A Practical and Theoretical view.
2) Characteristics of young children important to an assessment situation
3) Important Factors when interpreting Test Performance
4) Predictive Validity
5) Ethical issues
6) Theoretical requirements of an appropriate test
7) Infant and Pre-school tests
Young children are difficult subjects to assess accurately because of their activity level and distractibility, shorter attention span, wariness of strangers, and inconsistent performance in unfamiliar environments. Other factors that may affect a child's performance include cultural differences and language barriers, parents not having books to read to their child and a child's lack of interaction with other children. Consequently, assessment of infants, toddlers, and young children requires sensitivity to the child's background, and knowledge of testing limitations and procedures with young children.
Informal relaxed settings where the child can be as much at ease as possible are recommended when doing assessment. Assessing a child within the context of his or her community and the interacting social systems, and taking into account the family's needs, resources, and concerns affect both the evaluation and possible interventions.
When testing young children the examiner needs to take certain aspects into account. The young child's immature developmental status influences the responses to testing more than older children or adults. Other considerations, which are important, are the socio-physical environment and sensory integrity of the young child.
A young child is very active and has a short attention span. This can be difficult to deal with in a testing situation. The child I tested was 2 years of age, and I could not get her to sit still and listen to me for more than a few seconds. The world around young children is so interesting that they are so busy concentrating on other stimuli. The task at hand needs to be very interesting to keep the child busy and to be an adequate source of stimulation. I found that by changing the task that needed to be completed, the child would once again pay attention, but was quickly distracted, meaning that I had to keep the pace quick and try and make the tasks interesting for the child. Because children have a tendency to favour certain tasks over others, the test results can be influenced. The child is probably capable of completing a substantial amount of the test, but, because of the frequent movement and limited concentration, the test results can not always be accurate. I found that the child I was testing wanted to do what she desired, and was more interested in playing with the telephone, the television and any other gadgets that she found interesting. She seldom completed the tasks that were required of her, and looked at me with a big smile on her face, but did not answer any questions directed at her.
The developmental stage at which a child is functioning will also influence their behaviour in a test situation. The child I tested was very amused by her own behaviour, laughing often and looking at me blankly. Other children could be shy with strangers, stubborn and negative, and this all influences the test situation. The stage of cognitive development has an influence on what the child understands. I found that the child I tested was too young to understand many of the tasks I gave to her, especially naming as many animals as she knew. She is capable of naming the animal and it's parts when she can see the animal, but when asked to name as many animals that she could think of, she left the table.
Young children are not as able as older children to regulate their behaviour. Their emotions can change from one state to another within a matter of seconds, and the examiner has to be prepared for this. Luckily the child I tested remained content and even though she was unresponsive, she was happy to move around the room and explore what she found interesting. A child's temperament and affective state can influence cognitive ability, Vygotsky stated that "emotions and intellect are unified in a dynamic and meaningful system," and "emotions serve as the language of the young child, where verbal skills are lacking." (Grieve, 1992). The examiner needs to focus more intently on the emotions of the child, as they are meaningful and represent how the child is feeling. A child who is content but not interested in what the examiner is doing has just as much meaning as a child who cries and is frustrated.
Another important factor in the assessment situation is the child's physical and socio-cultural environment. The child's early development is a transactional process. The child is influenced by, and in turn influences the nature of the caregiving and social environment. Children's social relationships can have a considerable effect on their cognitive growth and this in turn influences the assessment situation. The child's ability to socially interact and engage with others is relevant to the assessment situation (Grieve, 1992). The child I assessed was comfortable around me and did not show any negative behaviour when her mother left the room. This would impact positively on the overall assessment, as she was willing to engage in an unfamiliar situation with me.
Culture is an important facet of the child's social environment as cultural beliefs guide and direct behaviour. Each culture has different value systems and these guide the individual. Different cultures also view concepts such as intelligence differently. For example, the Western cultures have placed much relevance on the concept of IQ, which has become a heavily coined term. Other cultures, for example Rural African Cultures see intelligence not as a number on a chart but focus on the abilities to perform certain skills that are necessary for family life and growth.
Children's cultural experiences can, therefore, have a huge effect on the way they view the assessment situation, and the way that they respond to tasks. Testing techniques developed for Western children do not always hold the same meaning for non-western children. Certain items on a task may be foreign to these children, and, therefore unfair. Unfortunately, there is no test which is "culture free" making the need for the examiner to be open and aware of cultural differences a very important part of the assessment situation. Children need to be viewed within their cultural context and when measuring cognitive abilities the examiner needs to be sensitive to cultural variation (Grieve, 1992).
