Evaluation of pedagogical quality in early childhood education: A cross-national perspective.

Abstract. In a comparative study between Germany and Sweden, observers from different countries and cultures make parallel and independent observations of the quality in early childhood education. For evaluation of quality, the observers use the Early Childhood Environment Rating Scale (ECERS), combined with a documentation of the perceptual process underlying the ratings of quality with the ECERS. The study focuses on how the concept of quality in ECERS is concretized in pedagogical processes in early childhood education, on how those pedagogical processes can be made visible and on the validity of evaluations of quality with the ECERS in cross-national comparative studies.

Quality has been an important aspect of the pedagogical debate throughout the 20th century. Today, maybe more than ever, modern society is characterized by constant change. Economic, social, and cultural relations are transformed through global markets, organizations, and media. This highly specialized society demands that the rising generation embrace its social and cultural inheritance through the learning of history, language, values, and knowledge, preferably after being both critically analyzed and improved. Furthermore, as societal boundaries shift, the young generation must be able to cooperate and communicate in a highly diverse environment. To live a good and meaningful life in this complex society, each child needs to develop and make full use of his/her own abilities and interests (Bartley, 1998). In most countries, the educational system is supposed to guarantee this achievement and ensure equality to all children in the society. Quality throughout the educational system is therefore an important issue for both the individual and the society.

In many countries, early childhood education has become the first step in the general education system. Research shows that the level of quality in early childhood education can have long-term effects on a child's attitudes towards further education and educational achievement (Andersson, 1992; Clarke-Stewart, 1987; Osborn & Millbank, 1987; Phillips, McCartney, & Scarr, 1987; Schweinhart & Weikart, 1980; Sylva, 1994). To meet the needs and the rights of the child, as well as the demands of society, the content of early childhood education and the pedagogical environment have to be organized in such a way that the surrounding world becomes visible, understandable, and comprehensible, thus creating optimal possibilities for the child to learn and develop (Bartley, 1998; Hultqvist, 1990; Pramling, 1994; Pramling Samuelsson, 1997, 1998; Sylva, 1994). To guarantee quality in early childhood education, methods to evaluate, describe, visualize, and improve various pedagogical processes must be made available.

However, knowledge about what constitutes quality, how it is concretized in pedagogical processes, how different levels of quality can be evaluated, and how data can be interpreted is rather limited. Therefore, the unique design of this project attempts to shed some light on these important aspects of early childhood education. The uniqueness of the project is that researchers (1) from different countries (Sweden and Germany) and cultures make parallel and independent observations of the quality in early childhood settings. To evaluate the presence of quality, the researchers use the Early Childhood Environment Rating Scale (2) (ECERS) combined with documentation of the perceptual process underlying the ratings of quality.

The study is embedded in a joint research project of quality in early childhood education between Germany and Sweden. Its focus is on how the concept of quality in EGERS is concretized in pedagogical processes in early childhood education, on how those pedagogical processes can be made visible, and on the validity of evaluations of quality with the ECERS in cross-national comparative studies. This study addresses these research questions: How can various pedagogical processes in early childhood education become visible? and How is quality embedded in various pedagogical processes in early childhood education, according to the ECERS?

A Pedagogical Perspective of Quality

This study emanates from a pedagogical perspective of quality that takes the perspective of the child and focuses on what is best for a child's learning and development. To live a good life today and tomorrow, a child needs an educational setting where he/she can, together with peers, learn and develop a multitude of competencies and skills, meeting the norms, values, beliefs, and attitudes held by society. According to this perspective, quality in early childhood education depends on structural aspects and pedagogical processes, as well as social and cultural values and traditions in the surrounding society. A pedagogical perspective of quality is based on years of research and practical experience in early childhood education settings, inferring that certain aspects of quality benefit a child's learning and development more than others do. To define what is meant by good quality, a shared understanding of what those qualities are--and an agreement on how they are rooted in pedagogical processes in early chil dhood education--is necessary in comparative studies of evaluations of quality (Andersson, 1995, 1999; Harms & Clifford, 1980, 1983; Karrby, 1992, 1997, 1999; Karrby & Giota, 1994, 1995; Karrby, Giota, Sheridan, & Daversjo-Ogefelt 1995; Sheridan, 1997; Sylva, 1994; Tietze, Cryer, Barrio, Palacios, & Wetzel, 1996; Tietze & Giota, unpublished manuscript).

A Common Instrument To Evaluate Quality

A shared understanding of quality and the use of common instruments are necessary to evaluate the quality in early childhood education in a comparable way. The Early Childhood Environment Rating Scale (ECERS), developed by Harms and Clifford in 1980, is a common and widely used instrument reflecting a broad understanding of quality in early childhood education. The rationale underlying ECERS is that common dynamics, materials, etc. must be present to make developmentally appropriate gains-- physically, socially, intellectually, and emotionally (Harms & Clifford, 1980, 1983).

The American version has been applied in various other English-speaking countries, including the United States, Canada, Bermuda, Bahrain, Australia, the Philippines, Hong Kong-Singapore, and the United Kingdom. In addition, translated national adaptations are available in Germany, Italy, Spain, Greece, and Sweden. In a number of studies throughout various countries ECERS has been used for research, as an instrument for comparative assessment, and for investigations of the relationship between process quality and the various conditions of early childhood education provision (Karrby & Giota, 1994, 1995; Phillips, Scarr, & McCartney, 1987; Rossbach, Clifford, & Harms, 1991; Scarr, Eisenberg, & Deater-Deckard, 1994; Whitebook, Howes, & Phillips, 1989). In Sweden, ECERS also has been used as an instrument for self-evaluation and as a base for development work and improvement of quality in early childhood education (Sheridan, 1997; Andersson, 1999).

ECERS consists of 37 items, which define different levels of quality in typical situations of early childhood education. The criteria of ECERS originated from years of research into children's learning and development and from practical experience. The criteria focus on basic and general conditions for a child to learn, such as being active, participation, communication, and cooperation with others, and the right for a child to develop his/her own interests and goals for learning. Beneath those criteria of quality lie assumptions of what constitutes quality in early childhood education and how different levels of quality are rooted in pedagogical processes.

