9.1 Throughout the course of the study small classes in some schools have done better than others, although from year to year the most successful and weakest schools have not always been the same. This suggests that an even bigger factor in pupil progress arises from the combination of the teacher (some are more effective) than others and the mix of pupils in the class (their prior attainment, home background, etc). Nevertheless school effects in the small classes do have some influence and it therefore seems worthwhile to identify the variables which appear to correlate with a school’s success. Accordingly, an aggregated score based on attainment means in the three core subjects and a combination of attitudes and motivation scores was constructed for each school using the data from the small classes. This yielded six ‘very successful’
or ‘high achieving’ and four ‘very weak’ or ‘low achieving’ schools.
9.2 The data from the Principals’ questionnaire were presented previously.
There, a distinction emerged between those school leaders who saw their role as establishing goals and expectations and providing necessary resources but then delegating responsibility for teaching and the curriculum to middle managers and those who took a more active part in promoting and participating in curriculum development and the teachers’ professional learning. A recent meta analysis of leadership dimensions (Robinson 2007) has also identified these three factors along with two others (regular evaluation of the curriculum and teaching, and an orderly and supportive environment) as having the greatest effect on student outcomes. But of the five leadership dimensions the greatest effect was when school leaders actively participated in curriculum development, undertook visits to the classroom and engaged in formal and informal professional development sessions. This produced a large effect size (0.84) compared to the other four dimensions where the range varied from 0.27 (orderly and supportive environment) to 0.42 (regular evaluations).
9.3 The corresponding scale to Robinson’s (2007) strongest leadership dimension, active participation in curriculum and teacher learning development, was labelled in the SCT study curriculum development leadership. Besides the
active participation of the school Principal this included the provision of various kinds of support for the staff members’ professional development (mainly through inter- and intra-school sharing and the availability of specialist help from outside agencies) and the creation of a collegial culture.
Table 9.1 examines the mean scores on this scale for Principals from the six high and four low achieving schools.
Table 9.1 Principals rating as an Active Curriculum Development Leader
School Mean s.d. N
High Achieving 4.67* 0.52 6
Low Achieving 4.00 0.01 4
* = p<0.05 large effect size
9.4 It can be seen that the Principals of the high achieving schools rated this kind of leadership more highly. Another factor which also correlated with being a successful school was the Principal’s experience. In the more successful schools Principals had been in post for, on average, 11 years and 10 months compared to colleagues in the less successful schools where the average was six years. Principals in more successful schools were also more concerned about the effects of large classes on effective links with parents, presumably because they saw such liaison as important. Finally, a further scale emerged concerning the Principal’s beliefs about the perceived benefits of smaller classes. This included improvements in pedagogy, ability to deal with the needs of the less able pupils, better relationships between teachers and their classes and improved teamwork among staff. High scores on the scale indicated that the Principal held strong beliefs about the positive impact of small classes. This scale consisted of 33 items and had an internal consistency (Cronbach alpha) of 0.91. In the event, however, only 27 Principals supplied a full set of responses, four of these being among the six most successful schools. The two remaining successful schools were among the top quartile with a mean greater than 4.0 on the 5 point scale. Of the four least successful schools, two were in the bottom quartile with a mean rating below 3.87 while the other two both had mean ratings of 3.97. Although the latter trend is not as clear cut as in Table 9.1 there is some support for the view that a strong conviction about the value of small classes, length of experience in post and, in particular, active involvement in the schools’ curriculum development (including the development of teaching and learning strategies) all contributed to successful implementation of SCT.
9.5 A further factor which appeared to differentiate between successful and unsuccessful schools was the additional support that teachers received in the form of help from a classroom assistant. The comparison is shown in Table 9.2. While the vast majority of teachers received no help, 24 of the 148 teachers in the successful schools (16.2%) had the services of a teaching assistant for varying periods each week compared to 13.0% in less successful schools. However, when these figures were investigated further, 20 of the 24 of these classroom practitioners were teachers of English. The likely inference, therefore, is that these high achieving schools belonged to the Primary Literacy Programme where classroom assistance was made available and the
extra help provided in the more successful schools was not due to additional support by the Principal for SCT.
