Table of Contents | Previous |
ATTACHMENT
ASSESSING EMOTIONAL AND SOCIAL COMPETENCE DURING PRESCHOOL YEARS
12.1. INTRODUCTION
Why do we need to assess preschoolers’ emotional and social competence? We have made clear throughout the previous chapters that we see universal social-emotional programming as a prime goal, so that every preschooler might be exposed to the salutary effects of their caregivers’ efforts to promote secure attachments, emotion knowledge, emotion regulation, and social problem-solving abilities. Nonetheless, we also urgently wish to reach those children who need us most. To meet both goals, psychometrically excellent assessment tools are important assets; that is, for all children it is important to document the changes wrought by social-emotional programming; as we related in earlier chapters on programming, evaluation research is often still needed. Further, when endeavoring to ameliorate risk processes and augment resilience processes of children already exhibiting social and emotional deficits, assessment is essential. Whenever possible, we must know the strengths as well as the weaknesses of each child, so that we may intervene appropriately.
We need to put these mandates into a broader perspective, however; the overall issue of assessment during early childhood, and its relation to school readiness and other decisions, is currently widely debated. Horton and Bowman (2002) have reported on expert opinion and state trends in preprimary assessment in general. They note that assessment during early childhood crucially needs scrutiny. Expanding early childhood education and child care enrollments, better scientific knowledge about early childhood development, and decisions about public spending, all oblige persons working with young children, and their parents, to carefully consider which assessment tools to use, as well as why and when to use them. We need to utilize assessments that yield the most-needed, developmentally grounded information, most economically and most ethically in terms of teacher, parent, and child time, effort, and attention.
Thus, in concert with NAEYC developmentally appropriate practice (Bredekamp & Copple, 1997), assessment should be integrated with the curriculum, beneficial to all parties, often based on ongoing teacher observation, primarily reliant on the child’s everyday activities, and pertinent to all learning and developmental domains. Data emanating from such assessment should, however, not be used for high stakes decisions, such as retention in kindergarten. Instead, assessment is performed to meet the needs of screening individual children to understand their strengths and weaknesses, to promote improved, individualized instruction, and to evaluate programming. Furthermore, no assessment tool can meet all of these needs, so that several tools are likely to be needed (Muenchow, undated).
The experts queried in Horton and Bowman’s (2002) Erikson Institute study suggested the following guidelines for preprimary assessment:
- Assessment couched within the program curriculum or, in anecdotal records, can and should occur almost daily. We want to know how children with whom we work are doing on a frequent basis so that we may seamlessly tailor and individualize our work with them.
- On a weekly basis, assessment can take place via teacher meetings, in which teachers discuss their unique views on children’s progress, and portfolios for each child, in which exemplars of this progress (or its lack) are documented.
- Parent evaluations and teacher checklists can be very useful if used on a quarterly basis.
- Individualized screening for disabilities and delays should be done yearly.
- Consultants, in the case of our concerns here, mental health consultants, can provide their important perspectives as needed.
- Standardized assessment tools completed by the children themselves are never indicated at this age level
Accordingly, better social and emotional assessment tools are sorely needed, to fit both the mandates we have put forward and the guidelines suggested in Horton and Bowman’s (2002) report. In 1996, we decried the then-current lack of assessment tools for measuring social and emotional development in preschoolers (Denham et al., 1996). Not only has there historically been a dearth of assessment tools, but also, those that do exist have often been hampered by a number of deficiencies. Some, like the Denver Developmental Screening Test-Revised, are severely enough limited in coverage to render them almost useless; others, like the Vineland Social-Emotional Early Childhood Scales (SEEC; Sparrow, Balla, & Cicchetti, 1998) are seriously lacking because, although they may include many important aspects of emotional and social functioning, they ultimately are too limited and include some very odd skills.
Nonetheless, as with programming, we consider that there are some “best bets” for assessment (see also Raver & Zigler, 1997). Coincident with the newfound interest in social and emotional competence and recently unearthed evidence of their value, a number of new, potentially useful SEL assessment tools have been developed or improved. We review these tools here, closely paralleling our earlier breakdowns of emotional and social tasks of early childhood, dividing them into those specifically addressing emotional competence, social competence, a combination of the two, or especially targeting screening for the presence of behavior problems. In our evaluation of each assessment tool, we consider whether important, salient issues and central developmental tasks in the social and emotional domains are appropriately captured. Finally, we make an initial effort at imagining “best practice” in preschool SEL assessment, considering in particular the fit of instruments reviewed here within Horton and Bowman’s (2002) guidelines.
12.2. EMOTIONAL COMPETENCE ASSESSMENT
Within the domain of emotional competence, our goals for assessment would be to understand group change and individual differences in the following:
- Attachment to caregivers: We would like to know the quality of children’s relationships with their caregivers—including, if possible, qualities of security, dependency, and conflict
- Emotional expressiveness: We would like to know the child’s enduring patterns of emotional expressiveness—does s/he exhibit a range of expressed emotions, or a smaller set, and with what frequency?
- Emotion knowledge: We would like to know how well the child has come to understand the names for emotions, the expressions corresponding to these names, and situations in which these emotions are common. We would also like to know whether the child is beginning to understand that others may have emotional reactions that differ from one’s own. Other aspects of emotion knowledge, such as of display rules and ambivalent emotions, are probably just emerging during the preschool age range, so that their assessment would not be too useful.
- Emotion regulation: We would like to know whether the child generally is able to regulate emotional expressiveness and experience, in order to continue activity and interaction with others. It would also be useful to be able to know which strategies a particular child uses for such regulation.
As well, all assessment tools should, as much as possible: (a) involve parents; (b) be culturally appropriate and able to accommodate linguistic needs of children from major language groups; (c) take developmental status of young children into account; (d) incorporate data from different sources over time; and (e) be easy to administer and easily understood (Muenchow, undated). Thus, in the discussion to follow, we attempt to converge upon a collection of assessment methods that will allow us to tap into these important domains in a developmentally appropriate manner.
12.2.1. Attachment To Caregivers
Three available measures of attachment to caregivers appear to have both psychometric and practical utility in highlighting the quality of adult-child relationships during this age period, from differing perspectives and viewpoints: (a) the Student-Teacher Relationship Scales; (b) the Attachment Q-Sort; and (c) attachment-related Narrative Story Stem Completions. We review each of these in turn.
12.2.1.1 Student-Teacher Relationship Scales
Pianta’s Student-Teacher Relationship scales (STRS; Pianta, 1997; Pianta & Nimetz, 1991; Pianta et al., 1995) provide benchmarks on teachers’ self-reported closeness, conflict, and dependency experienced with/from particular children. Psychometric properties are good, and the scales are quick for teachers to complete.
These relationship qualities persist across time and to some extent across teachers, also providing preschoolers, as noted in Chapter 3, with relationship stability they may not have built with their parent (Howes et al., 2000; Mitchell-Copeland et al., 1997). Thus, it could be important to document qualities of children’s relationships with their daycare or preschool teachers. In recent research, STRS scales were negatively related to externalizing behaviors in preschoolers; that is, children with whom teachers report closeness showed less aggression and other out-of-control behavior (Ramos-Marcuse & Arsenio, 2001).
