This is the old United Nations University website. Visit the new site at http://unu.edu

Assessing intellectual and affective development before age three: a perspective on changing practices

Abstract
Introduction
Myths concerning intelligence testing in early childhood
Principles of assessment
Interdependence of development
Multiple sources and multiple components
Assessment sequence
Child-caregiver relationships
Framework of typical development
Emphasis on organizing and functional capabilities of the child
Identify current and emerging competencies and strengths
Collaborative process
Assessment as the beginning of intervention
Reassessment as an ongoing process
Conclusions
References

Samuel J. Meisels and Sally Atkins-Burnett

The authors are affiliated with the School of Education at the University of Michigan in Ann Arbor, Michigan, USA.

This paper is based in part on Meisels SJ, Atkins-Burnett S. The elements of early childhood assessment. In: Shonkoff JP, Meisels SJ, eds. Handbook of early childhood intervention, 2nd ed. New York: Cambridge University Press, in press.

Abstract

Developmental assessment is an ongoing process of systematic observation and analysis, the purpose of which is to understand the child’s competencies and resources and the caregiving and learning environments most likely to assist the child in making the best use of his or her developmental potential. For many years young children have been assessed with normative instruments that focus primarily on evaluating their intellectual development. Many myths surround the use of these instruments with very young children, including the following: intelligence can be defined and measured with confidence; intelligence test data have diagnostic relevance; early intelligence tests have predictive value; such tests are useful for assessing young children with special needs; practitioners value and use these tests in their clinical practice; and IQ tests fulfill the legal purposes and intent of public laws. Each of these statements represents a misstatement of fact, and none of them has a strong evidentiary basis.

In place of the narrow classificatory function served by such tests, a variety of other approaches to assessing intellectual and affective development in the first three years of life are described. These instruments are consistent with a group of principles of responsive, develop-mentally oriented assessment that include recognizing the interdependence of development, understanding the importance of using multiple sources and multiple components in assessment, providing a meaningful sequence for assessment, respecting and evaluating child-caregiver relationships, basing assessments on a framework of typical development, emphasizing the organizational and functional capabilities of the child, focusing on the child’s current and emerging competencies and strengths, viewing assessment as a process of collaboration and the beginning of an intervention, and understanding that assessment and intervention are two elements of a larger whole. Assessments that are defined in terms of these principles assist us in meeting our most important goal: assisting all children to achieve their potential.

Introduction

For more than 60 years, researchers have sought to develop systematic ways of learning about the development of infants and young children. In Nancy Bayley’s first monograph concerning mental development, she attempted to specify those infant behaviours that could legitimately be tied to later mental functioning. To do this she posed such questions as “What specific behavior precedes later mental achievements? To what extent are these later achievements dependent on the earlier? Can we predict later development from early behavior? How do individual growth rates compare with the norm for a group of infants? To what extent are these rates affected by environmental conditions?” [1, p. 7]. The instrument that she developed, which has been revised twice since its experimental versions [2,3], was intended to answer these questions.

The Bayley Scales of Infant Development II (BSID-II) [3] are used primarily to sample the intellectual growth of infants and toddlers. The scales are most successful in sorting, categorizing, and ranking children according to demonstrable parameters of infant intelligence. Despite its widespread use, very few researchers, including Bayley, came to believe that what we can learn about infant functioning in the first few months and years of life provides us with accurate information about children’s later intellectual development or performance [4-6; but for a differing opinion, see refs. 7 and 8]. It seems that all we know with certainty is that the relationship between early-appearing, pre-verbal intellectual functions and typical intellectual performance that is manifested at school age and beyond is poorly defined. As Escalona and Moriarty noted, “When applied at age levels below 18 months, the term ‘intelligence test’ is misleading.... The true relationship between that which is measured by infant tests and that which we later call intelligence remains largely unknown” [9, p. 597].

Until recently, little attention has been paid to the assessment of affective components in infancy except by researchers. Although the BSID included an infant behaviour rating scale, it was not utilized by practitioners because of the absence of standardization data before the most recent version, and also because of limited understanding of the potential clinical applications of the information obtained from this scale. In the 1970s and 1980s, researchers examined temperament, focusing on the relationship between mother and child [10-14]. But again, most practitioners did not use information about temperament because of the absence of a developmental framework in which to place this research. More recent approaches to assessing infant affect (e.g., the Infant-Toddler Symptom Checklist [15]) focus on detecting deficits rather than identifying strengths, but in the social and emotional realm the presence of negative behaviours does not necessarily signal the absence of positive behaviours. Children sometimes have strong social skills while nevertheless exhibiting problem behaviours. It is important to assess the presence of both skills and problems within the context of the child’s overall development.

This paper will review the assumptions that underlie the use of the prevailing normative paradigm in assessment and will present research that supports different perspectives on evaluating children’s intellectual and affective growth and development. Integrated within a discussion of the principles of a responsive assessment model, we will present alternative approaches to assessing the intellectual and affective status of very young children. To begin, we will try to understand why the normative model of intelligence testing that is the foundation of testing in early childhood and beyond is inappropriate for the assessment of young children.

