DEFINITIONS: POTENTIAL AND ACHIEVEMENT

The very able are defined here as those who either demonstrate exceptionally high-level performance, whether across a range of endeavours or in a limited field, or those whose potential for excellence has not yet been recognised by either tests or experts. There is a distinction between the recognised gifts of children and those of adolescents and adults. The children's are usually seen in precociousness in comparison with others of the same age, and the adults' in productions based on many years of dedication to the chosen domain.

Aptitudes may range across different areas, such as intellectual, artistic, creative, physical and social, or be limited to one or two. But whatever the potential, it can only develop into exceptionally high achievement in circumstances which are rich in the appropriate material and psychological learning opportunities. Arguments about precise definitions and the identification of such children have been active for nearly a century, and will doubtless continue. However, it is educationally more productive (and more scientific) to look at achievements in terms of the dynamic interaction between individuals and their opportunities for learning throughout life. Children with potential for outstanding performance may need educational provision which is not provided by non-specialist schools.

In this report, the terms 'very able' or 'high ability' are used, as well as that troublesome word 'gifted', with its implications of gifts bestowed intact from on high. The term may be further modified with adjectives: moderately gifted, very gifted, highly gifted, profoundly gifted, seriously gifted, average gifted etc., suggesting the possibility of precise identification along a single spectrum of abilities, usually IQ. It is also claimed, without scientific evidence, that at the highest (or most profound) levels, such children are likely to be warped in personality and emotional development, and so should be treated differently from 'average' gifted children. Human abilities, though, however advanced in childhood, are always relative and inevitably change during life (Arnold & Subotnik, 1994).

Nevertheless, as almost all international researchers use the term gifted it would be verging on the deviant to avoid it. Genius is a description usually saved for adults, referring to the few who have made an innovative and lasting impact on a global scale, typically Albert Einstein, Marie Curie or Pablo Picasso. In the United States, though, it has been applied to small children (e.g. Terman, 1925-1929), although none of the sample of 1,572 actually became adult geniuses (Holahan & Sears, 1995).

Distinguishing high-level potential as distinct from measurable achievement is particularly difficult because the true potential of very able children, who are 'merely' working above average level, is easily missed. Research shows that the very able are not a homogeneous group, whether in terms of learning style, creativity, speed of development, personality or social behaviour. Consequently, there are perhaps 100 definitions of 'giftedness' around, almost all of which refer to children's precocity, either in psychological constructs, such as intelligence and creativity, but more usually in terms of high marks in school subjects (Hany, 1993). In formal school education, for example, social or business talents are rarely considered, and physical and artistic prowess are frequently seen as inborn potential which can be developed to excellence by coaching and practice. To some extent, the way a very able child is defined depends on what is being looked for, whether it is academic excellence for formal education, innovation for business, or solving paper-and-pencil puzzles for an IQ club.

Does giftedness last?

Evidence from follow-up studies shows that high test-scores or marks in school are not a reliable indicator of adult careers, except for those who continue in a similar path, such as teachers and academics (Freeman, 1991; Subotnik et al, 1993; Holahan & Sears, 1995). Using children's precocity as the prime identifying feature of high ability is probably responsible for its later apparent loss, often called 'burn out', which is usually due to the others catching up.

The Goertzels' (1978) study of 317 eminent adults found that two-thirds of them were not in any way precocious, and Gardner's (1993, 1997) case-studies of 11 world-changers found that even by the age of 20 only Picasso's future world status was apparent. Schools are effective, though: they can change children's perceptions about themselves, with notable effects on their lifelong achievements, to the extent that the earlier the child starts in formal education the more likely that child is to reach a high level of school achievement (Sylva, 1994). But overall, schools appear to have relatively less effect on the fulfilment of gifted potential than that of pupils of more average ability, possibly because they do not, on the whole, focus on the development of the pupils' special gifts (Subotnik et al 1993).

The major difficulty with follow-up studies of gifted children is that almost all begin with children chosen by extremely high scores on tests, usually either of IQ or school attainment. This restricts the range of abilities and achievements available for investigation, and so affects the reliability of any predictions. However, a unique study in California began with 130 one-year-olds of unknown potential, the only criterion being that they were healthy (Gottfried et al, 1994). Various measures of intellectual, physical and social development were made regularly until they were 9 years old. Those with an IQ of 130 or more on the Wechsler Intelligence test were designated gifted and compared with the others. The researchers concluded that giftedness is a developmental phenomenon, which can rise and fall over time: 'late bloomers' do exist and can be missed in a single testing. A rich educational environment is essential to develop intrinsic motivation, curiosity and love of learning, which Renzulli (1995) calls 'task commitment'.