We consider that, along with yielding information about a particular child’s status vis á vis attachment issues, this measure can be helpful to teachers in reflecting on their contributions to an affective relationship with the child. As a companion measure, the Teacher Relationships Interview (Stuhlman & Pianta, 2001) could be administered to teachers by mental health consultants, so that the caregiver/teacher could come to understand their own views of themselves as a secure base for children’s attachment, disciplinarian, teacher, and caretaker. Such information about both adult and child can be useful for ongoing program planning, as well as for yearly “status reports.”
Table 12.1. Example Items from the Student-Teacher Relationship Scale
| STRS Scale | Example Items |
|---|---|
| Closeness | I share an affectionate, warm relationship with this child; this child shares information about himself |
| Conflict | This child and I always seem to be struggling with each other; despite my best efforts, I am uncomfortable with how this child and I have gotten along |
| Dependency | This child reacts strong to separation from me; this child is overly dependent on me |
Note. Item content quoted with special permission of Publisher, Psychological Assessment Resources, Inc., 16204 North Florida Ave., Lutz, FL 33549, from the Student-Teacher Relationship Scale, by Robert C. Pianta, Ph.D. Copyright 2001, by PAR, Inc. Further reproductions prohibited without permission of PAR, Inc..
12.2.1.2. Attachment Q-Sort (AQS)
It is also useful to view the child’s attachment to parents, teachers, and caregivers, especially from an objective observer’s point of view. The Attachment Q-Sort is essentially an extended rating scale, using q-methodology (Waters & Deane, 1985), suitable for examining attachment-relevant child behaviors. That is, raters (either the adult in question—parent, teacher, or caregiver—or an independent observer) sort cards, upon which statements about children’s possible behaviors are written, into a fixed distribution of piles, depending on their similarity to the actual behaviors of the child in question. Measurement theory suggests that distinct advantages exist for rating scales—especially that raters can average their observations across contexts, with a commensurate decrease in error variance—resulting in scores that can be quite trustworthy. Further, q-sorts have special status even among rating systems, in that several errors that may afflict other types of rating scales, such as errors of leniency or central tendency, are alleviated by the force distributions required (see Figure 12.1).
There are two means of generating scores from this measure. In the first, the score distribution of all sorted cards describing the child in question is correlated with sorted cards describing “the optimally secure child” (as in Mitchell-Copeland et al., 1997; Waters & Deane, 1985). Criterion sorts exist for security, dependency, and sociability. In the second, scores for specific behaviors (with each card’s rating varying from 1 to 9, from “least like” to “most like” the child in question) are summed into scales to create categories of attachment organization that are conceptually consistent with organizational categories derived from other attachment assessments (Howes & Ritchie, 1999).
 |
| [D] |
To complete this measure, raters first become familiar with descriptions of attachment behaviors by reviewing the computer-based Attachment Q-Set Advisor (Waters, Posada & Vaughn 1994). Typically, training includes coding videotaped examples of visits with parents and young children, in order to assure interrater reliability. Then, observers generally observe the child and parent or caregiver for up to six hours, during which caregivers are encouraged to go about their usual activities and to treat the visitor as they would a visiting friend, neighbor, or helper. Observers make extensive notes during and after each visit and complete the AQS after the final visit.
Sample items include: “Child keeps track of adult’s location,” “If adult reassures, child will approach,” and “Child actively goes after adult if upset.” For security of attachment with teachers and caregivers, several items do not pertain; a subset of the 90-item Attachment Q-Sort can be used to describe children’s relationships with nonparental adults. Several items are not relevant to teacher attachment (e.g., those referring to bedtimes), so that these are omitted when observing nonparental attachment figures. These subscales and organizational categories are reliable and valid, and can be used adequately for preschoolers. In various studies (e.g., DeMulder, Denham, Schmidt, & Mitchell, 2000), AQS scores have been associated in theoretically meaningful ways with measures of behavior problems, social competence with peers, and teacher perceptions of child-teacher relationships. Thus, use of the AQS could give teachers, parents, and mental health consultants a picture of how the child compares to securely attached children on metrics that appear to be directly associated with SEL outcomes in the preschool classroom. Item analysis of important attachment-relevant behavioral exemplars on which a particular child scored low (e.g., keeping track of the presence of the adult) could yield specific behavior objectives for individual program planning. Thus, the AQS’ central value seems to lie in its use in individualizing programming, as well as in summarizing pre- and post-programming status, and in screening for insecure attachments. One dual disadvantage might be the need for training and the amount of time required for observation.
12.2.1.3 Narrative Story Completions: Children’s Views of Their Own Relationships
In this measure, trained experimenters administer six narrative story completions involving attachment themes, along with a warm-up story (Bretherton et al., 1990). The stories include, for example, separations from parents and child transgressions (e.g., spilling juice). Family figures and props are present for each narrative story. Codes for both structure (e.g., security) and content (e.g., aggression) are useful in evaluating narrative story completions. Each individual code is scored as present or absent for each story. A total score for each code is computed by taking the sum of its individual scores across all stories.
Initial use of this battery has indicated a potential for clinical usefulness (Oppenheim, Emde, & Warren, 1997); children with more negative parental representations in their narratives were rated as having more behavior problems, and their mothers rated themselves as having more psychological problems. Warren, Oppenheim, and Emde (1996) also found a positive relation between aggressive/emotionally negative narrative themes and ratings of children’s behavior problems; Ramos-Marcuse and Arsenio (2001) found a positive relation between children’s positive view of their attachment relationships and their social competence.
In another study (Denham, Blair, et al., 2002), 3-year-olds’ narrative story completion scores fit well enough with their AQS scores for security of attachment to mother, and to teacher, to form one “mega-composite” that was associated with emotional competence when interacting with peers. Children who scored as “more secure” on the mega-composite were more able to understand emotions of others, less likely to show anger toward peers, and more able to manage emotionally stressful events; they were also considered more socially competent by teachers and peers two years later, in kindergarten. As with the AQS, advantages of the narrative story completions are its use in individualizing programming, as well as in summarizing pre- and post-programming status, and in screening for insecure attachments. One disadvantage might be the need for highly trained testers/coders, and the need to videotape the administration of the measure.
12.2.2. Emotional Expressiveness
Two available measures of children’s emotional expressiveness appear to have both psychometric and practical utility, from differing perspectives and viewpoints: (a) Denham’s FOCAL Observation System; and (b) the Child Behavior Questionnaire (Rothbart, Ahadi, & Hershey, 1994). We review each of these.
12.2.2.1. Observed Emotional Expressiveness
Denham’s (1986) FOCAL Observation System accesses children’s happiness, sadness, anger, tension/fear, tenderness, and “other” emotions, and can be coded live to examine the emotions expressed by children during play. In this observational coding system, emotions are operationally defined by facial, vocal, and motor indices (Denham, 1986; Denham et al., 1990; Denham et al., 1997). Use of the measure is marked by good interobserver reliability. Both frequency and duration metrics are available.
Observation systems are expensive in terms of personnel time needed for implementation, but they can be invaluable. They can be utilized as follows: through observing a child for 12 or more 5-minute intervals across several weeks’ time, a profile of either the amount of time the child spends expressing each emotion, or the relative frequency of each. Is Tomas’ predominant emotion anger? Does Jimmy hardly ever show emotion except for the tension evidenced by his thumb chewing? Knowing these aspects of the children’s social and emotional functioning can aid teachers in tailoring their SEL programming. Further, the observational system could be used in pre- and post-programming documentation.
