Positive behaviour in the early years: research report

CHAPTER FOUR WHAT IS THE EXTENT AND NATURE OF BEHAVIOUR DIFFICULTIES AMONG CHILDREN IN EARLY YEARS AND EARLY PRIMARY SETTINGS?

4.1 Introduction

Whilst the project set out to focus on positive behaviour, to do so it was necessary to focus on the range of observable behaviours shown by children in the early years. The literature suggests that behavioural difficulties are often the result of a whole set of factors which include the individual, social circumstances, and institutions, such as education (Abdelnoor, 1999). By taking account of the complexity of circumstances that interrelate to create any difficulties a child may experience, the project sought to avoid a view that difficult behaviours are necessarily 'within child'. New concepts of well-being and involvement in learning and the learning environment, were introduced to settings in order to move away from such a deficit model of behaviour. The direct observations undertaken were in keeping with the work normally undertaken by early years staff.

Cooper (in Sanders & Hendry, 1997), in keeping with the bio-ecological systems theory of Bronfenbrenner, points out that all humans exist within a social context and so behaviour is a product of interactions between people, environment and the motivation of the individual concerned. Clearly, then, the issue of inappropriate behaviour is complex and cannot be entirely or solely related to the home environment, the community, the nursery or the school or teacher.

Measures were chosen to allow a bridging of all these environments, and so matched measures were used across the age strata, in pre-school and primary, and by practitioners and parents. These common measures included the Strengths and Difficulties Questionnaire, the Adult Strategies Questionnaire and the Transitions Questionnaires. Additionally parents completed the Daily Hassles Questionnaire and practitioners complete the Well-being and Involvement Scales. Descriptives, crosstabulations and correlations have been used to relate the data generated by each instrument.

Data was gathered from parental and teacher perspectives. Data from the Parental and Teacher returns is presented separately, but in the case of data that is collected from both parents and staff, through the same measures, eg the Strengths and Difficulties Questionnaires, where tables reflect matched data the same table title is used, and tables showing data gathered from parents are labelled 'a', whilst tables showing data gathered from teachers are labelled 'b': these appear in consecutive sections. An example is Table 4.2a Total Difficulties across SDQ Levels - parental perceptions of % of children per level (Normal, Borderline, Abnormal) is matched by Table 4.2b Total Difficulties across SDQ Levels - teacher perceptions of % of children per level (Normal, Borderline, Abnormal). Some data is presented sequentially to show similarities and differences in the views held by parents and teachers, in order to allow comparisons of view. Here since the numbers of returns vary between parents and staff, the percentage figures are the favoured means of comparison.

4.2 Parental perceptions of the nature and extent of Behaviour Difficulties

The highest return on any one of the single parent measures was 724 (Daily Hassles Questionnaire). Of the responding parents 8 % (n= 46) were fathers. 92% of responding parents were mothers (n= 517). On a basis of the common set (n=603), 3% of parents were under 21, whereas 11% were over 40 years of age. The mean age of the parents taking part in the study was between 22-30 and 30-40 although closer to the latter category (respective scores are 3 and 4; mean score is 3.73). In terms of parent returns the children were divided fairly evenly by gender, with 306 boys and 297 girls matched to the 603 parent returns. The mean age of participating children was 3.76 years.

It should be noted that where comparable data was gathered from parents and staff, tables have a suffix of (a), eg 4.3a for parental returns and a suffix of (b) for staff returns.

As shown in table 4.1 parents completing the Daily Hassles data considered that over half of the children had no behaviour difficulties (57.7%), 30.7 % were perceived to have minor difficulties, and 5.6 % fell into the categories of definite and severe difficulties.

Table 4.1 - What % of children are perceived to have behaviour difficulties (from minor to severe)? (Daily Hassles data)

Difficulties		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	No	348	57.7	61.4	61.4
	Minor Difficulties	185	30.7	32.6	94.0
	Definite Difficulties	32	5.3	5.6	99.6
	Severe Difficulties	2	0.3	0.4	100.0
	Total	567	94.0	100.0
Missing	0	20	3.3
	System	16	2.7
	Total	36	6.0
Total		603	100.0

Slightly more boys than girls presented within each category of difficulty, but this was not a significant difference. The gender balance was explored further by means of the Strengths and Difficulties Questionnaire (Table 4.4).

When asked by means of the Parent Strengths and Difficulties Questionnaire (P- SDQ) (Table 4.2a) how many children fell into each level of behaviour in terms of the four negative behavioural domains of emotional symptoms, conduct problems, hyperactivity and peer related problems, parents reported that 81.5% fell within the normal range, with 18.5% of children being seen as having borderline or abnormal behaviour - these figures sit within the reported approximately 20% of children perceived to be presenting with some difficulties at any one time. These results are comparable to previous studies which found that 15% of 3 year olds were considered by parents to present mild behavioural problems and a further 7% considered to present with moderate or severe behavioural difficulties (Richman et al., 1982), and Thompson and colleagues found 13% of 3 year-olds to be perceived by parents to have behaviour problems (Thompson et al., 1996). Approximately 6% -10% of 11 year olds were viewed by parents and teachers as showing significant emotional and behavioural problems in the Isle of Wight study carried out by Rutter and colleagues (Rutter et al, 1970) with rates of perceived problems found to be almost doubled in a similar study in an inner London borough (Rutter et al, 1975).

Table 4.2a Total Difficulties as perceived by parents across SDQ Levels - % of children per level (Normal, Borderline, Abnormal)

Total Difficulties		Frequency	Percent	Valid Percent	Cumulative Percent
Valid 18.5% borderline or abnormal	Normal	432	71.6	81.5	81.5
	Borderline	46	7.6	8.7	90.2
	Abnormal	52	8.6	9.8	100.0
	Total	530	87.9	100.0
Missing	System	73	12.1
Total		603	100.0

Only in the 'abnormal' level of behaviour did boys display more difficulties than girls: 12% of boys and 7% of girls presented with an abnormal level of difficulty in terms of the total average difficulties on the 4 negative domains of the SDQ.