Myths concerning intelligence testing in early childhood

The conceptual hold over the thinking of practitioners, policy makers, and the public regarding conventional models of assessment may be attributable to several influential myths or presumptions. Neisworth and Bagnato [16] identified six such myths that relate to the use of intelligence testing with children younger than age three. The first myth is that “professionals know what early intelligence is and agree it can be measured” [16, p. 2]. This view reifies intelligent behaviour by claiming to be able to specify the characteristics of such behaviour, although intelligent behaviour is defined in different ways at different ages in the first few years of life. Inasmuch as no adequate theory of intelligence has yet been propounded [17], the notion that we can agree on how to measure intelligence is circular: it assumes that we know what the construct is that we are measuring before we have even been able to define that construct.

The second myth regarding early childhood intelligence testing states that “research supports the measurability, reliability, and diagnostic utility of early intelligence testing” [16, p. 3]. There is no evidence to support this claim. Rather, intelligence scores in infancy appear to be more the sum of a child’s skills and behaviours in selected contexts, than a predictive index of future functioning and abilities. Because of different life experiences, equivalent IQ or developmental scores in different children may have different meanings. This suggests that the psychometric utility of such scores is very limited. Additional factors limit their diagnostic utility. For example, despite recent research emphasizing the importance of social-emotional development [18], most measures continue to emphasize cognitive functioning as the major determinant of development. The interdependence of domains is seldom afforded adequate attention.

The third myth is a corollary of the second: “Early intelligence tests have predictive validity” [16, p. 5]. This statement is challenged by the fact that measures of the home environment have demonstrated greater predictive power on later outcomes than have early intelligence tests [19]. In short, the tests do not have predictive validity; rather, the child’s context and what takes place in that context carries the weight of the prediction. Moreover, since the test content used in assessments differs markedly as the child grows older, children may demonstrate vast differences in behaviour between early and later displays of “intelligence.”

Neisworth and Bagnato’s fourth myth concerns the use of IQ tests with children with disabilities or special needs: “Standardized administration procedures provide a reasonable and representative assessment of young children with special needs” [16, p. 8]. Among the problems with this statement is the general absence of disabled children in the normative samples of nearly all intelligence tests [20], as well as the inappropriateness of some procedures or requests built into the tests that do not account for a particular child’s disability. As a result, they contend that these instruments “really measure the child’s disability rather than ability” [16, p. 8].

The fifth myth concerns how intelligence scores are used to identify children in need of early intervention services: “Psychologists and other practitioners value and use intelligence tests to identify young children in need of early intervention” [16, p. 9]. In a survey conducted by Bagnato and Neisworth [21], 43% of psychologists and other practitioners considered early intelligence tests to be virtually without value. Their survey also highlighted practitioners’ views concerning the serious flaws in form, content, and function of these tests.

The final myth claims that tests of early intelligence “fulfill the legal purposes and the intent of public law” in the United States [16, p. 12]. The basis of the federal laws in the United States that regulate services to young children and their families acknowledge that early childhood assessment is specifically intended to take into account a wide range of environmental factors and to rely on information from a variety of significant actors in the child’s life. Although clinical observations made during the administration may contribute to understanding the child, the test scores are of limited value, particularly when administered to an ethnically or culturally diverse group of children. In some states, the sole use of IQ tests with disabled minority students has been ruled illegal.

Overall, there appears to be a significant misalignment between early intervention programmes and the type of assessments (primarily intelligence testing) that are used most commonly for determining eligibility for these programmes. This misalignment can be seen in at least three fundamental areas. First, the conventional, norm-referenced assessments that are in widespread use primarily separate low-functioning from average- and high-functioning children. This is one of the purposes of diagnostic classification, and norm-referenced tests are judged by their accuracy in accomplishing this purpose [22]. Yet, many very young children cannot accurately be placed in conventional aetiological groups, categories, or types. Such classifications frequently have little utility. For example, we may know that a child is delayed in development, but information about the aetiology of the delay or the most reasonable ways of assisting such children in their development is not available from examination of such test scores and protocols [23]. Of greater value is to begin to treat these children non-categorically - as children in need of special services - and then gradually acquire the information that will enable more differentiated services to be created and provided [24].

Second, norm-referenced tests assume stability in human characteristics and a smooth curve concerning the predictability of these characteristics over time. Yet, the reality of working with young children and their families is not stability but change. For all but the most seriously impaired young children, we do not expect long-term predictability as much as the ability to forecast the next steps on the basis of experience with intervention [22]. A third misalignment concerns the manner in which the information is acquired conventionally in the assessment process. Traditionally, assessments are administered in single sessions, usually in standardized settings. But in order to understand development and the factors that contribute to it among very young children, it is essential to sample behaviour over time and to observe that behaviour in natural settings, noting the circumstances under which children are able to demonstrate skills and the occasions that particularly seem to challenge them. Otherwise, the likelihood of our being misled by the information acquired in the assessment process becomes at least as great as the possibility of acquiring insights into the development of young children and their families. It appears that if we are to be able to respond to Bayley’s questions of more than 60 years ago, we will need to rethink our methodology for learning about children’s growth and development.

