3.1 The concept of giftedness is so complex that it is difficult to arrive at a single satisfactory definition.1 In this chapter, the Committee has reviewed the most influential of the definitions of giftedness. These are important because the definition will influence the procedures used to identify gifted children.
3.2 The many definitions of giftedness which have been advanced differ in the emphasis placed on particular characteristics and behaviour of children. According to Dr Murray Print, Senior Lecturer in Education of the Western Australian College of Advanced Education, the various definitions of giftedness range from specific, precise, hard data definitions based on percentage scores or IQs to vague, generally-worded concepts emphasising student behaviour or even potential ability.2
3.3 A precise or 'unitary' definition of giftedness based on intelligence as measured by an individually administered test of general ability was developed by Lewis Terman. Terman defined the gifted as the top one per cent of the population in general intellectual ability as measured by the Stanford-Binet Intelligence Scale or a comparable instrument. The definition of giftedness which emerged from the work of Terman in the 1920s identified gifted children as those who excel in all areas of development and who are motivated, self-reliant and obtain high scores on intelligence and achievement tests.3
3.4 Under this traditional approach, it was necessary to decide only which intelligence quotient separated the gifted from the non-gifted. If gifted children could be identified reliably by tests of general intellectual ability, it followed that special educational provisions could be made to cater for the most able few per cent of the school population. For example, in 1932 and for many years after, tests of general ability were used to choose a group of children in New South Wales government metropolitan primary schools to be placed in special 'OC' or opportunity classes. The group consisted of those who scored in the top 2.5 per cent on those tests.
3.5 Some witnesses criticised the sole use of tests of general intellectual ability to define and therefore identify gifted children. The New South Wales Federation of Parents and Citizens Associations observed that a definition which regards giftedness as being confined to 'one or two per cent of the student population who demonstrate exceptional intellectual and academic ability' is not only erroneous but dangerous and should be discredited.4 It has also been argued that the traditional view of giftedness leads to an approach which uses arbitrary cut-off points, operates on a selective basis and assumes that giftedness is static and unvarying over time.5
3.6 Tests of general intellectual ability have been criticised as being biased in favour of white middle-class children and against poorer children from lower social class and children from minority racial or ethnic groups. It is claimed that children from the latter groups are not able to score as well as white middle-class children of comparable intelligence because of differences in language ability and cultural values.6 Tests of general intellectual ability have also been found to miss the under-achieving but bright child.7
3.7 Although research has shown that tests of general intellectual ability have limitations such as those described in paragraph 3.6, they have a role in identifying gifted children. Miriam Goldberg, Professor of Psychology and Education, Teachers College, Colombia University, was invited by the Commonwealth Schools Commission in 1978 and again in 1980 to survey provisions for the education of gifted children in Australia. In her report, Issues in the Education of Gifted and Talented Children in Australia and the United States, Professor Goldberg remarked that 'the evidence is overwhelming that the majority of those who, as children, achieved high scores on individually administered intelligence tests far more often than their lower scoring peers, achieve excellence in some intellectual or related field of endeavour'.8
3.8 The results obtained from tests of general intellectual ability are still widely used to define giftedness. Legislation in some States of the United States of America specifies IQ cut-off points for the purposes of funding of educational programs for gifted children. Children in Pennsylvania, for example, must be identified as having an intelligence quotient of 130 or above after completing an individually administered intelligence test to be eligible for special programs for gifted children.9
3.9 During the 1940s and 1950s, educators and psychologists began to question the validity of tests of general intellectual ability and, in particular, 'their power to choose in a valid and reliable way the children who are gifted'.10 This questioning was a result of further research on the various abilities which constitute 'intelligence'. For example, Professor J.P. Guilford, an eminent American psychologist, in his work on the 'Structure of the Intellect' proposed a model of intelligence which comprised 150 different cognitive abilities. Guilford's model included an array of creative talents and conceptional thinking not measured by tests of general intellectual ability. His work suggested that the traditional 'intelligence' test assessed only the ability to think convergently, that is, to deduce a single answer that is either right or wrong. Guilford proposed that tests should be constructed to assess potential for creative or productive accomplishment. These tests should also assess the ability to suggest many different potential answers, that is, the ability to think divergently.11
3.10 Many educational organisations, including the British Schools Council, the United States Office of Education, the New South Wales Department of Education and the Commonwealth Schools Commission, have advanced multi-faceted definitions of giftedness which incorporate concepts such as creative and divergent thinking, originality, sensitivity and improvisation.