Table 9.2 Availability of teaching assistants (TAs) Periods with a TA
each week
High Achieving schools
Low Achieving schools
Total number of teachers
0 124 40 164
1 2 2 4
2 18 1 19
3 0 2 3
4 2 1 3
6 1 0 1
12 1 0 1
148* 46 194
* = p<0.05 small effect size
9.6 When teacher variables are examined there were no differences in qualifications or in the average length of teaching experience between the most and least successful schools. Some of the teachers in the extreme successful and unsuccessful schools were also part of the observation sample and were therefore classified into the various teacher types described in section 7 of this report. Table 9.3 shows the breakdown of the various teaching types. Although the trend is for the more successful schools to have a greater proportion of individual and pair sustained enquirers and the less successful ones to have more group task monitors these differences are not statistically significant. Overall, the combined profile in the most successful and least successful schools does not differ appreciably from that in Table 7.1 where the total sample of all the teachers who participated in the observation study is presented. Here, in the schools at the extreme ends of the attainment range, 33 of the 110 teachers (30%) are individual and pair sustained enquirers compared to 30.1% in the whole sample and 22 (20%) are group task monitors (18.2% in the whole sample). Successful and unsuccessful schools have a smaller proportion of whole class instructors (26.4 % as against 30.1% overall) and therefore slightly more whole class questioners (23.6% as against 21.3%) but none of these differences reach significance level.
Table 9.3 % of teaching types in high and low achieving schools Teacher types High achieving
schools
Low achieving schools
Total sample
N % N % N %
Individual and pair sustained enquirers
26 36.1 7 18.4 119 30.1
Group task monitors 11 15.3 11 29.0 73 18.5
Whole class instructors 20 27.8 9 23.7 119 30.1 Whole class questioners 15 20.8 11 28.9 84 21.3
Total 72 100.0 38 100.0 395 100.0
9.7 A similar analysis can be conducted for pupils. This is shown in Table 9.4.
Comparing the overall totals with the proportion of pupils types across the whole sample (Table 7.6) it can be seen that there are more solitary workers (52.3% to 43.8%) and fewer attention grabbers or seekers (5.8% to 10.5%) and this comes about mainly because in the low attaining schools nearly three-quarters of pupils (71.9%) are of the solitary worker type compared to 42.6%
in the high achieving schools. As a result, there are considerably more intermittent workers (high=28.4%; low=8.3%) and more active collaborators (high=22.3%; low=15.6%) in the high achieving schools, although the figure for the active collaborators is not statistically significant. This result would appear counter-intuitive as both theory and empirical research (Carroll, 1963, Block, 1971, Harnischfeger & Wiley, 1978) predict that time on task is one of the major determinants of learning. One possible explanation is that the solitary worker category masks two different kinds of pupils. From the profiles in Table 7.4 solitary workers are more often working on the own when seated in a group (69.35%) or seated in front of the class listening to the teacher (17.14%). In the latter case pupils while attending to the teacher’s words may not understand what is being said and it may be that a higher proportion of this kind of solitary worker exists among the classes in the low achieving schools. This would contrast with pupils who while actively engaged when working alone were able to complete the task and gain increased understanding as a result. Alternatively, it may be that the major factor in determining pupil progress lie outside the classroom behaviour in factors such as teacher competence, home background, parental support and so forth. This would explain why the more successful schools appear to be able to tolerate a higher proportion of pupils with a poorer work ethic, the intermittent workers.
