DOCUMENT RESUME ED 461 408 PS 023 121 AUTHOR Moore, Gary T. Early Childhood Physical Environment Rating Scales, TITLE Dimensions of Education Rating Scales, Observation Schedules, and Behavior Maps for the Description and Measurement of Child Care Centers. PUB DATE 1994-00-00 NOTE 37p.; For closely related paper, see ED 377 950. PUB TYPE Book/Product Reviews (072) EDRS PRICE MF01/PCO2 Plus Postage. DESCRIPTORS *Day Care Centers; Early Childhood Education; *Educational Environment; *Evaluation Methods; Evaluation Problems; *Institutional Evaluation; *Physical Environment; Preschool Evaluation; *Rating Scales Infant Toddler Environment Rating Scale; Purdue Home IDENTIFIERS Stimulation Scale ABSTRACT This paper reviews four scales that have been used to evaluate the physical environment of, child care centers, and introduces five new scales that are designed explicitly to assess the physical environment of child care centers and their associated outdoor play environments. The psychometric properties, environmental content, and effectiveness of the Harms and Clifford Infant/Toddler Environment Rating Scale, the Home Observation for Measurement of the Environment Inventory, the Profile for Early Childhood Programs, and the Purdue Home Stimulation Inventory are discussed. The paper then discusses five sets of scales that have been developed to address the deficiencies of current physical environment scales: (1) the Early Childhood Center, Children, and Teacher Profiles; (2) the Early Childhood Teacher Style and Dimensions of Educating Rating Scales; (3) the Early Childhood Physical Environment Scales; (4) the Playground and Neighborhood Observation Behavior Maps; and (5) the Environment/Behavior Observation Schedule for Early Childhood Environments. In conclusion, the paper suggests the principle components for the development of a new, more comprehensive set of environmentally based scales for early childhood environments. (Contains 22 references.) (MDM) Reproductions supplied by EDRS are the best that can be made from the original document. Early Childhood Physical Environment Rating Scales, Dimensions of Education Rating Scales, Observation Schedules, and Behavior Maps for 00 c) the Description and Measurement of Child Care Centers' U.S. DEPARTMENT OF EDUCATION Office of Educational Research and Improvement EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC) *This document has been reproduced as Gary T. Moore received from the person or organization originating it. 0 Minor changes have been made to improve reproduction quality. Center for Architecture and Urban Planning Research ' Points of view or opinions stated in this docu- ment do not necessarily represent official OERI position or policy. University of Wisconsin-Milwaukee "PERMISSION TO REPRODUCE THIS MATERIAL HAS BEEN GRANTED BY G Abstract TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC)." It is well known that the quality of child care matters and that the quality of the physical environment matters. A series of scales have been developed for the description and measurement of the physical enviromnent of child care centers, for assessing teacher styles and philosophy of early childhood education, and for systematically observing children's cognitive and social behavior in any early childhood setting. This paper revieWs the need for such scales, offers a critical analysis of other scales currently available, and In reports on the Early Childhood Physical Environment Scales (EC-PES). conclusion, the paper suggests the principal components for the development of a new, more comprehensive set of environmentally based scales for early childhood environments. 41, My sincere thanks for assistance Paper submitted for consideration to Environnzent and Behavior. to Shari Sivakumaran and Nancy Genich. Running head: Early Childhood Physical Environment Scales. AVAILABLE BEST COPY 2 2 Early Childhood Physical Environment Scales The Need for Scales of the Physical Environment In much of the environmental and social science literature on child care, the construct of "environment" is most often limited to the effects of aspects of the social and organizational environment (e.g., amount and quality of adult interaction with children) and not the physical and/or designed environment. Conversely, those designing schools and child care centers tend to ignore the role of the social environment and often espouse, if unintentionally, an environmental deterministic position. In the child care field, a number of rating scales have been developed and are in wide use. Among the best known are the Ear 6, Childhood Assessment Profiles (Abbott-Shim & Sibley, 1992), the HOME Observation for Measurement of the Environment (Caldwell & Bradley, 1982, 1984), the various EarlY Childhood Environment Rating Scales (Harms & Clifford, 1980), the Purdue Home Stimulation Inventory (Wachs, 1990, 1994), and the Accreditation Procedures of the National Academy of Early Childhood Programs (National Association for the Education of Young Children, 1985). There are many similarities in structure and content among these scales. All have the explicit purpose of describing and evaluating different aspects of child care programs and centers. And yet most of them focus almost exclusively on the programmatic, social, or organizational aspects of the environmental quality of child care. Few pay attention to the physical designed environment of child care centers. 