Predicting adult excellence

In a survey of the research on prediction, Trost (1993) calculated that less than half of "what makes excellence" can be accounted for by measurements and observations in childhood. He found intelligence and other cognitive factors to be the most reliable indicators, but, given a high level of aptitude, the key to success lies in the individual's dedication.

Follow-up studies of gifted children began with Lewis Terman (1925-29) who by 1928 had selected 856 boy and 672 girl "geniuses", almost all from California, eventually producing more than 4,000 variables. There were, however, considerable flaws of procedure, e.g. "no private, parochial (religious) or Chinese schools" (Holahan & Sears, 1995, p. 11). The subjects, aged between 2 and 22, were mostly children of white university staff, sometimes of Terman's colleagues. The sample was collected over a period of 7 years. The children were first selected by teachers then tested, those with IQ 135 plus being taken into the sample. The general population was used for comparison. But even by 1928 a quarter of this sample of 'Termites' (as they came to be known) had vanished from the study and were simply replaced, and leavers continued to be replaced for years. Terman's "geniuses" were considerably above-average in every way, including height and leadership qualities, but then they did have above-average nourishment and education. The latest review of the 'Termites' in their seventies and eighties shows that they have not been noticeably more successful in adulthood than if they had been randomly selected from the same social and economic backgrounds regardless of their IQ scores (Holahan & Sears, 1995). Perhaps because of their relatively privileged upbringing, they had an exceptionally high level of self-confidence and the creativity to make plans and stick to them: they have lived longer and more healthily, and are still more actively involved than the general population.

In spite of the doubts about the research design, the Terman results are in accord with more scientific longitudinal work by Schaie (1996), who found that intellectual and perceptual abilities remain high for individuals who are active and open-minded, but in particular, those who feel satisfied with their life's accomplishment in mid-life are at a considerable advantage as they age. The Munich Longitudinal Study of Giftedness began in 1985 (Perleth & Heller, 1994). It has a sample of 26,000 children, identified on a wide variety of intellectual, personality and achievement tests, although personal interactions with the children have only recently been introduced. The team devised 30 identification scales, which disclosed a significant number of underachievers, who were characteristically found to be anxious, easily distracted and to have poor self-esteem.

In a review of 14 American and German follow-up studies of varying design and loss of subjects over time, Arnold and Subotnik (1994) pointed to several important factors in the conditions for the development of talent. They suggest an "inextricable link" between the identification of potential and timing due to age-related stages of development, so that accuracy in predicting achievement increases with the age of the sample studied. The child's own interests appear to be an excellent and often neglected indicator of adult attainment (Renzulli, 1995: Hany, 1996; Milgram & Hong, 1997). Accordingly, for the greatest reliability, information should be collected at different points in an individual's life, most reliably within specific subject areas in which the child shows promise and interest.

A 15-year follow-up of 82 'valedictorians' (the highest grade earners in high school) from 32 schools across Illinois showed that even such exceptional grades were not good long-term predictors of later high achievement (Arnold, 1995). The research was thorough, with each individual being given five or six interviews after leaving school. They had enjoyed all aspects of school, which they found to be a supportive social environment, and had used it efficiently to prepare for their future lives. Their major academic advantage was in their determination to better themselves. Neither boys nor girls felt themselves to be outstandingly clever nor had they been labelled as such; in fact, during their college years, the women continually lowered their estimates of their intelligence. None of this sample made outstanding progress in their careers (particularly the women), and at 26 years old many were disillusioned.

The long-term benefits of early special provision for the gifted are still far from certain. In spite of an initial higher measured achievement, the advantage tends to disappear over a few years (White, 1992). But these discouraging results have mostly been based on the old-style pattern of IQ-identified gifted children being given the only provision available at that time. Much knowledge has been accumulating and the future of education for the very able is undergoing major changes in outlook, which may well bring improved results for more children.