Before I tested the child, demographic information was obtained, so as to see whether the particular child was at a disadvantage to other children because of a different repertoire of items in the child's learning environment. As the child I tested was from an urban environment there was no reason to believe that she had not been exposed to the important items on the test. If the child had been from a rural environment where books and one on one attention was not available, I would have taken this into account when scoring the test (if scoring and interpreting results was required). A child of two does not always attend a pre-school, and is often left at home with a caregiver. This should be taken into account when interpreting test performance, as certain items on the test may be unknown to the child.
The child's atypical development is an important factor when testing and interpreting the scores. Motor handicaps are particularly important when assessing an infant, as most standardised tests focus on the child's motor development. This must be taken into account when selecting the appropriate assessment procedures. The presence of a handicap can affect the assessment performance. Disabled children tend to show a greater variability in day to day performance than non-disabled children. This could affect the interpretation of assessment scores, and should be taken into account by the examiner.
Certain tests are administered, so as to predict what the child's level of functioning will be in later childhood. The general finding is that infant test scores correlate positively but unimpressively with childhood test scores (Gibbs, 1990). Because infants develop at different paces, the predictive validity of infant tests is minimal. McCall (1976) states that there is essentially no correlation between the performance during the first six months of life with IQ after the age of five. The correlation between infant and school age test scores does not exceed .40. The correlation between pre-school tests and later IQ is more positive and meaningful. The older the child is when initial testing takes place, the stronger the relationship is with later IQ. Results of previous studies suggest that IQ becomes relatively stable by the age of 8, so if a child is tested at this age or later, then there is likely to be a strong correlation with scores on later tests.
On the basis of the performance of the child I tested, I would not be able to predict her level of functioning in five years time. She was disinterested in the test, and therefore, did not perform to her abilities. She hardly answered anything on the test; being distracted and more interested in other things around her. This would in theory give her a low-test score if the test had been a developmental assessment test. This does not reflect her capabilities at all, and thus I would not feel that it was an adequate prediction of her early childhood capabilities.
If it was accurate to predict that a child's development is relatively steady and continuos then one would be able to accurately predict later performance. Most people expect a certain degree of continuity in development that forms an individual's identity. An individual's IQ is expected to remain stable over their lifetime, and studies of older children support this view. Unfortunately, the ability to predict later mental development from infant test scores is very poor, particularly among infants who are developing normally. Children referred to clinics because of low functioning on infant test scores show greater predictive validity on infant test scores (Gibbs 1990). MacRae (1955) evaluated 40 infants and found moderately high correlations between their scores on infant tests administered prior to 12 months and their WISC IQ score at 5 years of age (Gibbs, 1990).
Every profession has distinct ethical obligations to the public. These obligations include professional competency, integrity, honesty, confidentiality, objectivity, public safety, and fairness, all of which are intended to preserve and safeguard public confidence (Gregory, 2000). Promoting ethical practices in assessment is considered to be a very important goal of the organisations involved in assessment. Codes are intended to increase the awareness of ethical practice among their memberships and to promote ethical uses of assessment in various contexts. Professional organisations publish formal ethical principles, which bear upon test use. Before testing takes place the tester needs to obtain informed consent from the parents or guardians of the child. The Standards Manual states that "informed consent implies that the test takers or representatives are made aware of the reasons for testing, the type of test to be used, and what testing information will be released and to whom" (Gregory, 2000 p. 576).
Consent was gained to test the child in my situation, and the mother of the child was made aware that the results of the test have no importance whatsoever, and are simply for the writing of an assignment. I went over the test with her before testing the child, and she was satisfied that the items could not harm her child in any way. The mother opted to leave while I worked with the child, as she believed she would be a distraction. A consent form was signed and in a professional situation this would be a requirement of testing a young child. All procedures, test items and reasons for testing were explained to the mother, leaving no doubt in her mind as to why I was using her child as part of my course.