3.11 The British Schools Council developed a widely-accepted definition of giftedness in 1971. The Council defined a gifted child as one 'who is outstanding in either general or specific ability in a relatively broad or narrow field of endeavour'.12 The Schools Council report of 1971 suggested that a useful cut-off point for intellectual giftedness is the top two or six per cent of the School population.13
3.12 In the United States, a multi-faceted definition of giftedness was adopted in the Marland Report of 1972. The report defined gifted and talented children in the following way:
Gifted and talented children are those identified by professionally qualified persons who by virtue of outstanding abilities, are capable of high performance. These are children who require differentiated educational programs and/or services beyond those normally provided by the regular school program in order to realize their contribution to self and society.14
The Marland Report further defined 'children capable of high performance' as including:
Those who have demonstrated any of the following abilities or aptitudes, singly or in combination:
(1) general intellectual ability,
(2) specific academic aptitude,
(3) creative or productive thinking,
(4) leadership ability,
(5) visual and performing arts aptitude,
(6) psychomotor ability.15
The Report's concept of giftedness applied 'to a minimum of three to five per cent of the school population'.16
3.13 The definition proposed in the Marland Report has be adopted in its entirety or in part by numerous education systems in the United States and elsewhere. By including a wide variety of abilities in the definition of giftedness, the Marland Report's definition has helped to defuse criticisms that the selection of gifted children is elitist. The definition, however, has also presented some problems.
3.14 Firstly, it refers to imprecise concepts such as 'creative thinking', 'psychomotor ability' and 'leadership ability'. What are creativity and leadership and how are they reliably measured? Does psychomotor ability merely encompass athletic or gymnastic prowess or does it also include fine motor co-ordination or talent in dance, which is usually included in the performing arts? Perhaps because of these difficulties, 'psychomotor ability' was not included in the definition of giftedness incorporated into United States legislation in 1978.17
3.15 Secondly, the definition used in the Marland Report does not include motivation or task commitment as an element of giftedness. Critics of the definition argue that one of the key factors characterising the work of gifted persons is the ability to be involved totally in a problem or area for an extended period of time.18 Thirdly, some researchers have suggested that the definition tends to be misinterpreted and misused because educators treat each of the six areas of abilities as individual independent categories and ignore the inter-relationships among the categories.19
3.16 Dr Braggett informed the Committee that in various countries, including Australia, the concept of giftedness has broadened since the Marland Report.20 This is reflected in the definitions used by State education authorities in Australia. These definitions may be found in Appendix 3. In 1977, for example, the Macdonald Report of the New South Wales Education Department described gifted children in the following terms:
Talented students are an atypical group possessing exceptional abilities and capable of outstanding performances.21
In 1980 the Commonwealth Schools Commission advanced another definition which also owed much to that used in the Marland Report. The Commission concluded:
Gifted students are those possessing to an outstanding degree demonstrated competence or potential in intellectual, creative and/or other abilities and needing different education or services beyond those provided by the regular schools.22
3.17 Multi-faceted definitions have expanded the concept of giftedness, but have introduced concepts which are difficult to measure objectively. As the definitions of giftedness move from the precise and 'conservative' to the imprecise and 'liberal' there is less emphasis on objective measurement of performance and potential and more reliance on the judgement of individuals.23 'Liberal' definitions introduce value judgements and the problem of subjectivity in measurement. For example, how can 'potentially valuable lines of human endeavour' be assessed objectively?