WAYS OF IDENTIFYING THE VERY ABLE

Each method of identifying the very able distinguishes a somewhat different group of children, with possibly different consequences for their self-concepts and education. For example, because a high IQ score is to some extent a measure of school-type achievement (see below) it is good at picking up children who will continue to do well at school, but it is not very helpful in identifying sporting prowess, which is decided by experts. The problem with any form of selection for special provision is that if promotion of the skills and talents needed by society is limited to an elite, however skilfully that elite is selected, there will inevitably be individuals whose potential contributions remain undeveloped or who must find other routes to fulfilment. And, because it is not possible to predict the kinds of talents that will be needed in the future, there has to be a wide variety of skills and outlooks available in any society. Research is presented here on the four most frequent methods of identification: 1 Intelligence; 2 Teacher recommendation; 3 Parent recommendation; and 4 Peer nomination.

1 Intelligence

Very high intelligence, as measured by IQ tests, is by far the most popular criterion for defining children as very able or gifted. However, after more than a century of use, this procedure is still highly contentious. Just one problem is that IQ testing is strongly influenced by belief systems which include social and moral values. An example is in the Stanford-Binet Intelligence Scale in the question "What's the thing to do if another boy/girl hits you without meaning to do it?". The correct response must involve forgiveness.

The concept of intelligence is changing

A wider and generally accepted view of intelligence is that it is an individual way of organising and using knowledge, which is dependent on the social and physical environment. Consequently, conventional methods of measurement, notably the IQ test reflecting the old ideas of relatively fixed capacities, are being replaced by measures which aim to distinguish the many components of intelligence, so that they can be presented as profiles of ability, such as in the British Ability Scales. Other standardised tests are available in specific areas, such as for musical or science aptitudes.

There is also growing influence of the Theory of Multiple Intelligences conceived by Howard Gardner at Harvard (Gardner, 1983). He argues that human abilities are best considered as at least seven distinct intelligences, only two of which, linguistic and logical- mathematical, fall within the usual definitions of intelligence. The other five, spatial (chess, painting etc), musical (playing or appreciation), bodily-kinaesthetic (sports and gymnastics), interpersonal (social skills) and intrapersonal (self-awareness), have usually been regarded as talents. Gardner has recently added to this list with naturalist intelligence (knowledge of the living world), spiritual intelligence (cosmic issues) and existential intelligence (ultimate issues). However, the evidence for their distinctiveness has not been replicated by other researchers, who usually find that these 'intelligences' overlap. Schools have now been set up in the USA where the children are taught via these intelligences, and in Britain the University of the First Age in Birmingham, which helps underachieving children, is basing its enrichment provision on these ideas (UFA, 1996).

In contradiction, Gagné (1995) of Montreal has proposed four aptitude domains (intellectual, creative, socio-affective and physical) and four talent fields (academic, technical, artistic and interpersonal). Their development, he suggests, depends on the learning context, normally the school, which should recognise the child's own efforts.

Sternberg (1993; 1997), at Yale, warns that it is important to be aware of what is culturally valued and what is not. This influences the way one shapes one's activities in everyday life, and must be part of any assessment of intelligence. His Triarchic Theory of Intelligence is concerned with applications of processes, which he says can be used in schools. For example, linguistic aptitude could be channelled into the child's natural preferences - analytic (as in criticism), creative (as in poetry) or practical (making advertising copy). For him, high ability is the development of one's natural aptitudes to an exceptional level. Renzulli (1995) emphasises task commitment, creativity and innate ability as indications of giftedness. Recent work (e.g. Subotnik et al, 1993) has also shown that high ability may take many different forms; it may appear in quite unexpected situations and at different points during a lifetime.

There is considerable evidence for biological differences between the highly able and other children. This implies that there is a limit to how far teaching can bring on a pupil. The IQ scores from studies of more than four hundred sets of identical and non-identical twins separated at birth were investigated in later life (Plomin et al, 1994: Bouchard, 1997). This work has discovered considerable genetic influence on intelligence, at about 70% the strongest correlation found for any psychological characteristic. Such studies have also highlighted environmental influences, notably that the younger the child the more potent the environment. But no specific gene for giftedness has yet been discovered. Lykken et al (1992) concluded from their separated twin research that specific talent appears to be an 'emergenic' trait, one which depends on a particular configuration of genes, so finely balanced that any small difference will result in distinct changes of behaviour. Indeed, we know that constant stimulation of the brain and its responses can change its fine structure and function for life, as shown by the use of visual techniques for helping dyslexics (Bakker, 1990). The sleep rhythms of intellectually gifted children appear to be different from the average; they have more REM (rapid eye movement) sleep (Grubar, 1985; Dujardin et al, 1990). Some psychologists, such as Eysenck (1995), have claimed that intelligence (the ability to learn) can be measured by speed of reaction. This would be a very helpful guide to teachers, if the experimental results were not so unreliable.

