervention classrooms through concealed randomisation. Overall, there were 25 research classrooms, 13 intervention classrooms were fitted with dynamic soundfield (278 learners) and 12 acted as a control (217 learners). All the participants completed the Achievement for Excellence (Interactive Computerised Assessment System) (AfE (InCAS)) suite of adaptive assessments at the start and end of the study. These were administered and marked by the CEM at the University of Durham who were blind to the intervention.

RESULTS: Comparisons of the pre-test and post-test Developed Ability and Mental Arithmetic module scores revealed that dynamic soundfield systems were primarily effective in classrooms with good C50 values. In the data handling subtest, the dynamic soundfield intervention produced significant improvements in classrooms with excellent acoustics for speech. In reading, the intervention classrooms categorised as fair and good for speech clarity demonstrated significantly higher learner scores in word recognition, comprehension and reading compared to the control. Only the good C50 intervention classrooms demonstrated a significant improvement in word decoding. Results from a series of two-way mixed ANOVAs showed that learners from the most deprived quintile exposed to dynamic soundfield in classrooms with good C50 values gained a significant benefit in the Developed Ability module and subtests. Learners from the least deprived quintile, in classrooms with excellent C50 values, showed a statistically significant improvement in the General Mathematics subtests of Data Handling and Measure, Shape and Space. Learners from both SIMD 1 and 5 quintiles demonstrated a significant benefit in the Reading module and Word Recognition subtests. Only SIMD 1 learners showed a significant advantage of dynamic soundfield in the word decoding subtest, which was primarily effective in good C50 classrooms. In the Non-Verbal Ability and Word Recognition subtests, there was a significant reduction in the gap between SIMD 1 learners in the intervention classroom and SIMD 5 learners in the control.

The views and experiences of learners towards classroom noise and the use of dynamic soundfield were explored quantitively: noise questionnaires were completed by all participants. Logistic regression model analysis revealed that learners exposed to dynamic soundfield could hear the class teacher more easily compared to the control. Overall, the results indicate that dynamic soundfield may be regarded as a cost-effective intervention strategy that could contribute to raising attainment and reducing inequality in Scottish education.