Re-Examining the Cognitive Scale of the Developmental Support in Childcare

Abstract

The Cognitive Scale of the Development Support in Childcare (CSDSC) developed in 2023 has adequate reliability and validity. However, it has some limitations in terms of its factor structure. Thus, we re-examined the structural factors, reliability, and validity of the scale using the same data (513 valid responses) as in the original study. In contrast to the two-factor model of the original scale, the revised scale has a one-factor structure. The scale’s Cronbach’s α value was 0.83. A confirmatory factor analysis used to assess whether a one- or two-factor structure was more appropriate for the scale demonstrated that the one-factor model was a better fit, and the revised scale had a higher degree of validity than the original one. The results indicate that the revised and shortened CSDSC has sufficiently high levels of reliability and validity, suggesting that the scale is appropriate for evaluating nursery teachers’ values regarding child development support in childcare.

1. Introduction

Inclusive early childhood education and care are global concerns. In recent decades, the awareness of the need for education systems to become disability-inclusive has increased worldwide [1]. Even in Japan, childcare and educational facilities have been practicing inclusive childcare/education. That is, infants and children with disabilities (or those who are suspected of having disabilities, including developmental disabilities) are cared for at the same facilities as those without disabilities. However, the number of children with suspected developmental disabilities is gradually increasing [2]. As the number of children requiring extra support for developmental problems increases, specialized knowledge and skills are required for childcare [3]. Having more than one child with special needs in a class has become a common scenario [4].

There are three main types of care facilities that are predominantly for preschool children in Japan: hoikuen (nursery schools), yochien (kindergartens), and nintei kodomo-en (certified childcare centers). These facilities are governed by various ministries and guidelines.

These guidelines were revised simultaneously in 2017, and the latest revisions emphasize the importance of child development support in childcare on the basis of each child’s characteristics and developmental process. For instance, in the guidelines for nursery schools, the Ministry of Health, Labor, and Welfare [5] claimed that nursery teachers should understand child development and provide care on the basis of each child’s developmental process. Similarly, the guidelines for kindergarten education state the following: ensure that instruction in line with developmental issues is carried out on the basis of the individual characteristics of each child by considering that early childhood development is achieved through diverse processes and interactions between various aspects of physical and mental elements and that the life experiences of each child are diverse [6].

Japanese nursery teachers must practice inclusive childcare and provide developmentally appropriate support for each child in terms of the importance of inclusive education and childcare and in accordance with the guidelines [7]. Japanese teachers’ beliefs and practices regarding developmentally appropriate practice (DAP) can be considered developmentally appropriate and synchronized with the core principles of a play-based curriculum, as it is known in the United States [8]. In the United States, DAP is an educational approach; it refers to teaching strategies that consider children’s ages, interests, abilities, and experiences to help them achieve challenging and achievable goals [9]. Additionally, as child-initiated activities (e.g., free play) are the focus of a DAP-based curriculum [10], Japanese teachers believe that play is all-encompassing and enhances all domains of development—cognitive, social, emotional, and physical [8]. However, respecting the diversity of children’s development is one of the contemporary issues for Japanese early education and care [11], and addressing the needs of each individual child is imperative [12].

Individual nursery teachers’ attitudes and values regarding childcare significantly impact childcare practices and methods of providing support [13]. For example, Allport defined an attitude as a mental and neural state of readiness, organized through experience, exerting a directive and dynamic influence upon the individual’s response to all objects and situations with which it is related [14]. As the definition emphasizes, attitudes are considered closely related to individuals’ behaviors. In the field of special needs, the attitudes of teachers toward children with special needs are an important part of effective teaching, and teachers with positive attitudes towards inclusion are more likely to act positively toward students with disabilities than other teachers with negative attitudes [15]. Thus, nursery teachers’ attitudes and values regarding child development support in childcare can influence child development support practices in many ways; however, little is known about this aspect.

Accordingly, the authors conducted semi-structured interviews with 11 female nursery teachers (four nursery school teachers, three kindergarten teachers, and four certified childcare center teachers) to clarify their qualitative understanding of child development support in early education and care [7]. The interview data were analyzed using the modified grounded theory approach (M-GTA) [16,17,18]. The analysis generated nine categories and 30 concepts. In terms of the object and method of support, the teachers’ understanding of child development support was divided into two types: direct and indirect. The former is directly provided by nursery teachers in the growth and development of children. The latter is indirectly provided to children through outreach and support from their environment, such as from parents, colleagues, and the local community.