3.18 Several of the multi-faceted definitions of giftedness refer to both demonstrated ability and potential ability. The definitions seek to include children whose exceptional abilities have not been developed. Professor Goldberg has observed that under-achieving children should be considered in terms of their potential, rather than their demonstrated performance. The difficulty with this approach is one of assessment, The solutions proposed to assess potential in many fields are less than satisfactory. Professor Goldberg considered that in all cases 'valid means of assessing potential abilities are simply not available'.24
3.19 A model of giftedness that has been extensively used in Australia was devised by Joseph Renzulli, Professor of Educational Psychology, School of Education, at the University of Connecticut. Renzulli suggests that there are three basic traits which characterise successful, outstanding individuals. These traits are:
3.20 Renzulli considers that gifted children are those possessing or capable of developing these traits and applying them to any potentially valuable area of human performance. The possession of only one set of traits does not imply giftedness. Rather, it is the interaction among the three which provides the necessary ingredient for creative or productive accomplishment. Each set of traits is an equal partner in contributing to giftedness. Superior ability without the spark of creativity or the will to persevere is unlikely to provide a high level of performance in any area.26
3.21 The first trait identified by Renzulli is 'above average ability'. Renzulli suggests that above average ability is not necessarily equated with high intelligence quotients. He considers that the most productive people are not always those who score in the top five per cent on tests of general ability. He does not, however, completely reject tests of general intellectual ability for determining above-average ability. He suggests that intelligence quotients should be seen in perspective. In Renzulli's model, intelligence or academic potential tests should be used only for initial screening purposes or to establish minimum performance levels. Renzulli advocates greater use of indicators of creative thinking, ratings of past accomplishments and ratings of creative production.
3.22 The second trait that Renzulli Claims characterises gifted children is 'task commitment'. This is defined as 'energy brought to bear on a particular problem (task) or specific performance area'.27 This notion of task commitment is perhaps best illustrated by the maxim that 'genius is one per cent inspiration, and ninety-nine per cent perspiration'. The third trait is creativity. The emphasis here is on original thinking, ingenuity and divergent thinking. Renzulli considers that creativity should be included as an integral feature of giftedness. He is concerned, however, that the identification of creative potential raises the 'haunting issue of subjectivity in measurement'.28
3.24 The term 'identification' is generally applied to. procedures used to screen gifted students from among the wider population or to select them for specialised programs. Such procedures can be broadly classified as 'objective' or 'subjective'. Although not all identification procedures are clearly 'subjective' or 'objective', the classifications are useful for discussing the procedures most commonly used.29 It should be noted that 'objectivity' has a specialised meaning in testing and should not be considered the opposite of 'subjectivity'.30
3.25 Objective procedures include group and individual intelligence tests, achievement tests, specific aptitude tests and creativity tests. Subjective procedures may include teacher nomination, parent nomination, peer nomination and self-nomination.31
3.26 Screening procedures used to identify gifted children in Australian schools usually involve the use of a test of general intellectual ability. Some tests may be given to a group whilst others are used individually. Group tests are relatively inexpensive and efficient to administer, whereas individually administered tests are more costly. Group tests require only specific and limited professional input. Individual tests must be administered and assessed by specially trained personnel.
