Abstract

The International Childhood Care and Education (I.C.C.E.) was an international project with the participation of four European countries (Austria, Germany, Spain and Portugal) and four states of the United States of America. It was a transnational cooperative study about diversity and quality of educational experiences of pre-school children in different socialization settings: family and pre-school centers. The general objectives of the research were to study the diversity and quality of educational experiences of pre-school aged children in different socialization settings and to analyse the impact of these experiences on children’s development. Among the different specific objectives of the study, in this paper only four objectives will be analysed:

(1) Identification of the structural quality characteristics of different types of pre-school centers;

(2) Identification of the quality features of centers;

(3) Analysis of the relationships between family’s characteristics and center’s characteristics;

(4) Identification in both settings of those characteristics predicting development outcome variables of children.

In a brief review, we enumerate two of the most important findings: (1) It was possible to identify family and center characteristics as predictors of children’s development (2) and to recognize the family variables as the most important predictors of child development.

Introduction

With the advent of widespread compensatory education in the U.S.A. and abroad, and a general preschool movement (Tietze & Ufermann, 1989), early childhood education became an issue of heightened international interest. Thus, systematic approaches to making evaluative comparisons of early childhood programs were initiated to determine how well Early Childhood Programs were actually meeting the needs of children, their families, and the larger society.

Despite these comparisons, questions remain about how programs really work, what kind of quality is provided and under what conditions, how Early Childhood Education (ECE) process quality is effected by the structural features of programs and what effect quality has on the developmental outcomes of children. An understanding of these mechanisms is essential for policy-makers, practitioners and researchers in the field.

I am going to present some results of I.C.C.E.- International Childhood Care and Education, a transnational cooperative study about the diversity and quality of educational experiences of preschool children in different socialization settings (family and kindergarten). In this study participate four European countries (Austria, Germany, Spain and Portugal) and four states of the USA: Colorado, California, North Carolina and Connecticut.

In this presentation only Portuguese data shall be discussed and also the most relevant, in special those concerning the settings (centers and families) and their impact on child development outcomes.

The theoretical background of the study can be synthetically presented in this phrase: “In the present research we are guided by an ecological perspective which considers child development as a result of the interaction between child and the different ecosystems in which he/she is involved,” (I.C.C.E., 1995).

In beginnings of the I.C.C.E. study the influence of the work of Bronfenbrenner (1979) was very important in order to conceptualize the importance of the “milieu”, in its impact on child development. For instance, Tietze & Rossbach, 1984, wrote, ”Bronfenbrenner states in his review on day care and preschool as contexts of human development that he couldn’t find any study on the effects of preschool on the child behavior at home”.

However, research conducted in Portugal, Spain, Germany and in the United States, in the scope of the I.C.C.E., revealed encouraging results, concerning the impact of quality characteristics of early childhood environments and developmental outcomes in children, (Lera, 1996, Tietze, Cryer et al., 1996, and Tietze, Bairrão et al., 1998).

Recently, Melhuish (2001) assumes that the theme of research on preschool education at the end of the second millennium has been the importance of quality of care. In this paper Melhuish emphasizes also the importance in development of preschool experience and why quality of care measurement is a major issue, not only for psychometric reasons, but also because the nature of measurement objectifies theoretical assumptions about what quality is.

More encouraging results appeared in the Cost, Quality & Child Outcomes, (1995), where child outcomes and their relation to center quality is demonstrated. The authors show children’s cognitive and social development are positively related to the quality of their child care experience. Compared to children in lower quality settings, children in higher quality classrooms displayed more advanced language development and pre-math skills, had more positive attitudes toward their child care experience, and have warmer relationships with their teachers. Also the study reveled that the quality of child care are positive associated whit developmental outcomes for children across all levels of mother’s education.

Rationale

As we mention before, the framework of the I.C.C.E. study is an ecological perspective following Bronfenbrenner’s ecological model of development, as it had been firstly presented in 1979, (Bronfenbrenner & Morris, 1998).