Table 4.3a - % children by age in e

age-years * Total SDQ negative domains Crosstabulation		Total Class			Total
		Normal	Borderline	Abnormal
age-years	0	1	1		2
		50.0%	50.0%		100.0%
	1	6		2	8
		75.0%		25.0%	100.0%
	2	16	2	3	21
		76.2%	9.5%	14.3%	100.0%
	3	122	26	22	170
		71.8%	15.3%	12.9%	100.0%
	4	173	10	15	198
		87.4%	5.1%	7.6%	100.0%
	5	113	6	9	128
		88.3%	4.7%	7.0%	100.0%
	6	1	1	1	2
		33.3%	33.3%	33.3%	100.0%
Total		432	46	52	530
		81.5%	8.7%	9.8%	100.0%
		100.0%	100.0%	100.0%	100.0%

The profile of frequency of difficulties by age revealed through analysis of the negative domains of the P- SDQ (Table 4.3a) shows the highest number of children presenting in the borderline and abnormal range in any age strata is 3 year olds. This ties in with parental comment that the older the child the more difficult parents can find their behaviour. One mother remarked "I try to instill good manners in my kids and explain to them why I am telling them off if they are being naughty. I try to be a good role model. As my kids get older I find it harder to stay calm when I am explaining things to them. I end up shouting which I then feel guilty about." Another mother said "I would like more help with his behaviour because I've tried everything."

The Strengths and Difficulties Questionnaire provides an insight into how parents view children's behaviour within each domain. In term of pro-social behaviour (Table 4.4a) boys and girls are evenly balanced within the normal levels of behaviour, and this is by far the largest grouping. Within borderline and abnormal levels of behaviour the numbers of boys presenting are somewhat higher than the number of girls. A correlational analysis of total difficulties and pro-social behaviour reveals a significant negative correlation at the >0.01 level (2 tailed). The higher the number of total difficulties presented by children, the lower their pro-social skills.

Table 4.4a - Parent perceptions of the relationship between pro-social behaviour and gender

Pro-social Class * gender Cross tabulation		gender		Total
		Male	Female
Normal	Count	243	245	488
	% within Pro-socialClass	49.8%	50.2%	100%
Borderline	Count	28	17	45
	% within Pro-socialClass	62.2%	37.8%	100.%
Abnormal	Count	14	9	23
	% within Pro-socialClass	60.9%	39.1%	100%
Total	Count	285	271	556
	% within Pro-socialClass	51.3%	48.7%	100.0%

4.2.1. Domains of Behaviour within the Parent SDQ

The tables that follow outline the parents' perception of their children's behaviour in each domain of the SDQ. The Strengths and Difficulties Questionnaire uses the terminology of 'normal', 'borderline' and 'abnormal'. We have used this terminology in order to maintain consistency in reporting the results and because alternative terms such as 'severe' do not overcome the concerns we as a team, and our readers may have about the negativity of some of Goodman's terminology. Our aim was to be able to highlight the extent of positive behaviours - to do so we had to embrace the notion of negative behaviours and the extent to which they may be troubling to parents, to staff and indeed within the peer group. The full features of all domains can be found in the Annex 3, and are also included in the discussion that follows.

Emotional Domain

Parents feel very positive about the emotional domain of young children's development. 87% of parents feel their children's development in this area is within the normal range (Table 4.5a). Parents are asked about the following behaviours in order to generate a score on the extent of difficulties-

• Often complains of headaches, stomach aches
• Many worries, often seems worried
• Often unhappy, downhearted or tearful
• Nervous or clingy in new situations
• Many fears, easily scared

Table 4.5a - Emotional Domain Parent SDQ

Emotional Class	Frequency	Percent	Valid Percent	Cumulative Percent
Normal	490	81.3	87.0	87.0
Borderline	37	6.1	6.6	93.6
Abnormal	36	6.0	6.4	100.0
Total	563	93.4	100.0
System	40	6.6
Total	603	100.0

Conduct Domain

However in terms of behaviour (conduct domain) the level of concern at borderline and abnormal levels includes over one third of cases: 38.5% (Table 4.6a). The 20% in the abnormal range does marry with the standard understanding of 20% of children experiencing difficulties at some time, however 38.5% outwith the normal range suggest a need for help and support for one third of responding parents. The following behaviours are considered in this domain- with differentiation on some items for younger children.

• Often has temper tantrums or hot tempers
• Generally obedient, usually does what adults request
• Often fights with other children or bullies them
• Often lies or cheats (in 3-4 version: often argumentative with adults)
• Steals from home, school or elsewhere (in 3-4 version: can be spiteful to others)

Table 4.6a - Conduct Domain Parent SDQ

Conduct Class	Frequency	Percent	Valid Percent	Cumulative Percent
Normal	344	57.0	61.2	61.2
Borderline	104	17.2	18.5	79.7
Abnormal	114	18.9	20.3	100.0
Total	562	93.2	100.0
System	41	6.8
Total	603	100.0

Peer relationships Domain

Concerns about children's peer relationships also affect parents (Table 4.7a) - here parents consider that 20.4% of children have difficulties either at a borderline (11.3%) or at an abnormal level (9.1%). When asked about the benefits of pre-school provision parents frequently respond in terms of the social benefits, clearly parents and practitioners have a role to play in supporting children in this area, as staff also consider that 19.8% of children have such difficulties.

• Rather solitary, tends to play alone
• Has at least one good friend
• Generally liked by other children
• Picked on or bullied by other children
• Gets on better with adults than with other children

Table 4.7a - Peer Relationships Domain - Parent SDQ

Peer relations class	Frequency	Percent	Valid Percent	Cumulative Percent
Normal	444	73.6	79.6	79.6
Borderline	63	10.4	11.3	90.9
Abnormal	51	8.5	9.1	100.0
Total	558	92.5	100.0
System	45	7.5
	603	100.0

Hyperactivity Domain

When asked to consider behaviour in the hyperactivity domain items parents are asked to score whether they consider their child: restless, overactive and unable to stay still for long; constantly fidgety or squirming; easily distracted with wandering concentration; thinks things out before acting or sees tasks through to the end with a good attention span. This range of behaviours leaves room for parents to make a positive response rather than simply focusing upon whether difficulties may be present or not. Nearly 80% of children are seen by their parents to be within a normal range in this domain. However we find 21.7% feel their children have some difficulties in this area, with 13.2% presenting with an unusual level of difficulty (Table 4.8a).

Table 4.8a - Hyperactivity Domain - Parent SDQ

Hyperactivity Class	Frequency	Percent	Valid Percent	Cumulative Percent
Normal	433	71.8	78.3	78.3
Borderline	47	7.8	8.5	86.8
Abnormal	73	12.1	13.2	100.0
Total	553	91.7	100.0
System	50	8.3
	603	100.0

4.2.2 Pro-social Relationships Domain

In the pro-social domain (Table 4.9a) which is not used to calculate the extent of overall difficulties, but rather to establish the extent of positive social behaviours, parents are asked to respond to the following statements -

• Considerate of other people's feelings
• Shares readily with other children
• Helpful if someone is hurt, upset or feeling ill
• Kind to younger children
• Often volunteers to help others

Table 4.9a - Pro-social relationships Domain - Parent SDQ

Pro-social Class	Frequency	Percent	Valid Percent	Cumulative Percent
Normal	488	80.9	87.8	87.8
Borderline	45	7.5	8.1	95.9
Abnormal	23	3.8	4.1	100.0
Total	556	92.2	100.0
System	47	7.8
	603	100.0

Despite specific concerns in other domains and high levels of concern in the conduct domain, 87.8 % parents report that their children respond positively towards others. A few children fall into the abnormal range (4.1%) and 8.1% are considered to be borderline, however overall the outcomes for this domain tell us that parents of young children recognize the ways in which their children are developing positive behaviours towards others.