To develop a prototype version of the scale to quantitatively measure nursery teachers’ understanding of child development support in childcare, a pilot study based on 30 concepts was administered to 42 nursery teachers as follows [19]. First, a set of draft questions was created. Then, the content validity of the questionnaire was examined by two former nursery school directors and one psychologist in the field of parenting support, and the questionnaire was completed in a pilot study by 42 nursery teachers at nursery schools, kindergartens, and certified childcare centers. The reliability of the scale was verified using the responses to the questionnaire (five-point Likert scale). After conducting an item analysis and item–total correlation analysis, Cronbach’s α coefficient was calculated for the 21 items (the final number of questions), and the result was 0.86, indicating sufficient internal consistency. Thus, an appropriate prototype version of the scale was developed for the pilot study.

A total of 513 valid responses obtained at 40 facilities for early education and care were statistically analyzed to examine the structure of the factors [20]. As a result, a cognitive scale called the Cognitive Scale of the Development Support in Childcare (CSDSC) was developed to assess values and attitudes. The scale has a 17-item, two-factor structure comprising the “recognition of direct and indirect supports in childcare” and “recognition of institutional childcare environment”. The Cronbach’s α value for the scale was 0.84, and the values for the subscales were 0.81 and 0.86. However, although the CSDSC has adequate reliability and validity, some limitations should be considered. For instance, a confirmatory factor analysis revealed that the model fit of the two-factor, 17-item structure can be considered within an acceptable range, but the degree of the data fit is poor, resulting in issues of factor validity. Additionally, the second factor of the CSDSC comprises two items, even though an exploratory factor analysis requires at least three items to correspond to each factor. The factor is unstable as a sub-scale given that it comprises two items. The scale also has a low total variance; the two factors with eigenvalues greater than 1 explained approximately 29.50% of the total variance.

These limitations should be overcome so that the scale can be used extensively in research and as a practical tool for measuring the value of child development support in childcare. Thus, the present study re-examined the structural factors, reliability, and validity of the scale using the same data (513 valid responses) as in the previous analysis and explained the changes made from the previous version of the CSDSC.

2. Materials and Methods

2.1. Participants

This cross-sectional study was conducted among Japanese childcare teachers employed at public or private nursery schools, kindergartens, and certified childcare centers in the northern part of a rural Japanese prefecture from September to December 2021.

Self-administered anonymous questionnaires were distributed to childcare teachers only when the facility managers or city governments (in the cases of public nursery schools) agreed to participate in the survey. In total, 589 teachers from 40 facilities consented to participate in the survey. From the responses, 513 (valid response rate: 87.1%) were included in the analysis after excluding those who were not certified childcare workers and/or certified kindergarten teachers and those who had missing values for at least one of the questionnaire items. The confidentiality of the study was indicated in the documents provided to the participants.

2.2. Measurements

2.2.1. Demographics Questionnaire

The questionnaire was developed by the authors. The participants were asked about their sex, age, years of experience, type of childcare facility, the age of the children in the class they were in charge of, and type of employment.

2.2.2. Cognitive Scale of the Development Support in Childcare

A prototype version of the CSDSC was developed and selected on the basis of semi-structured interviews [7] and a pilot study [19]. The prototype version of the CSDSC was a 21-item self-report measure used to assess nursery teachers’ understanding of child development support. Participants were asked to indicate their level of agreement or disagreement on a five-point Likert scale (highly disagree, somewhat disagree, neither agree nor disagree, somewhat agree, or highly agree), and their responses were scored as 1, 2, 3, 4, or 5.

2.3. Analyses

Statistical Methods for Evaluating Reliability and Validity

Reliability was investigated using a corrected item–total correlation and Cronbach’s α coefficient, both of which are indicators of internal consistency.