However, in our understanding, the more recent perspective of an ecological model of development presented by Bronfenbrenner & Morris, (1998), is more adequated to this type of research. In this perspective, human development is viewed as the product of four components and the relationship among them: process, person, context and time process, the primary mechanism for development, includes all interactions between the developing person and the environment. The influence of this process of development varies as a function of the characteristics of the others three components of the model, namely person (biological endowment and dispositions), environmental contexts (both immediate and more distant and including the relationships in those contexts), and time periods in which the processes occur.

However, the rationale of this kind of research is incompletely defined without refer the other major components of the ecological perspective:

• The site background;
• The family background
• And the characterization of the socialization settings: structure and process.

All these features or group variables, if the study is empirically conducted, are organized according to the proximity of the child and they are responsible for the developmental status of child. Tietze and Rossbach, (1984), Tietze, (1986), and Bairrão, (1995), defined these components and also present some graphic representation of the model. Let us now define some of these components.

A setting can be defined: as a relatively stable social unit (personal dimension; e.g. number of adults and children, their age an sex), which is normally associated with a specific locality (space-material dimension; e.g. size of the room, available outdoor space), in which relatively stable action patterns can be observed (action dimension; e.g. typical plays, participation in action) and which is anchored in a organizational, legal, economic and functional context (organizational anchoring; e.g. financial support, legal frame).

The definition allows for an empirically based description which goes beyond the labels, thus providing the basis for an intercountry and cross country comparison of care environments for young children (Tietze, 1986).

Figure 1 presents a conceptual framework with five major variable groups based on an ecological orientation. Following Tietze, 1986.

Fig. 1 – Major variable groups, organized according to their proximity to the child.

At the most distal level are site background variables representing characteristics of the larger community in which the family resides. Potentially influential characteristics include level of industrialization, degree of urbanization, average educational level, and so forth.

• At the next level in the model are family background variables thought to influence the child’s socialization setting. These variables include family structure, parental education level, socio-economic status, and so forth.

• At the next level appear setting structure variables, defining characteristics of the socialization environment thought to influence the process occurring in that environment. These variables include measures of available space per child, the ratio of adults to children, the age composition of the group of children, the socialization expectations of the parent or teachers, and so forth.

• At one level removed from the child are variables defining setting-process ˿ the socialization process that children experience in the various settings of their daily lives. These variables measure aspects of children’s interactions with people and thing in their environment.

• At the level of the child are variables defining the child’s developmental status ˿ the product of a socialization process interacting with concurrent characteristics of the child. The variables at this level include indicators of social competence, cognitive-linguistic development (growth, health, and nutritional) status, and pre-academic skills.

Other authors, besides those before mentioned, also refer structure and process and outcomes as important features in the study of child care characteristics.

The Present Study

Tietze, Cryer, Bairrão, Palácios and Wetzel, (1996) after a long review of literature, assume that “within individual countries, observed Early Childhood Education process quality has been examined in relationship to both structural variables and to child development outcomes. However, most cross-national comparisons of early care and education have not examined observed process quality. Rather, they have used a case study approach than typically consolidate compared information about the structures underlying national Early Childhood Education systems (ECE). Despite the increasing amount of information that has become available through those efforts, little is at present known about how the levels of ECE process quality compare between different countries...” (page 450).

So, our study was based on the past explorations involving process quality and to move a step toward in our understanding of process quality as a construct, a large cross-national study, the International Child Care and Education Project (ICCE) was initiated on 1993 in five countries: Austria, Germany, Portugal, Spain and the U.S.A. Some results of this study will be presented.

With regard to ECE, the participating countries have traditions, cultures and environments that are similar enough to apply the same instruments when assessing ECE process quality, but diverge enough to expect differences with regard to the actual process quality that is produced in their ECE programs.

The ICCE study was a consolidation of two comprehensive projects, the European Child Care an Education Study (ECCE), which is being completed in Austria, Germany, Portugal and Spain, and the Cost, Quality, and Child Outcomes in Child Care Centers Study (CQ&O), which is being carried out in the United States. Each of the two projects was designed with its own scope and purpose.

From this study some results are presented.

Method

Structure and Process as defining quality: Conceptual Rationale and Instruments.

