5. St Vincent: Quantitative survey

BACKGROUND:

Previous chapters concerning Trinidad and Barbados showed that females performed better in school than males. In the Trinidadian example, consideration began with a survey (confirming the superiority of female performance) and provided evidence of within-class actions that supported the performance of females. In Barbados, the project confirmed speculation that females performed better in schools than males; yet anomalies due to paternal and maternal occupation and with whom the child lives were identified. Some of the anomalies were explored in depth through classroom observations in a prestige secondary school and in a female-only secondary school. Both Trinidad and Barbados are among the islands in the Caribbean whose provision of schools allows all children to attend at primary and secondary levels.

Discussions within the Caribbean acknowledged that universal primary and secondary education was not characteristic of all the islands. Speculation was made concerning the participation and performance of females in situations where there was limited access to secondary school places. A quantitative study was deemed necessary to investigate participation and attainment levels in an educational system that allowed for universal primary education but restricted entry (due to limited places) to secondary schools. This study would investigate:

1) whether there was a higher proportion of females than males in secondary schools;

2) whether there was consistent superior female performance in both primary and secondary schools on within-class and national measures of school attainment; and

3) whether restricted entry to secondary schools affected children's expected career and educational opportunities.

St Vincent was identified as a island which displayed the differences in primary and secondary school places for children. Upon discussion with the Ministry of Education (St Vincent and the Grenadines) we agreed that a representative study could be undertaken in primary and secondary schools which replicated the previous Barbados study.

Setting the scene:

Who participates in schooling in St Vincent?

The school system in St Vincent is financed by the government and provides local primary schools throughout the island for all children. Entry to selective and stratified secondary schools is based upon levels attained on the Common Entrance Examination. While there are a number of secondary schools distributed around the island, the two high status schools (with established educational histories) are the male-only grammar and female-only high school. These schools are located in the main town, Kingstown.

The quantitative survey was undertaken with four specific age groups: Junior 2 (7 to 8 year olds), Junior 5 (11 to 12 year olds, the main primary school leaving group if the children had successfully passed their CEE), Form 2 (13 to 14 year olds) and Form 4 (16 to 17 year olds). Table 5.1 displays the island-wide distribution of males and females within these age groups (statistics taken from St Vincent and Grenadines, Ministry of Education, 1996).

Table 5.1: Island-wide distribution of males and females attending school within the age groups of the survey (percentage of sex type per age group in brackets)

AGE GROUP/ Sex	Junior 2	Junior 5	Form 2	Form 4
Male	1532(53.6)	1238(45.5)	754(41.8)	570(38.7)
Female	1328(46.4)	1481(54.5)	1050(58.2)	904(61.3)
Total	2860(100)	2719(100)	1804(100)	1474(100)

Total school enrolment for children in St Vincent and the Grenadines was 22149 in 1995/6, and there was a slightly greater number of males than females (11630 males and 10519 females) attending school in that year. Table 5.1 shows a proportional (and numerical) decline in school participation by both boys and girls. Based on this distribution, readers will see that there are fewer secondary school places for students, approximately 60% of the primary school cohort will gain entry to secondary school places. The decline within the secondary schools, though, was far more evident among males than females.

The Sample:

As with the Barbados quantitative survey, we wished the St Vincent sample to be representative, to provide information on the participation of girls and boys in schooling, to assess underlying reasons for within-class and national attainment and to question whether attainment was related to career and educational expectations of children. Because of the number and location of primary and secondary schools on the island, it was agreed with the Ministry of Education that the sample would be:

a. proportional with regard to educational district,

b. stratified with regard to ages of children (Junior 2, Junior 5, Form 2, and Form 4) and types of secondary schools (co-educational, single-sex),

c. focused to ensure a range of CXC performance of secondary schools, and

d. clustered to gain information from whole classes within the selected schools.

All of the schools in this survey were state-funded. (There are very few private schools in St Vincent (three primary schools and no secondary schools) and none were included in this sample.) Of the sixty state-funded primary schools in St Vincent, the survey drew upon twelve schools (20% of the total). Of the twenty-one secondary schools, the survey drew upon seven schools (33% of the total). Among the secondary schools focused upon, the top performing (by CXC Examination results) schools included the male-only, the female-only single-sex schools and a co-educational school. All other secondary schools were co-educational and presented a range of mid to low CXC performance. All of the primary schools were co-educational. Overall, data was recorded for 1379 children. There were 916 primary school pupils and 463 secondary students surveyed.

The survey profile of parental occupations from the St Vincent sample was slightly different from the reported Barbados sample, displaying a higher proportion of craft and elementary labouring work undertaken by fathers (Table 5.2) and by mothers (Table 5.3).