IQ tests

Intelligence is usually measured by IQ tests. High ability or giftedness is then determined by a score which is above a chosen cut-off point, usually at around the top 2-5%. But it is important to know which test has been used. The two most popular tests, the Stanford-Binet and the Wechsler Intelligence Scales, measure IQ somewhat differently. The Stanford-Binet is verbally biased while the Wechsler is mathematically biased, and the upper limits of each differ by 20 IQ points, which makes comparisons between scores at their upper ends difficult. But they are not even an adequate measure of intellectual giftedness because of the 'ceiling effect', whereby highest measures of the tests are too low to distinguish satisfactorily between the top few per cent. What is more, some children get every item right and could obviously score more highly were the test to accommodate them. Some psychologists calculate additional IQ points above the test's ceiling, even to beyond IQ 200 (e.g. Gross, 1993, Silverman, 1993), though there is no scientific basis for this. Much current research and theory, though, is indicating that intelligence, however it is defined and measured, is only part of the complex dynamics of exceptionally high-level performance, which must include such matters as self-esteem, support, motivation - and opportunity.

A very close positive relationship was found when children's IQ scores were compared with their rated home educational provision (Freeman, 1991). The higher the children's (Stamford-Binet) IQ scores, especially over IQ130, the better the quality of their educational support, measured in terms of reported verbal interactions and activities with parents, number of books and musical instruments in the home etc. Because IQ tests are decidedly influenced by what the child has learned, they are to some extent measures of current achievement based on age norms. The vocabulary aspect, for example, is dependent on having heard those words.

IQ tests can neither identify the processes of learning and thinking nor predict creativity (Cropley, 1995; Urban, 1995), and because they only measure a narrow band of intellectual behaviour they cannot predict other aspects of life, such as what career a person is likely to follow or how individuals will cope in social situations. In fact, drive and energy have often been found to be relatively more predictive of life success than high IQ in children (e.g. Albert, 1992; Holahan and Sears, 1995; Subotnik et al, 1993; Schaie, 1996). There are other kinds of intelligence tests, however, such as the more culture-fair Raven's Matrices, which do not contain specific learned material, such as words, and instead use pattern matching. Teachers can use these tests with groups, which are useful as a general guide to reasoning ability. They do not result in an IQ score, but offer a percentile (where the score is described as a percentage of all values lower than or equal to it). It must be noted, though, that they have an even lower 'ceiling' than the IQ tests, the upper limit being the 95th percentile, and so they cannot differentiate very well between the top 5% of children.

Nevertheless, in spite of the evidence pointing out the debits of the IQ score as a measure of an individual's all-round ability, it has been proved many times to be a valid and reliable measure of potential for achievement in school, the purpose for which it was designed. When the IQ score is taken into account along with the understanding of the test used and as a part of an assessment it provides a useful and reliable tool for identifying academic ability.

Can gifted performance be learned?

Studies of successful people brought Howe (1990) to the conclusion that "in the right circumstances almost anyone can" ... acquire exceptional skills (p.62). He argues that self-direction, self-confidence, a sense of commitment and persistence can effectively produce gifted performance. Indeed, without a cello, tuition and a family to back him Yo Yo Ma could not have become a great cellist, whatever his talent; but then, if one were to give other children the same provision, would they also turn out to be virtuosi?

Attempts to teach gifts have been carried out in American laboratory studies which for some years have been analysing the specific skills of expertise (Ericsson & Lehman, 1996). However, even in those strictly controlled conditions the trainees differed in the level of expertise they could reach, and the researchers found (as have many others) that motivation and practice (as any teacher knows) make a vast difference to results. The researchers concluded that the most important variable in gaining expertise is sufficient ability to gain a foothold in the learning process, and then to put in thousands of hours of learning and practice. This was also found in Holland ( Elshout, 1995) in a study of expertise in children solving physics problems. They were selected for intelligence (High and Low) and their expertise in physics (Advanced v Novice), resulting in four groups (n = 28). The difference between the Highs and the Lows was much larger in the Novice group than in the Advanced group. Experienced Lows outperformed Highs of less experience, indicating that experience compensated for lower intelligence.

