South Lanarkshire Behaviour Support

Evaluating Behaviour Support in Mainstream Secondary Schools in One Education Authority

(Work-in-progress)

George Head, Faculty of Education, University of Glasgow

Jean Kane, Faculty of Education, University of Glasgow

Nicola Cogan, Strathclyde Centre for Disability Research

Paper presented at the British Educational Research Association Annual Conference, University of Leeds, 13-15 September 2001

South Lanarkshire Behaviour Support

Background

In recent years, the Scottish Executive Education Department and its predecessor, the Scottish Office Education and Industry Department, have been instrumental in encouraging local authorities to pursue the government’s policies on social inclusion. Central to this policy is educational inclusion. Through its alternatives to exclusions initiative, the Scottish Office made money available to schools to reduce the number of exclusions by one third over a period of three years.

South Lanarkshire Education Resources took advantage of this scheme to set up their own behaviour support initiative at the start of the 1998 school session. The initiative was aimed at the authority’s twenty-one secondary schools and the money from the Scottish Office was used to pay for extra teaching pointage for each of the schools. This pointage was to be used specifically for the reduction of exclusions but each of the schools was free to use it as it saw fit.

The pointage was allocated to the schools by Education Resources on the basis of perceived need, based on figures supplied by schools themselves. Thus different schools received different pointages, the highest being 1.0 full-time equivalent (FTE) teacher. Some schools chose to use the extra FTE to directly address exclusions by allocating it to a specific teacher whose remit would become, wholly or largely, behaviour support. Other schools decided to spread their allocation amongst, say, the existing Guidance staff in the hope of reducing exclusions through increased monitoring of behaviour and pastoral care. Consequently, the different allocation of pointage led to different preferred strategies emerging from each of the schools.

In order to establish the effectiveness of the initiative, each of the schools was asked to identify six pupils receiving extra support as a result of their referral to the behaviour support services in the school. These pupils would be tracked throughout the three year duration of the scheme. In order to monitor a range of experiences, the schools were invited to consider two each of pupils for whom it was envisaged that intense support would be needed, two for whom a moderate amount of support would probably be adequate, and two who would require little support. In this way, they hoped to unearth a continuum of experience for pupils and staff. The schools were asked to provide two sets of information regarding the initiative. Firstly, they were asked to report on each of the pupils using a standard form drawn up by Education Resources. These were to be known as the Case Studies. In addition, on another standard form, schools were asked to report on each year’s behaviour support activity. These were to known as the Annual Reports.

In November 1999, South Lanarkshire Education Resources approached the researchers and asked them to evaluate the initiative.

Methodology

In order to evaluate the effectiveness of the initiative, Education Resources wanted four questions addressed

• What is working?
• Where are systems not working?
• What else is needed?
• Is this aspect of the whole strategy providing value for money?

In order to answer Education Service’s questions, criteria had to be created. These were developed through consultation with representatives from Education Resources and members of the research team. The researchers wanted to know what was different in terms of investment (finance, pointage and staff development); exclusions (numbers and percentage change); provision of support (changes in structure, use of pointage) and attitudes (of pupils, parents and teachers). Based on these criteria, the researchers would know that the initiative was working if:

• there was a reduction in the number of exclusions
• there was increased provision of support for pupils
• there was an increase in appropriate staff development
• pupils, parents and schools reported positively

In addition to the annual reports and case studies, Education Resources also provided data from Scottish Educational Establishments Management Information Systems (SEEMIS) for the year preceding the initiative and the first two years of the initiative. The SEEMIS data as presented to the researchers contained two main sections:

• the number of openings lost
• the number of exclusion incidents

These were subdivided by gender, giving four categories in all: openings lost, male and female, and exclusion incidents, male and female. The relationship between openings and incidents was not explicitly stated and for the purposes of the research it was assumed that ‘exclusion incidents’ indicated only those incidents which actually resulted in a formal exclusion from school. It was also assumed that the openings lost were a direct result of the incidents referred to in the SEEMIS data only. The SEEMIS data was then subjected to statistical analysis by the researchers.

The annual reports (n=21) were designed to examine behaviour support activity in the authority’s secondary schools. The reports were divided into six main sections:

• ways in which behaviour support is accessed within school
• the management and structure of behaviour support
• methods used
• effectiveness of each method used
• attitudes to behaviour support
• staff development

There was also a section in which respondents could make general comments. The annual report was completed by a single member of staff, usually the behaviour support teacher or a member of the management team responsible for behaviour support.

The case studies (n=116) were designed to examine the progress of pupils who had received extra support as a result of their referral to the behaviour support systems in schools. The case studies were also divided into six main sections:

• attendance rates
• the referral procedure
• reason(s) for exclusion(s)
• agreed plan of action / timescale / nature of support
• current level of attainment
• involvement in behaviour support

Again, there was also a section in which behaviour support staff could make general comments. The case studies were also completed by a single member of staff, often the same person who compiled the annual report.