Table 5.2: Paternal occupations of St Vincent sample

Value Label	Frequency	Percent	Valid Percent	Cum Percent
top managers	17	1.2	1.5	1.5
professionals	44	3.2	3.9	5.4
tech assoc prof	98	7.1	8.8	14.2
clerks	20	1.5	1.8	16.0
service/sales	150	10.9	13.4	29.4
skilled agric/fish	15	1.1	1.3	30.7
crafts	274	19.9	24.5	55.2
machine operators	16	1.2	1.4	56.6
elementary	302	21.9	27.0	83.6
unemployed	40	2.9	3.6	87.1
vague	112	8.1	10.0	97.1
immigrant	17	1.2	1.5	98.7
deceased	13	. 9	1.2	99.8
student	2	.1	.2	100.0
no occupation identified	259	18.8	Missing
Total	1379	100.0	100.0

Table 5.3: Maternal occupations of the St Vincent sample

Value Label	Frequency	Percent	Valid Percent	Cum Percent
top managers	7	.5	.6	.6
professionals	95	6.9	7.5	8.0
tech assoc prof	23	1.7	1.8	9.9
clerks	85	6.2	6.7	16.5
service/sales	109	7.9	8.6	25.1
crafts	39	2.8	3.1	28.2
machine operators	7	.5	.6	28.8
elementary	488	35.4	38.5	67.2
unemployed	148	10.7	11.7	78.9
housewife	195	14.1	15.4	94.2
vague	48	3.5	3.8	98.0
immigrant	18	1.3	1.4	99.4
deceased	6	.4	.5	99.9
student	1	.1	.1	100.0
no occupation identified	110	8.0	Missing
Total	1379	100.0	100.0

Information provided by the survey:

1. Who is succeeding in school?

This first question explored whether the superior attainment performance of females within the St Vincent sample was the same as found in Barbados. The comparisons drew upon the raw end-of-term scores from each class, the standard deviated (standardised) within-class scores for all children, and the CEE scores for the secondary school students. Analyses were undertaken for the whole sample and at each year level. Within-class scores included an average of core curriculum scores and subject scores in English, mathematics, science and social studies.

Within-class attainment: Drawing upon the 'raw' scores assigned by the teachers on the end-of-term examinations, girls were found to attain higher percentage grades and perform at a higher standardised level overall (Table 5.4). While girls performed at a higher level than boys, their performance was not consistently higher in all curriculum areas (a finding distinct from Barbados and Trinidad). Girls' attainment excelled in English and social studies. Their attainment was not significantly higher than boys in mathematics or the sciences.

Table 5.4: Raw and standard deviated scores for core subjects and significance of differences between boys and girls.

Core Course	Average Male Score	Average Female Score	Difference based on Raw Score	Difference based on S.D. Score
Average	52.32	54.70	F1,1249= 7.454, p<0.0064	F1,1228= 5.983, p<0.015
English	50.39	55.06	F1,1234= 21.355, p<0.0001	F1,1213= 19.750, p<0.0001
Maths	49.64	49.20	N.S.	N.S.
Integrated Science	55.06	56.77	N.S.	N.S.
Social Studies	54.84	58.70	F1,1161= 13.225, p<0.0003	N.S.

When the within-class attainments were explored by year in school, even fewer differences were found between the performance of boys and girls. There were no significant differences in attainment between boys and girls in any of the core curriculum subjects at the Junior 2 level. At Junior 5, girls attained higher raw scores in English, mathematics and social studies, although the standardised score only confirmed this result in English. At second form, boys attained higher raw scores in mathematics but this was not confirmed using the standardised scores and girls scored higher on the standardised scores in English. At fourth form, girls scored significantly higher in English, social studies and biology but the standardised scores only confirmed this result in English. The disparity between raw and standardised results is important; raw scores tell that girls were assigned higher scores than boys but within the classes the standardised scores did not show as many differences in performance between males and females as were found elsewhere in the Caribbean.

Differences between boys and girls were also explored with regard to the type of school attended. In these analyses, the average and English scores were used for comparison. Differences showed that girls attained higher within-class and standardised scores in primary schools (all of these schools were co-educational) and these differences were also found in the mid-attaining co-educational secondary schools. Differences between the attainments of boys and girls were not found among the low attaining and high attaining secondary schools (Table 5.5).

Table 5.5: Raw and standardised score performance of boys and girls by type of school attended (for co-educational schools only), drawing upon average scores and scores in English

Type of school	Average Male Score	Average Female Score	Difference based on Raw Score	Difference based on S.D. Score
Primary Average	51.13	35.27	F1,797=3.295, p<0.069	N.S.
English	49.61	54.06	F1,788=10.796' p<0.0011	F1,788=9.239/ p<0.0024
Secondary (low att)
Average	53.12	53.45	N.S.	N.S.
English	51.77	56.99	N.S.	N.S.
Secondary (mid att)
Average	49.48	54.39	F1,149=6.159, p<0.014	F1,144=15.092/ p<0.0001
English	41.28	49.21	F1,144=15.092, p<0.0002	F1,144=18.382, P<0.0001
Secondary (high att)
Average	51.08	50.97	N.S.	N.S.
English	48.00	52.59	N.S.	N.S.