Each item is rated individually on a seven-point scale using criteria from basic physical conditions, such as space and materials to teacher competence and ability to create an environment conducive to children's learning and development. The ratings are based on both easily observable conditions, such as room display and sufficiency of materials, and information obtained from interviews with the teachers and directors, as well as on impressions that must be interpreted or implied from observed communication and social interaction. In order to rate the quality of an early childhood setting with the ECERS, the total environment must be observed: space, equipment, material inside and outside, surroundings, documents, the atmosphere, interactions between teacher and children, the social climate, and the educational style. Therefore, one might regard the rating process with different assessment scales as a quantification of qualitative data: "Interestingly, although we term this data 'quantitative' they are attempts to measure 'qualitative' aspects of the teaching and learning process" (Pascal & Bertram, 1994, p. 11).

The decision to use ECERS in this study is based both on the fact that it assumes a child's perspective, and that the criteria are defined in such a way that ratings of quality in different early childhood education settings are comparable in a national as well as a cross-national perspective.

Validation Through Combined Methods

The best way to validate the results from a study is to triangulate multiple techniques (i.e., to use strategies in which different methods are combined in both the design of the study and in the analysis and interpretation of the data) (Wolming, 1998). In this study, measurement using the ECERS has been combined with documentation of the perceptual process underlying the ratings of quality with the ECERS. By using a quantitative as well as a qualitative approach to the collection of data, the analysis should therefore strengthen the validity of the results. What is important for the validity of interpretations of ECERS scores is that both the perceptual process underlying the ratings of quality and the measurement procedures are scrutinized in the context of the overall aim and setting of the study (Pedhazur & Pedhazur Schmelkin, 1991).

Method

Design

The core of the comparative project consisted of parallel and independent evaluations of quality, using ECERS conducted by two teams of trained observers, one from Germany and one from Sweden. Each team included three researchers. All of the observers had previous experience using the scale and had learned how to consider and evaluate the quality from the same criteria in diverse early childhood settings.

The 20 participating child care centers, 10 in each country, were randomly chosen. In Germany, the sample was a subsample drawn from a sample of centers in the states of Berlin and Brandenburg. In Sweden, the sample is from the cities of Gothenburg, Lerum, and Skovde. The ages of the children in the evaluated child care centers was between 1-5 years old in both countries.

Each child care setting was visited two times, in a period of two continuous weeks, by two 3-member observation teams representing both countries. Altogether, each unit was evaluated by each of the six observers according to a predetermined schedule with a systematic variation of combinations. A positive effect of this design was that the effects of memory of previous observations were eliminated, as no observer visited the same center twice. One serious threat to the validity of this study was the stress the pedagogues might have felt during observations. However, every child care setting was evaluated twice by two different teams and the results showed no difference between the scores from the first and second ratings (Tietze & Giota, manuscript).

The validity of the measuring process is one of the most important questions and was constantly discussed throughout the joint project. For example, the language barrier was one of the main topics of concern when the project was first conceived. After much discussion, the decision was for the researchers to translate to each other, rather than "guessing" at what seemed to occur in different situations.

ECERS and Scoring Rationale

National adaptations (3) of the ECERS to rate the quality in the 20 child care settings were used by both the German and the Swedish research groups (Karrby, 1989; Tietze, Schuster, & Rossbach, 1997). Evaluations of quality with the ECERS are based on both observations and interviews of pedagogues, which are then compiled on a scale between 1 and 7. The result is thereafter presented as an average score of the total sum. No statistically significant differences were found among teams of observers from the two countries, or between the Swedish and the German version of the ECERS. Two thirds of the explained variance is due to country and child care setting. These results indicate considerable differences in the average quality of the 20 units included in the sample, as well as among the 10 units in each of the two national samples. The ECERS' scores mean for the German child care centers is 3.8 on a seven-point scale; for the Swedish child care centers, it was 4.7. The ECERS scores for the German centers range from 3.0 to 5.1, and in Sweden from 4.0 to 5.2 (Tietze & Giota, manuscript).

Data Collection

The statistical analysis provided a rather coarse measure, which mirrors the average of the total quality within and between the sites in both countries as evaluated with the ECERS. Therefore, a qualitative approach also was undertaken to gain additional information from the different observers' perceptions of quality in participating child care settings--both underlying their systematic evaluations of quality by the ECERS and beyond them. In addition to the ECERS ratings, the observers gave the rationales for their scoring of selected items during an observation, and described them informally. A balanced design was developed, which ensured that each observer described the same subset of three items for a specific child care center.

Data Analysis

According to Munton, Mooney, and Rowland (1995), deconstructing concepts of quality is one way to visualize its underlying values, as well as the processes that went into production of those definitions. In this study, the researchers chose the opposite way. To describe and visualize observed pedagogical processes in early childhood education, they reconstructed the perceptual process, which underlies the evaluation of quality with the ECERS (Pedhazur & Pedahazur Schmelkin, 1991). Perceptions are, in this article, used as a wider definition to include the observers' thoughts, beliefs, values, experiences, and feelings.

As earlier described, ECERS dfines quality in early childhood education according to a scale of 1 to 7; the ECERS mean for the German child care centers is 3.8, and the Swedish child care centers is 4.7. What do these numbers mean for a person who is not familiar with the ECERS? Even for a researcher using the ECERS it can be very hard to comprehend and visualize how different levels of quality manifest themselves in various pedagogical processes in early childhood education. Therefore, to be visualized and analyzed, the observers documented the perceptual process underlying the ratings of quality with the ECERS. The qualitative analysis and the reconstruction of the perceptual process are based on the documentation of the observers' unique and shared perceptions of quality, in relation to the ratings of quality with the ECERS in the Swedish and German child care centers. This study focused on a country-specific perspective of quality, and therefore the perspective of the individual observer has been condensed to the perspectives of the German and the Swedish teams.

The documentation was analyzed in different stages to explore various themes and pedagogical processes in early childhood education, which are related to single items or to groups of items in the ECERS. Those themes are chosen to mirror different dimensions of quality and pedagogical processes in early childhood education as evaluated with the ECERS. Themes, such as inside space and equipment or a child's potential for language development, are related to several items in the ECERS. Special emphasis also is given to aspects of quality that are factored into the evaluation of several items, even though they are not directly rated in the ECERS, such as the presence of educational goals, values, and attitudes in regard to the importance of different practices, power relationships between children and adults, and relationships with parents. Some of these factors (for example, the presence of educational goals) are considered as indicators of the teachers' pedagogical awareness.