4.2.3 Impact of perceived difficulties on other aspects of behaviour

The data from parents who responded to the idea of such difficulties having an impact on friendships between children (Table 4.10a), suggests that whilst children's struggles with peer relationships are recognized, this has no impact for 59% - over half of the children, impacts only a little on the development of friendships for 33.2% of children, but for 7.8% there is a considerable impact. In terms of overall behavioural difficulties, parents do find an impact on home life, with 43% of children for whom this is low impact, and 18% of children's behaviour impacting quite a lot or a great deal on home life. With 43% in the niggles and constantly intruding low-level difficulties, this coincides with Munn's findings (2004) that for teachers it is the low-level behaviours that can be the most draining.

Parents also report that for 11.4%, behavioural difficulties impact on children's learning to a marked extent, whilst 37.3% note there is some impact. Equally in terms of leisure activities impact of perceived difficulties are felt a little for 33% of children, and more considerably for a kernel of 8.3% of children.

Table 4.10a - Impact of perceived difficulties on friendships

Impact of perceived difficulties on friendships	Frequency	Percent	Valid Percent	Cumulative Percent
Not at all	135	22.4	59.0	59.0
Only a little	76	12.6	33.2	92.1
Quite a lot	12	2.0	5.2	97.4
A great deal	6	1.0	2.6	100.0
Total	229	38.0	100.0
0	358	59.4
System	16	2.7
Total	374	62.0
	603	100.0

Turning to the possible links between children's levels of negative behaviour and parental or family factors we find no significant relationship between parental age and the level of hassle caused by their children's behaviours, but we do find a significant relationship (albeit at 0.05 level of significance- 2 tailed) between parents' age and perceived difficulties in the areas of their children's emotions, concentration, behaviour or being able to get on with other people (the younger the parent the higher the level of perceived difficulties). Discussions in the focus groups have suggested that the people seen by younger parents as most able to provide them with support in coping with their children's behaviours are their own extended family. This discussion is reported more fully later in the report, but there are implications in how professionals attempt to work with hard to reach parents if there is low trust for these women in terms of professional expertise. Here too is interesting to note that the younger the child the higher the score on perceived total difficulties ( SDQ parents): this correlates at a > 0.01 level of significance (2-tailed). Further girls tend to score higher on the pro-social scale, however this is significant at a 0.05 level only. Young parents may therefore be finding their young male children harder to handle than their young female children.

Strikingly, difficulties are seen by parents to have been present for over a year in 55% of children in the sample. A further 26.5% of children are reported to have difficulties for a 6 month period and 15.7% for 6-12 months. In other words there are families experiencing that their children have sustained difficulties. Whilst this calls for a sustained response, it should be noted that 46% find these behaviours only a little burdensome, whilst 17.2% find them either quite or very burdensome. Many behaviours are seen to be typical for the stage of development of the children, but where these are seen as of borderline or abnormal, there is a need for help and support.

Turning to the Daily Hassles data, there is a positive correlation between the Hassle Score and the perceived extent of burden it puts on parent/family as a whole at a <0.01 level of significance.

Overall, parents did not find dealing with their children's behaviour and needs to be a hassle. Only 1% of parents score a total over 70 on the hassles component; this means that they experience significant pressure in their parenting. Similarly the data indicates parents experience a high frequency of potentially hassling situations and events (6% score over total score higher than 50 on the frequency component). The mean total score on the hassles score is 39 (range of this scale is 0 -100), whereas 37 is the mean total score for the frequency component (range is 0 -80); respective standard deviations are 12.75 and 8.48. There is also a significant correlation between level of perceived hassle and frequency of which behaviour occurs. In the parent focus groups many parents said they found their children's behaviour could be embarrassing and a hassle in public, however 64% reported such hassle was low or very low when responding to the Daily Hassles Questionnaire. In the questionnaire returns only 17% report that public hassle is high or very high. Perhaps it is easier to share the hassles in a group situation, and more threatening to record this on paper.

4.3 Staff Perceptions of the Extent and Nature of Behaviour Difficulties

Parental perceptions of children's behaviour were matched by data collection with staff. Staff measures were labelled 'teacher' - but were completed by the range of staff who work in early childhood settings and not solely those who are registered teachers. The range of data gathered through use of the Strand A measures was extensive. This section starts with an overview from the Teacher Adult Strategies data, the remainder of this section presents evidence from the Teacher Strengths and Difficulties Questionnaires, Teacher Adult Strategies data and the Well-being and Involvement scales. Where there is comparable data with parental returns, the numbering of tables is matched but distinguished by using the suffix (a) for parental returns above, and the suffix (b) for the staff tables which follow. Over half the early educators participating in this study reported high confidence in working with young children presenting with behaviours that cause concern. They were able to identify a range of such behaviours. As with other data numbers of returns vary in relation to items completed and in the light of analysis of the number of complete cases in our final merged data file.

Table 4.11 - Teacher Adult Strategies responses - Behaviours causing some difficulty for staff, children or the setting as a whole (n = numbers of staff)

Difficulties		Difficulties behaviour	Difficulties concentration	Difficulties relationships	Difficulties emotions	Difficulties self-esteem	Difficulties toileting	Difficulties sleeping	Difficulties eating	Difficulties other
n	Valid	149	151	133	132	115	109	91	123	21
	Missing	19	17	35	36	53	59	77	45	147

Their responses show that difficulties in concentration in children as well as overall behavioural difficulties cause some concern. Relationships and children's capacity to cope with their emotions provide another area in which staff experience some difficulty in meeting needs and concerns. Staff responses to toileting, eating and sleeping difficulties accord with parental concerns in these areas, and given that parents say effective communication with their child's carers and educators makes a difference, it would seem these may be areas in which skilled workers could offer parents support, and less skilled need additional development.