3 garly Childhood Physical Environment Scales Scale2 The Harms and Clifford Infant/Toddler Environment Rating The Infant/Toddler Environment Rating Scale (ITERS), developed by Harms, Cryer, and Clifford (1990), is one of the best known and most widely used scales to assess quality child care. An adaptation of the Early Childhood Environment Rating Scale by the same cumulative group, it consists of 35 items organized into seven subscales, each measured on a It is intended for the assessment of the quality of center-based infant and 7-point scale. toddler care for children up to 30 months of age. It is based on a broad definition of child of space but care environments including not only what the authors call the organization It is as comprehensive as any scale also interaction, activities, schedule, and provisions. available for the overall assessment of child care. Psychometric properties. Many so-called scales are developed and promulgated in the environment and behavior field without adequate study of their psychometric qualities. Not moderately valid so the ITERS. Studies show the ITERS scale is moderately reliable and vis a vis other available scales and experts' opinions. For example, the interrater average % agreement = 47 and average % agreement + /-1 point on the scale = .78, while the average Spearman rank-order correlation among six judges r = .61, subscales r = .74, Test-retest reliability is r = .58 to .82, subscales average r = .69, overall scale r = .84. Criterion overall scale r = .79. The internal consistency using Cronbach's alpha = .83. validity is 83% across two judges in six classrooms on a simple high-low scale discriinination. Portions of this segment of the paper have been published previously as a test review in Children's 2 Environments (Moore, 1994b). 4 4 Early Childhood Physical Environment Scales And content validity is 49 to 82% against several comparison instruments, and 86% on expert evaluations of item importance (all statistics from Clifford, Russell, Fleming, Peisner, Harms, & Cryer, 1989). The particular ways in which these tests were conducted were far from the most robust that could have been conducted, and the resulting figures are hardly astounding. Furthermore, several standard reliability and validity tests were not conducted (parallel forms reliability, split-half reliability, inter-item correlations, standard error of measurement of centers over a number of parallel tests, and construct validity; cf. Ghiselli, Campbell, & Zedeck, 1981). So we may conclude that the scale is somewhat reliable and valid. But is it environmental? To assess the environmental content of the ITERS, two Environmental content. quantitative content analyses were conducted on the scale. Of the 35 items, 14 (40%) have some physical environmental content (environmental defined here as the designed or planned environment, not the social or organizational environment, i.e., that part of the total environment that can be manipulated architecturally). For example, the item "Furnishing for routine care" refers to numbers of pieces of furniture, comfort and support, storage, and child-sized, i.e., clearly a physical environmental item. On the other hand, "Nap" doesn't contain any reference to whether napping should be in separate nap rooms, in double-- functioning nap/play rooms, or in partially partitioned napping spaces. The scale is silent on this important physical environmental issue, as it is on many others. Of the 396 detailed descriptors that are used to score the scale items (e.g., "diapering done near source of hot water," or "nap is scheduled appropriately for each child"), only 35 5 Early Childhood Physical Environment Scales the physical or 8.8% have any physical environmental content that could help one assess environment of the facility. Some of these descriptors are very good, like requiring softness and cozy special areas for high scores on "Furnishings for relaxation and comfort" and But correlating the separation of activity areas from circulation with quality child care. there are very few of them, less than 10%, despite the term "environmental" in the scale title. Furthermore, in many places the environmental characteristics of a test item are confounded with behavioral use patterns and developmental objectives, that is, many items are double- or multi-barrelled. "Furnishings permit appropriate independence for toddlers (Ex. toddlers use small chairs...)." What is being assessed: The environmental characteristic (furnishings, i.e., the smallness of the chairs), the behavioral use pattern (that toddlers do or do not use small chairs, which could be influenced not only by the characteristics of the furniture but also by staff, whether games are spread out on the floor or on tables, and so forth), or the developmental objective (independence)? The scale is silent on many environmental characteristics related to quality care. For example, while briefly touching on the issues of zoning and safety of children, the 1TERS is silent on age-mixing, which is so much a part of many progressive programs, and on the pros and cons of different spatial layouts now known from the research literature to be important to child development. Under "Room arrangement," the scale uncritically assumes one overall organizational pattern for infant/toddler centers--the box-car arrangement of a double-loaded corridor with self-contained classrooms. The scale is very good about the necessity for a variety of play areas for infants and toddlers (art, music and movement, 6 Early Childhood Physical Environment Scales blocks, pretend play, sand and water play for toddlers), but again it is silent on the environmental characteristics that might facilitate these types of developmentally appropriate play activities. On the items measuring "Peer interaction," not one descriptor relates to the designed environment. Yet, all other things equal, twice the amount of children's peer interaction is found in modified open plan centers than open plan centers (Moore, 1987). Similarly, regarding "Caregiver-child interaction," where again no descriptor relates to the physical setting, significantly more caregiver involvement with children occurs in spatially well- defined activity settings than in moderately defined or poorly defined ones (Moore, 1986). The