In terms of validity, scaling, factorial, and construct validity were assessed. To assess scaling validity, ceiling and floor effects were examined for the prototype version of the CDSDC. The judgment condition was whether the values of the mean added/subtracted standard deviations were within the theoretically possible ranges. In this study, the Kaiser–Meyer–Olkin measure of sampling adequacy and Bartlett’s test of sphericity were used to assess the appropriateness of the factor analysis. Factorial validity was investigated via an exploratory factor analysis, using the maximum likelihood method with promax rotation. Promax rotation was used, which allows for the factor to be correlated, based on previous studies that showed inter-factor correlations in the factor structure of the scale. The number of factors was determined to evaluate the scree plot (scree test), eigenvalues (Kaiser criterion), and interpretable factor loadings. Items with factor loadings of at least 0.40 were retained. Regarding item selection, factor loadings of 0.40 and higher were considered meaningful; hence, items with loadings below 0.40 were excluded from further analyses [21].

To evaluate the construct validity, the factor structure of the revised CSDSC was subjected to a confirmatory factor analysis to assess the model fit. For fit indices, the chi-square test of model fit, goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), comparative fit index (CFI), and root mean square error of approximation (RMSEA) were calculated. The GFI and AGFI ranged from 0 to 1, with values above 0.90, indicating a good fit with the data [22,23]. CFI values of 0.90 and above and RMSEA values of 0.08 or less are considered appropriate ranges [24].

The content validity of the child development measure was evaluated by seven experienced practitioners and researchers.

IBM SPSS Statistics and Amos version 26 (IBM Corp., Armonk, NY, USA) were used for the statistical analyses.

2.4. Ethical Considerations

This study was conducted in accordance with the ethics committee of the Graduate School of Humanities and Social Sciences, Kumamoto University. To obtain informed consent, a letter describing the aims and procedures of the study was enclosed to assure participants that the survey was anonymous and voluntary. To protect individual privacy, we did not obtain any identifiers from the childcare facilities to which each participant belonged. In addition, they were assigned sequential numbers and converted into numerical values so that neither the facility nor the individual could be identified, thus ensuring respondents’ anonymity and confidentiality.

3. Results

3.1. Characteristics of Study Participants

Table 1. Characteristics of the participants.

	n = 513
	n	%
Sex
Female	498	97.1
Male	15	2.9
Age (year)
20 to 29	131	25.5
30 to 39	126	24.6
40 to 49	106	20.7
50 or above	150	29.2
Years of experience
Less than 5	103	20.1
5 to 15	274	53.4
16 or more	136	26.5
Type of childcare facility
Nursery school	358	69.8
Kindergarten	51	9.9
Certified childcare center	104	20.3
Age of the class in the participant’s charge
0 to 2 years old	267	52.1
3 to 5 years old	174	33.9
Others	72	14.0
Type of employment
Full-time employment	278	54.2
Part-time employment	235	45.8

presents the participants’ demographics. Of the 589 participants who provided data, 76 were excluded because they failed to meet the inclusion criterion of not being certified childcare workers and/or certified kindergarten teachers and/or had missing data in one or more of the questionnaire items. The final sample of 513 participants ranged in age from 20 to 50 years or above; 25%, 25%, 21%, and 29% were in their 20s, 30s, 40s, and 50s or above, respectively. Of these, 97% were female teachers, 70% were nursery school teachers, 20% were certified childcare center teachers, and 10% were kindergarten teachers. The participants included full-time (54%) and part-time workers (46%), and approximately half (53%) had between 5 and 15 years of experience in early education and care.

3.2. Factor Structure

Prior to conducting the factor analyses, the score distributions, floor and ceiling effects, and corrected item–total correlations were examined. No floor effects were observed, but a ceiling effect was detected in item 2 (“I wonder how to provide care to address each child’s challenges”). However, we proceeded with the analysis without excluding this item as it was not excluded in the previous study, and it was considered indispensable to nursery teachers’ understanding of child development support in childcare. Next, corrected item–total correlations were calculated to identify and remove less reliable items from the candidate items. The corrected item–total correlations for all items ranged from 0.01 to 0.59, with five items not meeting the cut-off level for item–total correlation (>0.3) [25]. These items were excluded from further analyses.

The Kaiser–Meyer–Olkin index for adequate sampling was 0.85, which exceeded the recommended value of 0.60. This indicates that the data represent a homogeneous collection of variables suitable for factor analysis. Bartlett’s test of sphericity was significant for the sample (χ² = 2211, 188, df = 120, p < 0.000), which supports the factorability of the dataset and demonstrates the existence of a non-zero correlation among the items. Overall, the dataset satisfied the basic requirements of the factor analysis.