Each observer analyzed his or her own documentation and thereafter compared the results within his or her own team and between teams. Thereby, the documentation was exchanged, discussed, analyzed, and interpreted during several occasions. A further analysis, in connection with the reconstruction of the perceptual process, was made independently by the authors of this article. The analysis shows a high agreement among the observers of how low or high quality is concretized in various pedagogical processes in early childhood education, but also that minor individual differences exist.

Results

To describe the evaluated differences of quality between child care centers in Germany and Sweden, the perceptual process was reconstructed with a focus on the quality of various pedagogical processes in early childhood education, using a nation-specific perspective.

The reconstruction of the perceptual process is presented from the German team's shared perspective of quality in the evaluated Swedish child care centers, and vice versa. In a variety of common situations and pedagogical processes in early childhood education, different themes are presented in relation to a description of low and high quality according to the criteria of analogous items in the ECERS. The researchers have chosen to present a documentation of five different themes reflecting various pedagogical processes in early childhood education. Each theme begins with a description of the criteria in ECERS, followed by a selection of the German team's documentation on early childhood education in the Swedish child care centers and then from the Swedish team's documentation on the German child care centers. In addition, the documentation is related to an average of both teams' numerical ratings of quality, and discussed from a viewpoint of validity as well as high and low quality in a cross-national perspe ctive.

Inside Space and Equipment for Children

Four of the items in ECERS evaluate the quality of the child care center's space, materials, furnishings, and display. Low quality is characterized by limited space, insufficient amount of furniture and materials for routine care and learning activities, a lack of an awareness of children's needs and interests, and improper or inappropriate use of materials. Characteristics of a high level of quality are not only the existence of physical resources, but also how they are made available and used to benefit each child's development and learning, as well as how they benefit the professional needs of the teachers.

The German Team's Documentation of the Swedish Child Care Centers. In the Swedish child care centers from this study, more than five rooms are used by one group of children. The main room, with a nearly complete set of kitchen equipment, is used for different kinds of activities, such as reading, painting, and playing with toys, as well as for breakfast, lunch, and preparing meals. In this room, chairs and tables are adapted to the height of the adults, together with special chairs--the so-called Trip-trap-chair--so that children can sit at the tables built for adults. The other rooms are more specialized and are used daily for art activities, dramatic play, and construction play; there also is a cozy room that serves as a place to read books, tell stories, hear music, and relax. Moreover, there is one bigger room often used by two groups of children for gross motor activities and circle times.

An additional difference from the German child care centers was the amount of space and availability of equipment for teachers and other adults. The structure of the rooms for the adults reflected, in a materialized manner, the concerns of respecting adults' mutual needs.

The Swedish Team's Documentation of the German Child Care Centers. In Germany, most of the units in a child care center consist of only one room, in which all activities take place. The Swedish team's first impression focused on the limited area of space and the scarcity of equipment and materials. Most apparent was the lack of materials and furniture for creative activities as well as the absence of water (e.g., a sink) in the unit. Nothing in the room gave the message that it is also a working place for grown-ups, as all the furniture and equipment was adapted for children. This first impression was, however, often followed by a feeling of admiration for how this group room, in spite of its restricted space, is planned with different activity corners, such as a construction area; a doll play corner; tables for games, meals, and creative activities; and a resting place. The material was often well-organized and in connection with the activity area.

In comparison, despite previous knowledge of the differences between physical conditions in German and Swedish child care centers, both teams were surprised by conditions in each other's settings. The German team focused on the amount of space and materials in the Swedish child care centers. Their documentation contained detailed descriptions of the available space, materials, and premises for the teachers. The Swedish team focused on the limitations and the lack of space and equipment in the German child care centers.

Both teams evaluated the quality of physical conditions higher in the 10 Swedish child care centers than in the German centers. For example, the numerical scoring of the furnishing and display for children, item 6, resulted in an average of 4.60 in the Swedish centers and 3.59 in the German; the scores for furnishings for relaxation and comfort, item 8, resulted in an average of 4.00 in the Swedish centers and 3.59 in the German. However, to rate these items, the observers also had to consider how the teachers promote and stimulate regular and independent use of physical resources, as well as to what degree the resources are used to provide a variety of learning experiences. Therefore, it is interesting that the highest score for both German and Swedish centers are for how the rooms are arranged (item 9: in Sweden, 4.82, and in Germany, 4.05). The lowest scoring is for furnishing for learning activities (item 7: in Sweden, 3.98, and 3.22 in Germany). The numerical ratings of items for physical resources, as well as the reconstruction of the perceptual process underlying those items, show that there is less space and a lower standard of material resources in the 10 German child care centers as compared to the Swedish centers.

Goals and Planned Activities

The ECERS item "schedule" describes the balance of structure and flexibility, as well as the transitions between activities. On a low level of quality, the routine care takes up most of the day. There is hardly any planned activities and the schedule is either too rigid or flexible. On a high level of quality, the schedule provides a balance between planned activities, routine care, and free play, both indoors and outdoors. The schedule also includes measures to meet children's individual needs.

The German Team's Documentation of the Swedish Child Care Centers. In Sweden, early childhood care and education is based on a structured schedule, with fixed time for daily routines and several planned activity periods for the whole group and for small groups. Different activities are often conducted simultaneously. At first glance, the authors were deeply impressed by the prevailing variety of planned pedagogical activities offered during the morning in the units visited. After a week of observation, more and more similarities were noted among the child care centers. For example, the book Mama Mu, a very famous Swedish picture book, was found again and again. Most of the pictures, rhymes, songs, and rhythmic movements, even dances derived from the book, were presented in differing ways during the gathering times. Other examples of circle themes depended on the season. Those circle times are conducted every day, and involve playful, creative, cognitive, and language activities, as well as fine and gross mot or activities.