12.2.2.2. Rated Emotional Expressiveness: Temperament
Three higher-order temperament factors pertinent to the assessment of emotional expressiveness and regulation have been isolated: (a) negative affectivity, (b) surgency; and (c) effortful control (Rothbart et al, 1994). Taken together, these factors comprise a child’s constitutional, individual pattern of self-regulation and reactivity, and are considered relatively enduring biological predispositions that are influenced over time by both maturation and experience (Rothbart et al, 1994).
How do these temperament dimensions map onto the construct of interest here—emotional expressiveness? Many children high on the temperament dimension of negative affectivity are easily angered in many situations. Others high on this temperament dimension are also anxious, fearful in new situations, and easily saddened. It is easy to see how this potent combination could make interacting with both peers and adults problematic.
Surgency is an aspect of temperament associated with extraversion, approach to novel stimuli, positive emotional expressiveness, activity, and high level pleasure. Hence, a child high on this dimension of temperament might be a lot of fun to be around-eagerly initiating contact with others, finding interesting things to do, sharing positive affect. On the other hand, there could be “too much of a good thing,” with children high on such a dimension possibly seen as irritatingly active and boisterous, risk-taking, and impulsive (Rothbart & Ahadi, 1994; Rothbart, Ahadi, & Evans, 2000).
For usage of the Child Behavior Questionnaire (Rothbart et al., 1994), we would choose parental report, in accordance with the conclusions of Rothbart and Bates (1998), because: (1) parents see a wide range of behavior; (2) recent measurement advances allow their reports even greater objective validity; and (3) most importantly, the social relationship aspects of child temperament are best captured in parental reports.
The CBQ is an instrument that assesses temperamental characteristics of children aged 3-8 years. Parents rate, on seven-point scales, how “true” 195 specific descriptive behaviors have been of their child over the past six months. The option of indicating that any item is “not applicable” to the child is also available.
The 15 CBQ scales yield the following temperament groups referring to emotional expressiveness (Rothbart et al, 1994): 1) negative affectivity and 2) surgency. Negative affectivity items involve discomfort experienced in over-stimulating situations, frustration, anger, and inability to soothe oneself, fearfulness, and sadness. The Surgency dimension includes active, approach, pleasure, and smiling scales. Internal consistency reliability is good for these scales, as is test-retest reliability. Their use can add to both parents’ and teachers’ knowledge of children’s expressiveness across many everyday contexts.
It is important to note that, although temperament is assumed to have a strong bio-behavioral component, it is not immune to modification via maturational, environmental, or relational means. As such, these scales, or internally consistent abbreviations thereof, could be useful perhaps on a quarterly basis for program planning and progress reporting, as well as at the beginning and ending of programming for evaluation purposes.
12.2.3. Emotion Knowledge
Denham’s Affective Knowledge Test (AKT; 1986) utilizes puppets to measure preschoolers’ developmentally appropriate understanding of emotional expressions and situations. Children’s understanding of emotion is assessed using puppets with detachable faces that depict happy, sad, angry, and afraid expressions. First, children are asked to both verbally name the emotions depicted on these faces, and then to nonverbally identify them by pointing. This procedure taps into their ability to recognize expressions of emotion.
Then, in two subtests of emotion situation knowledge, the puppeteer makes standard facial and vocal expressions of emotions while enacting emotion-laden stories, such as fear during a nightmare, happiness at getting some ice cream, and anger at having a block tower destroyed. Children place on the puppet the face that depicts the puppet’s feeling in each situation (Denham, 1986; Denham & Couchoud, 1990a; Denham et al., 1994). In eight situations, the puppet feels emotions that would be common to most people, such as those mentioned above. Finally, children are asked to make inferences of emotions in nonsterotypical, equivocal situations. This subtest measures how well children identify others’ feelings in situations where the “other” feels differently than the child. All the situations that the puppeteer depicts during this section of the measure could easily elicit one of two different emotions in different people, as in feeling happy or afraid to get into a swimming pool. Before the assessment, children’s parents report, via forced-choice questionnaire, how their children would feel; these responses determine the emotions expressed by the puppet. For example, if the parent reports that the child would be happy to come to preschool, the puppet is depicted feeling sad. Internal consistency and test-retest reliabilities are good (Denham, Caverly, et al., 2002; Denham & Couchoud, 1990a, 1990b; see also Dunn, Slomkowski, et al., for AKT relations with later indices of emotion knowledge at age six).
This measure appears to be especially ecologically valid, as it requires little verbalization and is performed during play. Scores on the AKT are slightly to moderately related to other indices of SEL. For example, researchers have found that children’s concurrent AQS attachment ratings are related to scores on the measure (Denham, Caverly, et al., 2002; Laible & Thompson, 1998); more secure children perform better on the AKT. Moreover, predominantly happier, less angry children also tend to perform better (Denham, 1986; Denham et al., 1990; Denham et al, 2003). Furthermore, AKT scores are related to other indices of SEL, such as moral sensibility and decision-making (Dunn, Brown, & Maguire, 1995), conflicts and interactions with friends (Dunn & Cutting, 1999; Dunn & Herrera, 1997). Finally, AKT scores are both concurrently and longitudinally related to peers’ and teachers’ evaluations of children’s social competence (Denham et al., 1990; Denham et al., 2003). Thus, knowing a child’s status on this measure can help teachers not only in knowing about his or her emotion knowledge, but also to prognosticate about skills to which the AKT is related. The AKT is easy to learn and to administer, children enjoy it, and it takes only about 20 minutes to perform. We consider it a useful assessment tool to document status and change in emotion knowledge as a key aspect of SEL; it has already demonstrated its usefulness in this role (Domitrovich et al., 2002; Shields et al., 2001).
12.2.4.1. Emotion Regulation
What do we need to know about young children’s abilities to regulation their emotions? We need to know at least two aspects of their emotion regulation: (a) the “end product”: do children have difficulty with regulating their emotions, leaving them vulnerable to extreme, long-lasting, and/or difficult to calm positive or negative emotions? and (b) the “process”: in their efforts to regulate either positive or negative emotions, what exactly do children do? What are their strategies? Developmentalists and early childhood educators vary on their focus when examining emotion regulation. In our review of instruments, we found one combined process and product assessment task, and two product questionnaires.
12.2.4.1. Emotion Regulation as Process and Product: An Analogue Task
Raver et al. (1999) have had success in using a modification of Mischel’s self-regulation task (Mischel, Shoda, & Rodriguez, 1989) with Head Start youngsters. In this task, children are asked to wait to open a tempting gift, until the examiner returns from retrieving something s/he forgot to bring to the testing room. The child’s behaviors are coded along continua reflecting their ability to regulate emotions, and the strategies that they use to do so (i.e., both “product” and “process” as indicated above). In Raver et al’s study, children’s use of self-distraction predicted peer and teacher reports of children’s social competence.
The ease of administration and coding of this assessment tool, its ecological validity, and its apparent power in describing emotion regulation process and product (not only during preschool, but predictively to adolescence; see Mischel et al., 1989), make it a viable candidate to include in our armamentarium. We see it as potentially useful for ongoing assessment during programming, as well as pre- and post-programming evaluation.
