Best Practices in Play Assessment and Intervention - CiteSeerX

102 downloads 40567 Views 117KB Size Report
based interventions, and progress monitoring. Play assessment and intervention meets the response-to- intervention (RTI) component of the 2004 reauthoriza-.
33

Best Practices in Play Assessment and Intervention

Lisa Kelly-Vance Brigette Oliver Ryalls University of Nebraska at Omaha OVERVIEW Play is enjoyable and motivating for children, and it is how they spend much of their time. Not only is play an important part of children’s daily routine, it is also a window into their developmental levels and a context where valuable teaching and learning can occur. Play as an assessment/intervention context is relatively new in the field of school psychology but is increasingly popular with practitioners and researchers because of the current emphasis on ecologically valid assessments, contextbased interventions, and progress monitoring. Play assessment and intervention meets the response-tointervention (RTI) component of the 2004 reauthorization of the Individuals with Disabilities Act. In fact, the National Association of School Psychologists’ (NASP) Position Statement on Early Childhood Assessment specifically states that play assessment is an appropriate approach to evaluating the needs of young children (NASP, 2005). The additional and powerful benefit of play assessment is that it also leads directly to interventions and progress monitoring in the same play context. Finally, play assessment has been suggested to be a culturally sensitive practice (Meisels & AtkinsBurnett, 2000). Play is ubiquitous in early childhood. Young children are constantly at play. Play can be conceptualized as an activity that is not a means to an end but an end itself (i.e., play for the sake of play), is motivating to the individual, and is associated with positive emotions (Tamis-LeMonda, Uzgiris, & Bornstein, 2002). Play assessment is when play is used as the context for evaluating a child’s current level of functioning and determining whether there are areas that require

intervention. Play intervention is when these interventions are conducted in the play context. For the purposes of this chapter, our focus will be on assessment and intervention in the cognitive domain. While play assessment and intervention can be used in other domains such as communication, motor, and social, it is beyond the scope and purpose of this chapter to include them. Important to the understanding of how play assessment and intervention came to the attention of school psychologists is the interest that basic researchers have had in the topic for decades, resulting in a vast empirical foundation (e.g., Piaget, 1962) that can be applied to the identification of children with specific intervention needs. Piaget’s extensive study of how children’s play changes with development is one of the earliest and arguably the most influential examples of such research. Specially, he chronicled the change from sensorimotor to representational play. Numerous descriptions of the typical developmental progression from exploratory to pretend play have been well documented and empirically supported (e.g., Belsky & Most, 1981; Elder & Pederson, 1978; Fenson, 1984; Fenson & Ramsay, 1980; Lyytinen, 1991; McCune-Nicolich, 1981; TamisLeMonda & Bornstein, 1996). While the specific terminology varies, what is described is a consistent and predictable progression from simple exploratory behaviors such as mouthing to complicated episodes of pretend play.

BASIC CONSIDERATIONS The use of standardized tests with young children has been criticized on several fronts. These types of tests 549

Best Practices in School Psychology V

serve eligibility purposes but do not provide information about how to develop or monitor appropriate interventions, they are not ecologically valid and do not assess the child in the natural context, and they are a potentially unmotivating and inappropriate means of gathering valid information about a young child (KellyVance & Ryalls, 2005). Further, many practitioners have expressed misgivings about the use of tests to determine young children’s current level of performance (Barnett, Macmann, & Carey, 1992) and would prefer to use alternative strategies such as play assessment (Bagnato & Neisworth, 1994). The origins of play assessment and intervention stem from psychoanalytic work with children at the turn of the twentieth century. Therapists used information from children’s play to determine their mental health needs and then used the play context during therapy. This practice continues in many different therapeutic approaches. More recently and for the reasons discussed above, school psychologists working in early childhood began using play assessment as an observational strategy in the 1980s (as discussed in Athanasiou, 2007). Linder (1990) brought even more attention to the procedures with the publication of her play assessment model and her subsequent play intervention model (Linder, 1993a, 1993b). Practitioners have been using play assessment and intervention, but the empirical support for its use has only recently emerged (e.g., Cherney, Kelly-Vance, Gill-Glover, Ruane, & Ryalls, 2003; Farmer-Dougan & Kaszuba, 1999; Kelly-Vance, Needelman, Troia, & Ryalls, 1999; Kelly-Vance & Ryalls, 2005; Kelly-Vance, Ryalls, & Gill-Glover, 2002; Myers, McBride, & Peterson, 1996). Fortunately, studies are confirming what practitioners have known intuitively, which is that this approach is a reliable and valid means of assessing young children (see below). Moreover, the assessment procedures can be used regularly to monitor the progress of children receiving interventions.