With regard to factorial validity, a factor analysis was performed using the maximum likelihood method with a promax rotation. The first eight eigenvalues for the factors were 4.85, 1.37, 1.22, 1.12, 1.06, 0.89, 0.81, and 0.76. The scree test criterion suggested retaining one or two factors; thus, analyses were conducted to examine the interpretability and utility of the one- and two-factor solutions. The one-factor structure was considered more meaningful given that the break between the steep slope and leveling off was between factor numbers one and two. In addition, as one of the limitations of the original CSDSC was that the two-factor solution contained a factor on which only two variables had primary loadings, only one factor was retained in the final analysis.

In terms of the item selection for the one-factor solution, factor loadings of 0.40 and higher were considered meaningful. Items with loadings below 0.40 on all factors were excluded from further analyses. After excluding 10 items with loadings below 0.40, the factor analysis was recomputed for the remaining 11 items to ensure that all items had factor loadings of 0.40 or higher. The final factor loadings of the items on the promax-rotated factors are presented in

Table 2. Factor loadings of the Cognitive Scale of the Development Support in Childcare and Cronbach’s alpha values.

Items	I
Factor I: A sense of difficulty in providing developmental support in childcare, α = 0.83
Due to staff shortage, I am unable to interact with each child on the basis of the individual development	0.74
A lack of manpower currently exists to provide support to each child in group activities	0.71
Due to the significant individual differences of children, constructing childcare plans and goals is difficult	0.60
Given that many children necessitate individualized attention, I wonder how to operate childcare	0.60
I struggle with fostering each child’s independence in group activities	0.58
I wonder how far I should let each child go on their own without my help	0.55
There is a disconnect in how parents perceive their child’s development	0.48
Only few specialized institutions can collaborate with the local community	0.46
I wonder how to provide care to address each child’s challenges	0.45
I experience negative emotions, such as irritability, when providing childcare	0.43
Disparities exist among the staff regarding how to assess and respond to children	0.42

Note: Factor loadings > 0.40 are presented. The items were translated into English for the manuscript.

. Standardized item loadings range from 0.42 to 0.74. Upon extraction, one factor accounted for 30.99% of the total variance in the measured variables (Table 2).

In Factor I, the questions were related to the nursery teachers’ sense of difficulty. Additionally, higher scores on the factor indicate a greater sense of difficulty in providing developmental support to each child in childcare; thus, the factor was named “a sense of difficulty in providing developmental support in childcare”.

3.3. Internal Consistency Reliability

Cronbach’s α was calculated to determine the internal consistency of the revised and shortened CSDSC items. The results confirm that the revised and shortened CSDSC has a high degree of internal consistency (α = 0.83).

3.4. Content Validity

Seven experienced practitioners and/or researchers in child development examined whether the concept name of the factor adequately expressed its meaning in common with the questions comprising the factor. They agreed that all 11 items comprising the factor represented a perceived difficulty in providing developmental support; thus, the factor was named “a sense of difficulty in providing developmental support in childcare”.

3.5. Construct Validity

To assess the construct validity, the revised and shortened CSDSC was subjected to a confirmatory factor analysis to examine differences in the goodness-of-fit between the one-factor model and the original scale (the two-factor model contained 17 items). This CSDSC was developed as an 11-item one-factor model.

Table 3. Data of confirmatory factor analysis models.

	χ2	df	p	GFI	AGFI	CFI	RMSEA
The original CSDSC
Two-factor model	526.28	118	<0.001	0.89	0.85	0.81	0.08
The revised CSDSC
One-factor model	161.10	43	<0.001	0.94	0.91	0.92	0.07

CSDSC: Cognitive Scale of the Development Support in Childcare; GFI: goodness-of-fit index; AGFI: adjusted goodness-of-fit index; CFI: comparative fit index; RMSEA: root mean square error of approximation.

exhibits that the revised and shortened CSDSC one-factor model provided more appropriate fit indices than the original CSDSC two-factor model. In the confirmatory factor analysis of the original CSDSC, assuming that there were two factors, the fit indices for the GFI, AGFI, CFI, and RMSEA were 0.89, 0.85, 0.81, and 0.08, respectively. The factor loading for each scale was significant (p < 0.001). When we conducted the same analysis assuming that there was one factor based on the results of the exploratory factor analysis, the indices were 0.94, 0.91, 0.92, and 0.07. Thus, the latter model had a better fit, meaning that it reached the recommended acceptable level of model fit (i.e., GFI, AGFI, CFI > 0.90, and RMSEA < 0.08).