Table 4.12 Total numbers of children for whom SDQ returns were received by age

Age Strata	Frequency	Percent	Valid Percent	Cumulative Percent
0-3	99	6.7	7.6	7.6
3-4	327	22.2	25.2	32.8
4-5	460	31.2	35.5	68.3
P1	411	27.8	31.7	100.0
Total	1297	87.9	100.0

With regard to the extent of behaviour difficulties (Table 4.12), when asked on the TSDQ for an overall judgement on difficulties in emotion, concentration, behaviour or being able to get on with people, as perceived by the staff, 63.3% of all children (n=742) were perceived to have no difficulties, and of the remaining 36.8%, only 3.1% (n=36) were considered to have severe difficulties. Comparing the extent of difficulties, it can be seen that there was no significant relationship between child age and behaviour difficulties. In relation to the specific TSDQ domains, children between the ages of 4-5 were perceived to show the most difficulties overall, with 37.6% (n=411) perceived as showing difficulties in the areas of emotional symptoms, conduct problems, hyperactivity and problems with peer relationships. Of the P1 children, 32.2% (n=354) showed some difficulties in these same areas.

The SDQ was administered in relation to 1476 children by members of staff in two different local authorities. The specific items completed on questionnaires varied, leading to some variations in returns from item to item. Of the members of staff completing the SDQ (Table 4.13), 6.4% were head teachers or centre managers, 40.7% were qualified teachers, and 50% were nursery assistants/nursery nurses. A further 2.9% returns were made by groups of staff on a joint basis.

Table 4.13 Responses to staff designation on T- SDQ child returns

Position held by respondents	Frequency	Percent	Valid Percent	Cumulative Percent
Manager/Head Teacher	66	4.5	6.4	6.4
Teacher/Nursery Teacher	423	28.7	40.7	47.1
Nursery Assistant/Nurse	520	35.2	50.0	97.1
Group of Teachers	30	2.0	2.9	100.0
Total	1039

The total difficulties across SDQ levels as perceived by staff are higher than those identified by parents. Staff consider that 24.5% of children show borderline or severe levels of difficulty, whereas parents consider 18.5% do so (Table 4.2a).

Table 4.2b Total Difficulties as perceived by staff across T- SDQ Levels - % of children per level (Normal, Borderline, Abnormal)

Total Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
24.5% borderline or abnormal levels	Normal	838	56.8	75.5	75.5
	Borderline	144	9.8	13.0	88.5
	Abnormal	128	8.7	11.5	100.0
	Total	1110	75.2	100.0
	Missing Cases	366	24.8
	Total	1476	100.0

The teacher adult strategies questionnaire provided 149 responses to the enquiry about staff levels of concern in relation to children's age. Levels of concern were very low in relation to babies under one year, rose in one to two year olds, and again in three and four year olds, with very few staff reporting a concern with the behaviour of 5 and 6 year olds. In percentage terms the highest levels of concern were reported by staff working with 3 and 4 year olds (Table 4.3b). Whilst staff reported concerns they also felt they had a range of strategies to employ. These are discussed further in the section on staff strategies. This is similar in percentage terms to the findings of the Teacher SDQ.

The profile of frequency of difficulties by age revealed through analysis of the negative domains of the T- SDQ (Table 4.3b) shows the highest number of children presenting in the borderline and abnormal range in any age strata is 33% of two year olds in each of borderline and abnormal behaviours (n=61). Overall as with the parental returns the majority of children are perceived to be within the normal range of behaviours.

Table 4.3b - % children by age in each T- SDQ level

Age-years Total SDQ negative domains Crosstabulation		Total Class			Total
age -years		Normal	Borderline	Abnormal
		451	67	67	585
		77.1%	11.5%	11.5%	100.0%
	0	3	1		4
		75.0%	25.0%		100.0%
	1	1		3	4
		25.0%		75.0%	100.0%
	2	5	5	5	15
		33.3%	33.3%	33.3%	100.0%
	3	95	36	25	156
		60.9%	23.1%	16.0%	100.0%
	4	170	19	16	205
		82.9%	9.3%	7.8%	100.0%
	5	109	16	12	137
		79.6%	11.7%	8.8%	100.0%
	6	3			3
		100.0%			100.0%
Total		838	144	128	1110
		75.5%	13.0%	11.5%	100.0%

In terms of pro-social behaviour and gender, staff consider slightly more girls than boys lie within normal levels of pro-social behaviour (Table 4.4b).

Table 4.4b - Staff perceptions of the relationship between pro-social behaviour and gender

Pro-socialClass * Gender Crosstabulation		Gender		Total
		male	female
		1	2
Normal	Count	341	373	714
	% of Total	32.6%	35.6%	68.2%
Borderline	Count	85	65	150
	% of Total	8.1%	6.2%	14.3%
Severe	Count	128	55	183
	% of Total	12.2%	5.3%	17.5%
Total	Count	554	493	1047
	% of Total	52.9%	47.1%	100.0%

According to staff within the borderline levels of behaviour there are more boys than girls, and in the abnormal range there are over twice as many boys as girls. As with the parental SDQ correlational analysis of total difficulties and pro-social behaviour, the T- SDQ also reveals a significant negative correlation at the >0.01 level (2 tailed). In the case of all children, the higher the number of total difficulties presented by children, the lower their pro-social skills

Of the 1047 returns relating to both pro-social behaviour and gender (Table 4.4b), 554 were in respect of boys and 493 were for girls. 84 forms were returned with no gender indicated. There were more girls reported to be in a normal range of pro-social behaviour, and 20% of boys presented as having more difficulties pro-socially in the case of both borderline and severe behaviours, whereas 11.5% of girls fall into this category in the view of staff.

4.3.1 Domains of Behaviour within the Staff SDQ

Overall returns on the five domains of behaviour numbered 1128 (Table 4.14). As with the Parent SDQ, four of the domains address negative behaviours, whilst one, pro-social behaviour addresses positive behaviours.

Looking at each of the four areas of emotional difficulties, conduct difficulties, hyperactivity and peer relationships it is interesting to note some variation according to the extent to which behaviours are perceived as being on the normal - abnormal range. The numbers of cases are summarised in Table 4.14.

Table 4.14- Numbers of cases on each T- SDQ scale by level of behaviour

Numbers of cases on each TDSQ scale by level of behaviour	Total Class	Emotion Class	Conduct Class	Hyper Class	Peer Class	Pro-social Class
	Count	Count	Count	Count	Count	Count
Normal	838	1076	978	865	959	754
Borderline	144	39	66	76	84	168
Severe	128	56	132	231	124	206

As with the parental returns the intention behind investigating negative behaviours as perceived by staff, is to establish the extent to which such behaviours are troubling, and to highlight the extent of positive behaviours.