12.2.4.2. Rated Emotion Regulation as Product: Teacher Ratings
Shields and colleagues (Shields & Cicchetti, 1997; Shields et al., 2001) also have a 24-item Emotion Regulation Checklist that taps both prevalent emotional expressiveness and the product aspect of emotion regulation; that is, it targets processes central to emotionality and regulation, including affect lability, intensity, valence, flexibility, and contextual appropriateness of expressiveness. Internal consistency for the emotion regulation and lability/negativity subscales is excellent (Shields et al., 2001). In terms of validity, the measure distinguishes well regulated from dysregulated children (Shields & Cicchetti, 1997). More specifically, with preschoolers, Shields et al. (2001) also found that overall emotion regulation at the start of the preschool year was associated with school adjustment at year’s end, whereas early emotional lability/negativity predicted poorer outcomes. Thus, we see this tool as potentially extremely useful for ongoing assessment during programming, as well as pre- and post-programming evaluation.
| Emotion Regulation Checklist Scale | Example Items |
|---|---|
| Lability/Negativity | Exhibits wide mood swings; is easily frustrated; is prone to angry outburts |
| Emotion Regulation | Is a cheerful child; responds positively to neutral or friendly overtures by adults; can say when s/he is feeling sad, angry or mad, fearful or afraid |
12.2.4.3.1. Rated Emotion Regulation as Process: Parent Ratings
Effortful control is an aspect of temperament associated with sensitivity to the emotional experiences of peers, which can lead to empathic and other prosocial responses, as well as to inhibition of aggressive impulses (Kochanska, 1993; Rothbart et al., 1994). More specifically, regulatory abilities in attention, in particular the ability to focus and shift attention voluntarily, and the ability to disengage attention from one’s own perspective to attend to another’s, are hallmarks of prosocial development (Kochanska, 1993; Rothbart et al, 1994). Thus, we would expect children higher on the effortful control dimension to be seen by teachers, observers, and peers alike as more socially competent.
The CBQ also yields this aspect of temperament (Rothbart et al, 1994). Effortful control encompasses scales measuring inhibitory control; maintenance of attentional focus during tasks; pleasure experienced during low intensity situations (e.g., looking at picture books); and perceptual sensitivity and awareness of external cues. Again, as for negative affectivity and surgency, internal consistency reliability is good for these scales, as is test-retest reliability. The instructions are generally clear and useful to raters, although the scales include many items and take some time to complete. As noted for the CBQ’s measurement of emotional expressiveness, its scales related to emotion regulation, or internally consistent abbreviations thereof, could be useful perhaps on a quarterly basis for program planning and progress reporting, as well as at the beginning and ending of programming for evaluation purposes.
12.2.4.4. Rated Emotion Regulation as Process: Strategies Reported by Parents or Teachers
Children’s coping behavior when faced with emotional situations with peers can be assessed with items developed by Eisenberg, Fabes, Nyman, Bernzweig, and Pinuelas (1994). Informants indicate on a 7-point scale, with 1 indicating “never” and 7 indicating “usually,” how often the child would engage in each of 12 general types of coping behavior when confronted with a problem situation. Item content reflects: Instrumental Coping (e.g., taking action to improve a situation), Instrumental Aggression (e.g., hitting), Emotional Intervention (e.g., crying to elicit help), Avoidance (e.g., leaving a problem), Distraction (e.g., keeping busy), Venting (e.g., crying to release frustration), Emotional Aggression (e.g., aggressing to release frustration), Cognitive Restructuring (e.g., saying “I don’t care”), Cognitive Avoidance (e.g., not thinking about the problem), Instrumental Intervention (e.g., getting help), Instrumental Support (e.g., talking to someone about the problem), and Denial (e.g., saying nothing happened). Based on the work of Eisenberg and colleagues (1994), data can be reduced to three summary scales:
- Emotional Venting (e.g., cries to release feelings/get help, solves problems/releases feelings through aggression).
- Constructive Strategies (e.g., getting emotional support or pragmatic assistance with the problem; solving the problem)
- voidant Strategies (e.g., using distraction, denying the problem)
Research suggests that parent’s completion of these scales is related to teacher’s evaluations of young children’s social competence (Denham & Blair, 2002).
These scales are very quick and easy for either teachers or parents to complete. They yield a snapshot of the child’s emotional coping strategies, and would thus be useful for individualizing programming as well as pre- and post-programming measurement.
12.3.1. SOCIAL COMPETENCE ASSESSMENT
Within the domain of social competence, our goals for assessment would be to understand group change and individual differences in children’s effectiveness in interaction, at the middle level of our model (see Figure 10.1). That is, we would like to know from varying perspectives—those of teachers, parents, and other children—whether some or all of the skills (i.e., those for which we can find assessment tools) at the lowest level of the model are attained.
Three available measures of preschoolers’ social competence appear to have both psychometric and practical utility, from teachers’ or peers perspectives and viewpoints: (a) the Social Competence/Behavior Evaluation; (b) the Penn Interactive Preschool Play Scales; and (c) peer sociometrics as tailored for use with preschoolers. We review each of these in turn.
12.3.1.1.Teacher Evaluations
12.3.1.1. Social Competence/Behavior Evaluation Short Form
The Social Competence/Behavior Evaluation (SCBE; LaFreniere & Dumas, 1996), which has been extensively normed with stratified samples of French Canadian and American preschoolers, is one possibility. It has internally consistent subscales for Externalizing, Internalizing, and Cooperation/Sensitivity. As such, it taps many of the components of social competence already outlined here (see also Denham et al., 1997). The measure also has been translated into Spanish.
| SCBE Scale | Sample Items |
|---|---|
| Aggression | Gets into conflicts with other children; opposes the teacher |
| Withdrawal | Doesn't talk or interact during group activities; avoids new situations |
| Cooperation/Sensitivity | Negotiates solutions to conflicts (note social problem-solving content); cooperates with other children |
The SCBE does, however, include many items that might be considered more purely “emotional” rather than “social” in nature. When these items are omitted, “purer” social competence scores can be obtained; Denham has created one overall score in her research by subtracting aggression and withdrawal scores from social sensitivity and cooperation scores. The “purer” scales are still associated with elements of children’s earlier emotional competence, including attachment, emotional expressiveness, emotion knowledge, and emotion regulation (Denham et al., 2003), and are extremely highly correlated with the original SCBE scales. Thus, we consider this measure to be an excellent candidate for yearly—or even quarterly—screening or ratings, as well as pre- and post-programming group results.
12.3.1.2. Penn Interactive Preschool Play Scales
“Play is an important vehicle for children’s social, emotional, and cognitive development, as well as a reflection of their development” (Bredekamp & Copple, 1997 p. 6). Consonant with this view, one measure that derives information on young children’s social competence, in context, is the Penn Interactive Preschool Play Scale (PIPPS; Fantuzzo, Sutton-Smith, Coolahan, Manz, Canning, & Debnam, 1995; McWayne, Sekino, Hampton, & Fantuzzo, 2002). This relatively new assessment tool has much to support it.
In this measure, informants, whether teachers/caregivers or parents, report on the rate of occurrence of developmentally appropriate behaviors within concrete, observable contexts in which preschoolers are actively engaged: their various play environments. The PIPPS offers an advantage to both teachers and parents: because young children’s play is so salient a part of their daily activities, informants have ample opportunities to observe it, and are likely to have the skills to understand and reliably complete a measure grounded in this phenomenon. Thus, informants are not required to list or describe behaviors—processes that are open to social desirability and other errors, both systematic and nonsystematic.