Qualitative data from the annual reports and the case study questionnaires was inputted into SPSS statistical database, using a coding system devised by the researchers. Each data set provided by the respondents was given an identification code to enable the researchers to interchange between quantitative data (withheld in the SPSS database) and qualitative data (withheld in the annual reports and case studies). Analysis was conducted of the sample characteristics for: the annual report and case study data; respondents’ comments on each of the six main sections detailed in the annual report and case study questionnaires; and additional comments made by respondents in addition to the comments made using the questionnaire format.

The statistical output generated for the analysis was based on data gathered from participants who responded to the questions in consideration using the format of the annual reports and case studies. However, data extracted from respondents who chose to express their opinions without strictly using the specified formats was examined qualitatively.

Statistical tests of significance (independent sample t-tests) were conducted to determine if any differences in opinion occurred on any of the variables of interest. Only results revealing statistically significant differences between the variables examined were reported. All statistical output was rounded off to the nearest decimal point.

Findings

Findings are organized according to the four evaluation criteria:

1. that there is a reduction in the number of exclusions
2. that there is increased provision of support for pupils
3. that there is an increase in appropriate staff development
4. that pupils, parents, teachers and schools report positively.

Information was drawn from three sources: SEEMIS data, annual reports from schools and case study reports. It became clear that this range of data was inadequate to enable full judgements to be made on all four criteria.

1Was there a reduction in the number of exclusions?

Two sets of statistics were available: the total of number of openings (half-days) lost through exclusion and the total number of incidents leading to exclusion. Both sets of statistics were divided into male/female incidents and exclusions. See Table 1.

Table 1: SEEMIS data: Secondary Schools, 1999

F openings lost1941(-1096)=-36.1% on 1997
(-1395)=-41.8% on 1998
F exclusion incidents243(-137)=-36% on 1997
(-199)=-45% on 1998
F ave openings/incident8.0(+0.9)=+12.7% on 1997
(+0.5)=+6.6% on 1998
M openings lost 9152(-9378)=-50.6% on 1997
(-7282)=-44.3% on 1998
M exclusion incidents1201(-967)=-44.6% on 1997
(-899)=-42.8% on 1998
M ave openings/incident7.6(-0.5)=-6.2% on 1997
(-0.2)=-2.6% on 1998
Total openings lost11093(-10474)=-48.6% on 1997
(-8677)=-43.9% on 1998
F openings lost17.5%(+3.4)=+24% on 1997
(+0.6)=+3.5% on 1998
M openings lost82.5%(-3.4)=-3.9% on 1997
(-0.6)=-0.7% on 1998
Total exclusion incidents1444(-1104)=-43.4% on 1997
(-1098)=-43.2% on 1998
F exclusion incidents16.8%(+1.9)=+12.8% on 1997
(-0.6)=-3.4% on 1998
M exclusion incidents83.2%(-1.9)=-2.2% on 1997
(+0.6)=+0.7% on 1998

The overall figures indicate a high success rate for the behaviour support strategy with exclusion incidents down by 43% in Year I. The overall number of exclusions (of whatever duration) was down. The data did not give an indication of the duration of exclusions, nor the number of exclusions experienced by individuals. However, when the decrease in exclusion incidents is linked to a similar percentage decrease in the number of openings lost in that same period (49% in Year 1) the suggestion is that the length of exclusion imposed had not altered. The correspondence in the decrease for both exclusion incidents and openings lost suggests that schools in this sample are not responding to pressure to reduce the impact of exclusions by shortening the timespan for each exclusion.

Fears have been expressed (Parsons, 1999; Munn et al, 2000) that schools might under-record exclusions as a response to the use of exclusions statistics as a performance indicator. The data available does not provide evidence of this. However, schools in the project did not seem to be addressing merely the overall exclusion statistics by curtailing the exclusion period and thus cutting the number of openings lost. It would seem that the exclusion ‘tariff’ attracted by incidents is unchanged. Schools would not seem to be excluding for shorter periods of time.

The gender breakdown of the statistics is consistent with findings elsewhere (Munn et al, 2000; Boyd and Hamill, 2000) and with national statistics produced for the first time by the Scottish Executive in July 2000 for the school session 1998/1999 (SEED, 2000). Nationally, for secondary schools the male:female balance is 4:1, with the ratio being much more emphatically towards boys in primary and special schools (Approximately 10:1). The South Lanarkshire data, though, indicates that although exclusions were down overall and were down for both boys and girls, the reduction was far less for girls. The behaviour support project would seem to have a differential impact on boys and girls.

2Is there increased provision of support for pupils with SEBD?

Some schools appointed or enhanced the time of a behaviour support teacher whose job entailed direct dealings with pupils regarding their behaviour. This could take the form of one-to-one working. In this context, the teacher can address both curricular and behavioural difficulties through a mixture of pedagogical and counseling approaches.

In other instances, support groups in which pupils and their support teachers could address a range of issues, some specific and personal, others more general, were set up.

In more general terms, the initiative was used to support already familiar behaviour support strategies including, co-operative teaching, behaviour monitoring schemes, liaison with parents, and individual target-setting.

This question raised issues of quantity and quality of support provided which were illuminated to some extent by the annual reports and by the case study reports.