Principles of assessment

A number of well-founded principles of early childhood assessment can be identified that offer a perspective that differs from that of conventional measurement. The Zero to Three Working Group on Developmental Assessment, a multidisciplinary group of professionals and parents, was convened in July 1992 to identify “problems and promising approaches in current assessment paradigms, policies, and practices” [25, p. 5]. The discussions initiated by this group led to the establishment of a set of principles that can be used to guide assessment of young children in both the intellectual and affective realms [26, pp. 17-25]. These principles are presented in table 1.

In general, the goal of early childhood assessment should be to acquire information and understanding that will facilitate the child’s development and functional abilities within the family and community. Developmental assessment in particular is “a process designed to deepen understanding of a child’s competencies and resources, and of the caregiving and learning environments most likely to help a child make fullest use of his or her developmental potential. Assessment should be an ongoing, collaborative process of systematic observation and analysis. This process involves formulating questions, gathering information, sharing observations, and making interpretations in order to form new questions” [26, p. 11]. The assessment principles listed in table 1 assume this definition of assessment. Below we will examine these assessment principles and offer examples of instruments and practices that utilize them, focusing on both affective and intellectual development of children below age three.

Interdependence of development

The child is a complete being - not a series of articulated skills, acquisitions, or elements - and the development of each area is dependent on other areas [27]. The child’s skill in naming a picture is an indication of sensory, cognitive, and motor abilities, as well as language acquisition. Underlying all of this is the emotional capacity that enables the child to relate to others and to organize his or her world. To consider only one area of development in isolation from the others leaves unrecognized the influence of the other areas and may impede our understanding of the child’s abilities and challenges. The child’s functional capacities should be examined in a variety of contexts in order to fully comprehend how the child integrates skills into his or her repertoire of behaviours and responses. In-depth examination of a child’s skills in a single area of development proceeds from the more complete picture of the child’s overall skills and knowledge acquisitions, recognizing the interdependence of the systems in development. Examples of assessment tools that focus on functional capacities in an integrated fashion include the Functional Emotional Assessment System (FEAS) [28], structured and non-structured play observations that are part of the Transdisciplinary Play-Based Assessment (TPBA) [29], and the Infant-Toddler Developmental Assessment (IDA) [30].

TABLE 1. Principles of responsive and developmentally oriented assessment

» Assessment must be based on an integrated developmental model

» Assessment involves multiple sources of information and multiple components

» An assessment should follow a certain sequence

» The child’s relationship and interactions with his or her most trusted caregiver should form the cornerstone of an assessment

» An understanding of sequences and timetables in typical development is essential as a framework for the interpretation of developmental differences among infants and toddlers

» Assessment should emphasize attention to the child’s level and pattern of organizing experience and to functional capacities, which represent an integration of emotional and cognitive abilities

» The assessment process should identify the child’s current competencies and strengths, as well as the competencies that will constitute developmental progression in a continuous growth model of development

» Assessment is a collaborative process

» The process of assessment should always be viewed as the first step in a potential intervention process

» Reassessment of a child’s developmental status should occur in the context of day-to-day family and/or early intervention activities

Source: ref. 26, pp. 17-22.

The TPBA is a comprehensive transdisciplinary approach to developmental assessment that is based on the premise that developmental functions are interdependent and children’s development is influenced by a variety of factors. TPBA is a functional approach to the assessment of young children with disabilities or those at risk for developmental delay, which actively involves the child, the child’s parents, and other professionals in a natural environment of assessment and intervention. TPBA is organized around the planning of a play session that is based on information about the child’s developmental status acquired from the parents. “Toys and materials that are appropriate to the child’s level are arranged to entice the child to play using various play strategies and developmental skills. One team member facilitates the child’s play to encourage the expression of optimal abilities” [29, p. 4]. Guidelines are provided for observing the cognitive, social-emotional, communication and language, and sensorimotor development of the child. Fundamental to this assessment is the ecobehavioural validity of its methods, materials, and techniques.

Because the assessment is planned with the family, and information is acquired from family members as well as from others who are familiar with the child, the baseline for the assessment is close to the family’s experience. Multiple opportunities for the child to interact with new and familiar materials are provided, and many observers are included in the assessment in order to capture as many perspectives as possible. Parents complete a pre-assessment inventory that provides valuable information about the child’s developmental level and skills and also assists the team in preparing an environment that will elicit the child’s optimal abilities. The assessment itself involves several phases: unstructured facilitation where the examiner follows and expands upon the child’s lead; structured facilitation in which the examiner attempts to elicit behaviours that were not spontaneous in the preceding phase; introduction of a peer in order to observe interaction among children; structured and unstructured play between the parents and the child; structured and unstructured motor play; and snack, which allows for screening of oral-motor difficulties as well as social and adaptive development. Throughout the observation period, a staff member discusses with the parent the representativeness of the child’s behaviours, some of the professional interpretation of behaviours, and the parent’s perception of the child’s behaviour. Observers are guided in their observations by questions that address both quantitative and qualitative aspects of the child’s behaviour rather than just the presence or absence of behaviour. Once the assessment is complete and the guidelines provided have been reviewed in terms of the child’s behaviour and accomplishments, transdisciplinary recommendations are developed and a programme planning meeting is convened to provide additional feedback for the child’s parents and others working with the child. This experience is designed to be highly respectful of the child, the child’s family, and the culture in which the child is being reared, and focuses on the interdependence of areas of development within the child.