The PIPPS yields three overarching scales: (1) Play Interaction—i.e., how creative, cooperative, and helpful children are during play; (2) Play Disruption—i.e., how aggressively and antisocially they behave during play; and (3) Play Disconnection—how withdrawn and avoidant children are in contexts where engaged play is more normative. These scales are internally consistent for both teachers and parents, and appear equally appropriate for low-income children of varying ethnicities, including African American and Hispanic (Fantuzzo, Coolahan, Mendez, McDermott, Canning, & Debnam, 1998; Fantuzzo & McWayne, 2002). In terms of validity, parents’ PIPPS scales are related to teacher PIPPS scales. As well, positive learning styles, the Social Skills Rating System, conduct problems, emotion regulation, and sociometric acceptance are also related in theoretically expected ways to the scales.
Because of these qualities and distinct advantages, we also recommend the PIPPS for yearly, or even quarterly screening or ratings, as well as pre- and post-programming group results. Whether the SCBE-30 or PIPPS is “better” is, however, something of a moot question. Both use simple vocabulary; both yield very similar subscales. Thus, the choice may be at the discretion of the user; we would recommend consideration of the PIPPS especially for the subpopulations whose needs were considered when it was developed.
12.3.1.2.Peer Evaluations
Even at an early age, it is important to consider how peers view the effectiveness of their classmates’ social interactions; their perspective is unique. Neither teachers nor parents can possibly see all social interactions in which children participate; furthermore, even if they could it remains important to get the point of view of those to whom the interactions are directed.
Sociometric measurement is a very robust indicator of peers’ assessments of preschoolers’ likeability (Denham & Holt, 1993; Denham et al., 1990; Howes, 1988). In these techniques, as used with preschoolers, children are asked to examine photographs of their classmates, and to indicate via a pictorial scale whether they “like a lot” “kinda like” or “don’t like” each person in the group of photographs. There also are other modifications possible in using the technique with preschoolers, including using only ratings of likeability rather than including “don’t like” ratings (e.g., Lemerise, 1997). In any case, the results of this process, summed across children, show how well liked the rated child is, overall.
Alternatively, preschoolers can nominate children who fit specific categories, although there is somewhat lower reliability for such assessments (Denham & McKinley, 1993; cf. Hymel, 1983). For example, to capture overall social status in each group, children can be asked to name an unlimited number of children who they “like a lot” and who they “don’t like very much.” As well, to identify aggression in their peers, they can be asked to name children who “start fights,” “yell and call other kids mean names,” “hit and push other kids.” From such nominations, social preference (number of times mentioned as liked), social impact (number of times mentioned), overt aggression (number of times mentioned as a fighter, etc.), and relational aggression (number of times mentioned as a name-caller) can be determined. Other positive qualities, such as “is fun to play with” can also be used for nominations. Many developmental studies in the last two decades support the reliability and validity of these procedures.
Despite some parents’ and teachers concerns, participation in sociometric measures involves no more risk (e.g., of “telling”—how one rated another in a mean way) than it does in children’s everyday social life. Researchers (e.g., Bell-Dolan, Foster, & Christopher, 1992; Bell-Dolan, Foster, & Sikora, 189; Bell-Dolan, Foster, & Tishelman, 1989) have found, following administration of sociometrics measures, no increase in negative interactions with unpopular peers, no increase of social withdrawal in less accepted children, and no expression of unhappiness or loneliness after participation in studies with sociometric measures. Most children appear to enjoy considering such issues and do not change their behavior, and may in fact benefit from discussion of such issues with researchers.
From the results of research on the use of sociometrics measures, Bell-Dolan and colleagues (see also Hayvren & Hymel, 1994, Ratiner, Weissberg, & Caplan, 1986) make the following recommendations for the administration of sociometric measurement:
- Use a distracter task after the sociometrics measure, for example, nomination of favorite songs or TV characters.
- Embed sociometrics within another task, so that the questions are not overemphasized.
- Do not administer right before times when children would be more likely to discuss their ratings or nominations—e.g., lunch, free time, or dismissal for the day.
- Carefully monitor effects of the measurement on children.
Sociometric techniques have much potential use in evaluating from the perspectives of the “consumers,” so to speak, pre- and post-programming results, as well as preschoolers’ yearly or semi-yearly social status. One way to assuage potential concerns that important adults in children’s lives may have about such measures would be to better educate them on the true nature of the measure, its ecological validity, and the importance of social status to concurrent and later functioning (Roff, 1990; Roff & Ricks, 1970). Such education should be embedded within better communication about SEL, and ways to attain it.
12.3.2. SOCIAL COMPETENCE/EMOTIONAL COMPETENCE “COMBINED ASSESSMENT”
Some assessment tools, notably many teacher checklists and/or curriculum- or programming-specific measures, tap constructs of both emotional and social competence in their comprehensive views of the child’s SEL. Five available “combined measures of children’s emotional expressiveness appear to have both psychometric and practical utility: (a) The Devereux Early Childhood Assessment; (b) the Battelle Developmental Inventory-2; (c) the Infant Toddler Social-Emotional Assessment; (d) the Minnesota Preschool Affect Checklist; and (e) “authentic” assessment tools, such as the Hawaii Early Learning Profile Preschool Assessment Strands. All are teacher and/or parent checklists, except for (d). We now review each.
12.3.2.1. The Devereux Early Childhood Assessment (DECA)
The DECA (LeBuffe & Naglieri, 1999; Yonamine, 2000) is a newly developed standardized, norm-referenced measure of resilience, completed by parents and teachers in a collaborative and supportive partnership. This measure is theoretically and psychometrically sound; subscales on initiative, attachment, self-control, and behavioral concerns are rated on a 5-point scale varying from “never” to “very frequently”. We like the DECA’s subscale demarcation, closely mirroring as it does our notions of SEL. Furthermore, its utility is being demonstrated. For example, the total resilience score (i.e., Initiative + Attachment + Self-control) is related to school readiness, as assessed by the Learning Accomplishment Profile-D cognitive and language scales (Devereux Early Childhood Initiative, 2001b).
The developers of the measure suggest that it be administered three times per year, with frequent updating allowing for periodic reviews of children’s progress. Teacher and parents complete the DECA Record Form to create a complete picture of the child. DECA results are shared and discussed with parents, and parents are included when planning strategies to help their child build protective factors and address behavior concerns. Teachers also can use these results to evaluate the effectiveness of strategies in supporting development of protective factors and minimizing or eliminating challenging behaviors. The DECA is already being used to document pre- and post-programming change (Devereux Early Childhood Initiative, 2001a, 2001c).
Another part of the DECA assessment system is the Reflective Checklist. The Reflective Checklists focus on the environment, daily programming, activities and experiences, supportive interactions, and partnerships with parents. In other words, teachers note what is working, or not, with each child and with groups of children. Strategies that foster childhood resilience are developed from the reflective checklists and DECA rating scales for both individual children and for the group. This process also supports the development of a partnership between the early childhood professional and parent working together to enhance the social and emotional health of the child.