The obvious difference from the German experience with circle time lay in the manner in which teachers were able to involve children without being intrusive and controlling. This form of teaching style often characterizes German full-group activities, in which each child has to remain as long as the activity lasts. It was evident that the Swedish teachers were aware of who would be inclined to distraction or boredom, and they chose a place next to those children and offered, if necessary, hugs, explanations, suggestions for alternative activities within the shared session, as well as verbalized perspectives for the time after the gathering. When teachers wanted a child to consistently join the circle, they made sure not to threaten shame, use corporal punishments, or isolate the child as a means of control.

The Swedish Team's Documentation of the German Child Care Centers. The Swedish team's impression of the German programs is that they have very little organized activity or gatherings during the day. The days appeared to proceed in a pattern of mostly free play, along with a few organized activities, such as painting or handicrafts. During the two weeks of data collection, no written planning of any kind was observed. There was no evidence that the teachers organized their programs according to the children's interests, previous experiences, or individual needs.

In Germany, planning seems to be conducted on a daily basis as the teacher chooses short activities (e.g., making lanterns for Halloween or visiting a playground). "Planning" around a theme seems to consists of different excursions and activities. In addition, the ongoing theme and related activities are not connected to expressed goals, competencies, or understandings the teachers wanted the children to learn and develop.

In order to rate this item, the whole structure of the day, the interactions between teacher and children, the teacher's awareness, the presence or absence of educational goals, etc., had to be taken into account. In ECERS, planned activities are often noted as integral parts of an item's score. Planned activities are evidently expressed, concretized and implemented differently in child care centers in Germany than in Sweden. The researchers interpret this finding as if the concept of planned activities has another meaning for teachers in Sweden than for German teachers. In Sweden, there is a tradition of long-time planning for the whole group and for small groups that supports each child's development and learning. For example, if the teachers have chosen the theme "Autumn," the theme is supposed to have a concrete goal about what aspects in nature they want the children to understand and what experiences are necessary in order to achieve this. The planning of the theme and its goals are, in most Swedish chi ld care centers, written down and well-displayed for parents and other interested parties.

The Swedish team customarily read about the goals of the theme, and the long- and short-term procedures as soon as they began a new child care unit. The presence of goals and related procedures were so taken for granted by the Swedish team that it was not even mentioned in the documentation of the Swedish child care centers. Over the two-week period of quality ratings in Berlin, the Swedish team gradually became aware that there was an absence of goals and written plans/schedules in the German centers. For the German team, it was the other way around. According to both teams' numerical scoring on schedule (item 26), the evaluated difference in quality between the 10 Swedish and the 10 German child care centers is obvious--5.15 for the former compared to 3.85 for the latter. The structure of the day had also become visible through the reconstruction of the perceptual process underlying the ratings of quality of item schedule. The documentation shows how this difference in quality is concretized in various peda gogical processes throughout the day.

Encouragement of Language Development In ECERS, four items are directly related and several others indirectly related to a child's potential to develop communication skills. According to the criteria in ECERS, a low-quality center is characterized by few materials and few planned activities that can help children to understand and develop language skills. Language is not used by the teachers to extend and encourage reasoning, but rather to control children's behavior and manage routines. A high-quality center is characterized by a diversity of materials, and by how they are presented and used to stimulate a child's language development. Important for a high rating of quality is the encouragement of expressive language throughout the day, the variety of daily planned activities for language development, and the teacher's competence in creating opportunities for children to verbally express different thoughts and ideas and expand on ideas. High quality also is marked by the teacher's competence in encouraging c hildren to reason and to use actual events and experiences as a basis for concept development, as well as the teacher's efforts to have an informal conversation with each child every day.

The German Team's Documentation of the Swedish Child Care Centers. Language development, especially the cultivation of the mother tongue, plays a major role in children's education. The Swedish teachers used many situations and activities in order to encourage children to communicate; for example, they often greeted children individually with an informal conversation at the beginning of each day. Furthermore, the teachers used mealtimes as opportunities for conversation.

There were also daily planned group activities, as well as small-group activities, to develop language abilities. These activities stimulated the exchange of information, question-and-answer sessions, and story telling. In the same way, the activities influence vocabulary, inflection, sentence structure, and articulation. The communication was based on running projects or chosen topics. The parents were able to support the activities for understanding or using language because they were informed about the topics and the chosen texts of songs, finger plays, or rhymes. All this information in written form was attached to an information board that parents referred to during the observations.

In classes with mixed-age groups, however, the very young children's needs were not being appropriately addressed according to their individual development, as teachers tried to better prepare the older children for school, particularly focusing on mathematics and language skills. The "school"-oriented teachers sometimes did not motivate the very youngest children, for whom many of the activities were too demanding.

The Swedish Team's Documentation of the German Child Care Centers. Language development seemed to be of minor interest in most of the 10 German child care centers. There were few books, mostly kept in inaccessible places, so that they were not readily available. During the visits, teachers seldom read books to the children, used materials or planned activities to encourage language development, or had dialogues with the children. However, in some of the high-quality units, the teachers communicated with the children throughout the day. They used different activities to make the children reflect, ask questions, and explain themselves, especially during creative activities and meal situations. In these centers the teachers sang songs with the children, told stories and rhymes, and encouraged the children to talk about different experiences.

By comparison, a child's language development appeared to be of greater importance in the 10 Swedish child care centers. In Sweden, there are more organized activities and materials to challenge children to use and develop their language. The numerical ratings of the language reasoning experiences (item 11) is 4.56 in Sweden and 3.89 in Germany; using language (item 12) was evaluated with an average of 4.67 in Sweden and 3.61 in Germany. The ratings on using learning concepts (item 13) was 4.51 in Sweden and 4.07 in Germany. However, from the German team's point of view, the encouragement of language development in the Swedish centers takes place more in a general sense than as a plan based on children's individual abilities, needs, and interests. The highest scoring in both countries is on informal use of language (item 14), which was 4.74 in Sweden and 4.08 in the German centers. These high scores are supported by the researchers' documentation, which describes how some teachers in German high-quality units have splendid interaction and communication with the children throughout the day.

Play

The ratings of low and high quality on the item "play" are, according to the criteria in ECERS, dependent on the materials provided, the time and space to play, pictures, stories, excursions, and experiences used to enrich and stimulate play, as well as on the interactions between the teacher and children during children's play.