Multiple sources and multiple components

The ability to take into account a variety of perspectives is essential to provide a complete view of the child’s strengths and capacities and the optimal means of promoting further development. Information can be obtained from a variety of contexts with different tools guiding the process and informing the assessment. All members of the assessment team must make evident to the others their perspective on the child and the underlying assumptions of that perspective. As each member shares an understanding of the child’s abilities, predispositions, and challenges, a more complete, informed, and multidimensional profile of the child emerges [31].

One example of an assessment that accounts for multiple sources and that is composed of multiple components is the Infant-Toddler Developmental Assessment (IDA) [30]. The IDA goes beyond traditional measures by addressing health, family, and social aspects of development as well as developmental dimensions. “The IDA is anchored in theoretical constructs and clinical perspectives which acknowledge the variety and interdependence of factors that influence the health and development of young children” [32]. The six phases of this assessment process provide a guide for team process, decision making, and how to include parents in the assessment. A team is formed of at least two professionals who function as developmental generalists and also contribute their own disciplinary expertise. This team may include social workers, developmental nutritionists, nurses, special educators, physicians, and physical, occupational, or speech therapists. The IDA helps these individuals organize information from multiple sources about the health, development, and social supports of the child and family. “The team shares responsibility for gathering, organizing, integrating, and synthesizing information, and for the problem solving inherent in all clinical work” [32, p. 19]. The team roles include family interviewer and primary family liaison, health reviewer, child evaluator, and assessment coordinator. The IDA is fundamentally an assessment that is premised on obtaining multiple perspectives on the wide variety of elements that comprise children’s performance, learning, and relationships.

Assessment sequence

Assessments should begin by establishing reliable, working alliances with the significant individuals in the child’s life. These individuals hold important information about the child and his or her capacities. Creating a reliable alliance with parents involves the use of sensitive interpersonal communication skills and the development of mutual trust and respect [33]. This calls for sensitive listening skills, responsivity to requests and concerns, openness to the family’s interpretations, and honesty in interactions. Mutual respect for the family also involves understanding the family’s strengths, challenges, and problem-solving strategies, as well as awareness and communication of the cultural assumptions undergirding assessment and professional recommendations. Hirshberg [34] describes precisely this kind of parent-professional relationship in his description of clinical interviewing. His critical insight is that human connectedness is essential for the process of assessment and intervention. This connectedness occurs at many levels: between parent and child, parent and clinician, and clinician and child.

After this relationship has been established, the assessment should be focused on practical outcomes. Assessment is not an end in itself. Rather, its goal is to obtain useful and accurate information about the child and the child’s nurturing environment, including the resources and obstacles inherent in that environment, in order to find or create the most optimal situation for supporting the child in meeting family goals. The validity of an assessment is determined in terms of its application.

Greenspan and Meisels [26, pp. 18-19] translate these ideas into a sequence of assessment as follows:

» Establishing an alliance with the parents, listening to their views of the child’s strengths and challenges, and discussing the issues to be explored in the assessment.

» Obtaining a developmental history of the child and an initial picture of the family’s experience: although basic information may be readily available, some insights may only emerge over time, as part of an ongoing relationship and working alliance with the parents.» Observing the child in the context of unstructured play with the parent(s) or other familiar caregivers.» If appropriate, observing interaction between the child and a clinician.» Making specific assessments of individual functions in the child, as needed.

» Using a developmental model as a framework for integrating all the data obtained from parents’ reports, direct observation, and other sources, and conveying and discussing assessment findings in the context of an alliance with the child’s primary caregivers, with the potential for starting an intervention process if needed.

Child-caregiver relationships

The interactions and relationship between child and caregiver form the foundation of the child’s ability to organize and respond to his or her world [35]. Parents are usually more skilled at reading and responding to their child’s cues than even the most skilled professionals. However, when the relationship between parent and child is strained or maladaptive and there is no substitute relationship, the long-term consequences for the child can be very negative [36]. Observations of interactions between the child and parent allow professionals to learn methods of intervention from the parent that have proven successful for the family and child, as well as ways in which the professional can offer support for more successful interactions.

Parker and Zuckerman [37] suggest that one of the goals of the assessment process should be to determine the level of involvement in the intervention process that is most beneficial for the family. For some families, a very active role in the intervention process is constructive. Other families are so overwhelmed by their own and their child’s demands that additional roles cannot easily be assumed and are often an additional burden that might seriously strain existing resources.

Assessment itself is an intervention in the lives of family members. Bailey [38] notes that every interaction with a family constitutes an assessment, and every assessment is itself an intervention. In the process of obtaining and sharing assessment information with families, professionals should communicate the range of development that is seen typically and should provide information about how to determine when a child might benefit from additional intervention. Greater awareness of developmental expectations for children of different ages may be all the intervention some families require. Although research clearly shows that the family relationship is central to children’s development [39-43], every family is different. There is no single intervention that is applicable to all families or all situations.