4. Discussion

The current study aimed to statistically analyze the structural factors, reliability, and validity of the original CSDSC. The original scale had several limitations, such as model fit and factor structure.

The re-analysis resulted in the development of a model (the revised and shortened CSDSC) comprising the following factor and 11 items: “A sense of difficulty in providing developmental support in childcare”. Higher scores indicate a greater level of difficulty in providing developmental support to each child during childcare. A confirmatory factor analysis was used to assess whether one- or two-factor structures were more appropriate for the scale. Overall, the fit indices in this study demonstrated that the one-factor model provided a good fit to the data and was better than the two-factor model. The Cronbach’s α coefficients for the revised and shortened CSDSC had sufficiently high internal consistency, similar to the original CSDSC in a previous study.

As mentioned above, previous studies have shown that nursery teachers’ childcare values significantly impact their childcare practices and methods of providing support [13]. Along these lines, nursery teachers’ attitudes and values regarding child development support in childcare can influence child development support practices in many ways. For instance, according to Matsuo et al. [26], Japanese kindergarten and certified childcare center teachers’ willingness to continue working is not influenced by the number of disabled children in a class but by the perceived difficulties associated with caring for the children. Furthermore, in a study on consciousness toward inclusive childcare, inexperienced caregivers and caregivers with no experience in caring for children with disabilities were more likely to struggle with practicing inclusive childcare [27].

In recent years, many studies have highlighted teaching as a highly stressful occupation. The turnover rate of nursery teachers in Japanese childcare facilities has been high, and Japan has been facing a shortage of kindergarten teachers [28]. A high teacher turnover rate can negatively affect the quality of preschool services, which can impact the community, especially children and families [29]. One of the important factors linked to high levels of turnover is job stress [30], and mental health is associated with teachers’ willingness to continue working [31]. The influence of “differences in attitudes on childcare” in daily childcare is significant for nursery school teachers and kindergarten teachers [32], and differences in attitudes toward childcare have the greatest influence on the perception of stress among relationships between childcare workers, which can be a major stress factor [33]. In addition to balanced work conditions, workers’ agreement with childcare and/or education policies in the workplace may reduce the turnover rate [26].

Therefore, differences in perceptions of child development support in childcare among nursery teachers do not only affect collaboration but also substantially influence daily childcare practices, parental support, and psychological health. The revised and shortened CSDSC can contribute to the identification of individual factors affecting childcare practices and the psychological health of childcare providers in today’s childcare settings. Thus, further research is warranted to identify individual and environmental factors that correlate with their values of developmental support in childcare and to contribute to their childcare practices. Moreover, the relationship between the value of child development support in childcare and stress must be clarified, and how the relationship differs depending on individual factors must be understood.

In terms of the limitations of this study, first, since all participants were recruited from a single prefecture in Japan, some regional characteristics may exist. Given that whether the results can be applied to nursery teachers in different prefectures remains unclear, caution must be exercised when generalizing the results. To this end, future research should examine the extent to which the results are robust across different participant samples. Second, due to the study’s cross-sectional design, all data were obtained at a single time point, and we cannot infer that the reliability of the results is repeatable. The reliability of the revised and shortened CSDSC was evaluated by calculating Cronbach’s α coefficient but not using a test–retest method. Third, this study did not consider additional statistical evidence such as construct validity and criterion-related validity; hence, future research should consider these aspects to obtain higher evidentiary support. Last, the revised and shortened CSDSC resolved the limitations of the original scale, such as the model fit and factor structure, except for a low total variance in the measured variables (30.99%), which should be taken up in future studies.

5. Conclusions

Despite the limitations mentioned above, this study demonstrated that the revised and shortened CSDSC has adequate levels of reliability and validity, making it more appropriate than the original CSDSC for evaluating nursery teachers’ values regarding child development support in childcare. This study suggests that the revised and shortened CSDSC is a reliable and valid scale for research on the value of child development support in childcare among nursery teachers in Japan.