I) - Centers

Structural Quality: Instruments used to assess structural conditions of Pre School Centers (PSC) included:

QSC (Questionnaire of Structural Characteristics):
        a) - QSC – T (Teachers) and
        b) - QSD (Directors)

Process Quality: The Early Childhood Rating Scale (ECERS)

The ECERS (Harms & Clifford, 1980) was developed in the United States to assess process quality of Early Childhood Education environments from a broad perspective. It has been widely used in child development research in the United States. (Phillips, 1987; Howes, Phillips et Whitebook, 1992; Culkin, Hellburn, Morris et Watson, 1990). However, the selection of the ECERS was due not only to the instrument’s wide use inthe United States, but also because of its consistency with the study’s conceptual framework and the fact that it had been used frequently in the participating European countries, as well as, in others where national versions of the scale were already developed or being tried-out prior to and independent of this study. (Kärby & Giota, 1994; Lima, Leal, & Bairrão, 1989; Palacios, Lera & Moreno, 1994; Rossbach, 1990; Tietze , Shuster & Rossbach, 1997).

Description of the scale

The ECERS is designed to give an overall picture of center-based surroundings created for young children. It is related to the quality of a specific classroom. The scale regards the term “environment” in a broad sense and includes such aspects as organization of space, equipment and activities to stimulate development, adult supervision and interaction, and time schedulle (Harms & Clifford, 1986). The scale consists of 37 items organized in seven apriori-subscales, according to similarities of contents and theoritical considerations:

I. Personnal Care Routines of Children (5 items)
II. Furnishings and Display for Children (5 items)
III. Language-Reasoning Experiences (4 items)
IV. Fine and Gross-Motor Activities (6 items)
V. Creative Activities (7 items)
VI. Social Development (6 items)
VII. Adult Needs (4 items)
VIII.

Each of the 37 items of the ECERS is scored on a 7 point scale from 1 (inadequate) to 7 (excellent).

II) - Families

Assessment of structural conditions and educational orientations was performed by the Questionnaire for the Family (QF). The QF was developed with general aim of obtaining a characterization of the family as one of the main settings where the child is socialized.

Family Process Quality

Process Quality in the family setting was assessed by the HOME. According to its authors, the Home Observation Measurement of Environment Scale (Caldwell & Bradley, 1984) assesses the quality and quantity of support and stimulation the child receives at home. The focus of the assessment is the child in his environment, the child as the recipient of stimulation received from objects and events, and the relationships in the family context.

The preschool version of HOME that was designed for 3-6 year olds was used in the ECCE study. It is made up of 55 items which the authors group in eight subscales: Learning materials, Language stimulation, Physical environment, Parent’s responses to the child: pride, affection, tenderness; Academic stimulation, Modeling and stimulation of social maturity, Variety of experiences, and Acceptance of the child.

The scale is used during a visit to the child’s home that can last between 30 to 60 minutes. During this time, the child must be awake, and at least one of the people who normally look after her must be present. All the items in the different versions of the HOME scale are scored in the same way (as “present” or “absent”). The addition of items scored as “present” of each subscale gives the total score. Thus the highest score is the same as the number of items in the HOME scale (55 items for the preschool version).

The psychometric properties of the scale are well established and show satisfactory values. Interobserver agreement is usually high, as well as internal consistency (higher than .80 in the scales total scores). The studies in which factorial analyses of the scale have been done show that the empirically derived factor structures do not always coincide with HOME subscales. Thus most people use the scale total score in analyses and depend less on the subscales scores.

III) - Child Outcomes Measures

For assessing the Adaptive Behavior, the Vineland Adaptive Behavior Scale (VABS) (Sparrow, Balla & Cicchetti, 1984) was used.

To assess cognitive abilities of children, the Peabody Picture Vocabulary Test (PPVT – R) was used. The PPVT – R is a widely used instrument to measure receptive vocabulary that correlates highly with other measures of intelligence and has adequate psychometric properties (McCallum, 1985). Administration and scoring procedures are standardized, and the subject’s score is compared to normative data. The material uses pictures plates to represent words increasingly more difficult. The test takes between 10 and 20 minutes.

Sample

Although Portugal has in 1994 the lowest ECE enrollment rates for 3 to 5 years old (35,5%), the coverage rate have increased considerably mainly after 1998. The enrollment rate is presently (2001) of 73,9%.