Several assessments focus specifically on these issues. For example, the Nursing Child Assessment Feeding and Teaching Scales [44, 45] were designed to highlight the interaction between parents and very young children and have been used in many different research and clinical settings. The scales are intended to capture the reciprocity of communication between child and parent and to explore the range of behaviours available to both members of the dyad. Parents are assessed on four subscales: sensitivity to the child’s cues, response to distress, fostering social-emotional growth, and fostering cognitive growth. Children are evaluated on clarity of cues and responsiveness to the caregiver. Performance on these scales has been associated with children’s language usage at ages three to five, child temperament, and psychosocial high-risk factors in children’s lives. The scales also discriminate among parents with different levels of schooling, those who are at risk for abuse as well as those who actually abuse their children, and those with high family stress. They can be of great utility to intervention programmes seeking to understand some of the difficulties that can potentially emerge between children and parents.

Framework of typical development

Early interventionists require a strong foundation in child growth and development. Growth in the early years is generally rapid and is accompanied by large variations in when and how children manifest different skills and behaviours. Cultural influences that may affect opportunities for learning may alter the arrival of developmental milestones. By viewing the development of all children on a continuum, those children who are born with disabilities or developmental delays can be viewed from the perspective of children who are not yet functioning as expected in given areas, rather than children who are unable to acquire the skills of typically developing children. Assessment frameworks that exemplify this view of development will provide more information for parents and interventionists because they will help us to see the child’s accomplishments within a normal continuum of accomplishments. They suggest a series of steps or experiences that must be provided, rather than a set of milestones that the child has failed to achieve.

One productive approach to assessing typical development is to observe the child in naturally occurring, unstructured play situations (as contrasted to the relatively structured play experiences that are part of the TPBA). Segal and Webber [46, pp. 215-24] describe nine benefits of play observations that relate to both the normal developmental information acquired in this setting and the opportunity to create a parent-professional partnership. The benefits of play observations can be summarized in terms of:

» providing an opportunity to assess the functional behaviour of a young child who either cannot or will not perform in a formal testing situation;

» enabling infants and toddlers, because of the flexibility and spontaneity of a play situation, to achieve a level of object or symbolic play that they may not demonstrate on a standardized assessment;

» providing important insights into temperamental variables;

» revealing aspects of the parent - child relationship that help explain the behaviour of the child;

» providing insights into numerous domains of development;

» giving clinicians special opportunities to learn effective play strategies from a child’s parent;

» suggesting ways of helping parents modify play strategies that are not fully effective;

» identifying strengths, coping skills, and risk factors that impact on a diagnosis and may be useful for designing a treatment plan;

» enhancing the parent-professional partnership.

Emphasis on organizing and functional capabilities of the child

As children learn to organize the world, they are increasingly able to learn about and from the world, and to take part in the world actively. Skills or behaviours with no functional application, learned and tested out of context, have no place in early intervention [47]. The goal should be to help children make meaning of their world and participate in it. Towards that end, assessment of discrete areas of functioning (e.g., auditory discrimination, visual-motor integration) should take place only to inform our understanding of the child’s struggle with a given area or to better learn about the resources the child brings to the learning situation. In short, knowledge of a child’s skills or abilities is only part of the picture. To complete that picture, we need to know how the child uses those skills and abilities, what motivates the child, what frustrates the child, and what brings the child satisfaction, as well as the availability of experiences for eliciting, supporting, and extending skills and abilities.

Greenspan’s [28] approach to assessing children’s development - particularly their emotional development - focuses directly on the functional capacities of the child. Called the Functional Emotional Assessment Scale (FEAS), this approach embeds assessment in the context of structured play interactions with the child’s caregivers. Among the core capacities of the child that are evaluated are “the child’s capacity for self-regulation, engagement, elaborating symbols and representations, and creating logical bridges or differentiations within his or her emerging symbolic world (‘emotional thinking’)” [18, p. 232]. Various levels of development are postulated in Greenspan’s approach, but it is in interactions with adults that the child is evaluated in terms of a number of specific expected primary emotional capacities.

The following six areas are incorporated into the FEAS:

» primary emotional capacities, such as those capacities characteristic of children of a certain age, including responding to a caregiver’s gestures with intentional gestures of his or her own, engaging in a complex pattern of communication, or imitating and copying another person’s behaviour;

» emotional range, incorporating gestures, touch, and speech used by the child to master his or her primary functional emotional needs;

» affective emotional range, which concerns the various affective themes used by the child to organize his or her play and relationships;

» associated motor, sensory, language, and cognitive capacities, including other developmental challenges not already included in the primary emotional capacities (such as ambulating, feeding, copying simple gestures, etc.);

» general infant tendencies, which refer to constitutionally and maturationally based capacities of self-regulation, attention, capacity to enjoy sights and sounds, touch, movement in space, etc.;

» overall caregiver tendencies, referring to the caregiving patterns that facilitate or impede the child’s growth and development.