3.27 Research indicates that group tests of general intellectual ability may not be reliable for the identification of gifted children. For example, in 1971, the United States Office of Education (USOE) found that as many as fifty per cent of all gifted children may remain unidentified if group tests alone are used. The USOE also found that in some cases, individual ratings on group tests tend to be higher for below-average students and lower for above-average students.32
3.28 Group tests have also been criticised because they neglect the thinking processes exercised in arriving at answers and because they measure convergent thinking skills only. Moreover, whilst group tests provide a measure of general intellectual ability, they give little information about specific talents and abilities. Group tests may also fail to recognise gifted students with reading difficulties, emotional or motivational problems or those who are culturally impoverished.33
3.29 Individually administered tests of general intellectual ability are more accurate than group tests in identifying the gifted.34 Scores obtained on individually administered tests were used for many years as proof of giftedness, with the cut-off point being usually two standard deviations above the mean. That is, a gifted child was one who had an IQ of 132+ as measured by the Stanford-Binet Intelligence Scale or an IQ of 130+ as measured by the Wechsler tests.35
3.30 Binet-type tests of general intellectual ability have a number of distinguishing features. The tests yield a general global measure of intelligence rather than an analysis of separate special abilities. Questions and tasks are grouped on the basis of difficulty, from easy to difficult, equating with age norms. Scoring is therefore tied to age norms.
3.31 Another test of general intellectual ability used to identify gifted children is the Wechsler Intelligence Scale for Children - Revised (WISC-R). In this test, questions and tasks are grouped into sub-tests. The results obtained on these sub-tests provide useful information concerning the different areas of mental functioning of the person being tested. Each of the Wechsler sub-tests has separate norms. A score on each sub-test may therefore be compared with the score for a group that is representative of the whole population. Scores on certain sub-tests may be totalled to give an overall score on a verbal scale. Similarly, the total score for the other sub-tests provides an overall score on a performance (non-verbal) scale. The total score on all sub-tests denotes a level of general mental ability and it is from this total that an overall intelligence quotient is obtained.
3.32 As mentioned in paragraphs 3.5 and 3.6, there has been much criticism of the use of tests of general intellectual ability for identifying gifted children. The Committee has referred earlier to the verbal and cultural biases of most tests and their inadequacies for measuring divergent thinking abilities or motivation. As indicated in paragraph 3.9, Guilford's work on the structure of the intellect recognised the importance of divergent thinking and the need to assess thinking processes other than those normally recognised by traditional intelligence tests. Divergent thinking generates a variety of Solutions to problems that lend themselves to multiple solutions. Tests such as the Torrance Tests of Creative Thinking have been devised to measure this ability.
3.33 Creativity tests are a relatively new development. Their use remains somewhat controversial because of problems regarding definition, validity and reliability. When creativity tests are used for identifying giftedness, information gathered by these tests is therefore generally supplemented by information from other sources.
3.34 Achievement tests are designed to assess how well a student has learned a particular content area or subject matter. Aptitude tests are designed to indicate whether a person has the ability to achieve in certain areas Such as music or mathematics. The best known example of an aptitude test in Australia is the Australian Scholastic Aptitude Test (ASAT). The ASAT is administered to Year 12 students in most Australian schools and is used to modify the marks gained by these students on tests of achievement. The test is intended to give some indication of a student's ability to succeed at tertiary level.
3.35 Some researchers have suggested that tests like the ASAT which are designed for older people, should be used as a means of identifying gifted children. Professor Stanley of Johns Hopkins University, Maryland, for example, who has studied mathematically precocious youth, has found that a test of this type is useful for the preliminary identification of children with gifts in mathematics. Professor Stanley's study of mathematically precocious youth uses the mathematics section of the Scholastic Aptitude Test (SAT). The SAT is employed as an entrance test for tertiary education in the United States. A similar project, a study of verbally gifted youth, uses the verbal section of the SAT to identify students with verbal gifts.
3.36 The fact that two students may have similar IQs gives little information about their specific abilities, nor does it indicate how they might achieve in specific subjects.36 Other 'objective' tests such as those discussed in paragraphs 3.32 to 3.35 may provide additional useful data. Procedures for identifying children for places in special programs will be further enhanced if additional information can be gathered about a student's pattern of abilities, current level of achievement and interests.