The responsibility for services for this age group is divided between two ministries - the Ministry of Education (ME) and the Ministry of Solidarity and Social Security (MSSS). Programs vary, with private for-profit as well as nonprofit program available, as well as full-or-part-day services. Some programs are free of charge (ME), and others require tuition.

In 1994, no consistently implemented curriculum was used in preschool programs. However the ME provide general orientations around which preschool teachers organize their work. Most programs emphasize the socio emotional development of the child and the importance of preparing children for elementary school.

Sampling procedures included three steps: (1) the selection of larger regions and sites within the regions, (2) the selection of Pre-School Centers (PSC), and (3) the selection of classrooms within the PSC.

The sample at this level, step 1, included: Greater Lisbon and the NUT (administrative unit) of the west zone, as well as Greater Porto and the NUT of Tamega. In step 2: 88 centers were chosen, including public, private non-profit, private for profit centers. In step 3, inside each sampled center or classroom with preschool age, children were randomly selected.

So the Portuguese sample is composed by 88 centers, including public, private non-profit, private for profit centers, where preschool age children were randomly selected. In our case 345 children with their families were chosen.

Portuguese data collection was finished in the spring of 1994.

Results (Multiple Regressions)

I) - The Quality of Preschool Centers (PSC) in Portugal assessed by the ECERS

Results of the ECERS in Portugal, as a measure of the Quality of Preschool Centers (PSC), show a quite homogeneous level of quality.

As above-mentioned, to measure the classroom quality (both structure and process aspects) trained observers used the ECERS. This scale permitted an evaluation of the quality of total childcare environments (structural aspects) as well as the more specific aspects of the relationship between teacher and child (process aspects). The overall index of center quality was scaled from 1 to 7, as follows, to conform to the scaling in the ECERS measures:

Inadequate: Health and safety needs of children are not met; no warmth or support from adults is observed, no learning is encouraged.

Minimal: Children’s basic health and safety needs partially met; adults provide a little warmth and support; there are few learning experiences.

Good: Health and safety needs are fully met; staff is caring and supportive of children; children are learning in many ways through interesting and funny activities.

Excellent: Everything is good; in addition, children are encouraged to became independent; the teacher plans for children’s individual learning needs; adults have close personal relationships with each child.

(Cost, Quality, and Child Outcomes in Child Care Centers, Public Report, second edition. Denver: Economics Department, University of Colorado at Denver).

On the basis of the results obtained in the Portuguese sample, we define the following intervals as indicators of different quality levels:

• Scores = > 3.0 and 3.6: Existing minimal conditions (Low), 1<3.
• Scores = > 3.6 and 4.6: Existence of sufficient conditions; (Medium), 3<5.
• Scores = > 4.6: Existence of developmentally adequate conditions; (High) 5 –7.

So, the results indicate a quite homogeneous level of quality. Most part of the Portuguese Preschool Centers (PSC) falls in the range of merely sufficient quality, between 3.6 and 4.6 (72%). In other words, sufficient conditions are present, but developmentally adequate conditions are rare (2%). However in our data there are no PSC scoring under 3. The worst scores are located between 3 and 3.59 and constitute 8% of the sample.

II) – Structural Aspects of the Preschool Centers (PSC) and its Impact on Quality

A regression analysis using ECERS global scores and structural aspects of Kindergarten assessed by Questionnaires (QSC – T and QSC – D) had been performed in order to a better understanding of the influence of structural variables in overall levels of quality as measured by the ECERS. In this analysis the number of PSC was of 66. The reason of this number deals with the fact that only in metropolitan areas we have all the three types of preschool provision: public, private non-profit and private for profit. So, the most influential variable explained 21% of the variance was: Centers’ area. This emphasizes the importance of the area in the ECERS scores: classrooms with larger areas have better quality.

In the regression analysis above referred, using ECERS global scores as criteria, the second most important variable was the Number of years in the center as a preschool teacher. This variable explained additionally 7% of the variance. We can conclude that classrooms with preschool teacher worked for a longer period tended to have better scores. The result stresses also the negative effect of high turnover and the need to stabilize the employment of preschool teachers.