Identify current and emerging competencies and strengths

The traditional model of assessment operates in terms of deficits by sorting and sifting children into different categories of disability or pathology. Identifying children’s competencies and examining how they achieve those competencies is an integral part of more recently developed assessments [25]. The child’s strengths and competencies alert us to the personal and ecological resources that a child may be able to call upon to meet developmental goals. They also aid us in fashioning interventions that make good use of those strengths and resources [30].

How a child manifests a particular skill or behaviour is at least as important as the mere presence or absence of the skill or behaviour [48]. For example, a child with motoric challenges may be able to walk but may have difficulty scanning the environment for obstacles or stopping and turning when necessary.

Many of the assessments described in this paper adopt this strength-oriented approach. This can be seen in the IDA [30], FEAS [28], TPBA [29], and naturalistic play observations [46]. Overall, these approaches begin with a recognition of what the child can do and with an attempt to understand the context in which the child is most familiar. The child’s functional ca- pacities and the child’s natural environment form the anchors from which it becomes possible to learn more about the child’s areas of difficulty in everyday functioning and relationships.

Collaborative process

Assessment of very young children should be premised on the quality of the working relationship between parents and professionals [35]. The professional’s job is not to promote his or her view of the child to the parent, but to join the parents and other professionals in viewing the child multidimensionally in a way that contributes to the generation of strategies that will help the child make developmental advances and organize his or her world more adequately.

A variety of parent report instruments and protocols for parent interviews are available to help inform the assessment process (see, for example, the Vineland Scales of Adaptive Behavior [49] and the Minnesota Child Development Inventory [50]). These instruments are useful for obtaining different perspectives on a child’s behaviours. Designed to be easily understood by most parents, they allow for active participation in the assessment process and give parents an awareness of the normative lens through which professionals view children. They provide a starting point in the conversation between parents and professionals.

Parents play a vital role in helping the professional understand how the familial and cultural contexts influence the child’s repertoire of skills. Professionals must maintain positions of cultural reciprocity when interpreting assessment findings and making recommendations to families [51; 52; Kalyanpur M, personal communication, 1996]. This means that the professional must identify the values inherent in professional interpretations and recommendations, determine if these values are congruent with the value system of the family, and explain to families the assumptions underlying these recommendations. Professionals must explicitly respect the cultural differences identified and, together with the child’s family, find the most effective means of adapting professional recommendations to meet the needs of the child in a manner that is culturally relevant [51, 52]. Parents are most apt to follow through on recommendations when parents and professionals hold similar perceptions of the child’s needs and strengths, the professional is perceived by the family as a caring individual, and information is presented clearly and precisely [53].

Overall, the key to a successful assessment goes well beyond “establishing rapport” - the first step described in most test manuals. Successful relationships require rapport as a necessary condition, but rapport is not a sufficient condition for learning about the needs and strengths of the child and the family. Other features of the relationship include respect, reciprocity, and flexibility. These characteristics are key to establishing a framework in which honest interactions can take place.

Assessment as the beginning of intervention

A complete assessment includes information about how to facilitate the child’s development and those supports that are needed to help the child exhibit desirable behaviours. When assessment occurs in isolation from intervention, particularly when it depends on traditional norm-referenced instruments, the outcome of assessment maybe confusing, misleading, and ultimately counterproductive. It is only by testing out the hypotheses uncovered during the assessment that we can fully evaluate the validity of an assessment. The intervention not only confirms or disconfirms the assessment hypotheses, it elicits new hypotheses and new information that are in themselves an assessment of current functioning [21].

The notion of “current functioning” is one that is extremely narrow. A child’s current functioning changes from moment to moment. The success of an assessment can be viewed in part as a function of its predictive invalidity. That is, as the information acquired from the assessment alters the context and content of the intervention, so the “current functioning” of the child changes and is transformed. Continuous assessment needs to be incorporated into the intervention so that the two functions become virtually seamless.

A critical new direction for early childhood assessment is that of a focus on utility. The treatment utility of an assessment refers to the degree to which an assessment is shown to contribute to beneficial treatment outcomes. The implications of this shift in outlook are very significant. Instead of asking whether a particular diagnosis is correct, we need to ask, “Is this assessment useful in practice?” [54, p. 964]. Instead of using assessments to sort children by achievement level or diagnostic category, we need to determine if assessments can help us do better, more appropriate, and more accurate interventions - information that we acquire through studying interventions. When assessment becomes fused with intervention, it becomes an iterative, cyclic process, rather than a static experience. To be effective - to have social utility - assessments must be closely tied to intervention.

Reassessment as an ongoing process

An important view represented here is that assessment and intervention should be interactive processes in which each informs the other. In the United States, reassessments are written into state and federal laws in order to prevent children from being assigned to special classes or programmes and then forgotten, never to make a transition to a more appropriate or less restrictive environment.