Emotional Domain

A higher percentage of boys 49% ( n= 533) were perceived to have normal levels of emotional behaviour as opposed to 42.6 % (n=463) of girls. Perceptions for borderline continued to show boys having fewer problems 1.5% (n=16) as opposed to 2.1% (n=23) in girls. However on ratings for severe behaviour boys were perceived to have slightly, but not significantly worse difficulties 2.6 (n=28) against 2.3 (n=25). Overall 8.1% of children are viewed by staff as having borderline or severe difficulties in the emotional domain (Table 4.5b).

Table 4.5b - Emotional domain Teacher/Staff SDQ

Emotion Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
8.1% borderline or severe levels	Normal	1076	72.9	91.9	91.9
	Borderline	39	2.6	3.3	95.2
	Severe	56	3.8	4.8	100.0
	Total	1171	79.3	100.0
	Missing Cases	305	20.7
	Total	1476	100.0

Conduct Domain

Similarly with issues of conduct, perceptions of staff showed that boys were more likely to display normal parameters of behaviour: 43% (n=470) of boys had normal conduct as opposed to 40% (n=437) of girls. Fewer boys had borderline difficulties, 2.7% (n=30) were considered borderline whereas 3.2% (n=35) of girls were. However, 7.2% (n=79) boys were considered to have severe conduct issues, whereas 3.8% (n=42) girls were considered in a similar light. Overall 16.8% of children overall exhibited borderline or severe levels of difficulty in the conduct domain. (Table 4.6b)

Table 4.6b - Conduct Domain Teacher/Staff SDQ

Conduct Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
16.8% borderline or severe levels	Normal	978	66.3	83.2	83.2
	Borderline	66	4.5	5.6	88.8
	Abnormal	132	8.9	11.2	100.0
	Total	1176	79.7	100.0
	Missing Cases	300	20.3
	Total	1476	100.0

Peer Relationships Domain

Difficulties with peers again showed boys in a positive light with 43% (n=466) showing as normal against 38.8% (n=421) for girls. However more boys showed more borderline difficulties 4% (n=43) against 3.2% (n=35) of girls. More boys 6.4% (n=69) were classified as having severe difficulties whereas 4.6% (n=59) of girls were considered to have severe difficulties in this area. Overall 208 children or 14% were deemed to show some level of difficulty in peer relationships (Table 4.7b).

Table 4.7b - Peer Relationships Domain - Teacher/Staff SDQ

Peer relationships Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
17.8% borderline or severe levels	Normal	959	65.0	82.2	82.2
	Borderline	84	5.7	7.2	89.4
	Abnormal	124	8.4	10.6	100.0
	Total	1167	79.1	100.0
	Missing Cases	309	20.9
	Total	1476	100.0

Hyperactivity Domain

The picture alters in the classification of hyperactivity issues. In every classification of hyperactivity the percentage of boys exceeded that of girls. Those boys who were considered in the normal range amounted to 35.7% (n=388) against 38.4% (n=418) girls. Boys were slightly ahead with 3.8% (n=41) in borderline cases, whereas girls had 2.5% (N=27). However the number of boys considered to have serious hyperactivity was more than double the number of girls. 13.6% (n=148) were considered to have severe issues of hyperactivity, whereas 6.1% (n=66) of girls came into this category. Overall taking all children together, 26% of children show some level of hyperactivity, and in 19.7% of these cases it is severe (Table 4.8b).

Table 4.8b - Hyperactivity Domain - Teacher/Staff SDQ

Hyperactivity Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
26.2% borderline or severe levels	Normal	865	58.6	73.8	73.8
	Borderline	76	5.1	6.5	80.3
	Abnormal	231	15.7	19.7	100.0
	Total	1172	79.4	100.0
	Missing Cases	304	20.6
	Total	1476	100.0

4.3.2 Pro-social relationships Domain

Reflecting on the areas of the SDQ addressed in the pro-social domain, 33.2% of children overall emerge as having some problems in the view of early years staff (Table 4.9b). This contrasts sharply with parental perceptions: staff consider nearly three times as many children to display serious difficulties in pro-social behaviour than their parents do. Part of the purpose of pre-school provision is to provide children with a widened social network and to encourage their play and cooperation with others, as a base for future inclusion.

In relation to staff perceptions, previous studies have reported that 17% of 4-7 year olds were perceived by teachers as having mild behaviour difficulties, with a further 16% viewed as having definite behavioural problems (Tizard et al, 1988), and 22% of 7 year olds were perceived by teachers to show some difficulties in behaviour, with 14% considered to present serious problems (Davie et al., 1972).

Table 4.9b - Pro-social relationships Domain - Teacher/Staff SDQ

Pro-social Class	Level	Frequency	Percent	Valid Percent	Cumulative Percent
33.2% borderline or severe levels	Normal	754	51.1	66.8	66.8
	Borderline	168	11.4	14.9	81.7
	Abnormal	206	14.0	18.3	100
	Total	1128	76.4	100
	Missing Cases	348	23.6
	Total	1476	100

4.4 Summary of parental and staff perceptions of behaviour using the SDQ behavioural domains

Table 4.15 shows the responses of parents and practitioners in relation to the different behavioural domains within the SDQ, and here there are some differences of view. Parents felt very positive about the emotional domain of young children's development with 87% feeling their children's development in this area to be within the normal range. This was reflected in the responses of the practitioners which placed around 90% of the children within the normal range in the emotional domain.

Table 4.15 - Comparison of responses of parents and practitioners on the SDQ in placing children in the normal range (approximate percentages)

Domain	Emotion	Pro-social	Conduct	Peer-relationships	Hyperactivity
Parents	87%	88%	60%	80%	80%
Practitioners	90%	66%	80%	80%	74%

In the pro-social domain 88% parents reported that their children respond positively towards others while practitioners placed 66% of the children within the normal range. In the conduct domain, while practitioners placed about 80% of children within the normal range parents considered a lesser proportion (60%) of their children to be in this range, with around 20% being perceived in the borderline range and 20% in the abnormal range.

In the peer relationships domain, about 80% of parents indicated that they considered their children to be in the normal range, and similarly practitioners placed 80% of children in this range. In the hyperactivity domain, 80% of parents considered their children to be in the normal range, while practitioners placed about 74% of children in this range, with about 20% being placed in the abnormal range.

4.5 Further analysis of Staff Adult Strategies data

Further analysis of the Staff Adult Strategies Questionnaire returns allows reflection on how children's behaviours are seen in relation to concentration, relationships, emotions, self esteem, eating and sleeping.

It is worth reflecting on the level of concern expressed here about children's concentration as it links to the section on well-being and involvement and is elaborated by the tables that follow. These tables show staff perceptions on the degree of challenge they experience in different aspects of child behaviour highlighted and investigated through the Adult Strategies questionnaire.