Training and Equipment While the specific materials and procedures required for different versions of play assessment and intervention vary, all forms are similar enough that general guidelines concerning equipment and training can be provided. With regard to equipment, one of the advantages of play assessment and intervention is that no specialized materials, forms, or equipment are required. All types of play assessment and intervention involve children playing with toys or other play materials (e.g., crayons 550

and paper). While some forms of play assessment do require a particular set of play objects (e.g., Fewell as cited in Athanasiou, 2007), others can be conducted in any toy room or early childhood classroom (e.g., KellyVance & Ryalls, 2005; Linder, 1993a; Ryalls et al., 2000) as long as the available toys are diverse enough to elicit a wide range of play behaviors, including complex pretend play, and are attractive to both boys and girls. Some specific considerations regarding toy choice will be discussed below. Although no specialized equipment is necessary to conduct play assessments or interventions, some training and experience is necessary; however, the nature of this training depends on the RTI tier in question. Certainly, knowledge about child development in general, and the developmental course of play in particular, is necessary at all three RTI tier levels. A thorough understanding of a tiered RTI system and the specific problem-solving approach used with children who have more intense needs is necessary in utilizing play assessment to guide the intervention process. With regard to the informal assessments that occur at Tier 1, the school psychologist needs to make sure the personnel and parents involved have a general understanding of how and when play changes with development. At Tier 2 or 3, most forms of play assessment can be considered observational coding systems, and thus some experience using such systems is beneficial. Familiarity with the specific coding system is obviously required, as is some training and experience identifying particular examples of success or failure or assigning specific codes to play behaviors (e.g., substitution). As with any observational coding system, such experience is necessary to ensure the reliability of the assessment across observers and across play sessions (Salvia, Ysseldyke, & Bolt, 2007). The nature and amount of training depends on the particular form of play assessment involved. With respect to the type of training required for actual intervention implementation, all interventions are grounded in the context of play and thus require no specialized education or training above and beyond the specific interventions themselves.

BEST PRACTICES Types of Play Assessment Unlike other forms of early childhood assessment, relatively little activity has occurred in the development of play assessment techniques. Of those that exist, only three have been described in enough detail to use in Chapter 33, Volume 2

Play Assessment and Intervention

practice. These three approaches are described in the following section.

Play Assessment Scale The Play Assessment Scale (PAS) is designed to evaluate the developing skills of children from 2 to 36 months (Fewell as cited in Athanasiou, 2007). This 45-item scale is developmentally sequenced and is organized into eight age ranges and toy sets so that only a portion of the items are rated for each child. Children are first observed in spontaneous play followed by a facilitated play session, and their play behaviors are coded according to the scale so that a play age can be determined. The play age is composed only of those behaviors observed in spontaneous play. A basal/ceiling approach is used and a conversion chart allows the rater to convert the raw score to the child’s play age. Anyone familiar with the PAS can administer the items and complete the rating scale, but Fewell stresses that the individual rating the child must thoroughly understand the individual PAS items and the play behaviors that would justify that a child possesses the skill reflected in each item.

Transdisciplinary Play-Based Assessment The most thoroughly described play assessment technique, Transdisciplinary Play-Based Assessment (TPBA), was developed by Linder (1990, 1993a). Her transdisciplinary approach captures the essence of collaboration in that early childhood service providers and parents work together to find out information about a child’s developmental levels and link it directly to interventions. The team approach is conducted in what she called an ‘‘arena’’ format where individuals from various disciplines observe the child in free play and document the play behaviors. The adults communicate their findings with one another throughout the observational period allowing for a shared perspective on the child’s skill level. General guidelines for conducting the data gathering sessions were published in her ground-breaking book (Linder, 1990). Prior to the play session, the professionals contact the child’s caregiver to find out their concerns and preliminary information on the child’s functioning. The structure of the session takes into account any information obtained from the parents. The observational team consists of a play facilitator who engages with the child, a parent facilitator who is responsible for discussing the process with the parent(s), Chapter 33, Volume 2

the evaluators, and a person to operate the video camera. A large play area such as a classroom is recommended for the session, but any setting with a variety of toys can be used. Initially, the child plays alone and no structure is imposed upon the activities. The play facilitator can participate but not initiate any of the play. Facilitation occurs in the second phase where the adult attempts to engage the child in activities that have not yet been observed and may be in the child’s repertoire. It is essentially a testing phase. Following the two phases involving adults, the child is then observed playing with another child and then interacting with a parent. A phase of motor play then occurs adhering to the same sequence as the initial play activities. The session ends with a snack, and feedback is provided to the parent(s). The entire session lasts between 60 and 90 minutes. Unique to the Linder model are the detailed coding guidelines. She provides developmental coding schemes for cognitive, social–emotional, communication and language, and sensorimotor development. The applicable ages vary, but, in general, the coding guidelines apply to young children up to 60–72 months of age. Each general category of development (e.g., cognition, communication, motor skills) is subdivided into more specific categories labeled subdomains. These coding guidelines are used to describe a child’s strengths and their area of intervention need. Specific suggestions for intervention are derived from the assessment results (Linder, 1993b).