There were 116 case study reports from 21 secondary schools of pupils who had been referred to behaviour support during the first year of the project. Of the 116, 99 (85%) were boys and 17 (15%) were girls. The case studies were not evenly spread across year groups. The majority of pupils were in their second year (43%), followed by third-year pupils (28%), first year (21%), fourth year (7%) and only one pupil in fifth year (1%).

The spread of the case studies is taken to be representative of the age-spread of referrals to behaviour support. If so, it raises some interesting issues. It could be assumed that the peak time of second year for referrals indicates that pupils experiencing difficulty take some time to manifest these difficulties in secondary school. And yet, when it has been noted that primary school experience is a very strong predictor of difficulty in secondary school (Furlong, 1985) , it is surprising that behaviour support is not used in ways which are more proactive. The decline in numbers of pupils referred from third year onwards raises the possibility of other forms of school exclusion. Older pupils may self-exclude and so an analysis of exclusions data coupled with an analysis of attendance data might provide some explanation for the relatively low number of referrals in fourth year.

The annual reports attempted to gather schools’ views of the impact of a range of behaviour support systems. Questions were asked about:

- the range of methods used to support young people with SEBD and

- the effectiveness of each of those methods.

School respondents were asked to score each method on a five-point scale. Table 2 indicates the range of methods and the frequency of use of each method. Table 3 indicates the perceived effectiveness of each method.

Methods used by schools
Table 2 / individual
(one to one)
support / group work / co-operative
teaching / liaison with
parents/
guardians / other
methods / individual
target setting / daily
assessment
sheets for
targets
frequency / 18 / 18 / 15 / 11 / 11 / 9 / 9
percent / 86% / 86% / 71% / 52% / 52% / 43% / 43%
Effectiveness of each of the methods used
Table 3 / Co-operative
teaching / daily
assessment
sheets for
target set / individual
target setting / liaison with
parents/
guardians / other
methods / individual
(one to one)
support / group work
very effective / 33% / 0 / 0 / 45% / 37% / 13% / 13%
effective / 60% / 80% / 67% / 20% / 25% / 47% / 47%
neutral / 7% / 20% / 33% / 35% / 38% / 40% / 33%
ineffective / 0 / 0 / 0 / 0 / 0 / 0 / 7%

The most frequently used methods were those undertaken outwith the ordinary classroom: one-to-one support and group work. Of the methods used to support pupils within the classroom, cooperative teaching was the most commonly used with methods two (target –setting and monitoring, and daily report sheets) used by fewer than half of the respondents. The number of respondents citing liaison with parents as a method used ( 52%) is perhaps surprisingly small. Evidence coming from other sources in the evaluation suggests that this might reflect the great difficulties secondary schools have in developing partnerships with the families of this group of pupils in particular.

With regard to perceptions of effectiveness, cooperative teaching received the strongest endorsement from respondents with 93% rating it as either effective or very effective. The two most frequently used methods, one-to-one support and group work were both rated as effective by 60% of respondents but group work was the only method to be classed as ineffective (by 7% of respondents).

The range of methods used across the sample of schools and within a large number of individual schools suggests a pragmatic approach to supporting pupils with SEBD. On the one hand, it is viewed as something best directed towards individuals experiencing difficulty (through one-to-one sessions or in small groups) while it is also constructed as support for teachers in the context of the classroom and the curriculum. The variety of responses favoured by schools could reflect awareness of the diverse range of difficulties experienced by pupils and a desire to respond appropriately to that range.

Each school has embraced a range of approaches to behaviour support, but the range is different in each school. Later stages of the evaluation have tried to identify the relative effectiveness of strategies as schools themselves adapt and develop their initial strategies.

3Is there an increase in appropriate staff development?

The annual reports provided information about staff development within the project. In most cases (88%) ‘staff development’ was taken to mean participation in formal courses by those identified as ‘behaviour support teachers’. Courses attended included those on counselling skills, dyslexia and managing challenging behaviour. A few schools saw behaviour support as permeating the structures of the school and staff development as beneficial for all teachers.

The data about staff development perhaps more than any other aspect of the data gave insights into the different models of behaviour support which were emerging as the project progressed. These models existed on a continuum from behaviour support as a separate and distinct strand in school provision (very often in an identifiable place) to behaviour support as permeating all aspects of learning, teaching, curriculum and pastoral care.

4 Do pupils, parents and teachers report positively?

This criteria pointed most clearly to the inadequacy of the data available at the end of the first year. The case study reports included questions intended for pupils receiving behaviour support, their peers, their parents and their teachers from across the curriculum. Almost invariably these sections were completed by a member of staff with a strong involvement in the behaviour support initiative. Sometimes it was clear that this person was acting as a scribe and was indeed reporting the views of others but, elsewhere, it seemed likely that the behaviour support teacher was reporting his/her assumptions of others’ experience.

Views given are wholly positive about the project. Not surprising, perhaps, when all schools had been allocated additional staffing as a result of their participation and when there was a desire to retain that staffing. However another factor in the very positive feedback could be the strong sense of ownership over the project felt by schools. This would vindicate South Lanarkshire’s decision to devolve to schools the responsibility for designing strategy in accordance with each school’s perceptions of what was needed.