The German Team's Documentation of the Swedish Child Care Centers. Free play is, in many German child care centers, the prevailing activity of early childhood education. Opportunities for free play were very limited in Sweden. In Sweden, free play seems to occur before breakfast, between the planned activities before noon, and during the outdoor activities. The children use all the space available, both inside and outside, for their play. They especially use the smaller rooms for block building, role play, and paintings, and they use the multi-purpose room for imaginative or gross motor play. The teachers move around between the rooms, actively fostering and widening the self-selected activities or moderating the joint activities of the children.

The Swedish Team's Documentation of the German Child Care Centers. In Germany, free play seems to be the main activity in early childhood education. According to the observations, the children mostly played by themselves throughout the day. The teacher is in the same room as the children, but usually only interacts with the children during their play when conflicts occur. Each unit has a fully equipped doll play corner and materials for block play.

In comparison, the documentation and the reconstruction of the perceptual process underlying the ratings of quality on play show that time for play differs between the German and Swedish child care centers, as did the teacher's role in children's play. In Germany, the children have more time to play by themselves, but with less material, support, and stimulation from the teacher. In Sweden, there is less time for continuous play, but the teacher interacts more with the children during play and uses the play as a learning opportunity. In both countries, historically, play has been an important role in early childhood education. Despite this fact, one of the lowest-scoring items in both countries is play (item 25). The scoring for the Swedish centers is 4.08, while it is 3.42 for the German centers.

The Social Climate and the Educational Style

This theme is based on a group of four ECERS items in which the item "tone" focuses on the general impression of the quality of interaction. A low quality is characterized by an atmosphere in which both children and teachers seem to be strained, irritable and angry, and in which children often cry and physical contact is principally used for control. A high quality of interaction is characterized by a calm but busy atmosphere. The teachers and children seem to be relaxed and happy. Adults show warmth in physical contact and a mutual respect exists among the adults and children. The teachers also seek to prevent problems by careful observation and skillful intervention, and through the development of the children's social skills.

The German Team's Documentation of the Swedish Child Care Centers. The overall impression of the social climate and the educational style in the Swedish centers is characterized by an interaction between teachers and children that promotes open-mindedness, self-reliance, and ego-resilience in their autonomous and social activities, as well as in structured learning situations. The atmosphere in the groups is very pleasant. The children are not aggressive or in conflict with peers as often as they are in the German child care centers. The children joined the adults' proposals for gatherings or small-group activities and appeared to be fully interested. They participated cooperatively, even though they sometimes were not truly interested by the offer to join. Social trust and trust in themselves seemed highly developed among the children. In general, they seemed emotionally more balanced and happier than children in the Berlin early childhood education centers.

The Swedish Team's Documentation of the German Child Care Centers. There was great variation in the social climate and the atmosphere among the 10 German child care centers. In some units, there was a great deal of conflict between the children; as a result, they often cry. In others, there is a pleasant atmosphere in which the children help and support each other. The atmosphere and educational style vary from no control, to harsh control and obedience, to an atmosphere with mutual respect between children and the teacher. In the latter, the teacher created a social and educational climate in which the children were encouraged to ask questions, investigate, and learn.

The Swedish team detected differences in the personal relationship between teacher and child, teacher and parent, and between countries. In Berlin, this contact seems to be more formal; hence, there is an observable "distance" between them. Between the parent and the teacher, this distance is expressed in subtle ways. The parent and teacher rarely use first names, the conversation between them is very proper and reserved and the parent seldom goes into the children's group room. There is less body contact (e.g., hugs) between the teacher and children, and fewer dialogues or conversations took place between them, compared to the Swedish centers.

The social climate and the educational style in the German child care centers observed can be characterized by three interaction styles between the teacher and the children: the "laissez faire," the authoritarian, and the democratic. The democratic climate seems to be found where the teachers from "the old east" (i.e., the former East Germany) have combined their working methods with pedagogical approaches from "the old west" (i.e., the former West Germany). Some of these teachers have succeeded in creating something new. The interaction with the children is very good and they appear to have awareness and a visible "pedagogical structure" in their work. In the centers in Sweden, the educational style, social climate, and atmosphere, as compared to Germany's, are more democratic and homogenous.

The numerical scoring is supported by the documentation, which shows that the evaluated quality ranking on the atmosphere, the educational style, and the interaction between teacher and children are higher in Sweden. The ratings of free play (item 29) are 4.58 in Sweden and 3.94 in Germany. On supervision (item 27), the ratings are 5.22 in Sweden and 4.54 in Germany. In Sweden, the teacher not only interacted more often with the children during free play and organized activities, but also possessed a more affirming and warm approach. The ratings on tone (item 32) are 5.10 in Sweden and 4.21 in Germany. In comparison to the German child care centers, Swedish children are divided into various grouping patterns, depending on the content of the activity and the children's ages. The ratings on group time (item 30) are also higher in Sweden (5.20) then in Germany (3.78).

Summary and Conclusion

An Evaluating Method To Visualize Pedagogical Processes

The result of this study is based on a qualitative approach used to concretize and visualize pedagogical processes in early childhood education and to validate the numerical ratings of quality with the ECERS, as well as with the interpretations of the outcome. In addition to ratings of quality with the ECERS, the observers gave the rationales for their scoring of selected items during an observation and described them informally. The perceptual process underlying the ratings of quality was thereafter analyzed and reconstructed from a country-specific perspective. Through the reconstruction of the perceptual process, an overall picture of the respective country's child care centers' profiles and levels of quality appears, as does differentiating characteristics of quality in a cross-national perspective.

The interpretation--based on both the numerical ratings and the documentation--indicates that the observers both rate and describe a quite similar picture of the different levels of quality in and among child care centers in Sweden and Germany. The analysis shows that the documentation in general is more or less based on the descriptions of the criteria on different levels of quality in the ECERS. Many of the descriptions explain which parts of the evaluations are based on observations and which on interviews with the teachers. However, the amount and precision of the documentation vary both between observers and situations. Common among all the observers is that they provided more detailed descriptions if there was: unusual space, atypical material conditions, or unusual activities on the part of the children and the teacher. The analysis of the documentation also indicates that each of the team members showed slightly different habits of perception and that the observers sometimes focused on different aspects of quality. For future use, the evaluation method should, therefore, include more structural aspects of quality that all observers can document. In this study, the researchers chose to analyze the data through a country-specific perspective. However, the documentation also can be analyzed from the perspective of the observer (e.g., Karrby, 1999) as a comparison between high- and low-quality child care centers, or by using combinations of these perspectives.