Table 9.4 % of pupil types in high and low achieving schools Pupil types High achieving
schools
Low achieving schools
Total
N % N % N %
Solitary workers 84 42.6 69 71.9 153 52.2
Intermittent workers 56 28.4 8 8.3 64 21.9
Active collaborators 44 22.3 15 15.6 59 20.1
Attention grabbers 13 6.7 4 4.2 17 5.8
197 100.0 96 100.0 293 100.0 9.8 There is some support for the latter view in that there is a significant
difference between the high and low achieving schools in the proportion of children born in Mainland China. In Cohort 1, two-thirds of pupils (66.1%) attending low achieving schools were born in Hong Kong, whereas the corresponding figure for the most successful schools was 86.4%. In Cohort 2 the corresponding figures were 65.2% and 85.5% respectively. Across all schools taking part in the SCT study the percentage of children born outside Hong Kong was just under 20%. These differences are statistically significant (1% level) and give rise to a small effect size. It follows from the earlier analyses that place of birth is closely linked with social economic factors and
also with the degree of support that parents give to their children in their studies after school. In the first year of the study a parental support index was constructed which included such information as educational level achieved by both parents, family income, home resources available for learning (space, computer use, internet availability etc.) library use and frequency of outside educational visits. Table 9.5 displays the results for both Cohort 1 and 2.
Table 9.5 Parental support in high and low achieving schools
School Cohort 1 Cohort 2
Parental support index
mean
s.d. N Parental support index
mean
s.d. N
High achieving
28.35** 5.33 647 25.59** 4.74 611 Low
achieving
24.45 5.82 168 22.16 4.16 113
** = p<0.01 (medium effect size)
For both cohorts parental support is significantly greater in the high achieving schools (1% level). Furthermore, pupils entering the high achieving schools are more likely to have experienced kindergarten education. In the more successful schools 99% of pupils had attended kindergarten compared to 80%
in the least successful schools.
9.9 The final part of the analysis in this section concerns the relationship between the teachers’ main specialism and the subject that they were teaching in the small classes. In all schools across three years (P1 to P3) the trend is for just over a third of teachers who were trained in Chinese to find themselves mainly teaching mathematics (34.6%). Given the research evidence pointing to the strong links between a teacher’s subject expertise and pupil attainment, it is therefore somewhat surprising that 39.9% of teachers trained in mathematics end up mainly teaching either Chinese or English. In the more successful schools however the trend is not so marked and only 26% of Chinese trained teachers are required to teach mathematics, although 39% of mathematics trained teachers mainly teach the other two subjects (viz. Chinese and English).
In the group of less successful schools, however, 50% of Chinese trained teachers teach mathematics while 42.8 % of mathematics teachers teach either Chinese or English (although these percentages are based on low numbers).
Overall, these differences are significant (5% level) and yield a small effect size. Returning to the puzzling features of Table 9.4 these poorly qualified mathematics teachers may be able to sustain effective control and hence have a high proportion of the pupils on task (i.e. more solitary and fewer intermittent workers) but their lack of subject knowledge may result in pupils failing to gain sufficient understanding.
9.10 In summary, therefore the more successful schools tend to have more experienced Principals who have stronger expectations as to the possibilities of SCT seeing it, in particular, as a means for promoting improved attainment and motivation, for coping with pupil diversity and for fostering better relationships between teachers and pupils. This, in turn, leads to these
principals taking a more active role in curriculum development and in the professional development activities concerned with improving the effectiveness of SCT. They are more likely to deploy resources in school to provide additional non contact time for teachers to participate in intra- and inter-school activities. Whether by reputation or location these successful schools attract more Hong Kong born children which results in higher attendance at kindergarten and greater parental participation in school activities coupled with a higher level of support for these pupils outside school, whether it involves completing homework, undertaking educational visits or borrowing books from the library. These schools also succeed in attracting more specialist mathematics teachers and therefore have fewer teachers with a mainly Chinese initial training qualification teaching the subject. The above factors combined seem to exert a greater influence on a school’s performance than do the other variables considered such as class size and teaching approach.
Although there is a trend indicating that in the classes of the more successful schools there are more sustained cognitively challenging interactions between the teachers and individual pupils and a preference for working in pairs rather than groups, these differences fail to reach statistical significance. The least successful schools, in fact, succeed in having more pupils ‘on task’ but this engagement doesn’t appear to be matched by equivalent levels of knowledge and understanding.
10. Relative performance of schools with high proportions of disadvantaged