CEE, a national comparison

Descriptive statistics indicate that girls were likely to attain at higher levels than boys at both primary and secondary school levels generally. The CEE scores taken at the end of the primary school present a very different picture. These scores were collected for all second and fourth form children in the sample. The scores represent only those who were successful on the examination and gained one of the limited number of places in secondary schools. There was virtually no difference in the average CEE scores for boys and girls; average for boys was 171.63 and average for girls was 171.66. While the scores for the boys and girls were nearly the same, the number of children that completed the CEE by sex tells a different story. Data for males was collected from 107 boys and data was collected from 253 females. Two points can be made from this data: first, students who successfully completed the CEE gained near-equivalent scores although there were nearly 2.5 females who were successful for every male; and second, although starting secondary schools with nearly equal CEE scores, girls were still more successful within their classes - especially in English.

2. What was the distribution of male and female pupils in each of the year levels: was there evidence of male drop-out or were girls restricted entry to secondary schools?

Overall, there were more girls than boys attending school in the sample and at each age level (Table 5.6). Within the primary schools there were slightly more girls than boys at each of the year levels. Within the secondary schools there was a dramatic drop off in the number of boys attending at the second and fourth forms. The drop off in male presence at secondary schools mirrored the national statistics. The drop off also shows that selective entry to secondary education was not dominated by equality (in number of boys and girls attending), nor was it dominated by males usurping secondary school places that characterise other developing countries (Brock & Cammish, 1997).

Table 5.6: Total number of males and females in the sample and distribution at each year level surveyed (percentage by year group in brackets)

SEX	YEAR IN SCHOOL				Row Total
	Junior 2	Junior 5	Form 2	Form 4
male	154	262	80	71	567
	(44.9)	(45.9)	(33.6)	(31.6)	(41.2)
female	189	309	158	154	810
	(55.1)	(54.1)	(66.4)	(68.4)	(58.8)
Column	343	571.	238	225	1377
Total	(25.3)	(42.1)	(17.6)	(16.3)	(100.0)

Focusing on boys only, we questioned whether consideration of parental occupation helps to provide an explanation for the dramatic drop-off of male participation at secondary school. Tables 5.7 and 5.8 display that most of the drop-off took place among boys whose parents worked in crafts and unskilled occupations. Boys whose parents worked in skilled and professional/managerial occupations were more likely to stay-on in secondary school.

Table 5.7: (Grouped) Paternal occupations of boys in school by year level excluding incomplete figures from form 4 (row and column percentages in brackets)

FATHER OCC.	EAR IN SCHOOL			Row Total
	Junior 2	Junior 5	Form 2
craft/elem/unemp	81	146	29	256
	(31.6)	(57.0)	(11.3)	(69.2)
	(68.6)	(73.7)	(53.7)
clerk/service/sk	16	26	15	57
	(28.1)	(45.6)	(26.3)	(15.4)
	(13.6)	(13.1)	(27.8)
professional/man	21	26	10	57
	(36.8)	(45.6)	(17.5)	(15.4)
	(17.8)	(13.1)	(18.5)
Column	118	198	54	370
Total	(31.9)	(53.5)	(14.6)	(100.0)

Table 5.8: (Grouped) Maternal occupations of boys in school by year level excluding incomplete figures from Form 4 (row and column percentages in brackets)

MOTHER OCC.	YEAR IN SCHOOL			Row Total
	Junior 2	Junior 5	Form 2
manager/prof	13	22	16	51
	(25.5)	(43.1)	(31.4)	(11.9)
	(10.5)	( 9.3)	(23.5)
clerk/service	12	31	21	64
	(18.8)	(48.4)	(32.8)	(15.0)
	( 9.7)	(13.1)	(30.9)
craft/unskilled/	83	154	15	252
	(32.9)	(61.1)	( 6.0)	(58.9)
	(66.9)	(65.3)	(22.1)
housewife	16	29	16	61
	(26.2)	(47.5)	(26.2)	(14.3)
	(12.9)	(12.3)	(23.5)
Column	124	236	68	428
Total	(29.0)	(55.1)	(15.9)	(100.0)

3. What other explanations can be provided for school-based success and failure?

3. What other explanations can be provided for school-based success and failure?

Within-class and CEE results and parental occupation: The Barbados survey showed that there were significant differences in attainment due to parental occupation, especially that of the mother, and that these differences accounted for more of the variance in attainment scores than the sex of the child. A