Levels of concern about behaviour, concentration, relationships, emotions and feelings, self-esteem, toileting, sleeping, and eating and appetite are shown in tables 4.16- 4.22 below.

Out of 151 staff responding to this question on children's concentration, only 16% (n=25) of staff had no concerns. 48% (n=73) had a few concerns, and 35% (n=53) had quite a lot of concern about children's capacity to concentrate (Table 4.16).

Table 4.16 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties in concentration by age

Difficulties- concentration			yes, a little	yes, quite a lot	no	Total
Strata	0	Count	1			1
		%	1.4%			0.7%
	1 years	Count	10	8	16	34
		%	13.7%	15.1%	64.0%	22.5%
	2 years	Count	17	14	2	33
		%	23.3%	26.4%	8.0%	21.9%
	3 years	Count	21	11	3	35
			28.8%	20.8%	12.0%	23.2%
	4 years	Count	17	14	3	34
		%	23.3%	26.4%	12.0%	22.5%
	5 years	Count	6	6	1	13
		%	8.2%	11.3%	4.0%	8.6%
	6 years	Count	1			1
		%	1.4%			0.7%
Total		Count	73	53	25	151
Total		%	100.0%	100.0%	100.0%	100.0%

When staff considered their levels of concern about children's competence or level of difficulty in the area of relationships, 56% (n=75) recorded a little concern, 15% (n=20) recorded considerable concern, and 29% (n=38) felt no concern about children's relationships with others (Table 4.17).

Table 4.17 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties in relationships by strata

Difficulties in relationships			yes, a little	yes, quite a lot	no	Total
Strata	0	Count			1	1
		%			2.6%	0.8%
	1 years	Count	16	3	15	34
		%	21.3%	15.0%	39.5%	25.6%
	2 years	Count	17	5	6	28
		%	22.7%	25.0%	15.8%	21.1%
	3 years	Count	22	3	3	28
		%	29.3%	15.0%	7.9%	21.1%
	4 years	Count	13	8	10	31
		%	17.3%	40.0%	26.3%	23.3%
	5 years	Count	6	1	3	10
		%	8.0%	5.0%	7.9%	7.5%
	6 years	Count	1			1
		%	1.3%			0.8%
Total		Count	75	20	38	133
		%	100.0%	100.0%	100.0%	100.0%

Reflecting on whether children's capacity to cope with their emotions and feelings is concerning, 53% (N=70) of respondents felt a little concern about this, 26% (n=34) felt quite a lot of concern and 21% (n=28) staff felt no concern about this aspect of children's conduct.

Table 4.18 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties with emotions and feelings by strata

Difficulties-emotions			yes, a little	yes, quite a lot	no	Total
Strata	0	Count	1			1
		%	1.4%			0.8%
	1 years	Count	13	7	13	33
		%	18.6%	20.6%	46.4%	25.0%
	2 years	Count	20	5	5	30
		%	28.6%	14.7%	17.9%	22.7%
	3 years	Count	16	10	3	29
		%	22.9%	29.4%	10.7%	22.0%
	4 years	Count	12	10	5	27
		%	17.1%	29.4%	17.9%	20.5%
	5 years	Count	8	1	2	11
		%	11.4%	2.9%	7.1%	8.3%
	6 years	Count		1		1
		%		2.9%		0.8%
Total		Count	70	34	28	132
		%	100.0%	100.0%	100.0%	100.0%

Across the sample the most concerns felt in terms of expression of emotions and feelings lay with the 4 year olds - here there would be an expectation that typically children are beginning to be able to express themselves in appropriate ways (Table 4.18).

Table 4.19 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties with self esteem by strata

Difficulties self-esteem			yes, a little	yes, quite a lot	no	Table
Strata	0	Count			1	1
		%			2.6%	0.9%
	1 years	Count	9	5	17	31
		%	15.3%	27.8%	44.7%	27.0%
	2 years	Count	14	3	7	24
		%	23.7%	16.7%	18.4%	20.9%
	3 years	Count	16	2	4	22
		%	27.1%	11.1%	10.5%	19.1%
	4 years	Count	14	7	6	27
		%	23.7%	38.9%	15.8%	23.5%
	5 years	Count	6	1	3	10
		%	10.2%	5.6%	7.9%	8.7%
Total		Count	59	18	38	115
		%	100.0%	100.0%	100.0%	100.0%

A third of staff, 33% (n=38), have no concerns about children's levels of self-esteem, but 51% (n=59) have some concern, and 15% (n=18) have quite a lot of concern (Table 4-19). Evidence shows the link between positive self-esteem and success in learning, and is an important area for focus by early years personnel. Data from this study indicates a positive sense of well being for many children, but not all, and it is those with low self-esteem and a low sense of well-being who are likely to need additional support.

Tables 4.20, 4.21 and 4.22 show staff perceptions of children's difficulties in the areas of toileting, sleeping, and eating and appetite. 40% of staff (n=44) express a little concern about children's toileting (Table 4.20).

Concerns are at their highest in terms of 1 and 2 year olds where it would be expected that there would be such a focus. However these concerns persist with 3 and 4 year olds, but in relation to a smaller number of staff (8%, n=9), and the majority of staff (51%, n=56) lack concern about children's toileting.

Table 4.20 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties with toileting by strata

Difficulties-toileting			yes, a little	yes, quite a lot	no	Total
Strata	0	Count			1	1
		%			1.8%	0.9%
	1 years	Count	6	6	14	26
		%	13.6%	66.7%	25.0%	23.9%
	2 years	Count	16		10	26
		%	36.4%		17.9%	23.9%
	3 years	Count	12	1	11	24
		%	27.3%	11.1%	19.6%	22.0%
	4 years	Count	9	2	14	25
		%	20.5%	22.2%	25.0%	22.9%
	5 years	Count	1		6	7
		%	2.3%		10.7%	6.4%
Total		Count	44	9	56	109
		%	100.0%	100.0%	100.0%	100.0%

Focusing on sleep issues, 72% (n=66) of staff have no concerns at all (Table 4.21). Provision for the very youngest children will attend to the need for rest for children, but for many practitioners concerns would be less in terms of their own service, and more in terms of how disrupted sleep or parental reporting of difficulties at home with sleep management might impact on the day-to-day experience of children. 26% of staff respondents do therefore report some concern (n=19 report a little concern, and only 6 report considerable concern).