Finally, classroom staff also are encouraged to reflect and improve upon their own skills in working with families by using strategies such as: (a) learning about each child’s family, culture and community; (b) using children’s home languages at the program; (c) establishing an ongoing system for exchanging information about each child with his or her family; and (d) giving families information about typical developmental skills and behaviors of young children.
| DECA Subscales | Sample Items |
|---|---|
| Initiative | Try or asks to try new things or activities |
| Attachment | Trust familiar adults and believe what they say; seek help from children/adults when necessary |
| Self-Control | Keep trying when unsuccessful (act persistent); calm herself/himself down when upset |
| Behavioral Concerns | Destroy or damage property, fight with other children |
12.4.2 Battelle Developmental Inventory (BDI)
This promising norm-referenced, standardized instrument (Newborg, Stock, & Wnek, 1984) has a new revision under way, for which Denham was a consultant on the Personal-Social Domain, which includes social/emotional items. Tracking of individual progress and evaluation of instructional programs, our foci in this chapter, are both included in the BDI’s purposes.
The BDI’s comprehensiveness, standardized test scores, empirically based age placement of its approximately 130 items, behaviorally anchored item descriptions, and improved, easier administration and scoring, all maximize its usefulness. Each skill item chosen for the new revision for the Personal-Social (and all other domains) has gone through a rigorous process of judgment on how critical or important it is to a child’s development. The BDI is designed to be used by teachers, diagnosticians, and multidisciplinary teams. Its authors consider it useful for screening and or for more in-depth assessment of specific nonhandicapped or handicapped children’s strengths and weaknesses for programming, as well as to help demonstrate the effects of programming. Use of a transdisciplinary assessment format also is possible.
Items on adult interaction, expression of feelings/affect, self-concept, peer interaction, coping, and social role are included. For example, topic areas include showing appropriate affection toward people, pets, or possessions, using adults appropriately to help resolve peer conflict, recognizing the feelings of others, and recognizing the basic similarities of all children. Excellent reliability data for the last version (before the current revision) are reported, with very small standard errors of measurement and high test-retest reliabilities.
Because some items can be scored via interview or observation methods, or structured format items can be corroborated via these methods, it behooves the examiner to gather all possibly relevant data on Personal-Social items before scoring. The examiner also must apply, when there are disagreements among these sources of data, a standard set of decision rules.
12.4.3. Infant Toddler Social-Emotional Assessment (ITSEA)
The ITSEA is a new adult-report measure for 12- to 36-month-olds. As such, it applies only to the lower ages in the range upon which we are reporting here, but it is nonetheless very promising as a tool to identify important aspects of SEL. For the age range with which we are concerned here, scales for Attention, Compliance, Prosocial Behavior, Peer Interaction, Empathy, Emotional Positivity, Task Mastery, and Emotional Awareness are included, all rated on a 3-point scale. Contrasting Externalizing Behavior (e.g., Activity, Peer Aggression, Aggression/Defiance, and Negative Emotionality Scales) and Internalizing Behavior (e.g., Inhibition/Separation Difficulties, Depression/ Withdrawal Scales) factors are also included. For all, internal consistency and test-retest reliabilities are good to excellent. Validity evidence is also encouraging (Briggs-Gowan & Carter, 1998; Carter & Briggs-Gowan, 1993; Carter, Briggs-Gowan, & Kogan, 1999). For example, relations with dimensions of temperament, attachment, emotion regulation, and coping, as well as age and gender, are as expected.
In short, the ITSEA relates as expected to several of the SEL constructs discussed here, such as attachment, emotional expressiveness and emotion regulation. In this respect it is capturing much that interests us in furthering the SEL of the younger preschoolers in the age range we cover in this volume. It should be of value for summarizing pre- and post-programming status, and to screen for possible delays.
| ITSEA Scale | Subscale | Example Items |
|---|---|---|
| Social-Emotional Competence | Attention | Can sit for 5 minutes while youread a story; can pay attention for a long time (not including TV) |
| Compliance | Follows rules; is easy to take care of | |
| Prosocial Peer Interaction | Is liked by other children; shares toys and other things | |
| Emotional Positivity | Laughs easily or a lot; is affectionate with loved ones | |
| Empathy | Is worried or upset when children cry; tries to help when someone is hurt | |
| Emotional Awareness (2-year-olds only) | Talks about own feelings; is aware of other people’s feelings | |
| Mastery Motivation (2-year-olds only) | Wants to do things for self; is curious about new things | |
| Externalizing Behavior Problems | Activity | Is restless and can't sit still; goes from toy to toy faster than other children his/her age. |
| Aggression/Defiance | Acts aggressive when frustrated; is disobedient or defiant | |
| Peer Aggression | Fights with other children; is mean to other children on purpose | |
| Emotional Negativity | Often gets very upset; cries a lot | |
| Internalizing Behavior Problems | Inhibition/Separation | Is very clingy; is shy with new people |
| Depression/Social Withdrawal | Seems withdrawn; seems very unhappy, sad, or depressed. |
12.4.4. Minnesota Preschool Affect Checklist
The Minnesota Preschool Affect Checklist observation schedule was first presented in Sroufe et al. (1984). This instrument, adapted by Denham and colleagues for Denham and Burton (1996), includes 53 items organized into “mega”-scales for positive and negative affect, inappropriate affect, positive and negative involvement (e.g., impulsivity, aggression, wandering, social isolation), peer skills, and empathy/prosocial behavior. Thus, many elements of emotional competence discussed in this volume, as well as some elements of social problem solving (e.g., deals with frustration by verbalizing the problem), and numerous relationship skills (as considered here in the model of social competence), are tapped by the MPAC.
| MPAC Scales | Exemplars of behaviors observed |
|---|---|
| Expression and regulation of positive affect | Displays positive affect in any manner-facial, vocal, bodily; shows ongoing high enjoyment (30 sec. or more) |
| Expression and regulation of negative affect | Uses negative affect to initiate contact, to begin a social interaction with someone; uses face or voice very expressively to show negative affect |
| Inappropriate affect | Expresses negative affect to another child in response to the other’s neutral or positive overture; takes pleasure in another’s distress |
| Productive involvement in purposeful activity | Engrossed, absorbed, intensely involved in activity; independent—involved in an activity that the child organizes for himself |
| Unproductive, unfocused use of personal energy | Wandering; listless; tension bursts |
| Lapses in impulse control | Context-related, physical, interpersonal aggression; inability to stop ongoing behavior; becomes withdrawn |
| Positive management of frustration | Promptly expresses, in words, feelings arising from problem situation, then moves on; shows ability to tolerate frustration well even if does not verbalize |
| Skills in peer leading and joining | Successful leadership; inept attempts at leadership; smoothly approaches an already ongoing activity |
| Isolation | No social interaction continuously for 3 minutes or more |
| Hostility | Unprovoked, physical, interpersonal aggression; hazing, teasing, or other provocation or threat |
| Prosocial response to needs of others | Interpersonal awareness-behavior reflecting knowledge or awareness about another person; helping behavior |
Trained observers watch children’s behaviors for 5-minute intervals, noting the presence of items. Previous research has shown good interobserver reliability for “mega”-scales, and concurrent validity with other indices of young children’s SEL (Denham, Zahn-Waxler, et al., 1991; Sroufe et al., 1984). Although training for observation using the MPAC is somewhat time-consuming, the detailed description of the child’s SEL across as few as four five-minute periods makes it a worthy candidate for use as pre- and post-programming assessment. For example, in Denham and Burton (1996), several of these scales, notably skills in peer leading and joining, showed change across preprogram to post-program periods, with those showing pre-measure deficits especially benefiting from the program.