Play in Early Childhood Evaluation System The Play in Early Childhood Evaluation System (PIECES) is the most recently developed and extensively investigated approach to play assessment (Cherney et al., 2003; Kelly-Vance, Gill, Ruane, Cherney, & Ryalls, 1999; Kelly-Vance, Needleman, et al., 1999; KellyVance & Ryalls, 2005; Kelly-Vance et al., 2002; Ryalls et al., 2000). The PIECES grew out of empirical work originally based on Linder’s TPBA cognitive development assessment guidelines. The PIECES approach is both similar to and dissimilar from Linder’s TPBA approach. With regard to assessment procedures, like TBPA, the PIECES assessment involves observation of a child engaged in free play. The PIECES is also similar to Linder’s TPBA in that the assessment can be conducted in any setting with any toy set as long as the toy set is large and varied enough to elicit a wide range of behaviors. 551

Best Practices in School Psychology V

In the PIECES assessment, children play freely for 30–45 minutes and can be videoed. They are not required to produce any specific behaviors with any specific toys. The procedures are dissimilar from Linder’s TPBA in that the child usually plays alone and without any facilitation on the part of adult observers. Parents and a session facilitator are present and free to praise the child and to repeat what the child says, but are instructed not to ask questions or suggest new play behaviors. Because the PIECES focuses exclusively on cognitive development, a multidisciplinary team of observers, while certainly allowable, is not necessary. With regard to the coding guidelines, Linder’s TPBA and the PIECES are similar in that the guidelines are broken down into multiple scales examining different domains of cognitive development, although the specific domains differ. The PIECES coding scheme contains one core subdomain (exploratory/pretend play) as well as several supplemental subdomains (see Table 1). The supplemental subdomains (e.g., problem-solving skills and planning, categorization, quantitative skills) are adapted versions of several of Linder’s subscales. The PIECES coding guidelines differ from TPBA in that the 13-item sequence that makes up the exploratory/pretend play core subdomain was drawn not from Linder but instead from the extensive empirical literature on the development of play (Belsky & Most, 1981; Fenson, 1984, Lyytinen, 1991; Tamis-LeMonda, Bornstein, Cyphers, Toda, & Ogino, 1992). In the PIECES coding system, every play behavior produced by the child can be classified on the core subdomain whereas the supplemental subdomains are reserved for specific types of behaviors (such as trial-and-error problem solving, sorting, and drawing), which may or may not occur in any given play session. The information obtained from this coding procedure is then compared to norms for typically developing

children to determine if the child has specific areas that require intervention. Interventions are targeted at the core and/or supplemental subdomain skills found to be discrepant from peers.

Research on Play Assessment Emerging evidence for the validity of various forms of play assessment has been reported in several studies. Support for the validity of PAS was established by Eisert and Lamorey (1996), Fewell and Rich (1987), and Finn and Fewell (1994). In these studies, PAS was found to correlate with standardized and nonstandardized measures of cognition, communication, motor skills, and adaptive behavior. Two published studies have reported on the validity of Linder’s TPBA model. The Myers et al. (1996) study of social validity found that parents and early childhood professionals preferred TPBA model to more traditional information yielded from standardized tests. Kelly-Vance, Needelman, et al. (1999) found a high correlation between scores from the Bayley Scales of Infant Development II and the cognitive development guidelines from Linder’s TPBA. In addition, one unpublished dissertation examined the concurrent and content validity and the test–retest and interrater reliability of TPBA and found it adequate (Friedli as cited in Athanasiou, 2007). A study of a more general form of play assessment found adequate validity when comparing a brief play assessment scale to measures of development (Farmer-Dougan & Kaszuba, 1999). Thus, while there is a smattering of publications on play assessment measures, much more work needs to be conducted to meet the current standards of reliability and validity, particularly in light of the fact that such techniques are explicitly listed as an alternative assessment method in the NASP Position Statement on Early Childhood Assessment (NASP, 2005).

Table 1. Core Subdomain and Supplemental Subdomains Assessed in PIECES Subdomain Core . Exploratory/pretend play Supplemental . Problem-solving skills and planning . Categorization skills . . .

Quantitative skills Drawing skills Seriation abilities

552

Levels

Description

13

Ranges from mouthing and simple manipulation to extended sequences of pretense

12 19

Ranges from simple search for a hidden object to complex means-end problem solving Ranges from simple combinations based on perceptual similarity to complex classification based on nonperceptual features Ranges from counting 1–5 to recognizing and labeling all numbers 0–9 Ranges from scribbling to drawing and/or copying complex forms and shapes Ranges from understanding of basic dimension terms (e.g., big) to seriation of four or more objects