Table 4.21 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties with sleeping by strata

Difficulties-sleeping			yes, a little	yes, quite a lot	no	Total
Strata	0	Count			1	1
		%			1.5%	1.1%
	1 years	Count	7	3	21	31
		%	36.8%	50.0%	31.8%	34.1%
	2 years	Count	6	2	14	22
		%	31.6%	33.3%	21.2%	24.2%
	3 years	Count	1	1	16	18
		%	5.3%	16.7%	24.2%	19.8%
	4 years	Count	4		9	13
		%	21.1%		13.6%	14.3%
	5 years	Count	1		5	6
		%	5.3%		7.6%	6.6%
Total		Count	19	6	66	91
		%	100.0%	100.0%	100.0%	100.0%

Children's physical well-being, diet and its relationship to overall health have an increasingly high policy profile. Here half of staff respondents have no concern at all about children's eating and appetite (51%, n=63). The other half are divided between a little concern and higher levels of concern. 42% (n=51) of staff express some concern right across the pre-school years, and a small number 9% (n=9) have a higher level of concern. These higher levels of concern most likely relate to particular children, given the low numbers (Table 4.22). Most children attend nursery for a part-day and it may be that concerns about eating and appetite surface more in terms of full-time children where meals are part of the daily routine.

Table 4.22 - Numbers/ % of staff expressing levels of concern/lack of concern about children's difficulties with eating and appetite by strata

Difficulties - eating			yes, a little	yes, quite a lot	no	Total
Strata	0	Count			1	1
		%			1.6%	0.8%
	1 years	Count	11	4	20	35
		%	21.6%	44.4%	31.7%	28.5%
	2 years	Count	11	1	13	25
		%	21.6%	11.1%	20.6%	20.3%
	3 years	Count	13	1	14	28
		%	25.5%	11.1%	22.2%	22.8%
	4 years	Count	12	1	11	24
		%	23.5%	11.1%	17.5%	19.5%
	5 years	Count	4	2	4	10
		%	7.8%	22.2%	6.3%	8.1%
Total		Count	51	9	63	123
		%	100.0%	100.0%	100.0%	100.0%

4.6 Well-being and Involvement

The extent and nature of behaviour difficulties was also reflected in the whole class monitoring approach to observing levels of well-being and involvement on the five point scale developed by Laevers (Centre for Experiential Education, Belgium). The staff in participating settings were trained in this approach at events held in each local authority. One local authority also invested in 'A Box Full of Feelings' which provides staff with follow-up strategies. These intervention packs were used in settings following the first round of well-being and involvement.

4.6.1 Well-being

The first round of screening for well-being included 1230 children. Each child was assigned a score on a five-point scale with 1 being low and 5 being high. Respondents could also indicate if a child was between levels and this is reflected in the table 13 that follows. Characteristics of well-being that were explored with practitioners included the following - when children are -

• Vocal
• Feeling safe
• Feeling comfortable in themselves
• Feeling stimulated & interested
• Feeling well physically
• Being together with others
• Enjoying life
• Self-regulating
• In a stream of experiencing
• At ease
• Being spontaneous
• Being open to the world & accessible
• Expressing inner rest & relaxation
• Showing vitality & self-confidence
• In touch with feelings & emotions
• Having feelings accepted, acknowledged by others (From Laevers)

In interpreting the data presented in Table 4.23, showing the levels of well-being in the first round of screening in December 2005 and January 2006, 26% children were scored '3' at the mid-point of the scale. 13.7% were perceived by staff to be at 2.5 or below- the monitoring approach would indicate intervention with all of these children (n=167), and continued monitoring and support for those on the level 3. Overall 60% of children were perceived by their early educators as being at least 3.5 or above, with 30% at level 4 (n= 370) and 25% hitting level 5 (n= 306). These high levels of well-being reflect positively on the early years settings, and this relates well to the environmental ratings conducted by the research team in all of the settings. Whilst not drawing a causal relationship, it may be interesting for staff teams to reflect on those children who fall below these high perceived levels of well-being, and the slightly lower scores evident in the "activity" area of practice. It is possible that by attending to appropriate activities for these children, levels of well-being could be raised overall, however other factors are also at play.

Table 4.23 - well-being levels in 1 round of screening. December/January 2006

First round of Well being		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	?	7	0.5	0.6	0.6
	1	31	2.1	2.5	3.1
	1.5	2	0.1	0.2	3.3
	2	109	7.4	8.9	12.2
	2.5	18	1.2	1.5	13.6
	3	325	22.0	26.5	40.1
	3.5	30	2.0	2.4	42.6
	4	370	25.1	30.2	72.8
	4.5	28	1.9	2.3	75.0
	5	306	20.7	25.0	100.0
	Total	1226	83.1	100.0
Missing	0	5	.3
	System	245	16.6
	Total	250	16.9
Total		1476	100.0

When we examine levels of well-being by age, we find for example that 23% of 0-3 year olds score a level 3 on well-being and 35% of P1 children score a level 5 on well-being.

Table 4.24 - The relationship between well-being and child age (n= 1230) (1 round)

Well-being1 * Age Crosstabulation		Age				Total
		0-3	3-4	4-5	P1
Well being Level	0	1	1	3		5
	?		2	4	1	7
	1	1	11	6	13	31
	1.5		1	1		2
	2	12	34	34	29	109
	2.5	5	7	6		18
	3	21	93	118	93	325
	3.5	7	6	12	5	30
	4	21	100	141	107	369
	4.5	9	8	11		28
	5	17	42	114	133	306
Total		94	305	450	381	1230

At the same time we find a significant negative correlation between levels of well-being in the first round and the Social Deprivation Indices: higher levels of well-being are related to low decile levels of social deprivation ( Annex 4, p.169). There is also a significant correlation (at the 0.01 level - 2-tailed) between the first and second rounds of well-being: if level of well-being is high in the first round then this is likely to be high in second round: in other words high levels of well-being are being sustained in many of these early childhood settings for many children. However, no significant correlation was found between social deprivation and levels of involvement, despite the fact that there is a significant correlation between well-being and involvement: if levels of well-being are higher, it can be expected that levels of involvement will be higher too. We also found a significant positive correlation between levels of well-being and parental age: if the parent is older, levels of well-being are higher: this fits with findings from the total difficulties score on the P- SDQ, where it was found that the younger the parent the higher the level of perceived difficulties in the areas of their children's emotions, concentration, behaviour or being able to get on with other people.