However, reevaluation can have another meaning that is more functional and potentially even more critical for the overall growth and development of children. Reevaluation can serve as a time to reflect on the effect of the intervention. Every intervention provides some of the information that is needed to create a new, and more differentiated intervention. The metaphor that may be most powerful here is that of a moving target and of successive approximations to that target. Children’s development is a moving target of skills, knowledge, experiences, dispositions, and personality variables. Every intervention alters the child in some way - sometimes for the better, as when the child breaks through to a new skill, and sometimes for the worse, as when the child’s motivation to learn is diminished by continuing experiences of failure and frustration. Reevaluation on a continuing basis is essential if parents and professionals are to understand what they should try to do next with the child. Information about the child’s prior history is useful but quickly loses its power and relevance with very young children. Constant infusions of new assessment information, acquired in the process of intervention, are essential to maximize the relationship between the child, the child’s family, and professionals.

Conclusions

Assessments of very young children serve a variety of purposes. They can:

» determine eligibility for publicly supported services such as special education services;

» inform families about the range of their children’s development, assisting them in determining where a child falls within that range, and helping them understand the uneven progress that children may make;

» help craft individualized family service plans that take into account both family and child factors as well as the greater ecology in which this family resides;

» evaluate the effectiveness of interventions;

» be used as documentation during assessment and intervention to help a family to appreciate their child’s progress and plan future interventions.

Fundamentally, assessment is justified by its ability to guide intervention. In order for assessments to provide a firm basis for interventions with young children, it is essential that information about children’s strengths, and family beliefs and goals, be included in the information gathering that constitutes the assessment process. As progress is documented in the course of intervention, family members will be able to recognize their children’s growth and will be able to appreciate the changes in their relationship with their children. In short, the conditions for more adequate assessment are central to helping us achieve our goal of helping all children achieve their potential.

References

1. Bayley N. Mental growth during the first three years. Genet Psychol Monogr 1933;14:1-92.

2. Bayley N. The Bayley scales of infant development. San Antonio, Tex, USA: Psychological Corporation, 1969.

3. Bayley N. The Bayley scales of infant development - II. San Antonio, Tex, USA: Psychological Corporation, 1993.

4. Bayley N. Development of mental abilities. In: Mussen PH, ed. Carmichael’s manual of child psychology. New York: Wiley, 1970:1163-2109.

5. McCall RB. Early predictors of later IQ: the search continues. Intelligence 1981;5:141-7.

6. McCall RB, Hogarty PS, Hurlburt N. Transitions in infant sensorimotor development and the prediction of childhood IQ. Am Psychol 1972;27:728-48.

7. Fagan JF, McGrath SK. Infant recognition memory and later intelligence. Intelligence 1981;5:121-30.

8. Fagan JF, Singer LT, Montie JE, Shepherd PA. Selective screening device for the early detection of normal or delayed cognitive development in infants at risk for later mental retardation. Pediatrics 1986;78:1021-26.

9. Escalona SK, Moriarity A. Prediction of school-age intelligence from infant tests. Child Devel 1961;32: 597-605.

10. Bates JE, Freeland C, Lounsbury ML. Measurement of infant difficultness. Child Dev 1979;50:794-803.

11. Carey WB. Measuring infant temperament. J Pediatr 1972;81:414.

12. Hubert NC, Wachs TD, Peters-Martin P, Gandour MJ. The study of early temperament: measurement and conceptual issues. Child Dev 1982;53:571-600.

13. Rothbart MK. Measurement of temperament in infancy. Child Dev 1981:52:569-78.

14. Thomas A, Chess S. Temperament and development. New York: Brunner/Mazel, 1977.

15. DeGangi GA, Poisson S, Sickel RZ, Wiener AS. Infant/ Toddler Symptom Checklist. San Antonio, Tex, USA: Psychological Corporation, 1995.

16. Neisworth JT, Bagnato SJ. The case against intelligence testing in early intervention. Topics in Early Childhood Special Education 1992;12(1):1-20.

17. Sternberg RJ. Death, taxes, and bad intelligence tests. Intelligence 1991;15:257-69.

18. Greenspan SI. Assessing the emotional and social functioning of infants and young children. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: National Center for Infants, Toddlers, and Families, 1996:231-66.

19. Sameroff AJ, Seifer R, Barocas B, Zax M, Greenspan SI. IQ scores of 4-year-old children: social-environmental risk factors. Pediatrics 1987;79:343-50.

20. Fuchs D, Fuchs LS, Benowitz S, Barringer K. Norm-referenced tests: Are they valid for use with handicapped students? Exceptional Children 1987;54:263-71.

21. Bagnato SJ, Neisworth JT. A national study of the social treatment “invalidity” of intelligence testing for early intervention. School Psychol Q 1994;9:81-102.

22. Meisels SJ. Charting the continuum of assessment and intervention. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: National Center for Infants, Toddlers, and Families, 1996:27-52.

23. Meisels SJ, Anastasiow NJ. The risks of prediction: relationships between etiology, handicapping conditions, and developmental outcomes. In: Moore SG, Cooper CR, eds. The young child: review of research. Washington, DC: National Association for the Education of Young Children, 1982:259-80.

24. Meisels SJ, Dichtelmiller M, Liaw FA. Multidimensional analysis of early childhood intervention programs. In: Zeanah CH, ed. Handbook of infant mental health. New York: Guildford Press, 1993:361-85.

25. Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996.

26. Greenspan SI, Meisels SJ. Toward a new vision for the developmental assessment of infants and young children. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:11-26.

27. Emde RM, Biringen Z, Clyman RB, Oppenheim D. The moral self of infancy: affective core and procedural knowledge. Dev Rev 1991;11:251-70.

28. Greenspan SI. Infancy and early childhood: the practice of clinical assessment and intervention with emotional and developmental challenges. Madison, Conn, USA: International Universities Press, 1992.

29. Linder TW. Transdisciplinary play-based assessment. Baltimore, Md, USA: Paul Brookes, 1993.

30. Provence S, Erikson J, Vater S, Palmeri S. Infant-toddler developmental assessment: IDA. Chicago, 111, USA: Riverside Publishing Company, 1995.

31. McCune L, Kalmanson B, Fleck MB, Glazewski B, Sillari J. An interdisciplinary model of infant assessment. In: Meisels SJ, Shonkoff JP, eds. Handbook of early childhood intervention. New York: Cambridge University Press, 1990:219-45.

32. Erikson J. The infant-toddler developmental assessment (IDA): A family-centered transdisciplinary assessment process. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:147-68.

33. Turnbull AP, Turnbull HR. Families, professionals, and exceptionality: a special partnership. 3rd ed. Upper Saddle River, NJ, USA: Merrill-Prentice Hall, 1996.

34. Hirshberg LM. History-making, not history-taking: clinical interviews with infants and their families. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:85-124.

34. Weston DR, Ivins B, Heffron MC, Sweet N. Formulating the centrality of relationships in early intervention: an organizational perspective. Infants and Young Children 1997;9:1-12.

35. Williamson GG. Assessment of adaptive competence. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:193-206.

36. Parker SJ, Zuckerman DS. Therapeutic aspects of the assessment process. In: Meisels SJ, Shonkoff JP, eds. Handbook of early childhood intervention. New York: Cambridge University Press, 1990:350-70.

37. Bailey DB. Issues and perspectives on family assessment. Infants and Young Children 1991;4:26-34.

38. Barnard KE, Morisset CE, Spieker S. Preventive interventions: enhancing parent-infant relationships. In: Zeanah CH, ed. Handbook of infant mental health. New York: Guildford Press, 1993:386-401.

39. Crnic KA, Greenberg MT, Ragozin AS, Robinson NM, Basham RB. Social interaction and developmental competence of pre-term and full-term infants during the first year of life. Child Dev 1983,54:1199-210.

40. Rauh VA, Achenbach TM, Nurcombe B, Howell CT, Teti DM. Minimizing adverse effects of low birthweight: four-year results of early intervention program. Child Dev 1988:59:544-53.

41. Sameroff AJ. Models of development and developmental risk. In: Zeanah CH, ed. Handbook of infant mental health. New York: Guilford Press, 1993:3-13.

42. Sameroff AJ, Fiese BH. Transactional regulation and early intervention. In: Meisels SJ, Shonkoff JP, eds. Handbook of early childhood intervention. New York: Cambridge University Press, 1990:119-49.

43. Barnard KE. What the Feeding Scale measures. In: Sumner GS, Spietz A, eds. NCAST: Caregiver/parent-child interaction feeding manual. Seattle, Wash, USA: University of Washington NCAST Publications, 1994:98-121.

44. Morriset CE. What the Teaching Scale measures. In: Sumner GS, Spietz A, eds. NCAST: Caregiver/parent-child interaction feeding manual. Seattle, Wash, USA: University of Washington NCAST Publications, 1994:53-80.

45. Segal M, Webber NT. Nonstructured play observations: guidelines, benefits, and caveats. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:207-30.

46. Goodman JF, Pollack E. An analysis of the core cognitive curriculum in early intervention programs. Early Educ Dev 1993;4:193-203.

47. Meisels SJ. Designing meaningful measurements for early childhood. In: Mallory BL, New RS, eds. Diversity in early childhood education: a call for more inclusive theory, practice, and policy. New York: Teachers College Press, 1994:205-25.

48. Sparrow SS, Balla DA, Cicchetti DV. Vineland scales of adaptive behavior scales. Circle Pines, Minn, USA: American Guidance Service, 1984.

49. Ireton H. The child development inventory manual. Minneapolis, Minn, USA: Behavior Science Systems, 1992.

50. Barrera I. Thoughts on the assessment of young children whose sociocultural background is unfamiliar to the assessor. In: Meisels SJ, Fenichel E, eds. New visions for the developmental assessment of infants and young children. Washington, DC: Zero to Three: The National Center for Infants, Toddlers, and Families, 1996:69-84.

51. Harry B. Developing cultural self-awareness: the first step in values clarification for early interventionists. Topics in Early Childhood Special Education 1992;12(3):333-50.

52. Human MT, Teglasi H. Parents’ satisfaction and compliance with recommendations following psychoeducational assessment of children. J School Psychol 1993;31(4):449-67.

53. Hayes SC, Nelson RO, Jarrett RB. The treatment utility of assessment: a functional approach to evaluating assessment quality. Am Psychol 1987;42:963-74.