WELLBEING
The association between teachers ’ leadership style and student performance
Dr Paul Parham , Mathematics teacher
Introduction and background
The positive impacts of inclusive leadership in different educational and workplace settings have attracted increasing attention in recent years ( Moss , 2019 ). Moss et al . ( 2016 ) undertook the first large empirical study examining the effects of managers ’ leadership styles on employee productivity , wellbeing and motivation across eleven large organisations ( n = 966 ), and reported a strong , statistically-significant positive correlation ( Pearson r = 0.87 ). This study was repeated by Moss ( 2019 ) in two Higher Education Institutes ( HEIs ), one in the UK and one in Norway ( n = 183 ), examining the association between the leadership style of employees encountered by students at different levels within the HEI and self-evaluated student performance , reporting a very similar finding ( r = 0.82 ). However , the impact of inclusive leadership on students within a secondary education environment has not previously been investigated in the UK or internationally . Thus , this research aimed to ( a ) examine whether the strong correlations between leadership style and performance observed in industry and HEIs also exist within Sevenoaks School , and ( b ) provide meaningful insights into how evidence-based teaching practice within the School may be enhanced . The previous edition of Innovate outlined the proposed study and plans for analysis ; in this article , the results are presented and discussed .
Materials and methods
The same validated survey instrument used in the Moss et al . ( 2016 ) and Moss ( 2019 ) studies was used , which took the form of an online anonymous multiplechoice questionnaire that students completed either at the start of lessons or as homework . The survey collected demographic data on students ’ gender , nationality , year group and ethnicity , as well as asking 34 questions regarding students ’ assessments of their mathematics teachers ’ inclusive leadership styles ( with teachers ’ inclusive leadership score being defined as the sum of the scores across all 34 questions ) and 12 self-reflection questions about their own productivity , wellbeing and motivation in mathematics as a result ( with student performance defined as the sum of the scores across all 12 questions ). The responses to the 46 non-demographic questions were on a Likert scale and student participation was on a parental opt-out basis . Ethical approval for the study was obtained from
Buckinghamshire New University Ethics Committee on 11th January 2019 . Data collection commenced on 26th March 2019 and completed on 3rd August 2019 .
Results
Two parents opted-out of the study , which resulted in a population of 1090 students . In total , 721 students completed the questionnaire ( response rate = 66.1 %) with a mean time for questionnaire completion of 10.7 minutes . Thirty-two students did not complete one or more questions in the questionnaire and these surveys were subsequently removed , resulting in 689 students being included in the final analysis . The response rate by year group and demographic characteristics of respondents are shown in Figures 1 and 2 , respectively , while the best scoring and worst scoring questions across all respondents are listed in Box 1 .
Figure 3 shows a scatter plot of student performance , the dependent variable , against teachers ’ inclusive leadership score , the independent variable ; both variables were treated as continuous and a clear linear relationship is evident from the scatterplot . Outlier analysis identified 12 student surveys and all statistical analysis was subsequently undertaken including ( n = 689 ) and excluding ( n = 677 ) these responses to assess the sensitivity of statistical inference to the presence of outliers . Student performance scores were not normallydistributed , although teachers ’ inclusive leadership scores were normally-distributed upon removal of outliers , and hence both Pearson r and Spearman p correlation coefficients were calculated for comparison . These were found to be r = 0.84 and p = 0.82 when outliers were included in the analyses , and r = 0.81 and p = 0.81 when outliers were excluded , indicating a strong positive correlation ; teachers that adopted more inclusive teaching styles saw improved student performance , and these results were found to be statistically significant at the 1 % level . It was also found that there was very little difference between r and p values throughout the correlation analyses , and the impact of outliers was furthermore found to be very minimal ; hence , only values of r with outliers removed will be reported hereafter in the interests of brevity .
Given that student performance was found to be very strongly positively correlated with inclusive teaching styles , this association was further analysed by the three components of student performance ; productivity ( r = 0.72 ), wellbeing ( r = 0.75 ) and motivation ( r = 0.76 ) were also all individually strongly positively correlated with teachers ’ behaviours . Table 1 shows a further stratified analysis of these correlations according to the four demographic variables , with all r-values found to be statistically significant at the 1 % level .
36