similar set of analyses were undertaken on the St Vincent data. For this analysis, the occupations of the father were broken down into three groupings: professional/managerial, clerk/skilled, and craft/unskilled. The occupations of the mother were broken down into four groupings: professional/managerial, clerk/skilled, craft/unskilled, and housewife. Using the raw within-class attainment scores in each of the core curriculum subjects and CEE many significant differences were found with regard to the occupations of father and mother. The higher the occupational level (and educational background) of the father, the higher the average and core curriculum and CEE scores for the child. This statement is made with regard to averaged grades in each subject and most of the curriculum subjects at each age level (Table 5.9) and is replicated using the standardised attainment scores. Focusing on the magnitude of differences between the different year levels in school, readers will note that differences between children of professional and craft background at secondary schools were not as wide as found in the primary school years. While there were significant differences between children from different paternal backgrounds in the secondary schools, the selection to gain entry into secondary school obviously withdrew a number of the (potentially) lower attaining children from the sample. Analyses for occupation of the mother paralleled those of the father; this was not surprising as there was a significant correlation between occupations of parents (rho= 0.358, p<0.0001). A more focused view of Table 5.10 shows that the general relationship between maternal occupation and within-class results did not remain the same at all year levels; the relationship between maternal occupation and subject attainment virtually disappeared at the second and fourth forms (Table 5.10). A possible explanation for the non-significant differences in within-class attainments explained by maternal occupation may be the gradual drop-off in school participation of children from the craft and unskilled parental occupations. The remaining children in school will be supported by a more educated and skilled home background, as well as being the children who have succeeded in their earlier years of schooling.

Table 5.9: Within-class attainment averaged by (grouped) paternal occupation

SUBJECT/ Yr.Level	AVERAGE	ENGLISH	MATHS	SCIENCE	SOC.ST.
Overall
prof..	61.90	61.87	58.94	62.76	64.78
clerk..	54.92	54.65	50.01	56.78	58.69
craft..	52.13	51.27	48.21	54.52	55.09
Signif.	F2,885=24.319, p<0.0001	F2,877=21.691, p<0.0001	F2,878=18.473, p<0.0001	F2,813=10.616, p<0.0001	F2,823=16.852, p<0.0001
Junior 2
prof..	70.77	73.63	67.33	73.52	68.59
clerk..	59.15	59.82	54.39	64.03	58.17
craft..	60.48	60.95	55.85	63.38	61.47
Signif.	F2,190=5.363, p<0.0054	F2,189-5.885, p<0.0033	F2,190=4.060, p<0.0188	F2,188-3.595, p<0.0294	N.S.
Junior 5
prof..	57.22	56.02	57.21	57.09	59.75
clerk..	49.01	50.35	44.39	51.22	51.63
craft..	46.76	45.72	44.09	49.14	48.90
Signif.	F2,404=10.909, p<0.0001	F2,399=8.795, p<0.0002	F2,404=13.314, p<0.0001	F2,388=4.351, p<0.0135	F2,382=8.428, p<0.0003
Form 2
prof..	65.90	65.67	59.71	67.16	71.10
clerk..	62.08	58.67	60.49	60.19	69.77
craft..	58.44	56.61	55.18	58.91	62.65
Signif.	F2,144=5.585, p<0.0046	F2,143=4.127, p<0.0181	N.S.	F2,143=3.551, p<0.0312	F2,128=4.149, p<0.0179
Form 4
prof..	58.62	58.00	53.97	56.06	64.38
clerk..	53.59	52.07	43.18	55.25	63.61
craft..	51.70	50.41	43.54	54.26	59.32
Signif.	F2.138=5.265, p<0.0063	F2,137=5.259, p<0.0063	F2,131=4.694, p<0.0107	N.S.	F2,117=3.501, p<0.0334
Average CEE scores
prof..	186.45	N.A.	N.A.	N.A.	N.A.
clerk..	173.44
craft..	166.35
Signif.	F2,236=6.237, p<0.0023

Table 5.10: Within-class attainment averaged by (grouped) maternal occupation

SUBJECT/ Yr.Level	AVERAGE	ENGLISH	MATHS	SCIENCE	SOC.ST.
Overall
prof..	60.58	60.26	56.31	63.60	64.33
clerk..	56.50	54.98	52.46	57.27	61.88
craft..	51.64	51.16	46.96	55.00	54.84
h/wife	54.09	52.62	50.44	55.09	57.66
Signif.	F3,1083=14.3290, p<0.0001	F3,1072=9.905, p<0.0001	F3,1075=9.925, p<0.0001	F3,979=7.108, p<0.0001	F3,1016=13.201, p<0.0001
Junior 2
prof..	73.24	73.16	69.95	76.42	73.42
clerk..	66.40	63.95	63.95	67.09	70.62
craft..	59.99	61.12	54.57	63.74	60.45
h/wife	53.89	52.12	47.68	61.08	54.50
Signif.	F3,206=6.868, p<0.0003	F3,205=4.596, p<0.0039	F3,206=5.470, p<0.0012	F3,205=3.345, p<0.0199	F3,203=6.195, p<0.0005
Junior 5
prof..	58.21	57.92	55.47	60.13	59.95
clerk..	50.12	48.18	47.88	52.56	52.11
craft..	46.20	45.60	43.08	49.38	48.19
h/wife	51.85	51.16	49.97	51.23	55.31
Signif.	F3,495=9.317, p<0.0001	F3,488=7.099, p<0.0001	F3,495=8.261. p<0.0001	F3,477=4.386, p<0.0046	F3,469=7.359, p<0.0001
Form 2
prof..	61.27	60.79	56.07	62.41	65.68
clerk..	61.12	59.45	56.31	59.53	70.02
craft..	59.92	58.52	55.11	60.94	66.15
h/wife	56.65	53.76	54.63	58.07	61.19
Signif.	N.S.	N.S.	N.S.	N.S.	N.S.
Form 4
prof..	55.25	54.91	49.44	59.06	62.39
clerk..	54.09	54.10	46.63	54.74	61.89
craft..	51.57	49.20	41.92	56.83	61.14
h/wife	56.02	54.82	48.23	56.00	61.25
Signif.	N.S.	F3,177=3.337, p<0.0207	N.S.	N.S.	N.S.
Average CEE scores
prof..	182.44	N.A.	N.A.	N.A.	N.A.
clerk..	179.33
craft..	162.31
h/wife	171.79
Signif.	F3,296=6.249, p<0.0004