Even for high-level artistic performance it has been said that "formal effortful practice is a principle determinant of musical achievement" (Sloboda et al, 1996, p. 287). In an earlier contradictory statement, though, when reporting work in Britain with school-age talented musicians in Manchester, he found that "the best students had done less formal practice in their early years than had the average students" (Sloboda, 1993, p.110). Rather, he found that the best students had received more praise then the others, and their parents had made them feel 'special'.

Cultural and family attitudes have a considerable effect on high-level achievement. For example, Berry (1990) found highly significant geographical and religious differences between Nobel prize-winners. In proportion to their numbers, Jews were heavily over-represented, in certain subject areas 50 times more. Zuckerman (1987) suggested that as 75% of Jewish Laureates came from lower socio-economic backgrounds, it could not have been social advantage which produced that excellence, but rather, in line with other research, the cultural influence of the family's drive for success. Indeed, in their late adulthood, the most successful of the Terman sample were distinguishable neither by IQ nor by earlier school achievement but by family background, notably the aim for success (Holahan & Sears, 1995). In many Pacific Rim countries, as well as Russia, measured intelligence is largely ignored and success is attributed to sheer effort, hence the growth of out-of-school crammers. His investigations led Flynn (1991) in the US to conclude that the culture of hard work is probably responsible for so many Asian-American (usually meaning Pacific Rim) youngsters' greater school and work success than their higher IQ classmates. In fact, Hess & Azuma's (1991) in-depth research showed that American children needed much more help and praise than Japanese children in their motivation to learn.

2 Teacher recommendation

Naturally, teachers' judgements of their pupils' abilities affect their expectations and treatment of them. This can be seen in the organisation of learning groups and selection for examinations, which will in turn affect the pupils' attitudes to education. When pupils move to more challenging work, both teacher and pupil expectations are increased (Good, 1996). We do not know, though, how any teacher's personal conceptions or stereotypes of giftedness affect his or her actual teaching.

There can be wide variation between teacher judgements and objective measures. Individually, teachers' attitudes towards the very able vary greatly; some feel resentment while others overestimate their all-round abilities, as was found in a Finnish-British survey (Ojanen & Freeman, 1994). But teachers have been found to judge the highly able consistently, in that they will continue to pick the same kind of children (Hany, 1993). In England, Bennett et al (1984, p.215), found that 40% of potentially high achievers had been underestimated by their teachers, and Nebesnuik (from Eyre 1997, p. 17) "showed a significant discrepancy between the assessment of able pupils by their primary teachers and subsequently by year 7 teachers" [pupils aged about 11-12]: the primary teachers often chose children as able by their ways of working rather then their cognitive ability. Using IQ as his criterion of giftedness, Tempest (1974) in Southport, UK, found that out of 72 six year-olds identified by their teachers as gifted, only 24 had IQs of 127 or above and seven had IQs of under 110. But two children with reading ages six years in advance were not nominated as gifted. He concluded that teacher recommendation alone was not reliable.

Hany (1997), in Germany, investigated how 58 secondary school teachers judged giftedness, using a rating scale of 60 suggested traits. The teachers were seen to be biased in their judgements in certain respects. They did not fully consider the basis of comparisons or non-obvious characteristics, in that they would choose pupils who were most like their expectations. Creativity was not often seen as an aspect of giftedness, and emotionally the gifted were often expected to be playful, arrogant, uncontrolled and even disturbed. The teachers often kept a mental image of a gifted pupil who would have exceptionally good logical reasoning, quick comprehension and intellectual curiosity in combination with good school grades. Individual gifted pupils were often vividly remembered by teachers, who would use those characteristics to identify others. But when teachers were offered guidance by the researchers on how to identify the gifted in their own subjects they found it very helpful, particularly because of their day-to-day knowledge of the pupils.

How teachers perceive and thus identify the gifted has been seen to vary considerably between different cultures. The estimations of more than 400 secondary teachers in Germany and 400 in the USA were compared with 159 in Indonesia. The German teachers estimated 3.5% of children as gifted, the Americans 6.4% and the Indonesians 17.4% (Dahme, 1996). Even within one country, the USA, percentages of the child population identified as gifted vary between 5% and 10% across the states, although most now have legislation and financial support for 'gifted education' (OERI, 1993). It is to be expected that the definitions and special facilities provided by educational authorities would have some effect on teachers' choices.