Grieve (1992) states that there are certain theoretical requirements that a test must meet in order for a test to be appropriate. Assessment should reveal more than what is directly observable, as mentioned earlier, the child's response is not always an accurate representation of what the child is capable of. A test should not equate ability with a single quantitative score, the test should look farther than what the child has scored and allow flexibility in the examination of a child's problem solving ability and reasons for a given response. As the child is distracted and active a test is not always an accurate representation of the child's ability, but the test should give some indication as to the potential ability of the child, as most tests are based on this assumption. A test should be criterion-based rather than normative, as each child develops differently and every response may have a different meaning. Every child perceives the world differently from another, and a test should allow for this. Tests should be based on a sound theory of cognitive functioning and the assessor should have a sound knowledge of developmental sequences, and bear in mind the characteristics of the developmental status of the particular child being tested. A test should be as culturally fair as possible in order to be reliable, and not rely on verbal skills alone, and should be sufficiently flexible to meet the individual needs of each particular child (Grieve, 1995).
As many children perform better in a comfortable situation, with the caregiver or parents in close contact, a test should allow for this if necessary. The assessment situation should be non-threatening and enjoyable for the child and depend on the examiner's considerable skill to establish rapport and maximise the child's performance.
Infant scales and pre-school tests measure different components of intellectual ability. Infant tests measure the developmental progress of babies and children focussing on areas such as gross-motor, fine-motor, language development, adaptive behaviour, personal-social behaviours, sensori-motor intelligence, and so on. Pre-school tests focus more on the cognitive abilities of the child. The most widely used infant measures are:
Batelle Developmental Inventory, (BDI) (Birth to 8). This is a Standardised/Norm-Referenced test. 341 items assesses Personal-social, adaptive, motor, communication cognitive domains. There are 22 subdomains (e.g., coping, peer interaction, attention, memory, expression of feelings). The Battelle merges norm-based, curriculum-based and adaptive features into a flexible instrument. It provides a good example of curriculum referencing and linking assessment, intervention and evaluation beyond a curriculum. It is helpful in identifying a child's patterns of strengths and weaknesses. Some caution is needed in interpreting scores because of the small number of items in most subdomains (Ritter, 1995).
The Bayley Scales of Infant Development-II (1 to 42 months) is a Standardised/ Norm referenced test. Broad content coverage includes--Mental Scale, Motor Scale and Behaviour Rating Scale. The Psychomotor and Mental Scale each yield an Index Score, with a Mean of 100 and a standard deviation of 16. Mental and motor ages can be estimated from the norm tables provided.
The Bayley is designed to be presented with a parent present. A Behaviour Rating Form is also included. Items are playful and enjoyable for both child and experienced assessor. Several items (e.g. peg board) are scored at more than one age level, depending on the child's performance. This is strength of the measure, in that it limits the number of items administered, and these items can sometimes substitute for a preliminary screening. However, the assessor needs to be very familiar with the test in order to use this feature effectively. The Mental Scale is correlated with the motor scale, and appears to underestimate the mental development of children with motor problems. The Behaviour Rating form provides the assessor with information on the child's social/emotional behaviour. It is useful for evaluating the reliability of scale scores, in light of rapport between assessor and child (Ritter, 1995).
Pre-school children show a wide variety in emotional maturity and respond differently to the examiner. For some children the examiner can be sure that the test results reflect an accurate level of cognitive functioning, but with other children, this is not so. A child can be shy and non-responsive, but this does not show a lack of skill, but rather a lack of co-operation. Pre-school tests need to be approached with caution to avoid the negative consequences of labelling or overdiagnosis (Gregory, 2000).