12.4.5. “Authentic Assessment”
Although the field of developmental/school psychology still leans toward norm-referenced assessment (which may have its place when appropriate norms are available), criterion-referenced assessment can be preferable for many objectives (Bagnato, Neisworth, & Munson, 1997). After all, in the classroom or at home we probably care whether a child attains a certain social-emotional objective, perhaps more so than how that child compares to others. In any case, the key is to link assessment and programming. Each community, program, and child serve as their own controls, with previous performance as the point of reference to construct slopes of expected and actual performance profiles. Such assessment can be intimately tied to the lessons and curricula being used.
12.4.5.1 The Hawaii Early Learning Profile
The HELP Strands Preschool Version is an example of social and emotional authentic assessment tools (Vort Corporation, 1999) that can enable us to plan for intervention. This criterion-referenced assessment tool includes important criterion-referenced objectives on social-emotional scales for attachment/separation/autonomy, development of self, expression of emotions and feelings, learning rules and expectations, social interactions and play, social language, and personal welfare/safety.
One of HELP’s most applauded features is its high number of specific skills and allied intervention strategies. The HELP Strands report qualitative descriptions of emotional and social competence areas, along with approximate developmental age levels, which can be utilized to substantiate need for intervention, but is not equivalent to a score on a norm-referenced scale. The density of skills represented helps families and educators with curriculum planning, identification of strengths and weakness, and monitoring of children’s progress in small, incremental steps.
All of the items appear to have excellent content validity. Level, style, ability to interaction, customary behaviors, unique description of child and planning objectives as well as specific and detailed connections to classroom practices and family involvement. The assessment takes advantage of the child’s spontaneous behavior (e.g., reaction to new places and people, parent-child interaction, behavior during transitions between activities). Credit codes are well differentiated.
Thus, the HELP strands are designed for use with young children who are delayed, have disabilities, or are considered at risk. However, the skills it enumerates are listed in chronological order within the age at which the skills are generally acquired, so that the scale may also be of use to track the ongoing progress of nonhandicapped preschoolers, both individually and for groups. Our main concern with this assessment tool is the weight given to developmental milestones that do not “fit” within the emphases communicated in this volume. Of course, it was not the authors’ goal to do so! However, we would like to see more specific assessment goals pertaining to emotional expressiveness, emotion knowledge, emotion regulation, social problem solving, and relationship skills. This said, it should be noted (see Table 12.7) that the HELP Strands Preschool Version in fact includes some very good exemplars of these domains.
| HELP Strand | Example skills |
|---|---|
| Attachment/adaptive skills |
|
| Responsibility/Rules |
|
| Social Interactions and Play |
|
12.4.5.2.1. DECI Strategies
The DECA Program can also be seen as a comprehensive means of authentic assessment for both social and emotional competence. In implementing the DECI program, early childhood teachers and care providers are urged to conduct running record observations of the children in their classrooms, for several reasons. This type of observation provides accurate, objective and complete information without including the observer’s judgments or biases. These observation recordings can provide data on specific aspects of attachment, emotional expressiveness and emotion regulation, as well as initiative and behavioral concerns, which can be shared with parents and used by professionals and specialists from different disciplines. The following guidelines are suggested by the Devereux Foundation (Devereux Early Childhood Initiative, 2001d) for such running record observations (see also Chapter 5):
- Conduct several 10 to 15 minute observations, rather than observing for 20 to 30 minutes at a time.
- Observe several children at a time while they are engaged in small group activities.
- Target one child per day.
- Arrange the environment to create clear observation paths.
- Wear an “observation hat” to let children know you are busy for a few minutes.
- Plan time into your daily schedule to observe. Note the observer, target child(ren), time and place.
- Develop and use a shorthand system so you can write quickly and capture more detail.
- Plan for more adult support (aides, volunteers, parents) during observation periods. Provide on-going training on conducting running records.
12.5. ASSESSMENT OF BEHAVIORAL PROBLEMS
Even though we heavily emphasize the development of the positive in this volume, it is nonetheless true that, as a particular child’s ratio of resilience to risk processes decreases, more and more we need to fully examine and document his/her behavioral difficulties in detail. Thus, we need to explore assessment tools for behavioral problems—both internalizing and externalizing—as well as those for the emotional and social competence aspects of SEL. Overt behavioral problems have important bearing on whether and how children can acquire both emotional and social competence.
A number of possibilities exist to evaluate evidence of the intensity of preschoolers’ challenging behaviors (i.e., externalizing, internalizing, or “other”): (a) the Child Behavior Checklist (CBCL)/2-3 (Achenbach, 1987); (b) Feil, Walker, and Severson‘s (1998) Early Screening Project (ESP); (c) Hresko, Miguiel, Sherbenou, and Burton‘s (1994) Developmental Observation Checklist System (DOCS); (d) Merrell’s Preschool and Kindergarten Behavior Scale, 1994; and (e) Sinclair, Del-Homme, and Gonzalez‘s, (1993) measure, for example. Feil et al. (1998) and Sinclair et al. use gating procedures, in which presence of a criterion number of screening items scored “present” leads to further assessment, and finally to the specified need for services. The Feil, Walker, Severson, and Ball (2000) measure has been shown to be appropriately applicable to the multicultural milieu of Head Start. As well, the ITSEA, SCBE, DECA, MPAC all address the issue of behavior problems.
However, although these measures all appear to have good psychometric qualities and some practical utility, if teachers or program administrators deemed a separate behavior problems scale necessary and potentially useful, we would advocate the use of the Adjustment Scales for Preschool Intervention (ASPI; Fantuzzo, Bulotsky, McDermott, Mosca, & Lutz, 2002; Lutz, Fantuzzo, & McDermott, 2002). These scales are based on contextualized items actually developed with preschool early childhood teachers. Caregivers are directed to choose contextual descriptions of behaviors (if any) that have applied to the child within a time period of one month.
For example, for a question “How does this child cope with new learning tasks?” choices include “won’t even attempt it if he/she senses a difficulty,” “approaches new tasks with caution, but tries”; for a question “How is this child at free play/individual choice?” choices include “engages in appropriate activities, “rather loud but not disruptive,” “disturbs others’ fun.” Thus, the rater does not have to consider the child’s behavior out of context. Other contexts include how the child helps with classroom jobs, answers questions, talks with the teacher, takes part in games with other children, behaves in the classroom, respects other people’s belongs, and who the child has as companions. Contextualizing the actual behaviors of children within specific situations helps early childhood teachers to accurately report behavioral difficulties, which they sometimes are reluctant to do for fear of unfairly labeling ones so young.
Items yield five internally consistent major problem scales across their contexts—aggression (22 items), withdrawn-low energy (18 items), socially reticent (12 items), oppositional (10 items), and inattentive/hyperactive (11 items). A secondary factor analysis has yielded two overarching undercontrol and overcontrol scales, which coincide well with current diagnostic thinking. The original five scales show good concurrent validity with the PIPPS. For example the two tests form robust aggregates of interpersonal disruption (a combination of the ASPI aggression and PIPPS play disruption scales), under active disconnection (a combination of the ASPI withdrawn, reticent with PIPPS play disconnection and lack of play interaction scales), and oppositional disconnection (a combination of ASPI oppositional and PIPPS play disconnection scales). Thus, if particular pressing needs for using a behavior problems scale exist within a program, this measure would be our recommendation. Given our predilection for promoting SEL rather than focusing on problems, and the ASPI’s relationship with the PIPPS, it is also possible that its use would be sufficient.