Parental occupation by type of school attended: Occupation of parents was an important factor in the explanation of the presence of particular children in stratified types of secondary schools in both Trinidad and Barbados, and a similar expectation was held for St Vincent. Analyses focused solely on secondary schools, as this is where stratification within the school system occurs. Tables 5.11 and 5.12 clearly show that children whose fathers and mothers were employed in managerial or professional positions were most likely to attend the top attaining schools in the sample. The difference in distribution between children from various parental occupations within secondary school types was not significant for paternal occupation but was highly significant for maternal occupation (X² (6)=19.242, p<0.004). Table 5.11 shows a higher proportion of children from a professional/managerial background attended the high attainment secondary schools.

Table 5.11: Distribution of students in types of secondary school by grouped paternal occupation (percentage of occupation by school types in brackets)

SEC. SCHOOL TYPE	FATHER'S OCCUPATION			Row Total
	prof/ man	clerk/ skill	craft/ unskill
sec.-low att	8	12	39	59
	(12.1)	(18.8)	(23.5)	(19.9)
sec.-mid att	15	23	51	89
	(22.7)	(35.4)	(30.7)	(30.1)
sec.-high att	43	29	76	148
	(65.2)	(45.3)	(45.8)	(50.0)
Column	66	64	166	296
Total	(22.3)	(21.6)	(56.1)	(100.0)

Table 5.12: Distribution of students in types of secondary school by grouped maternal occupation (percentage of occupation by school types in brackets)

SEC. SCHOOL TYPE	MATHER'S OCCUPATION				Row Total
	prof/ man	clerk/ skill	craft/ unskill	house-wife
sec.-low att	10	17	50	16	93
	(15.5)	(18.9)	(32.9)	(19.5)	(24.0)
sec.-mid att	16	29	54	26	125
	(25.4)	(32.2)	(35.5)	(31.7)	(32.3)
sec.-high att	37	44	48	40	169
	(58.7)	(48.9)	(31.6)	(48.8)	(43.7)
Column	63	90	152	82	387
Total	(16.3)	(23.3)	(39.3)	(21.9)	100.0)

Raw scores and CEE scores by type of secondary school: The difference in social background of those attending the various types of secondary schools is shown in a slightly different way when considering whether the different types of secondary school provide differential feedback to their students. Table 5.13 clearly displays that the higher attaining schools awarded the highest within-class scores to their students. Also, the students in the top attaining schools scored higher on their CEE examinations. This finding shows a significant differentiation on the part of schools and teachers. Within-class scores were assigned by each teacher, based on the work that the class covered in each subject over the term. Theoretically, each type of school could and should provide the same range and average scores for its students. The table clearly shows that the high attaining school provided the highest scores for students (Scheffe post hoc analysis) and there was little differentiation between the mid and low attaining schools.

Table 5.13: Average end-of-term, core curriculum subject and CEE scores for types of secondary school

Type of school	Average score	English	Maths	Science	Social Studies	CEE
High Att	61.03	61.83	55.08	62.02	66.88	191.16
Mid Att	52.54	46.27	51.07	50.52	63.56	158.73
Low Att	53.36	55.59	39.67	60.85	59.59	160.04
Diff.	F2,449=29.232, p<0.0001	F2,443=63.587, p<0.0001	F2,438=23.339, p<0.0001	F2,351=22.493, p<0.0001	F2,393=8.875, p<0.0002	F2,339=62.356, p<0.0001

Even within the top attaining secondary schools, there was a significant difference between the three schools that made-up this grouping. The three schools included the high status male-only grammar school, the high status female-only school and a co-educational school. CEE scores show that girls attending the female-only school attained the highest scores, followed by the male-only and co-educational schools. The school giving the highest within-class scores to students was the female-only school. These scores were closely followed by the male-only school. Both of the single-sex schools gave higher scores to their students than the co-educational school (Table 5.14).