In an investigation into mathematics teachers' attitudes towards able pupils in 500 English comprehensive schools, the researchers presented them with suggested lists of the characteristics of mathematically able children (Chyriwsky & Kennard, 1997). The teachers generally agreed with these, but also expressed concern that such pupils were both educationally underchallenged and frequently deterred by peer-group pressure. The teachers often felt hindered by constraints on time and material resources in teaching bright pupils, and that any available extra provision was targeted towards the least able.

Research in Oxford used experimental checklists for teachers in 11 comprehensives to identify the top 10% of third-year pupils (year 9) in their own areas of expertise: physics, mathematics, English and French (Denton & Postlethwaite, 1985). However, these check-lists were up to 22 items long, which teachers found off-putting, and this resulted in a low response rate. Further, the research did not make use of objective criteria, such as standardised test scores or the pupils' eventual O-level results. Instead, only the teacher-set classroom tests were used, which, as they were marked by the same teachers, could be self-fulfilling. The best items for accurate teacher prediction of test marks made use of the more obvious characteristics, such as verbal ability for English tests, and mathematics for mathematics tests. In French and physics, clear errors of teacher judgement were found, more noticeable at the start of the school year than at the end when the pupils were better known to the teachers. Both this Oxford and the German studies (Hany, 1995) were in accord that teachers mostly judged high potential by general rather than specific abilities. Both, though, found that high abilities were subject-specific, and suggested that this focus was the most likely to produce the most accurate predictions.

In Russia, teachers' opinions, intelligence tests and a creativity test were used to select 73 children aged 7-8 as gifted (Sheblanova, 1996). Both this group and a random control group of 76 children were tested yearly for three years and their results compared. Large intellectual differences were found between the groups, although only 54-60% of teachers' ratings had predicted these. The teachers were good at selecting the high-level achievers, but had difficulty in identifying high-level intellectual and creative ability. Interestingly, although there were no differences in scores between the sexes on intelligence or creativity, the Russian teachers still considered the girls to be more gifted, whereas in Israel, using check- lists, teachers chose two boys for each girl (Shahal, 1995).

A multiple criteria identification procedure for primary school selection of the very able was devised in Jordan (Subhi, 1997). The research involved 25 schools with 4,583 third-graders (8-9 year-olds) of whom 217 (4.74%) were identified in a wide trawl using standardised tests for intelligence, creativity, achievement, mathematical skills and task commitment (all translated into Arabic), as well as peer, self, teacher and parent nomination. Teachers' nominations were found to have missed 50 pupils with IQs above the cut-off point of IQ 130. The highest IQ children were the least likely to nominate themselves. But in the peer-nominations more boys than girls were nominated, even when the girls had equally high IQs. This is the case not only in the Middle East, but in Midwestern USA (Peterson & Margolin, 1997). There, teachers from two middle schools were asked to recommend children for a gifted programme. The teachers appeared to treat "giftedness" as if it were an absolute and universally agreed set of characteristics, and were sure of their decisions. But Latino children, and those from other minority groups, were found to have been passed over.

The above research on teacher nomination of very able pupils is pertinent to the Revised National Curriculum in Britain, which allows some flexibility in deciding at what depth to teach, although the possibility of extension tests to assess more intense teaching has been abandoned for the stated reason that: "the cost of extension papers was not justified by their use ... most [simply] drew on the programmes of study for the next key stage." (DfEE/SCAA, p.11). Consequently, as Standard Attainment Tasks (British SATs are also known as 'end-of-key-stage-tests' and used for the 4 key age-related stages) only measure the three core subjects of English, maths and science: "The Secretary of State proposes to place sole reliance on teacher assessment [my italics] for the award of levels to exceptionally able pupils" (DfEE 1996, p.4). However, teachers are not being given any help in assessing their most able pupils - with all the potential problems of misjudgement outlined above.