10 21 11

Chapter 33, Volume 2

Play Assessment and Intervention

This lack of research motivated us to begin a systematic examination of the psychometric properties of play assessment. We began with the development of coding guidelines. Our goals in the development of these guidelines were to develop a system of assessment that would thoroughly evaluate a child’s skills in the cognitive domain, would lead directly to the development of appropriate interventions to be implemented in the context of play, and would easily allow for monitoring the effectiveness of these interventions. As discussed above, we began with the most popular form of play assessment, Linder’s TPBA, and adapted her coding guidelines to include information drawn from extensive basic research on play. The resulting PIECES guidelines have been evaluated empirically and found to have high interrater reliability (Kelly-Vance & Ryalls, 2005). In addition, this high interrater reliability was achieved with a relatively simple training procedure. That is, individuals with a background in observational techniques can be trained to accurately and reliably use these guidelines to assess play behavior with as little as a half day of training (Kelly-Vance, Gill, et al., 1999; Kelly-Vance et al., 2000). In addition to work developing psychometrically sound coding guidelines, we have also worked to develop a play assessment procedure that is both practical, from the perspective of practitioners and families, as well as valid and reliable. Our procedure, again initially based on Linder’s technique, was briefly described above. In developing this procedure, we have empirically examined a number of session characteristics and identified a process that is effective as well as easily and accurately reproduced. We have investigated numerous procedural aspects including toy selection (Cherney et al., 2003), facilitation versus nonfacilitation (King, McCaslin, Kelly-Vance, & Ryalls, 2003), the familiarity of the context (Drobny, 2003), and the impact of the presence of a peer (Fredenberg, 2004; Mendez, 2005). We have also examined other important attributes such as test–retest reliability (Kelly-Vance & Ryalls, 2005), correlations with other standardized measures (Fiscus, 2006), and use with special populations (Gill-Glover, McCaslin, Kelly-Vance, & Ryalls, 2001; Johnson, 2005; Kelly-Vance, Needelman, et al., 1999).

How to Conduct a Play Assessment Because the PIECES is the only approach that has been empirically tested to this degree, we believe it to be best practice at this time. We have focused on developing a Chapter 33, Volume 2

procedure that is both valid and reliable and is practitioner friendly. While we believe this method to be best practice, the model allows for flexibility in ensuring that the procedures are appropriate for each individual child. The basic structure described below should be used, but adaptations for specific referral concerns should be taken into account.

Session Characteristics Context One of the many positive characteristics of play assessment is that it can be conducted in a variety of settings such as home, daycare, preschool, or laboratory. We have consistent results when play assessments conducted in the home were compared to those in a laboratory setting (Drobny, 2003); however, the child’s comfort level should be considered when selecting the location and interpreting the findings. Selection and Organization of Toys When selecting toys for play assessment it is important to make sure that the toys are appropriate for the age, assumed developmental level, and gender of the child being assessed. While it is commonly recognized that children of different ages/developmental levels may choose to play with different types of toys, it is important to keep in mind that boys and girls also have preferences and these preferences may have an impact on the outcome of the assessment. By 2 years of age, children display preferences for certain types of toys (O’Brien & Huston, 1985). That is, boys and girls tend to prefer to play with toys that are stereotypically consistent with their own gender. While both boys and girls show this tendency, research has shown that this tendency is more pronounced for boys (Cherney et al., 2003). Specifically, Cherney and colleagues found that while boys played with predominantly male stereotyped toys (e.g., cars and trucks), girls played predominantly with neutral toys (e.g., puzzles and cash register). More importantly, for both boys and girls, it was found that the male stereotyped toys elicited the simplest forms of play and the female stereotyped toys (e.g., dolls and kitchen set) elicited the highest levels of play. Thus, when conducting a play assessment the school psychologist should ensure that a variety of toys that are gender appropriate are available and encourage children to play with toys that have the potential to elicit more complex forms of play. In addition to the types of toys present, the arrangement of the toys must also be considered. For 553

Best Practices in School Psychology V

the purposes of play assessment, it is important that all toys are visible and accessible to the child. In addition, similar toys (e.g., dolls and doll-related toys) should be grouped together to encourage thematic play. Finally, while the playroom should be neatly organized, it need not be pristine, and, in fact, a mild level of disarray may be helpful to encourage certain types of behaviors. Specifically, problem-solving behavior is often observed when children attempt to assemble puzzles and construct train tracks, so leaving them somewhat unassembled is recommended.

Facilitated Versus Nonfacilitated Play Some play assessment approaches involve significant levels of facilitation and/or modeling, often requiring the reproduction of very specific play behaviors. We believe, however, best practice involves a nonfacilitated, free-play situation with minimal direction. Our rationale is twofold. First, eliminating facilitation from the format helps ensure a more standardized session, which removes facilitator bias and increases the likelihood of obtaining high levels of reliability across raters, children, and sessions. Second, there is empirical evidence that facilitation either has no effect or, conversely, has a negative impact and children’s ratings are actually lower after adult direction (King, et al., 2003). One exception to this guideline concerns questions about skills that may not have been demonstrated during the session or potentially emerging skills. It is appropriate to attempt to elicit behaviors described in certain supplemental subdomains, such as drawing and/or counting, that do not spontaneously occur for many children. Similarly, if there are questions about the upper limits of some emerging behaviors it is acceptable to attempt to elicit such behaviors, particularly when developing interventions. Instead of facilitating play in a traditional sense, parents and play team members may participate in child-directed play and verbal praise. In addition, adults should redirect children (e.g., ‘‘What else can you play with?’’) who continue to play with the same toy in the same manner for 5 minutes or longer. This is an attempt to vary the play behaviors during the session but not to encourage any specific type of play. Presence of Peers Another form of social interaction involves the presence of a peer in the play assessment process. This approach is appropriate for evaluating social interactions but may have limited utility when determining cognitive functioning. Mendez (2005) and Fredenberg (2004) analyzed 554

the impact that same-age peers had on a child’s cognitive functioning and found that it did not improve the level of play. In fact, evidence suggests that different characteristics in peers elicit different types of play. Given that our goal is to standardize play assessment procedures, it would follow that peer interactions are not a necessary component of the session.