ECERS

The criteria in the ECERS focuses upon those aspects of quality that, according to research and practical experiences, are regarded as essential for a child's learning and development in a broader sense. The criteria in ECERS consist of skills and abilities that are important for a child to develop, coupled with the manner in which they are carried out, which results in different levels of quality in early childhood education. According to ECERS, a child's learning process and creation of knowledge are better developed if the early childhood atmosphere is open and positive; if the environment is safe, healthy, and stimulating; and if a child can take an active part in different processes and activities, with professional teachers who know how to create opportunities for a child to develop skills and competencies by using actual events and experiences as a basis for learning. High quality, according to ECERS, is therefore in accordance with those skills and abilities that throughout the educational system, in most European countries, are agreed upon to be important for child development. Everyday life skills are critical, such as the ability to cooperate; be responsible, active, creative, communicative, flexible, reflective; solve problems; take initiative; think critically; and learn how to learn (European Union, 1996).

Reconstructing the Perceptual Process

The reconstruction of the perceptual process underlying the ratings of quality with the ECERS verifies that the criteria in ECERS reflects pedagogical processes in early childhood education, as well as different levels of quality, in a distinct and modulated way. The comparative study between Germany and Sweden shows that the pedagogical quality is higher in the 10 Swedish child care centers as compared to the 10 German child care centers, when evaluated with the ECERS. The results also indicate considerable differences in the average quality of the 20 child care centers included in the sample as well as among the 10 units from each of the two national samples (Tietze & Giota, manuscript).

According to the reconstruction, Swedish child care centers are, in relation to the German ones, marked by very good personnel, space, and material conditions. A quality child care center is not only a good place with good conditions for children and their parents, but also one with good working conditions for teachers. Care and education in most of the Swedish centers are based on a structured and balanced schedule, with fixed time set aside for daily routines, and several planned activity periods for the whole group, as well as for more homogeneous subgroups (e.g., by age, interests, etc.). Fostering social skills and language development plays a major role in education. In Sweden, the atmo-sphere in the groups is often very pleasant and more democratic than in Germany. An OECD report (Utbildningsdepartementet, 1999), also found the Swedish approach to be democratic.

The reconstruction shows a far greater variation in and among the German child care centers than in the Swedish centers. This variation is to be found in the external as well as in the internal structure of the child care center. The observed pattern of the external structure in the participating German child care centers can be roughly generalized as follows: The physical conditions are rather poor as compared to Swedish standards. There is little space and sparse equipment and materials. In the centers visited in Berlin, the children stayed in the larger group most of the day. There was no small-group organization in an effort to meet children's individual interests, experiences, and needs. What especially caught the Swedish observers' attention was the variation of teaching style between low- and high-quality child care centers.

In low-quality child care centers, the educational style was either too loose or too controlled. There was a lack of many important aspects of quality, such as a written plan or stated competencies and understandings that the teachers believed the children should learn and develop during the daily events. Characteristic of high-quality units was the teachers' splendid interactions with the children, and sustained communication with the children throughout the day.

The differences seem to stem from different traditions, educational orientations, space, and material and staff conditions. In addition to differences in the physical resources, Sweden also has a rather long tradition of national goals presented as recommendations (Socialstyrelsen, 1987, p. 3). In 1998, the state mandated an early childhood national curriculum for children ages 1 to 5 (Ministry of Education and Science in Sweden, 1998a), linked to the curriculum for children between the ages 6 to 16 (Ministry of Education and Science in Sweden, 1998b). Since 1998, early childhood education has been considered to be the first step of the educational system under the jurisdiction of the Department of Education. The aim of these curricula is to lay foundations for lifelong learning. Ever if learning is emphasized, it is not a question of formal and teacher-directed learning. The children are seen as active players in their education, taking initiative and creating their own knowledge (Sommer, 1997). The nature of the goals in the Swedish curriculum may be described as qualitative, which means that the focus of a child's learning is on change and development of understanding. In Germany, there is no national curriculum and German preschool programs overemphasize less formal teaching methods, while academically oriented activities are deliberately excluded. Compared to the criteria in ECERS, a scoring on the highest level of quality requires more planning and teacher inputs/scaffolding than what is prescribed in the German preschool programs (Tietze, Cryer, Barrio, Palacios, & Wetzel, 1996).

However, despite obvious cultural differences, there appears to be a core of basic elements common among Western countries that are recognized as being necessary for children's positive development and learning (Tietze et al., 1996). When quality definitions are scrutinized, the themes of these core elements appear repeatedly with only a variation in details. Common elements that are considered important are safety, healthful care, developmentally appropriate stimulation, positive interactions with adults and peers, encouragement of individual emotional growth, and positive relations with other children (Tietze et al., 1996).

Country borders do not limit experiences, knowledge, values, norms, and theories of learning. Modern information and communication technology makes it possible for people to share and communicate different approaches, both within and between nations and cultures. Among the Western industrialized countries, based on their long-lasting exchange, a consensus regarding quality early childhood education has emerged.

Future Research Into Quality

The main task for early childhood education is to promote children's learning towards the overall goals and values of society. To find out if, and in what way, a child care center has promoted children's learning and what children have learned by being there, we have to evaluate. The methods used by a center to enhance learning must be critically analyzed and revised, or new ones developed, in accordance with new research and changes in society concerning its values, culture, traditions, and overall goals.

The original version of the ECERS has been extended with 20 new items covering literacy, mathematics, science and environment, and diversity. The extension was developed by Sylva, Siraj Blatchford, Taggart, and Colman (1998). The original ECERS has also been revised; ECERS-R was published in 1998 (Harms, Clifford, & Cryer, 1998). It is important to emphasize that the ECERS is only one way to look at quality and that it can be combined with other evaluative methods, such as the Leuven Involvement Scale for Young Children and the Adult Style Observation Schedule for Early Childhood Education (Laevers, 1994), and those that use observations to build a profile of a teacher's style of interacting in the child's learning process (Pascal & Bertram, 1994).