4.6.2 Involvement

Normally children who show high levels of well-being also show high levels of involvement. Features or indicators of child involvement are -

• Concentration The attention of the child is directed toward the activity. Nothing can distract the child from his/her deep concentration.
• Energy The child invests much effort in the activity and is eager and stimulated. Such energy is often expressed by loud talking, or pressing down hard on the paper. Mental energy can be deduced from facial expressions which reveal 'hard' thinking.
• Complexity and Creativity This signal is shown when a child freely mobilises his cognitive skills and other capabilities in more than routine behaviour. The child involved cannot show more competence - he/she is at his/her very 'best'. Creativity does not mean that original products have to result, but that the child exhibits an individual touch and what she/he does furthers his/her own creative development. The child is at the very edge of his/her capabilities.
• Facial Expression and Posture Nonverbal signs are extremely important in reaching a judgment about involvement. It is possible to distinguish between 'dreamy empty' eyes and 'intense' eyes. Posture can reveal high concentration or boredom. Even when children are seen only from the back, their posture can be revealing.
• Persistence Persistence is the duration of the concentration at the activity. Children who are really involved do not let go of the activity easily; they want to continue with the satisfaction, flavour and intensity it gives them, and are prepared to put in effort to prolong it. They are not easily distracted by other activities. 'Involved' activity is often more prolonged but it can be dependent on the age and the development of the child.
• Precision Involved children show special care for their work and are attentive to detail. Non-involved children gloss over such detail, it is not so important to them.
• Reaction time Children who are involved are alert and react quickly to stimuli introduced during an activity e.g. children 'fly' to a proposed activity and show prolonged motivation and keenness. ( NB. Involvement is more than an initial reaction.)
• Language Children can show that an activity has been important to them by their comments e.g. they ask for the activity repeatedly. They state that they enjoyed it!
• Satisfaction The children display a feeling of satisfaction with their achievements.

These indicators are used as observation guidance for staff, rather than items to be scored. Different children have different indicators of their own involvement in the learning environment. In this study staff and parents report that children's levels of concentration can cause them concern. By looking closely at the extent of children's involvement we can gain insight into their relationship with their learning environments. The recent report from HMIe, Improving Scottish Education (2006) recommends that pre-school staff "address the learning needs of individuals, particularly with regard to those who require additional support in their learning" (p.10) and in the primary schools section "The quality of pupils' learning experiences is still too variable and too often lacks relevance, engagement and excitement." (p.24). In this enterprise attending to children's involvement, concentration and engagement in learning stands to promote positive behaviour and equip children to participate and initiate in the activities offered to them.

Interrogation of results of the staff perceptions of young children's involvement was based on two rounds of observation by staff, four months apart. In the first round 1,208 children were involved in the whole class monitoring process for involvement. With the indictors in mind, staff observed 95 children in strata 1, 0-3 years; 309 children in strata 2, 3-4 years of age; 405 children in strata 3, 4-5 years of age, and 399 primary one children.

In the staff view 230 (19.1%) of children were at level 2.5 or lower in terms of their involvement in the early years setting. 362 (30%) are at the midpoint score of 3.5 and 3, whilst 613 (50.9%) of children in round 1 were experiencing high levels of involvement.

Table 4.25 - Number and age of observed children- involvement

Strata		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	1	95	7.9	7.9	7.9
	2	308	25.8	25.9	33.8
	3	404	32.6	32.7	66.5
	4	398	33.5	33.5	100.0
	Total	1205	99.8	100.0
Missing	System	3	.2
Total		1208	100.0

That half the child sample are at this level of involvement in their early education is a good thing, that the other half are average or below is a real cause for concern. Parents also raise concerns about children's levels of concentration. If so many young children are potentially disengaged in their learning practitioners need scope to develop learning environments that engage all children. The ECERS scores suggest that the key area for intervention is in terms of the activities on offer for children. These scores support a view that variety, levels of choice and an enrichment of imaginative and creative play opportunities would go some way towards addressing these issues. Children overall are experiencing higher levels of well-being than involvement according to the staff who work with them, it would therefore appear that shifts in provision and opportunities for learning would allow children to engage more fully.

Further it is likely that as the school year goes on, children's levels of involvement increase. Staff perceptions suggest that levels of involvement are a little higher in the second round of whole class screening, however this should be approached cautiously as there were fewer respondents in the second round, and this may be associated with settings where returns were lower.

Table 4.26 - Levels of involvement round 1

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	?	3	0.2	0.2	0.2
	1	50	3.9	4.2	4.4
	1.5	2	0.2	0.2	4.6
	2	155	12.1	12.9	17.4
	2.5	20	1.6	1.7	19.1
	3	318	24.8	26.4	45.5
	3.5	43	3.4	3.6	49.1
	4	358	27.9	29.7	78.8
	4.5	14	1.1	1.2	80.0
	5	241	18.8	20.0	100.0
	Total	1204	93.9	100.0
Missing	0	5	.4
	System	73	5.7
	Total	78	6.1
Total		1282	100.0

Table 4.27 - Levels of involvement round 2

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	?	1	0.1	0.1	0.1
	1	21	1.6	2.5	2.7
	1.5	1	.1	.1	2.8
	2	76	5.9	9.2	12.0
	2.5	10	.8	1.2	13.2
	3	253	19.7	30.6	43.8
	3.5	23	1.8	2.8	46.6
	4	262	20.4	31.7	78.3
	4.5	13	1.0	1.6	79.9
	5	166	12.9	20.1	100.0
	Total	826	64.4	100.0
Missing	System	456	35.6
Total		1282	100.0

In relation to age in the first round data across all age strata more children have levels of involvement at the 3 point and higher on the scale, than presented at 2.5 or below, but within the lower scored groupings 25% of 0-3 year olds, 26% of 3-4 year olds, 13% of 4-5 year olds and 18% of Primary 1 children are experiencing levels of involvement at 2.5 or below. On previous speculations about the link between well-being and the quality of environments these figures would seem to suggest that attention needs to be given to the relevance and appropriateness of activities for 0-3s and 3-4s. The number of 4-5 year olds with lower levels of involvement is less, but this figure rises again on entry to primary school. Linking this insight to what we have learned about transitions provides evidence for increased attention to children's involvement at transition to school.

Table 4.28 - Levels of involvement by age (Involvement 1 round)

Involvement1 * Age Crosstabulation		Age				Total
Involvement level		1	2	3	4
	?		2	1		3
	1	2	17	12	19	50
	1.5	1	1			2
	2	18	48	34	55	155
	2.5	5	12	2	1	20
	3	20	95	109	93	317
	3.5	12	16	11	4	43
	4	18	86	131	123	358
	4.5	6	2	6		14
	5	13	29	96	103	241
Total		95	308	402	398	1203

Summary

The focus of this project was on perceptions rather than interventions. Taking part in the assessment of children's well-being and involvement focused the attention of many of the staff on the importance of these concepts. In many settings staff were keen to build on the information they had generated about levels of well-being and involvement. Positive changes in children's well-being and involvement over time cannot be assumed as many factors may be involved. Raised awareness of staff will be important in creating such change.