Describing Children’s Play Children’s cognitive levels can be described by using the PIECES coding guidelines. Practitioners should familiarize themselves with the guidelines and use it to analyze a child’s skills and areas that may need intervention. The play behaviors are documented by videotape (or live play if videotape is not available). Specific examples of play should be noted as they apply to the core and supplemental subdomains. Then the behaviors are compared to the coding guidelines and the child’s highest level of play is noted for all subdomains. In addition, practitioners may also want to document the type of play that was seen most frequently in each subdomain. Finally, the child’s age equivalents from the PIECES subdomains are compared to his or her chronological age and discrepancies are noted.

Best Practice in the Use of Play Assessment Data in an RTI Model Given the requirement of systematic, data-based practice of RTI, play assessment and intervention is a logical choice for school psychologists working in early childhood settings. Play assessment and intervention have several features that make it appealing in an RTI model of practice. The developmental sequence described in play assessment coding schemes allows practitioners to monitor progress in a logical and empirically validated manner, and data can be collected regularly and with minimal environmental modification. Play assessment is one piece of data that can be collected and integrated with other assessment approaches to gain a complete picture of the child’s level of performance. In contrast to standardized tests that can be difficult for parents to comprehend and find useful, the information obtained from a play assessment is easy for parents to understand. Providing interventions in the play context is also a logical format, and early childhood teachers and parents of young children are arguably more likely to demonstrate treatment adherence if they find the process credible. Chapter 33, Volume 2

Play Assessment and Intervention

RTI’s conceptual three-tiered model can be applied to play assessment and intervention (see Table 2). The first tier parallels general education in the school-aged population. In early childhood, this tier applies best to families, child care centers, parent resource facilities, and community agencies. School district personnel and educational service units can disseminate general information about child development, how to observe children’s play to determine where they are functioning, how to promote cognitive development through appropriate play activities, and who to contact if a child is not meeting developmental milestones. The goal at this tier is for families and early childhood caretakers to have access to the information and implement simple, basic procedures of play assessment and intervention. Research indicates that mothers have a basic understanding of the order of play development but know less about specific developmental milestones in play than milestones in other domains such as language and motor development (Tamis-LeMonda, Chen, & Bornstein, 1998; Tamis-LeMonda, Shannon, & Spellmann, 2002). If a parent or childcare worker has concerns about a child’s development, contact can be made to the school district or educational service unit, and Tier 2 interventions will be considered. In Tier 2, a formal play assessment is conducted by a school psychologist, and the child’s strengths and areas of intervention need are determined. Based on this play assessment data, interventions are then developed collaboratively with the professional who conducted the play assessment and the individual(s) responsible for intervention implementation. Possible intervention providers include parents, preschool teachers, and child care providers. Given their knowledge of data-based decision-making and intervention development, school psychologists are an obvious choice for the role of consultant with these cases. This role includes interpreting assessment data, scripting the intervention, and monitoring the progress of the intervention. If a child is in a preschool or daycare center, small group interventions may be appropriate and the consultant could develop a protocol for several children with similar

needs. Otherwise, individual interventions may be required. Incidentally, the frequent use of individual intervention at this level contrasts what generally occurs with school-aged children where small group interventions are most common at Tier 2. In both individual and small group interventions, a goal is determined and the child’s progress is regularly monitored by the service provider. At this tier, progress is determined by using the play assessment coding scheme and periodically evaluating the child’s rate of progress toward a goal. Because of potential for rapid change in early childhood, frequent monitoring is recommended. Tier 3 is a more intense version of Tier 2 and may require that a child spend a trial placement in an early intervention program. The intervention is more intensely implemented, progress monitoring data are collected more frequently, and a team of early childhood professionals should be included to consider whether the child is eligible for special education services. At both Tiers 2 and 3, a play assessment coding scheme provides a benchmark for where a child should be functioning and can help teams evaluate a child’s discrepancy from peers, rate of progress, and amount of support needed to ensure progress. If the child continues to be functioning significantly below his or her peer and is not progressing at a rate that would indicate he or she may catch up, early intervention services through special education should be considered and a full evaluation conducted. The child’s response to the play intervention should be a major consideration in the eligibility decision.

Case Illustrations As can be gleaned from the information above, play assessment and intervention can be effectively used in an RTI model at all three tiers. In this section we will present two case examples to illustrate more specifically how this process would unfold. The first case will exemplify the first tier and describe how information about play can help a parent better understand his or her child’s development and determine whether the child is progressing appropriately. The second case will address the second and third

Table 2. Play Assessment and Intervention in a Three-Tiered RTI Model Approach and intervention

Tier 1

Tier 2

Tier 3

Assessment approach Intervention approach Service delivery approach Intervention agents

Screening General skills Informational Parents

Monthly or bimonthly Small groups Consultation and/or direct services Parents, teachers, childcare providers

Intense frequency (weekly or more) Small group or individual Consultation and intensive direct services Parents, teachers, childcare providers, and special education support staff

Chapter 33, Volume 2

555

Best Practices in School Psychology V

tiers by describing how play assessment and intervention can be used in early intervention services.