The above methods evaluate programs, while the Teacher Belief Scale examines the teacher's philosophy regarding developmentally appropriate practices as outlined by NAEYC (Burts, Hart, Charlesworth, Fleege, Mosley, & Thomasson, 1992). The Q methodology approach is subjective in design and lets the respondent sort items into categories of either developmentally appropriate or inappropriate. It allows the participant's unique perspectives about various phenomena to emerge and to determine if the subjective structure of attitudes is static or changing over time (Ernest, draft; Sexton, Snyder, Wadsworth, Jardine, & Ernest, 1998). The Q method also has been used to explore the subjective beliefs of teachers and family members, and to investigate if they share views concerning developmentally appropriate practices (Ernest, 2000).

The different approach to evaluate quality with the ECERS that has been used in the present comparative study clearly visualizes how different levels and aspects of quality are concretized in pedagogical processes in early childhood education in a national as well as a cross-national perspective. To learn from each other, various pedagogical processes in early childhood education must become visible in order to be reflected upon, critically analyzed, and improved. This approach to evaluation of quality with the ECERS can thereby lead to a greater understanding of what constitutes quality in early childhood education and how it is concretized in various pedagogical processes--an understanding that is necessary if we want to discuss, compare, and learn about quality issues with each other. ECERS, combined with a documentation of the rationales for scoring, can therefore be a powerful tool in the development of teachers' awareness and professionalism as well as insightful for researchers, administrators, and politicians seeking to guarantee the right of the child to quality early childhood education. Early childhood education should be an integrated whole in which society's cultural values and goals permeate what goes on in the practical work.

Notes

(1.) A German research team at the Free University of Berlin (Wolfgang Tietze, Kathe-Maria Schuster, Marita Stahnke, and Martin Schlattmann) and a Swedish research team at the University of Gothenburg (Gunni Karrby. Joanna Giota, Sonja Sheridan, and Anette Daversjo Ogefelt).

(2.) Published in 1980 by Thelma Harms and Richard Clifford.

(3.) The German version was published by Tietze, Schuster, and Rosbach in 1997 and consists of 37 items. Interrater agreement was assessed as kappa = .85 and the internal consistency of the scale, alpha .94 was found. The Swedish version was translated by Karrby in 1989 and consists of 28 items (24 are original items and 4 are combined items). An interrater agreement ranges from .83 to .96 percent of the total agreement, or a difference of one point. The internal consistency ranges from .89 to .96. (For further information about ECERS adaptations, see Tietze and Giota, 1999.)

References

Andersson, B. E. (1992). Effects of day care on cognitive and socioemotional competence of thirteen-year-old Swedish school children. Child Development, 63, 20-36.

Andersson, M. (1995). A Swedish version of the Early Childhood Environment Rating Scale (ECERS). Stockholm: Lararhogskolan.

Andersson, M. (1999). The Early Childhood Environment Rating Scale (ECERS) as a tool in evaluating and improving quality in preschools (Studies in Educational Sciences, 19). Stockholm: Institute of Education Press.

Bartley, K. (1998). Barnpolitik och barnets rattigheter. Child policy and children's rights (Monograph from the Department of Sociology, No. 67). Goteborg: Goteborgs Universitet: Department of Sociology.

Burts, D. C., Hart, C. H., Charlesworth, R., Fleege, P. O., Mosley J., & Thomasson, R. H. (1992). Observed activities and stress behaviors of children in developmentally appropriate and inappropriate kindergarten classrooms. Early Childhood Research Quarterly, 7(2), 297-318.

Clarke-Stewart, K. A. (1987). Predicting child development from child care forms and features: The Chicago Study. In D. A. Phillips (Ed.), Quality in child care: What does research tell us? (Research Monographs of the National Association for the Education of Young Children, Vol. 1, pp. 21-42). Washington, DC: National Association for the Education of Young Children.

Ernest, J. M. (Draft). Family members' and teachers' subjective beliefs about developmentally appropriate practices. Manuscript in preparation. Dept. of Learning and Instruction, SUNY at Buffalo, New York.

European Union. (1996). Council for cultural co-operation. Strasbourg: Education Committee.

Harms, T., & Clifford, R. (1980). The Early Childhood Environment Rating Scale. New York: Teachers College Press.

Harms, T., & Clifford, R. M. (1983). Assessing preschool environments with the Early Childhood Environment Rating Scale. In B. Spodek (Ed.), Studies in Educational Evaluation, special issue, Evaluation in Early Childhood Education (Vol. 8, pp. 261-269).

Harms, T., Clifford, R., & Cryer (1998). The Early Childhood Environment Rating Scale, revised edition (ECERS-R). New York: Teachers College Press.

Hultqvist, K. (1990). Forskolebarnet: En konstruktion for gemenskapen och den individuella frigorelsen. The preschool child: A construction for community and the individual liberation. Stockholm/Stehag: Symposion Bokforlag.

Karrby, G. (1989). A Swedish research version of the Early Childhood Environment Rating Scale (ECERS). Gothenburg, Sweden: Gothenburg University, Department of Education.

Karrby, G. (1992). Kvalitet i pedagogiskt arbete med barn: Nya vagar inom barnomsorgen. Quality in pedagogical work with children: New ways in early childhood education. Stockholm: Allmanna Forlaget.

Karrby, G. (1997). Behall och utveckla god kvalitet i forskoleverksamhet. In Socialstyrelsen (Ed.), Roster om den svenska barnomsorgen- En antologi. Maintain and enhance good quality in early childhood education. In Socialstyrelsen (Ed.), Voices about Swedish child care. An anthology (pp. 93-99). Stockholm: Socialstyrelsen.

Karrby, G. (1999, September). Conceptions of quality in early childhood education: An analysis of rationales for the rating of quality with the ECERS. Paper presented at the EECERA 9th European conference in Helsinki, Finland.

Karrby, G., & Giota, J. (1994). Dimensions of quality in Swedish day care centers: An analysis of the Early Childhood Environment Rating Scale. Early Child Development and Care, 104, 1-22.