ITERS scale is also silent on a number of other environmental issues that child care directors, architects, and other designers are confronted by each time they consider the facility program for a new or renovated center. Among these are location, size of center and ways to decentralize extremely large centers, spatial density, scale, image, circulation, Clifford et al. seem to agree with this and character of the outdoor activity areas. assessment, also making the point (p. 24) that only "one or two" of the ITERS items correspond to the physical environment. Other Early Childhood Environment Rating Scales The same lack of physical environmental content is true also of the other available early childhood scales. Early Childhood Physical Environment Scales 7 Home Observation for Measurement of the Environment Caldwell and her colleagues (e.g., Caldwell & Bradley, 1982, 1984) developed the well-known and widely used Home Observation for Measure of the Environment (HOME) Inventory. The two versions, while focused on infants and preschool age children in home settings illustrate the lack of environmental content. For infants, the subscale of "Organization of Physical and Temporal Environment" contains items like "When mother is away, care is provided by one of three regular substitutes" and "Someone takes child to grocery store at lest once a week." These items, while undoubtedly important to the total environment of the home for children, are better conceptualized as part of the interpersonal or temporal environment, not the physical designed or planned environment. The older preschooler version under the subscale "Physical Environment" has better items such as "Building has not potentially dangerous structural or health defect" and "There is at least 100 square feet [9 mi of living space per person [or child--differs in different reports on their work] in the house." A variation of this scale is being developed for use in a current large-scale US national study of the quality of child care. Called the Early Childhood Child Care HOME Inventory, the "Physical Environment" subscale now has seven items in it, including items relating to "Outside play environment appears safe," "Neighborhood is aesthetically pleasing," and "Rooms used by children are not overcrowded by furniture." While this is an improvement over the early versions of the HOME Inventory, even this scale is silent on a whole range of other physical environmental features judged from other research to be developmentally advantageous. 8 Early Childhood Physical Environment Scales Bradley (1982, 1985) summarized a series of studies of predictive or criterion validity showing that the HOME Inventory administered at 6, 12, and 24 months was a better predictor of children's intellectual development at age 3 on the Stanford-Binet Intelligence Test than were the Bayley Scales of Infant Development, quite an impressive finding. The Physical Environment subscale, for instance, showed multiple correlations of 0.40, 0.39, and 0.41 at 6, 12, and 24 months with the Stanford-Binet IQ. One wonders how much better the predictive validity might have been if a wider and richer variety of physical environmental variables had been included. Profile for Early Childhood Programs Shim and Shibley (1992) have developed a Profile for Early Childhood Programs that The Learning Environment 17-item is explicitly intended for use in child care centers. subscale calls for a variety of activities to be available for children (e.g., art materials, science materials, language materials), five items deal with the arrangement of the classroom and definition for space (e.g., "At least three partitions are used to form physical boundaries at least three learning areas"), and two items pertain to whether or not the classroom environment reflects the child as an individual (e.g., "An area exists in the room where one or two children may choose-to work alone"). This scale is the most environmental of those "on the market." But while having some items pertaining to the physical environment of the "classroom," it also, like the others, is silent on a whole range of other important issues like the neighborhood context through which the child moves, the site, the organization of the building as a whole (e.g., open plan, 9 9 Early Childhood Physical Environment Scales size is dealt closed plan, or modified open/closed plan), the size of the center (and whether outdoor with by decomposition into houses, pods, or modules, or not), and the quality of the activity areas. Purdue Home Stimulation Inventory The Purdue Home Stimulation Inventory (Wachs, 1990, 1994) is an attempt to measure the physical and social parameters of the child's home environment. As stated by its author, which social- two physical environment subscales refer to the "physical stage upon interpersonal interactions take place, rather than upon these interactions themselves" This is the clearest differentiation between social and physical (Wachs, 1990, p. 1). environmental constructs of any of the early childhood scales available even in the research literature. Several items are very interesting, and more But what of the items themselves? physically environmental than most other scales, like "Stimulus shelter" away from noise and people, language materials "in places where child has free access to them" and small manipulables "within reach," "rooms to people ratio," "audio-visual responsive toys," "if the physical setup of the child's home is such that the child's movements are restricted within the home" (with examples for coding given), and "room decorations." This scale, much more than any of the others, is theoretically driven. It derives from environment in early a strong view about the importance of the stimulus properties of the childhood development (Wohlwill & Heft, 1987) and especially the importance of the active 10