Case 1 Ms. Austen, a mother of three young children, contacted Springfield Elementary School to find out more about child development and, more specifically, whether her children were on the appropriate developmental trajectory. Ms. Austen then met with the school psychologist who served Springfield Elementary. She stated that her main concern was whether she was providing the appropriate learning environment for her children to develop necessary skills and be ready for kindergarten. The school psychologist suggested that, by watching her children play, Ms. Austen could learn about how her children were functioning. She was asked to keep a record of the type of play that her children frequently engaged in as well as any newly emerging play activities. One month later, she returned to the school to share her observations with the school psychologist. The PIECES coding scheme was discussed with Ms. Austin and compared to her observations (see Table 3), at which time it was determined that the children were functioning within the normal range. This reassured her that she was providing appropriate activities. The session concluded with the school psychologist giving Ms. Austen a handout that listed developmentally appropriate activities that she could do at home with the children. Case 2 Ms. Jennings visited the pediatrician for her 3-yearold son Jeremy’s annual examination. During the

appointment the pediatrician interviewed Ms. Jennings, who had concerns that her child might not be demonstrating age-appropriate cognitive skills. The pediatrician gave Ms. Jennings pamphlets on typical developmental milestones for 3-year-olds and referred her to the school district for screening. Ms. Jennings contacted the school, and the early childhood consultant went to the house to conduct a screening of Jeremy. This screening consisted of an observation of Jeremy during 30 minutes of free play followed by an interview with Ms. Jennings. The early childhood consultant determined that Jeremy’s pretend play was below the expected level and the consultant referred him to the early childhood school psychologist. The school psychologist conducted a more thorough play assessment of Jeremy in his preschool and then classified his behaviors using the PIECES coding scheme. In total, the play behaviors observed in the preschool were consistent with both how Ms. Jennings reported that Jeremy plays at home and what the early childhood practitioner reported seeing in the initial play assessment. In the exploratory/pretend play subdomain, Jeremy’s skills were predominantly at the level of simple single behavior pretend play acts. An example of this type of play was when Jeremy pretended to eat a cookie and then had his teacher pretend to eat the cookie. Active other-directed acts were emerging but less common, indicating skills at the 24-month level. Jeremy demonstrated this skill when he was playing with a toy dog and made it bark. In contrast, age-appropriate pretend play at Jeremy’s age should have involved short sequences of three pretend play acts such as making the dog bark,

Table 3. Examples of Ms. Austen’s Descriptions of Her Children’s Play and Corresponding PIECES Codes Children Brady (4 months)

Mother’s description . . .

Elliot (28 months)

. . . . .

Lydia (51 months)

. . .

556

Plays with feet and chews on toes Bats at toys Grabs toys/anything and puts them in his mouth Makes snakes with Play-Doh; slithers them around and says ‘‘sssss’’ Counts to 14 Pretends to be a cat; crawls around and meows Puts on sunglasses, grabs a bag, and says, ‘‘Bye Mommy, I go work.’’ Lines up toy cars by size Makes pretend cookies with Play-Doh; offers cookies to guests Draws birthday cake, adds candles, sings birthday song Plays doctor: gives shots, looks in throat and ears, gives medicine, listens to heartbeat, performs surgery

PIECES subdomain and level . . . . . . . . . . .

Exploratory/pretend Exploratory/pretend Exploratory/pretend Exploratory/pretend

play: play: play: play:

mouthing simple manipulation mouthing active other-directed act

Quantification skills: counts to 5+ Exploratory/pretend play: self-directed act Exploratory/pretend play: multischeme combination Categorization: sorts objects by size Exploratory/pretend play: multischeme combination Exploratory/pretend play: multischeme combination Exploratory/pretend play: event

Chapter 33, Volume 2

Play Assessment and Intervention

Chapter 33, Volume 2

Figure 1. Case 2: Pretend play progress monitoring data. 4

Number of Pretend Play Acts

feeding the dog, and then petting the dog. Jeremy’s baseline for this type of pretend play was zero. The goal was that, after 6 weeks of intervention, Jeremy would demonstrate three-step nonfacilitated pretend play acts in a half-hour play session. With regard to the supplemental subdomains, Jeremy was observed to be functioning somewhat below the expected level on categorization skills. Specifically, Jeremy showed some deficiencies in sorting and nesting objects. The school psychologist followed up on this observation and found that prompting play with nesting cups and a nested puzzle did not improve Jeremy’s performance. The goal established based on assessment was that, after 6 weeks of intervention, Jeremy would spontaneously or with facilitation successfully nest a fiveitem series of nesting cups and complete a nested puzzle depicting common shapes (e.g., circle, square, and triangle). Jeremy’s performance on the other supplemental subdomains (problem-solving skills and planning, quantification, drawing, and seriation) was at ageappropriate levels indicating no need for intervention. Intervention for both target areas involved modeling and adding to Jeremy’s spontaneous play behaviors. For example, if Jeremy pretended to feed a toy dog a cookie, the interventionist would add to Jeremy’s behavior by modeling an additional play act, such as pretending to pour the dog some milk. In addition to modeling, direct teaching of sorting and nesting behaviors was also used to increase age-appropriate categorization behaviors. Both the preschool teacher and the mother were trained to implement the interventions. Training for the teacher and mother was conducted by the school psychologist in one 30-minute session. The school psychologist was in contact with the teacher and mother on a weekly basis to address concerns and monitor intervention integrity. To monitor progress the school psychologist observed Jeremy in a free-play situation once a week. At the end of 6 weeks of intervention the data were reviewed and it was determined that Jeremy had met the goal in the core subdomain of pretend play (see Figure 1) but not in the supplemental subdomain of categorization (see Table 4). Thus, in this case play assessment and intervention led to improvement in one of two target domains. While the goal was not met in the second subdomain, both the parent and teacher believed that Jeremy had shown improvement over the course of the intervention and chose to continue implementing the interventions. Because of the progress shown, it was determined that no referral for early childhood special education services was necessary.