Karrby, G., & Giota, J. (1995). Parental conceptions of quality in day care centers in relation to quality measured by the ECERS. Early Child Development and Care, 110, 1-18.

Karrby, G., Giota, J., Sheridan, S., & Daversjo-Ogefelt, A. (1995, October). Metod for bedomning av kvalitet i forskoleverksamhet - Forskning och utvecklingsarbete. Method to evaluate the quality in early childhood education: Research and development work. Paper presented at the Congress "Kvalitet i kommuner och landsting," Goteborg.

Laevers, F. (1994). The innovative project Experiential Education and the Definition of Quality in Education. In F. Laevers (Ed.), Defining and assessing quality in early childhood education. (Studia Paedagogica, No. 16.) Belgium: Leuven University Press.

Ministry of Education and Science in Sweden. (1998a). Curriculum for preschool. Stockholm: Fritzes.

Ministry of Education and Science in Sweden. (1998b). Curriculum for the compulsory school, the pre-school class and the after school centre. Stockholm: Fritzes.

Munton, A. G., Mooney, A., & Rowland, L. (1995). Deconstructing quality: A conceptual framework for the new paradigm in day care provision for the under eights. Early Child Development and Care, 114, 11-23.

National Association for the Education of Young Children. (1991). Accreditation criteria and procedures. Washington, DC: Author.

Osborn, A. F., & Millbank, J. E. (1987). The effects of early education: A report from the Child Health and Education Study. Oxford, UK: Clarendon Press.

Pascal, C., & Bertram, T. (1994). Effective early learning: Developing a methodology for quality evaluation and development in preschool settings. Worcester: Worcester College of Higher Education.

Pedhazur, E. J., & Pedhazur Schmelkin, L. (1991). Measurement, design and analysis. Hillsdale, NJ: Lawrence Erlbaum Associates.

Phillips, D., McCartney, K., & Scarr, S. (1987). Child care quality and children's social development Developmental Psychology, 23, 537-543.

Phillips, D. A., Scarr, S., & McCartney, K. (1987). Dimensions and effects of child care quality. The Bermuda study. In D. A. Phillips (Ed.), Quality in child care: What does research tell us? (Research Monographs of the National Association for the Education of Young Children, Vol. 1, pp. 43-56). Washington, DC: National Association for the Education of Young Children.

Pramling, I. (1994). Kunnandets grunder. Provning av en fenomenografisk ansats till att utveckla barns satt att uppfatta sin omvarld. The basis of knowledge. A phenomenografic approach to develop children's conceptions of the world. Goteborg: Acta Universitatis Gothoburgensis.

Pramling Samuelsson, I. (1997). Att lara och utvecklas i forskolan. In Socialstyrelsen (Ed.), Roster om den svenska barnomsorgen: En antologi. To learn and develop in preschool. In Socinlstyrelsen (Ed.), Voices about Swedish child care: An anthology / SOS-report 1997:23 (pp. 29-36). Stockholm: Socialstyrelsen.

Pramling Samuelsson, I. (1998, August). A curriculum for 1-5 year olds in Swedish preschool. Paper presented at the OMEPs' 22nd World Congress, Kopenhamn.

Rossbach, H. G., Clifford, R. M., & Harms, T. (1991). Dimension of learning environments. Cross-national validation of the Early Childhood Environment Rating Scale. Paper presented at the AERA Annual Conference, Chicago.

Scarr, S., Eisenberg, M., & Deater-Deckard, K. (1994). Measurement of quality in child care center. Early Childhood Research Quarterly, 9, 131-151.

Schweinhart, L., & Weikart, D. (1980). Young children grow up: The effects of the Perry Early Childhood Education Program on youth through age 15 (Monographs of the High/Scope Educational Research Foundation, No 7.). Ypsilanti, MI: The High/Scope.

Sexton, D., Snyder, P., Wadsworth, D., Jardine, A., & Ernest, J. M. (1998). Applying Q methodology to the investigations of subjective judgments of early intervention effectiveness. Topics in Early Childhood Special Education, 18(2), 95-107.

Sheridan, S. (1997). Kvalitetsbedomningar med the Early Childhood Environment Rating Scale-En jamforelse mellan en extern utvarderares och forskolepersonals bedomning med ECERS. Evaluations of quality with the Early Childhood Environment Rating Scale A comparison between external evaluations and preschool teachers self-evaluations of quality with the ECERS. Gothenburg, Sweden: Gothenburg University: Department of Education.

Socialstyrelsen. (1987:3). Pedagogiskt Program for forskolan. The Pedagogical Program for Preschool. Allmanna rad fran Socialstyrelsen. Stockholm: Allmanna forlaget.

Sommer, D. (1997). Barndomspsykologi. Utveckling i en forandrad varld. Childhood psychology. Development in a changed world. Stockholm: Runa.

Sylva, K. (1994). School influences on children's development. Child Psychology and Psychiatry, 35(1), 135-170.

Sylva, K., Siraj Blatchford, I., Taggart, B., & Colman, P. (1998). The Early Childhood Environment Rating Scale: 4 curricular subscales. London: Institute of Education.

Tietze, W., Cryer, D., Barrio, J., Palacios, J., & Wetzel, G. (1996). Comparisons of observed process quality in early child care and educational programs in five countries. Early Childhood Research Quarterly, 11, 447-475.

Tietze, W., & Giota, J. (Unpublished manuscript). Cross-national assessments of process quality by the ECERS: A study on methodological issues.

Tietze, W., Schuster, K. M., & Rossbach, H.-G. (1997). Kindergarten-Einschaetz-Skala (KES). Deutsche Fassung der Early Childhood Environment Rating Scale von Thelma Harms & Richard M. Clifford. Neuwied: Luchterhand.

Utbildningsdepartementet. (1999). OECD-country note. Early childhood education and care policy in Sweden. Stockholm: Utbildningsdepartementet. Ministry of Education and Science in Sweden.

Wolming, S. (1998). Validitet: ett traditionellt begrepp i modern tid. Validity: A traditional concept in post-modern times. Pedagogisk Forskning i Sverige, 3(2), 81-103.

Whitebook, M., Howes, C., & Phillips, D. (1989). Who cares? Child care teachers and the quality of care in America. Oakland, CA.: Child Care Employee Project.