Table 5.14: Average end-of-term, core curriculum subject and CEE scores for high attaining secondary schools

Type of school	Average score	English	Maths	Science	Social Studies	CEE
Male	62.84	63.93	62.33	61.68	72.50	190.49
Female	67.25	67.42	62,33	71.91	73.46	213.63
Co-ed.	50.98	52.35	38.42	52.09	56.22	159.19
Diff.	F2,185=56.891, p<0.0001	F2,185=35.929, p<0.0001	F2,185=61.304, p<0.0001	F2,156=37.981, p<0.0001	F2,73=21.155, p<0.0001	F2,163=95.294, p<0.0001

While students attending the female-only school were assigned the highest average within-class scores of the top attaining schools, we questioned whether girls in the top attaining co-educational school attained higher than boys in that school. Before this analysis was undertaken, it was noted that there were very few males in the co-educational school, girls outnumbered the boys by a ratio of ten to one. A review of the raw and standardised scores showed that there were no significant differences in the attainments of boys and girls in this school.

Pre-school attendance: Attendance of pre-school contributed a significant amount of the variance in the within-class scores for secondary students in Barbados and similar analyses were undertaken on the St Vincent data. The survey data showed that a very high proportion of children stated that they had attended pre-school; nearly 65% said that they had attended a pre-school. Overall, if children attended a pre-school before starting primary school, they were more likely to attain higher within-class scores (calculated on the raw data, F1,1212= 13.675, p<0.0001; and on the standardised scores F1,1212= 4.604, p<0.0321). Similar findings were shown in the three main curriculum subjects of English, mathematics and science, but significant differences were not found for social studies. The analyses were undertaken separately at primary school and secondary school levels. Findings showed significant differences for primary schools (raw scores: F1,784= 18.054, p<0.0001; standardised: F1,784= 4.834, p<0.028), but there were no significant differences found within the secondary school population (although the numerical averages for pre-school attenders was higher than non-attenders). The higher raw and standardised within-class scores for pre-school attenders may be explained by the greater likelihood that those children who attended pre-school came from a privileged home background; parents who worked in professional or managerial occupations were more likely to send their children to pre-school than other parental occupations (father X² (2)=8.011, p<0.018; mother X² (3)=15.480, p<0.0015). There was no difference in pre-school attendance between boys and girls in the sample.

With whom does the child live? A child's success in school in Barbados was partially determined by parental background and this was seen to affect type of school attended. A further home-based contribution to educational success was with whom the child lived. In St Vincent, most children lived with the mother or mother and father (Table 5.15). These results provided a comparable percentage of children who lived with the mother only and father only but a smaller proportion of the sample lived with both parents and a larger proportion lived with grandparents(see Table 4.1.21).

Table 5.15: With whom does the child live?

	Frequency	Percent
mother only	523	37.9
father only	51	3.7
both parents	513	37.2
grandparents	148	10.7
other relatives	83	6.0
guardian	40	2.9
(missing)	21	1.5
Total	1339	100.0

In Barbados, children who lived with both parents attained at higher levels within-class and nationally. A similar result was found among the St Vincent sample (Table 5.16). Children who lived with both parents attained higher within-class scores and this was substantiated using standardised scores. These findings were consistent for all children and among separate analyses for boys and girls. The children who attained, on average, the lowest in schools were those who lived with the mother only. When focusing on the CEE scores, a different picture is presented. There were no significant differences between the scores when compared to the different 'living with' situations. There are some indications that boys living with grandparents and girls living with both parents did better than others on the CEE. Readers should remember that CEE scores represent an elite group of children who passed their primary school leaving examination, thus this was a much smaller sample than found in the whole survey.

Table 5.16: Averages and differences on within-class and national scores for children living with mother, father, both parents and grandparents

CHILDREN/ Scores	ALL CHILDREN	MALES ONLY	FEMALES ONLY
Common Entrance Live with:
mother	171.39	172.61	170.88
father	152.40	158.33	149.86
both	173.19	164.83	176.14
grandparent	169.33	178.83	166.62
Significance	N.S.	N.S.	N.S.
Standardised Live with:
mother	-0.1381	-0.2284	-0.0794
father	-0.0117	0.0102	-0.0379
both	0.1314	0.0279	0.2086
grandparent	-0.0351	-0.2351	0.1070
Significance	F3,1107= 5.854, p<0.0006	F3,452= 2.644, p<0.048	F3,650= 4.009, p<0.0076
Raw Average Live with:
mother	52.29	50.99	53.19
father	52.31	50.98	53.98
both	55.60	53.98	56.84
grandparent	52.89	50.95	54.39
Significance	F3,1128= 4.042, p<0.0072	N.S.	F3,650= 2.975, p<0.0311

A review of the relationship between parental occupations and with whom the child lives showed that there was no significant difference between the range of paternal occupations and with whom the child lives. For maternal occupation, though, there was a significant difference (X² (9)= 45.618, p<0.0001), showing that maternal occupation when living with both parents was most likely to be managerial/professional or housewife. Other 'living with' situations showed lower proportions of these occupations for mothers.