3

2

1

0

Baseline Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Weeks Goal Line

Trend Line

SUMMARY The existing empirical literature indicates that there are numerous benefits to using play assessment and intervention with young children. Because the assessment is conducted in the natural environment of play, it is motivating and elicits the highest level of a child’s functioning. The process offers flexibility while remaining Table 4. Case 2: Categorization Progress Monitoring Data Week 1 2 3

4

5 6

Spontaneous (S) or facilitated (F) play

Play behavior with nesting cups and shape puzzle

F F F F F S F F S F S S S S

Nested two cups Placed circle in puzzle Nested three cups Placed circle in puzzle Nested four cups Placed circle in puzzle Placed square in puzzle Nested five cups Placed circle in puzzle Placed square in puzzle Nested two cups Placed circle and square in puzzle Nested three cups Placed circle and square in puzzle

557

Best Practices in School Psychology V

psychometrically sound. The results of a play assessment lead naturally to interventions that can be implemented and monitored by many different individuals in various settings with a minimal amount of training. As exemplified in the case studies, this method can be used by parents and educators in an RTI framework.

REFERENCES Athanasiou, M. S. (2007). Play-based approaches to preschool assessment. In B. A. Bracken & R. J. Nagle (Eds.), The psychoeducational assessment of preschool children (4th ed., pp. 219– 238). New Jersey: LEA. Bagnato, S. J., & Neisworth, J. T. (1994). A national study of the social and treatment ‘‘invalidity’’ of intelligence testing for early intervention. School Psychology Quarterly, 9, 81–102. Barnett, D. W., Macmann, G. M., & Carey, K. T. (1992). Early intervention and the assessment of developmental skills: Challenges and directions. Topics in Early Childhood Special Education, 12, 21–43. Belsky, J., & Most, R. K. (1981). From exploration to play: A crosssectional study of infant free play behavior. Developmental Psychology, 17, 630–639. Cherney, I. C., Kelly-Vance, L., Gill-Glover, K., Ruane, A., & Ryalls, B. O. (2003). The effects of stereotyped toys and gender on play assessment in children aged 18–47 months. Educational Psychology, 23, 95–106. Drobny, J. M. (2003). The use of play assessment across settings for preschool age children. Unpublished manuscript. Eisert, D., & Lamorey, S. (1996). Play as a window on child development: The relationship between play and other developmental domains. Early Education and Development, 7, 221–234. Elder, J. L., & Pederson, D. R. (1978). Preschool children’s use of objects in symbolic play. Child Development, 49, 500–504. Farmer-Dougan, V., & Kaszuba, T. (1999). Reliability and validity of play-based observations: Relationship between the play behavior observation system and standardized measures of cognitive and social skills. Educational Psychology, 19, 429–441. Fenson, L. (1984). Developmental trends for action and speech in pretend play. In I. Bretherton (Ed.), Symbolic play: The development of social understanding (pp. 249–270). Orlando, FL: Academic Press. Fenson, L., & Ramsay, D. S. (1980). Decentration and integration of the child’s play in the second year. Child Development, 51, 171–178. Fewell, R. R., & Rich, J. S. (1987). Play assessment as a procedure for examining cognitive, communication, and social skills in multihandicapped children. Journal of Psychoeducational Assessment, 2, 107–118.