12.6. SUMMARY AND CONCLUSIONS: RECOMMENDATIONS ABOUT PRESCHOOL SOCIAL-EMOTIONAL ASSESSMENT
How should all this information be integrated? In Table 12.8 we made an initial attempt to consider how teachers, parents, and mental health consultants can work together as a team to focus on each child’s SEL. Columns of this table represent the divisions we have created in this chapter to capture the elements of SEL—attachment, emotional expressiveness, emotion knowledge, emotion regulation, social competence as evaluated by peers and adults, combined social-emotional assessment, “authentic” assessment, and behavior problems. Rows of the table correspond loosely to the usage of assessment tools mentioned by Horton and Bowman (2002), from daily observation of the child and her/his attainments in SEL programming, to weekly teacher meetings and child portfolios, to quarterly teacher and parent checklists, and yearly screening and program evaluation. Rows are also included for potential mental health consultant’s contributions, and to indicate those assessment tools that are research-based in nature. We have filled the cells of the table with those tools mentioned in this chapter that appear to fit each row-column intersection. Entries within each cell include those assessment tools that we consider would fit each goal. It is not our desire to dictate what measures can or should be used. Much flexibility is built into this table. We would, however, prescribe the following:
- Teachers and caregivers should become attuned to each child’s way of demonstrating attachment, emotional expressiveness, emotion knowledge, emotion regulation, social competence, and possible behavior problems. This attunement includes knowing what to look for, remaining observant, and taking note of everyday occurrences in the preschool classroom. Thus, we recommend completion of the HELP Strands Preschool Assessment Tool, as well as making careful anecdotal records or journals about each child, most likely by utilizing the DECA system for running records or something very close to it.
- Sometimes other measurement schemes can also be useful on a daily basis. For example, even if FOCAL observations of emotion were not routinely completed for an entire classroom, the teacher who knows the general system can note in her/his journal or running record that “Andy showed much anger today, but was very tender toward Becca when she fell of the swing.”
- At weekly teacher meetings and for inclusion in children’s portfolios, the HELP Strands Preschool Assessment Tool and the aforementioned anecdotal/running records also form important part of such conferencing. Also, the BDI and/or DECA could be informally completed and compared to earlier versions to show a snapshot of current SEL functioning.
- On a quarterly basis, more structured input may be secured, via, for example, the BDI, DECA, ITSEA, SCBE-30 or PIPPS, and questionnaires on the process of emotion regulation and on behavior problems. The AKT could be administered · quarterly, as well, perhaps by a mental health consultant.
- For yearly or semi-yearly assessment, basically the same mix of measures could be used, with the addition of Feil’s ESP as a screener for behavior problems. The point here is for teachers and early childhood administrators to choose a full complement of measures, which meet their needs and the needs of the children.
- We add notes on where a mental health consultant, whose participation is becoming more common in Head Start and elsewhere, could add his/her expertise. For example, where needed this professional could administer the Teacher Relationships Interview, and the contribution of the mental health consultant could be considerable in completing measures requiring more time than already-busy teachers might have, such as the AQS, Narrative Story Completions, FOCAL Observations, AKT, Raver’s Delay of Gratification Task, and sociometrics.
- Finally, we acknowledge which measures are mostly as yet research-based.
We are cognizant of the needs of teachers and parents, and do not wish to overburden them, even in the service of something as pressing as children’s social-emotional development; such encumbrances could potentially backfire. With the National Board for Professional Teaching Standards (NBPTS), we agree that any assessment should be administratively feasible, professionally acceptable, publicly credible, legally defensible, and economically affordable.
However, this potential system of record keeping need not be too onerous. Observation and use of the HELP Strands could be merged seamlessly with the daily activities of a classroom. Quarterly questionnaires for both parents and teachers could take no more than 60 minutes per child; for teachers, we would advocate choosing the DECA, BDI, PIPPS, SCBE or ITSEA, or at most a pair of these tools, depending on how they fit within the programming and philosophy of the classroom/center. The ASPI or ESP might be added if more extensive data on behavioral problems were needed than supplied by the DECA(6), and the Shields and Eisenberg emotion regulation material would take no more than ten minutes each. Parents could complete the temperament scales and the DECA on a quarterly basis.
In sum, we have found one or more assessment measures for each aspect of SEL that we have considered so important in this volume. We encourage teachers, parents, and others to view these measures together and decide what combination can best be tailored for the needs of the children in their care and the programs they are implementing. Thus, with some effort, we can learn much about young children and move toward maximizing their emotional and social competence.
| Type of Assessment | Attachment | Emotional Competence Assessment | Social Competence Assessment | Behavior Problems | ||||
|---|---|---|---|---|---|---|---|---|
| Emotional Expressiveness | Emotion Knowledge | Emotion Regulation | Peer Evaluation | Teacher Evaluation | Combined/Authentic Social/Emotional | |||
| Daily-Via Curriculum Or Anecdotal Records | HELP Strands; DECA Running Records | HELP Strands; DECA Running Records; Coding “FOCAL” during daily observations and interactions | HELP Strands | HELP Strands; DECA Running Records; Coding “FOCAL” during daily observations and interactions | Teacher Anecdotal Records of Peer Interaction (DECA Running Records?) | HELP Strands | DECA Running Records | Teacher Anecdotal Records (DECA Running Records?) |
| Weekly-Teacher Meetings or Portfolio | HELP Strands; shared anecdotes or journals (DECA Running Records?) | |||||||
| Quarterly-Parent and/or Teacher Checklist | STRS | CBQ: Negativity and Surgency Scales, BDI, ITSEA |
BDI, ITSEA | PRODUCT: CBQ: Effortful Control Scales; PROCESS: Shields et al., Eisenberg et al. | -- | PIPPS or SCBE | BDI, DECA (ITSEA if age level more appropriate) | ASPI or Portions of ITSEA, SCBE, DECA |
| Yearly-Screening or Pre-Post Program Evaluation | AQS | Shields' Emotion Regulation Checklist Liability/Negativity, BDI, ITSEA | AKT, BDI, (ITSEA if age level more appropriate) | PRODUCT: Shields et al. | Sociometric Ratings Twice Yearly | PIPPS or SCBE | BDI, DECA (ITSEA if age level more appropriate) | ESP, ASPI |
| Mental Health Consultant | AQS, Teacher Relationship Interview, Narrative Story Completions | FOCAL Observations; MPAC | AKT (Quarterly or bi-annually) | PROCESS AND PRODUCT: Raver's Delayed Gratification Task; MPAC | Sociometric Ratings Twice Yearly | -- | -- | ASPI Possibly CBCL 2/5 or Feil's ESP, MPAC |
| Research-Based Assessment Tools | AQS, Narrative Story Completions | FOCAL Observations; MPAC | AKT | PROCESS AND PRODUCT:Raver’s Delayed Gratification Task, MPAC | Sociometric Ratings | -- | MPAC | MPAC |
| Table of Contents | Previous |