3.37 Teachers should be in a good position to identify gifted children. They are able to observe and to test children many times and thus should be able to identify students with outstanding gifts if those gifts are displayed in the classroom. Research has found, however, that teachers who have not been trained to identify gifted children generally are not able to identify them.37
3.38 The South Australian Association for Gifted and Talented Children informed the Committee that experiments conducted over the past twenty years have shown that when teachers rely on their own judgement alone, they fail to identify a majority of gifted children. The Association observed that in one experiment, teachers nominated only six per cent of a pool of gifted students whom they were asked to assess.38 Researchers have found that teachers are unable to identify young gifted children. One study found that only four per cent of kindergarten children identified by objective procedures as gifted were judged to be so by their teachers. About ninety six per cent of the kindergarten students nominated by teachers as being gifted were of average ability.39 The efficiency of teacher identification of gifted children appears to increase with the age of the children. At the secondary level, teachers untrained in identification techniques have correctly chosen as gifted over forty per cent of gifted students.40
3.39 Teachers who have not been trained in identification procedures apparently identify as gifted those children who are high achievers, who are attentive, quick to answer, neat in work, compliant and polite. Teacher nominations of gifted children frequently do not include gifted children who are poor achievers, whose work is untidy, or who are disruptive in the classroom. Teacher nomination may miss under-achievers who have learnt to hide their abilities in order not to be regarded as different. Similarly, under-achievers who have motivational problems could be missed.41
3.40 Research indicates that teachers nominate children more effectively when they have been trained in the field of gifted education.42 Checklists of the typical characteristics and behaviour of gifted children may be improve teachers' identification of these children.
3.41 Some associations for gifted children informed the Committee that parents are usually reliable judges of the children's abilities. Mr Nicholas Lethbridge, President of the Gifted and Talented Children's Association of Western Australia, for example, remarked that 'parents seem to know their own children far better than teachers'.43
3.42 There is research support for the observations made by parents. An American study which monitored the effectiveness of teacher and parent identification found that parents were more conservative than teachers when nominating their child as gifted. They nominated fewer children and were much more accurate in their choices. The study found that parent opinion a potentially useful source of worthwhile information in the identification of gifted children.44 The commonly-held belief that parents tend to over-estimate their children's ability is not always the case. Another American study found that parents from higher status neighbourhoods tended to be less accurate in nominating children who would qualify for gifted programs than were parents from lower status neighbourhoods.45 The study found that parents who are themselves high achievers tend to regard their gifted children as 'normal' or 'average'.
3.43 Peer nomination is increasingly seen as a promising method of identifying older gifted children, especially to supplement information obtained from other sources. Checklists to gather peer information are sometimes used to identify gifted children. A simple way of obtaining information from a child's peers is to ask students questions such as 'Whom would you ask for help in maths?' or 'Who are the three students in the class who are best at English?' or 'Who is really interested in computers?' Information provided by a gifted child's age peers is also valuable because it may reveal insights and details of abilities and interests not available from other sources.
3.44 Gifted children themselves may provide useful information about their abilities. Samples of art work, writing and other project work are obvious examples.
3.45 Marks or grades given for class work or scores on standardised achievement tests can assist in the identification process. School records of past achievement may also be useful. A gifted child may also show evidence of talent through creative writing, artwork, musical performance or skill in sport. Children who regularly score well above their age or grade level may be gifted. High marks obtained at school can be misleading however, as they may reflect perceptions of appropriate classroom behaviour rather than actual ability. Moreover, consistently low marks are not necessarily an indication of average or below-average ability as gifted students who are frustrated and bored may well score poorly on class tests.
3.46 The Committee's review of the evidence suggests that identification should be a continuous process which does not rely on a single measure, but is as comprehensive as possible. It should include the whole class or the whole school. The process should ideally be conducted by a team, to lessen subjective elements in the assessment. Objective measures are preferred, but should be complemented by subjective information. No matter what identification procedures are used, some talented children will be overlooked. Nevertheless, if screening for giftedness were a routine and regular feature of school life, many more gifted children would be identified and thus given a greater opportunity to achieve their potential.