558

Finn, D. M., & Fewell, R. R. (1994). The use of play assessment to examine the development of communication skills in children who are deaf-blind. Journal of Visual Impairment & Blindness, 88(4), 349–357. Fiscus, L. (2006, March). The play in early childhood evaluation system: Play-based assessment. Poster session presented at the annual meeting of the National Association of School Psychologists, Anaheim, CA. Fredenberg, M. (2004). The effects of peer interactions on play assessment: Do peers influence the cognitive skills observed in three-year-old children? Unpublished manuscript. Gill-Glover, K., McCaslin, L., Kelly-Vance, L., & Ryalls, B. O. (2001, April). A comparison of play-based assessment and the Bayley Scales of Infant Development-Second Edition with Spanish- and Englishspeaking children. Poster session presented at the annual meeting of the National Association of School Psychologists, Washington, DC. Johnson, J. (2005). The usefulness of the Play in Early Childhood Evaluation System (PIECES) in children with autism. Unpublished manuscript. Kelly-Vance, L., Gill, K., Ruane, A., Cherney, I., & Ryalls, B. O. (1999, April). Play-based assessment: Applying a reliable coding system. Poster session presented at the annual meeting of the National Association of School Psychologists, Las Vegas, NV. Kelly-Vance, L., Gill, K., Schoneboom, N., Cherney, I., Ryan, C., Cunningham, J., et al. (2000, March). Coding play-based assessment: Issues, challenges, and recommendations. Poster session presented at the annual meeting of the National Association of School Psychologists, New Orleans, LA. Kelly-Vance, L., Needelman, H., Troia, K., & Ryalls, B. O. (1999). Early childhood assessment: A comparison of the Bayley Scales of Infant Development and a play-based technique. Developmental Disabilities Bulletin, 27, 1–15. Kelly-Vance, L., & Ryalls, B. O. (2005). A systematic, reliable approach to play assessment in preschoolers. School Psychology International, 26, 398–412. Kelly-Vance, L., Ryalls, B. O., & Gill-Glover, K. (2002). The use of play assessment to evaluate the cognitive skills of twoand three-year-old children. School Psychology International, 23, 169–185. King, J., McCaslin, L., Kelly-Vance, L., & Ryalls, B. O. (2003, April). Facilitated vs. nonfacilitated play behavior of children. Poster session presented at the annual meeting of the National Association of School Psychologists, Toronto, ON. Linder, T. W. (1990). Transdisciplinary play-based assessment: A functional approach to working with young children. Baltimore: Brookes. Linder, T. W. (1993a). Transdisciplinary play-based assessment: A functional approach to working with young children (2nd ed). Baltimore: Brookes.

Chapter 33, Volume 2

Play Assessment and Intervention

Linder, T. W. (1993b). Transdisciplinary play-based intervention: Guidelines for developing a meaningful curriculum for young children. Baltimore: Brookes. Lyytinen, P. (1991). Developmental trends in children’s pretend play. Child: Care, Health and Development, 17, 9–25. McCune-Nicolich, L. (1981). Toward symbolic functioning: Structure of early pretend games and potential parallels with language. Child Development, 52, 785–797. Meisels, S. J., & Atkins-Burnett, S. (2000). The elements of early childhood assessment. In J. P. Shonkoff & S. J. Meisels (Eds.), Handbook of early childhood intervention (pp. 231–257). New York: Cambridge University Press. Mendez, K. (2005). Play assessment: The effects of peer interaction on children’s cognitive capabilities. Unpublished manuscript. Myers, C. L., McBride, S. L., & Peterson, C. A. (1996). Transdisciplinary, play-based assessment in early childhood special education: An examination of social validity. Topics in Early Childhood Special Education, 16, 102–126. National Association of School Psychologists. (2005). Position statement on early childhood assessment. Bethesda, MD: Author. Retrieved March 7, 2007, from http://www.nasponline.org/about_nasp/ pospaper_eca.aspx O’Brien, M. O., & Huston, A. C. (1985). Development of sextyped play behavior in toddlers. Developmental Psychology, 21, 866–871. Piaget, J. (1962). Play, dreams and imitation in childhood. New York: W.W. Norton.

Chapter 33, Volume 2

Ryalls, B., Gill, K., Ruane, A., Cherney, I., Schoneboom, N., Cunningham, J., et al. (2000, March). Conducting valid and reliable play-based assessments. Poster session presented at the annual meeting of the National Association of School Psychologists, New Orleans, LA. Salvia, J., Ysseldyke, J. E., & Bolt, S. (2007). Assessment in special and inclusive education (10th ed.). Boston: Houghton Mifflin. Tamis-LeMonda, C. S., & Bornstein, M. H. (1996). Variation in children’s exploratory, nonsymbolic, and symbolic play: An explanatory multidimensional framework. In C. Rovee-Collier & L. Lipsitt (Eds.), Advances in infancy research (pp. 37–78). Norwood, NJ: Ablex. Tamis-LeMonda, C. S., Bornstein, M. H., Cyphers, L., Toda, S., & Ogino, M. (1992). Language and play at one year: A comparison of toddlers and mothers in the United States and Japan. International Journal of Behavioral Development, 15, 19–42. Tamis-LeMonda, C. S., Chen, L. A., & Bornstein, M. H. (1998). Mothers’ knowledge about children’s play and language development: Short-term stability and interrelations. Developmental Psychology, 34, 115–124. Tamis-LeMonda, C. S., Shannon, J., & Spellmann, M. (2002). Low-income adolescent mothers’ knowledge about domains of child development. Infant Mental Health Journal, 23, 88–103. Tamis-LeMonda, C. S., Uzgiris, I. C., & Bornstein, M. H. (2002). Play in parent–child interactions. In M. H. Bornstein (Ed.), Handbook of parenting (2nd ed.). Mahwah, NJ: Erlbaum.

559

Best Practices in School Psychology V

560

Chapter 33, Volume 2