 

Research-based methods by which teachers can recognise very able pupils Use the outcomes of particular tasks set for pupils, rather than simply test scores or work-books. In smaller groups, recognition can be helped by subject-orientated discussion. Children's abilities change over time; some may develop later than expected while others may show a sudden drop in achievement. Because human memory is unreliable and easily affected by current impressions, teachers should keep careful records in checklist format. Focus on particular aptitudes, rather than apparent general intelligence or positive attitudes to school work. Personal interaction between teacher and pupil is important to find non-obvious talents. A second observer from time to time is a help. Check that choices are not biased by social class, ethnicity, gender or handicap etc. Watch for bias ­ make sure that all pupils have access to the facilities available. Recognise pupil efforts and techniques which attempt to overcome handicaps. Use as many assessment procedures as possible. Watch out for motivation and interests as clues to potential. Consider more pupils than are immediately obvious. Positively seek variety ­ look through the range of subject areas for pupils with diverse talents. Take more than one opinion, perhaps a group decision; but remember that groups can be dominated by strong characters and that a majority decision may not always be right. Lively activity outside school and little interest inside school can indicate gifted underachievers. Consult the pupils themselves.

Checklists

Checklists of the supposed characteristics of highly able children vary considerably, and some of the items can be confusing. Some, rather than being specific to aptitudes, may be socio-cultural. For example, a child asking a lot of questions can either be seen as gifted or as attention-seeking, or perhaps lives in a home where questioning is encouraged rather then one where children are encouraged to work things out for themselves.

One list may ask the teacher to look out for dedicated seriousness, while another suggests a keen sense of humour: although in a rare survey of experimental work, humour has not been found to have any relationship with creativity or intelligence (Galloway, 1994). While one list may point to a tendency to perfectionism (and thus procrastination), another will describe the highly intelligent as speedy decision-makers. Some see introversion as typical features of the high IQ child, although there is no reliable evidence of any personality features being associated with IQ.

American lists, in particular, sometimes suggest a higher level of morality and leadership in the gifted, for both of which there is little evidence when social class, home support etcetera are recognised. Indeed, in Finland, Merenheimo (1991) showed that "self-confidence and moral development of students aged 13-16 was firmly anchored to the influences of microcultures" (p.116). Although there is a positive statistical relationship between scores in tests of IQ and tests on morality (such as Kohlberg's Moral Development), a survey of international research by Adreani & Pagnin (1993) could not find any recognisable relationship between IQ and actual behaviour. In another survey of research into moral giftedness, the American, Rothman (1992) pointed out that "IQ explains but little in the development of moral reasoning" (p.330). He suggested that the higher scores of high-IQ children on moral reasoning tests result from their enhanced social interaction with adults. It is as though the intellectually gifted know what they should answer on the tests, but may or may not choose to abide by their answers.

Checklists (even official ones) can mislead. For example, one British LEA publishes this list (given here in full) of the characteristics of the able child, which mostly asks the teacher to look for signs of emotional distress, which is quite contrary to the evidence (see page 27).

• "Prefers friendship with older pupils or adults.

• Excessively self-critical.

• Unable to make good relations with peer groups and teachers.

• Emotionally unstable.

• Low self-esteem, withdrawn and sometimes aggressive."

(NIAS, 1994, p. 15, quoting George, 1992)

Another misleading characteristic often presented in checklists is poor-quality sleep, which has no scientific support. In fact, Terman reported that the most gifted in his sample slept longer and more soundly than the others. When Browder in Germany (in Perleth, 1993) managed to compare the sleep of children whose giftedness was unknown to their parents with non-gifted children, there were no differences in pattern or quality. Freeman (1991) found that children's length of sleep, as reported by parents, was directly related to their age, and not to intelligence or achievement. However, the children who were seen as gifted by their parents were described by them as sleeping badly and having more emotional problems than children - of equal measured ability- whose parents did not see them as gifted. Much of the stereotype of gifted children with sleep problems appears to come from parents.

The best that can be said of checklists is that they may stimulate teachers to think about the identification of the very able: the worst is that potentially high-achieving children who do not fit with the opinions of those who devise the lists will be missed. The most reliable research-based criteria which distinguish the highly able and could reasonably be used in a checklist are listed here (From Shore in Montgomery, 1996; Freeman1991).

 

A research-based checklist for very able pupils Memory and knowledge ­ excellent memory and use of information. Self-regulation ­ they know how they learn best and can monitor their learning. Speed of thought ­ they may spend longer on planning but then reach decisions more speedily. Dealing with problems ­ they add to the information, spot what is irrelevant and get to the essentials more quickly. Flexibility ­ although their thinking is usually more organised than other children's, they can see and adopt alternative solutions to learning and problem-solving. Preference for complexity ­ they tend to make games and tasks more complex to increase interest. They have an exceptional ability to concentrate at will and for long periods of time from a very early age. Early symbolic activity ­ they may speak, read and write very early.