Who helps with homework? Simply identifying with whom the child lives does not tell about the help that a child receives with school work. Children were asked whether anyone helps them with homework. Most children stated that they received help with their homework (81.4%). Children were more likely to receive help with homework when they attended primary school than secondary school (approximately 90% received help in primary school while 65% received help at secondary). There were no parental occupation differences between those children who received help or not. Table 5.17 reveals that mothers, sisters and brothers were most likely to help with homework within the family; fathers only helped twenty-one percent of their children. Within the table, we were able to explore whether males or females worked with friends on their homework, and found that sixty-two percent of those who received help from a friend were female. Using tutors to help with homework may appear, logically, to be the prerogative of children of professionals and managers. In this sample, though, the largest proportion of children who received help from a tutor (approximately 63%) came from a craft and unskilled home background.

Table 5.17: Who helps the child with homework

Who helps with homework	Percentage receiving help
Mother	48.6%
Friend	36.2%
Relatives	33.9%
Sister	32.5%
Brother	22.7%
Father	21.3%
Tutor	15.4%

Dividing children's within-class attainment into three groups of high, middle and low attainers and comparing this grouping to help with homework revealed that there were no differences between those who received homework help from mothers, brothers, sisters, relatives or friends. High attainers were more likely to receive help from their father than other groups. Low attainers were more likely to receive help from a tutor than other groups.

Attitudes to future education and occupation: The Ministry of Education in St Vincent requested that the children surveyed be asked to respond to three questions about the role of education in their lives. Children were asked what occupation they aspired to, whether and why education was important and to what level they expected to rise in education. Descriptively, children answering these questions showed extremely high expectations for future occupation and education. Table 5.18 displays that two-thirds of children thought that they would, eventually, rise to a professional or managerial occupation. Table 5.18 contrasts sharply with the actual parental occupations of the sample (Tables 5.2 and 5.3) which displays only 13% of parents in these professional positions.

Table 5.18: Occupational expectations of children in the sample

Value Label	Frequency	Percent	Valid Percent	Cum Percent
top managers	18	1.3	1.4	1.4
professionals	602	44.3	46.3	47.7
tech assoc prof	249	18.3	19.2	66.8
clerks	118	8.7	9.1	75.9
service/sales	79	5.8	6.1	82.0
skilled agric/fish	3	.2	.2	82.2
crafts	123	9.1	9.5	91.7
machine operators	1	.1	.1	91.8
elementary	11	.8	.8	92.6
unemployed	1	.1	.1	92.7
vague	95	7.0	7.3	100.0
	58	4.3	Missing
Total	1358	100.0	100.0

In a similar manner, many children displayed high expectations for their educational careers (Table 5.19). Nearly sixty percent of children stated that they expected to go to university and a further fourteen percent expected to attend technical college.

Table 5.19: Educational expectations of the sample children

Value Label	Frequency	Percent	Valid Percent	Cum Percent
primary	55	4.1	4.1	4.1
form 3	16	1.2	1.2	5.3
CXC	54	4.0	4.0	9.3
form 6	93	6.8	6.9	16.2
university	796	58.6	59.2	75.4
tech college	190	14.0	14.1	89.5
evening class	141	10.4	10.5	100.0
	12	.9	Missing
Total	1358	100.0	100.0

In a sense, the occupational and educational expectations appear unrealistic - the thoughts of those who have not yet entered the real and competitive world of higher education and jobs. In another sense, the children were very focused on the importance of education. When asked if education was very important, unimportant or if they were unsure, ninety-eight percent stated that education was very important and only half-a-percent stated that it was unimportant. When asked why education was important, over two-thirds of the children stated that it was for the development of knowledge and skills; another thirty percent stated that education would help to gain a job and less than ten percent identified social aspects of education (developing and seeing friends) as important.

The unusually high job and educational expectation could have been accounted for by young children's unrealistic images of their futures. When Tables 5.18 and 5.19 were broken down into primary and secondary school levels, very few differences were found between children at these levels. Children whose mothers or fathers worked in professional and managerial positions tended to identify these positions for their future employment. Children whose parents worked as clerks, skilled or service positions showed job expectations that ranged from professional through sales and service positions. A large majority of the children whose parents worked in craft and unskilled positions (nearly two-thirds of the children) expected to work in managerial and professional positions, and this finding characterised children at both primary and secondary school. A similar profile of educational expectation was found among primary and secondary school children. Many of the children expected to go to university, and this included over 55% of the children from craft and unskilled parental background.

When comparing job and educational expectations between boys and girls, significant differences were found between the sexes on each question. Girls showed a slightly higher expectation for professional and managerial positions and a higher expectation for clerk and service positions (Table 5.20) while more boys saw future positions in crafts and elementary labouring positions (X² (2)= 95.102, p<0.0001). Table 5.21 displays that girls also had a slightly different and higher expectation for then-education (X² (6)= 18.406, p<0.005). The educational aspirations of boys and girls appear very similar

except for those who expected to leave school before CXC examinations and those who aspired to university. Boys were more likely to see themselves as leaving school before CXC examinations and a greater proportion of girls aspired to university than boys.