The most common suitable pre-school
test known by Schakel (1986) as "the big 4" are the:
" Wechler Preschool and Primary Scale of Intelligence (WPPSI-R)
" Stanford-Binet: Fourth Edition (SB:FE)
" Kaufman-Assessment Battery for Children (K-ABC)
" McCarthy Scales of Children's Abilities (MSCA)
The Kaufman-Assessment Battery for Children (K-ABC) (2 ½ to 12 ½) is a standardised / norm referenced test. It is scored using standard scores, percentile ranks, stanines and age equivalents. Separate percentile tables are provided for several socio-cultural sub groups. The K-ABC is a multi-subtest battery, which facilitates inter-area comparisons. A Composite Mental Processing Score is calculated from the Sequential and simultaneous areas. A non-verbal score can be calculated from appropriate Mental Processing subtests. The statistical information is very complete and includes a choice of confidence bands. This type of information, the clear format for the test instructions, and the norm charts make it comfortable for the assessor to administer and score. K-ABC is not modelled after the Binet (most standardised tools are). It is based on a theoretical framework that looks at sequential and simultaneous processing. The child needs to be able to imitate the assessor and to respond to pictures and cards with adult in a "testing" situation (Ritter, 1995)
The Stanford-Binet: Fourth Edition
(2 to adult) is a Standardised/Norm-Referenced Test. It is used to measure overall
mental skill and provide a profile of mental functioning in four areas: verbal
reasoning (vocabulary comprehension, absurdities, verbal relations), abstract/visual
reasoning (pattern analysis/block design, form copying; matrices, paper folding
and cutting), quantitative reasoning (quantitative number series, equation building),
and short-term memory (sentence memory, bead memory, digit memory, object memory)
The Standford-Binet is a completely revised measure that is more like a whole new battery rather than a revision of previous items. It attempts to retain the advantages of the older Stanford-Binet tests (e.g. one basic test for ages 2 through adulthood), while also providing scoring patterns and interpretations similar to the Wechsler tests. (Vacc, 1995)
The Stanford-Binet consists of 15 sub-tests grouped into four areas. Each area yields a Stanford Age Score (SAS) with a mean of 100 and a standard deviation of 15. Scores from all four areas are combined into a Composite Score, also with a mean of 100 and a standard deviation of 15. The composite Score is equivalent to full scale IQ. Each sub-test within the group of sub-tests yields a standard type score, with a mean of 50 and a standard deviation of 8. The assessor needs to take care in reporting the child's results in order to avoid confusion between the scales used for the area and the sub-test scores. The manual presents suggestions for abbreviated batteries tailored to specific testing needs.
This version has advantages over the earlier more traditional Stanford-Binet L-M: the area scores facilitate comparisons between strengths and needs; a "routing test" is administered first and that score determines the starting point for each subsequent sub-test. For the pre-school child these advantages are offset by the absence of the playful activities that made the earlier versions attractive to young children and their assessors. However, this edition is appropriate for young children who have developed the ability to work with an adult in a standard assessor-child interaction style. A training tape and computer assisted scoring are available (Vacc 1995)
These are just two examples of Pre-school tests. They differ from Infant tests in that the tests are more complex, focussing more on cognitive skills rather than sensory motor skills. These tests also have a greater predictive validity, as the child is more likely to be developing at a fairly steady pace. Before assessing a child the examiner should know the developmental history of a child and assess whether the test is suitable for the particular child. Some test items may be too difficult for a particular child, or may not be culture fair, as the items may not be in the child's repertoire. There may also be a language barrier and the child may have difficulty understanding the instructions given to him or her. This would make the test unfair for the child, and would yield incorrect results.
Testing children is a difficult process in which many factors have to be taken into account. In the diverse societies in which many of us live each and every child has to be closely examined for cultural and social differences. The examiner has a huge responsibility to both the parents and the child to make the test as fair as possible. As assessment has a large role in society today in screening children for neurological impairment, mental retardation, and school readiness, it is essential that the examiner follows testing procedures properly and efficiently. Testing children is an arduous process where consent must be gained from the parents of the child, and all the information concerning the testing process, and results have to be carefully explained.
Young children are so active an alert
that the testing situation may become difficult, and drawn out. As I found out,
testing a child is not as easy as many people might think, and requires patience,
control and much persuasion. Children are constantly exploring their worlds,
and an examiner may not be an interesting part of this world, forcing the child
to find more interesting stimuli. It is important for an assessor to chose a
suitable test for each particular child, and to explore documents from parents
and significant others that may aid the assessment situation.
Assessment is important if there is any sign that a child is not developing normally and at the correct pace. Early detection of problems is critical and tests may aid in this detection.
Gibbs, E.D. & Teti, D.M. (1990). Interdisciplinary assessment of infants: A guide for early
intervention professionals. Pp.4-10, 77-88. Baltimore: Paul H. Brookes.
Gregory, R.J.. (2000). Psychological
Testing: History, Principles, and Applications (3rd ed.)
Boston: Allyn & Bacon.
Grieve, K.W. (1992). Play
based assessment of the cognitive abilities of young children.
Pp.5.6-5.21. Unpublished doctoral thesis, Unisa, Pretoria.
Ritter, S.H. (1995). Assessment of Preschool Children. New York: ERIC DIGEST.
Vacc, N.A. (1995). Testing Children. www.psycpage.com
|Click on these buttons to visit our journals|
All pages copyright ©Priory Lodge Education Ltd 1994-2004.