The variables more linked to a “Good Quality” of preschool centers were The characteristics of the region and the hours spent by preschool teachers in activities planning, respectively 7% and 4%.

III) – Relation Between Family and Centers Quality

Structural process quality of the families, as measured by the Family Questionnaire and by the HOME (Home Observation Measurement of Environment, HES) was analyzed on the basis of four different groups of kindergartens attended by the children (public metropolitan, public non metropolitan, private non profit and private for profit centers).

The families of children attending private for profit centers scored significantly higher on the HES than families of children who attended public non-metropolitan centers that obtained the lowest scores. These results indicate that children belonging to families with different characteristics attend the four groups of centers. It should be remembered that many children attending public non-metropolitan centers are also attending the lowest quality centers, (particularly in the interior of the country).

To analyze the possible relationship between educational quality of the family and educational quality in the centers, all children were divided into four groups (four quartiles) constituted on the basis of their HES results. Afterwards, it was undertaken the analyze of the ECERS scores in each group. No significant differences were found between the four quartiles, although there was a trend to find better ECERS scores in the third and fourth quartiles (those quartiles were constituted by children belonging to families with better educational quality).

IV) – Characteristics of Centers and Families and its Impact on Children Development

To explore the relationship among the above mentioned characteristics of centres and families, and children’s development, several analyses were undertaken using as child development measures the score obtained by the children in the P.P.V.T.) and in the V.A. B.S.

First, it was possible to identify a set of variables that characterize the family setting, or quality, as the stronger predictor of child development.

Second, there was found a significant association between several centre characteristics and child development, although with a smaller impact than the family variables. This comparison gave expected results that confirm the family as the primary socialization setting, and the most important, with a powerful impact on children’s development.

Analyses were then undertaken to detail some of these relationships:

1. To distinguish language development from adaptive behaviour;
2. To distinguish children with parents with low level of education from other children;

National analyses have shown that educational quality of the families (as measured by the H.E.S.) is the better predictor of child development. This relationship is stronger with P.P.V.T., a measure of receptive language, than with the V.A.B.S., a measure of adaptive behaviour (social and personal independence of children). The strong relationship between the H.E.S. and the P.P.V.T. does not disappear even if we consider the group of children with less educated parents (children with both parents with four or fewer years of schooling).

In such a homogeneous sample, one might expect restricted ranges in the H.E.S. scores. But in fact, H.E.S., scores showed a substantial variation, suggesting that children with parents with a lower educational level, could belong to families with different educational quality levels.

Quite surprisingly, educational quality of Pre-School Centres, as measured by the ECERS, did not have a significative impact on child development.

However, it is interesting to note the effect of the number of years of pre-school centres attendance on children’s adaptive behaviour. 4-year-old children in their second year of attendance of P.S.C, show best results in adaptive behaviour than 4-year-olds in their first year of attendance.

This relationship is visible in the global sample but is stronger in the groups of children with parents with lower educational status.

This lack of strong evidence of the impact of Pre-school setting in children’s outcomes could be linked with the restricted range of ECERS scores in the Portuguese sample and the consequent lack of variance. The “Medium Quality” of most Portuguese pre-school centres might be not so strong enough to have a more evident impact on children’s development.

Conclusions

(1) In the General Sample The Quality of Family, as The Home Environment Scale assesses it, is the best predictor of child developmental outcomes. The relationship between the H.E.S. and developmental outcomes is stronger in measures of language development than in measures of social adaptation.

(2) In what concerns Family Quality (H.E.S.) and language measures (PPVT), this relationship is maintained for children’s group from families of low income.

(3) The P.S.Cs. quality has low impact on children’s outcomes. This is due probably to the fact that Portuguese P.S.Cs. had overall conditions (values between 3.6 and 4.6 in a 7 points scale).

It means that it is necessary to improve “Good Quality” or “Hight Quality”, in order to have a more important impact in children’s developmental outcomes.

(4) However, it is interesting to verify that the number of years that children of pre-school attend P.S.Cs., has a significative impact in adaptive behavior of children. In fact, this is verified for the global sample

(5) Given the “compensatory” role of pre-school education (P.S.C.) in children’s development, a Higher Level of Quality must be developed, mainly for those children belonging to families with low income or in disadvantage. If we think also about children with special education needs, this problem is still more crucial.