Table 5.20: Male and female job expectation (grouped occupations) with percentages of occupation by sex in brackets

SEX	EXPECTED OCCUPATION			Row Total
	man./ prof.	clerk/ service	craft/ elemen.
male	326	42	99	467
	(69.8)	(9.0)	(21.2)	(38.8)
female	543	158	36	737
	(73.7)	(21.4)	(4.9)	(61.2)
Column	869	200	135	1204
Total	(72.2)	(16.6)	(11.2)	(100.0)

Table 5.21: Male and female educational aspirations (percentages of educational aspiration by sex in brackets)

SEX	EDUCATIONAL ASPIRATION					Row Total
	primary	form 3	CXC	form 6	university
male	32	8	30	38	295	542
	(5.9)	(1.5)	(5.5)	(7.0)	(54.4)	(40.3)
female	22	6	25	56	502	803
	(2.7)	(.7)	(3.1)	(7.0)	(62.5)	(59.7)
Column	54	14	55	94	797
Total	(4.0)	(1.0)	(4.1)	(7.0)	(59.3)

	tech college	evening class	Row Total
1	81	58	542
male	(14.9)	(10.7)	(40.3)
2	109	83	803
female	(13.6)	(10.3)	(59.7)
Column	190	141	1345
Total	(14.1)	(10.5)	(100.0)

4. From the range of significant results provided thusfar, what is the relative contribution of each result to the overall performance of pupils and students in school? Throughout these analyses a number of significant results have been presented to explain the performance of children within-class and on the national CEE. Significant results showed that girls attain at higher levels than boys, that children whose parents worked in managerial and professional positions attain at higher levels than others, pre-school attendance was associated with higher attainment, status of secondary school attended affects attainment, whether or not the child lived with both parents and if the father helped with homework also affect attainment. Stepwise regressions were undertaken to ascertain the amount and significance of each of these variables in explaining average end-of-term results, standardised end-of-term results and CEE scores. These results are conducted separately for primary and secondary school children.

Primary schools: Working with the standardised within-class scores, the regressions provided little information of significance. In total, about six percent of the variance was accounted for and this percentage was composed of two factors: occupation of the father (3.4% adjusted R square) and whether the child lived with both parents (a further 2.4% adjusted R square). Using the raw, non-standardised, within-class scores only 2% of the variance was accounted for and occupation of the mother contributed this amount. The most obvious conclusion is that success in primary school was not accounted for by the sex of the child, but was most likely to be affected by the occupation (and educational background) of the child's parents.

Secondary schools: Remembering that entry to secondary school was limited to only sixty percent of the primary school population, we expected that regressions may be very different from those found in Trinidad (Jules & Kutnick, 1990) and Barbados (Chapter 4.1). Stepwise regression on the raw within-class scores showed that six percent of the variance was accounted for by two factors; the occupation of the father (3.7% adjusted R square) and whether the father helped with homework (a further 2.4% adjusted R square). Using the standardised within-class scores, a smaller amount of variance was explained: help with homework by the father (2.3% adjusted R square) and sex of the student (a further 1.4% adjusted R square). Similar analyses were undertaken that drew upon co-educational schools only; this analysis excluded the two (single-sex) top prestige schools from the sample. Co-educational results mirrored the standardised variance, but the variance of the raw scores showed that father helping with homework accounted for 1.2% (adjusted R square), sex of the student accounted for another 1.1% (adjusted R square) and school type accounted for another 1.0% (adjusted R square). Focusing on the CEE scores, it must be remembered that only those children who were allocated a secondary school place provided a score - this in itself established these children as an elite group. The only significant contribution to the CEE variance was type of secondary school attended, and this contributed a large 29.3% (adjusted R square) of the variance. In surveying the range of regression results for secondary schools, a picture of the successful child shows one who is placed in the high attaining secondary schools, who receives help from father and is female.

SUMMARY POINTS:

1. The survey confirmed the drop-off in student numbers between primary and secondary schools that would be expected in St Vincent, but showed that the drop-off was more dramatic among males than females. Economic constraints, such as those found in Sub-Saharan Africa, did not preclude females from participating in education through secondary schooling and having high educational ambitions.

2. Girls generally performed at higher attainment levels throughout the survey, although the differences between their attainments and male attainments were not as dramatic as those reported from the Barbados survey.

3. Attainment results were highly stratified by parental (both paternal and maternal) occupation, attendance of pre-school and type of secondary school attended; this was seen most clearly in the secondary school regression.

4. Children who lived with both mother and father were most likely to attain at the highest levels. High attainers were most likely to receive help with school work from their fathers. Alternately, we can speculate about the existence of a vicious circle of failure among males. Males are most likely to drop-out of school, and often attained at low levels. These males did not have the support of living with both parents and low attainers are unlikely to receive help with school work from fathers. School disaffection and drop-out is likely to be repeated within the families of these males who are unable to achieve the high occupational positions and educational support correlated with educational success for their children.