(6) As Bronfenbrenner and Zigler stressed, pre-school education is not enough. For a real difference of in development, mainly for atypical groups, family and pre-school must develop in a relatively linked rhythm.

BIBLIOGRAPHY

Bronfenbrenner, V. (1974), “Is Early Education Effective?” A Report on Longitudinal Evaluation of Preschool Programs. Vol. 2. Washington, DC: Department of Health Education and Welfare.

Bronfenbrenner, V. (1979), “The Ecology of Human Development”. Cambridge, M.A. Harvard University Press.

Bronfenbrenner, U.; Morris, P.A. (1998), “The Ecology of Developmental Process”. In W. Damon and R.M. Lerner (Eds), Handbook of child psychology. Vol. I. Theoretical models of human development. New York, John Wiley & Sons, Inc., 993-1028.

Caldwell, B.M.; Bradley, R.M. (1984), Home Observation and Measurement of the Environment. Little Rock, AK: University of Arkansas of Little Rock, College of Education, Center for Child Development and Education.

Cost, Quality & Child Outcomes Study Team (1995), Cost, Quality, and Child Outcomes in Child Care Centers, Public Report, second edition. Denver: Economics Department, University of Colorado and Denver.

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

Lera, M.J. (1996), “Education under five in Spain: A study of preschool classes in Seville”. European Journal of Psychology of Education, 2, 139-150.

Lima, I.M.P.; Leal, T.; Bairrão, J. (1989), “A escala de avaliação do ambiente em educação infantil. Contributos para o estudo da qualidade de vida das crianças dos 3 aos 6 anos em contextos formais de educação”. In J. Cruz, R. A. Gonçalves e P. Machado (Eds), Psicologia e Educação. Investigação e intervenção. Porto: Associação dos Psicólogos Portugueses.

Melhuish, E.C. (2001), “The quest for quality in early day care preschool experience continues”. International Journal of Behavioral Development. I, 1-6.

Peisner-Feinberg, E.S.;; Burchinal, M. R.; Clifford, R. M.; Culkin, M. L. et al. (2000), The children of the cost, quality and outcomes study go to school: Technical report. Chapel Hill: University of North Carolina at Chapel Hill, Frank Porter Graham Child Development Center.

Rossbach, H.G. (1990), “Assessing the Quality of Kindergarten Environments with the Early Childhood Environment Rating Scale”.In: Pieters, J.M.; Breuer, K.; Simons, P.R.J. (ed.), Learning Environments. Contributions from Dutch and German Research. Berlin: Springer Verlag, 1999, 77-90.

Sparrow, S.S.; Balla, D.A.; Cicchetti, D.V. (1984), Vineland Adaptive Behavior Scales. Survey Form Manual. Circle Pines, MN: American Guidance Service.

Tietze, W. (1986), IEA: Preprimary Project. The Preprimary Project – An Overview (policopiado).

Tietze, W.; Bairrão, J.; Leal T.; Rossbach, H.G. (1998), “Assessing quality characteristics of center-based early childhood environments in Germany and Portugal: A cross-national study” European Journal of Psychology of Education, 2, 283-298.

Tietze, W.; Cryer, D.; Bairrão, J.; Palacios, J.; Wetzel, G. (1996), “Comparisons of Observed Process Quality in Early Child Care and Education Programs in Five Countries”. Early Childhood Research Quarterly, II, 4, 447-475.

Tietze, W.; Rossbach, H.G.; Ufermann, K. (1989), “Child care and early education in Federal Republic of Germany”. In P.P. Olmstedt & D.P. Weikart (Eds), How nations serve young children: Profiles of child care and education in 14 countries (pp. 39-83). Ypsilanti, M.I.: The High/Scope Press.

Zigler, E. (1997), “The Promise of Early Intervention to Enhance the Life Outcomes of Children in Poverty”. In J.L. Paul; M. Churton; H. Rouselli-Kostoryz et al. (Eds.), Foundations of Special Education. London: P. H. Brookes.