This is an open access article under the CC-BY-SA license. REiD (Research and Evaluation in Education), 6(1), 2020, 1-9 Available online at: http://journal.uny.ac.id/index.php/reid Developing instruments for measuring the level of early childhood development *1I Wayan Gunartha; 2Tajularipin Sulaiman; 3Siti Partini Suardiman; 4Badrun Kartowagiran 1Faculty of Language and Arts Education, Institut Keguruan dan Ilmu Pendidikan PGRI Bali Jl. Seroja, Tonja, Denpasar Timur, Kota Denpasar, Bali 80235, Indonesia 2Faculty of Educational Studies, Universiti Putra Malaysia Persiaran Masjid, 43400 Serdang, Selangor, Malaysia 3Faculty of Teacher Training and Educational Sciences, Universitas Ahmad Dahlan Jl. Ahmad Yani (Ringroad Selatan), Tamanan, Banguntapan, Bantul, Yogyakarta 55166, Indonesia 4Faculty of Engineering, Universitas Negeri Yogyakarta Jl. Colombo No. 1, Karangmalang, Depok, Sleman, Yogyakarta 55281, Indonesia *Corresponding Author. E-mail: w.gunartha@yahoo.com Submitted: 11 January 2020 | Revised: 17 April 2020 | Accepted: 20 April 2020 Abstract The aims of the study were to: (1) develop a set of instruments to measure the level of early childhood development (kindergarten group B), and (2) assess the quality of the developed instruments. This study is developmental research. The samples of the study were the students of kindergarten group B. The developed instrument was a set of questionnaires. Instrument testing was carried out in three stages with the number of subjects increased on each stage. The validity analysis of the questionnaire used confirmatory factor analysis (CFA). The reliability estimation of the questionnaire used composite reliability. The results of the study are in the form of instruments for measuring the level of early childhood development, which consists of an instrument to measure religious morality, social-emotional, language, cognitive, and physical-motor development. Based on field study, all instruments have a good fit model, construct validity, and reliability that meet the academic requirements of early childhood education. Keywords: instrument development, early childhood, kindergarten, potential How to cite: Gunartha, I., Sulaiman, T., Suardiman, S., & Kartowagiran, B. (2020). Developing instruments for measuring the level of early childhood development. REiD (Research and Evaluation in Education), 6(1), 1-9. doi:https://doi.org/10.21831/reid.v6i1.21996 Introduction Early childhood is the most important and fundamental beginning period. Therefore, early childhood is often called the Golden Ages. This period is also often called a sensi- tive period, the period of play, the critical period because this period will affect the fu- ture lives of the child. According to Woolfolk (2007, p. 23), approximately one month after conception, human brain development has begun. Neuron cells appear with incredible speed, i.e. 50,000 to 1,000,000 per second for approximately three months. When born, we have had about 100 to 200 billion neurons, and each neuron has about 2,500 synapses. Synapses that are unused or not getting stimu- lation from the environment will be trimmed (synaptic pruning). Berk (2007, p. 121) added that the complexity of connections between neurons will determine the child's level of intelligence. The same thing was said by Miller and Cumming (Rushton, 2011, p. 92). https://creativecommons.org/licenses/by-sa/4.0/deed.id https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran 2 - Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 ISSN: 2460-6995 (Online) All of the aforementioned statements show that the growth and development of the brain (synapses) are determined by the stimuli or stimulus provided to the child and the ac- tivities undertaken by children. Thus, children who are growing and developing must be activated by providing a variety of stimuli and appropriate activities. This is the important role of early childhood education (ECE) as a form of physical and psychosocial stimulation whether at home or in early childhood institu- tions, besides nutrition and health care. Hence, according to Valentine, Thomson, and Antcliff (2009, p. 196), in Australia, even early education and parenting has got priority policy. An appropriate stimulation, as well as nutrition, parenting, and also health services at an early age, will develop all of children’s potential, including their physical, cognitive, language, art, social-emotional, self-discipline, religious values, self-concept, and also self-re- liance, will develop optimally. Therefore, early childhood education is expected to contribute significantly to the improvement of human resources (HR) quality, which will make our nation to be high quality and full of compe- titiveness in the future. The importance of stimulation obtained by children in early childhood education insti- tutions is also proved empirically by many experts. Samuelsson (2011, p. 109) in her re- search on the role of early childhood educa- tion said that learning at an early age has an influence in the future, for example, the suc- cess of the school, as well as the attitude and attention, will be formed early on. Mann and Reynolds (2006, p. 153) concluded that pre- school intervention correlates with a reduc- tion in the incidence, frequency, and severity of the delinquency at age 18 years. Ashiabi (2007, pp. 205–206) states that a lot of advantages can be attained to let chil- dren playing with other children. For exam- ple, sociodramatic playing can improve a child's ability to imagine before acting, taking a role, empathy, altruism, and also emotions and rules understanding. Moreover, negotia- tion and problem-solving skills also increase, such as the ability to work cooperatively with others, share, self-control, and working with the group. In other words, sociodramatic play can enhance children’s social and emotional development optimally. A similar study was conducted by Beard and Sugai (2004, p. 408). By the importance of early childhood education, the government's attention to de- veloping early childhood education becomes greater. Since 2000, Early Childhood Educa- tion (ECE) started to become a central issue in education, including in Indonesia, even more, Erman Syamsudin, Director of Early Childhood Development (Ministry of Nation- al Education of Republic of Indonesia, 2011, p. vii), stated that early childhood education is one of the priority programs of national edu- cation development. Early Childhood Educa- tion (ECE) services are expected to nurture, grow, and develop the whole early childhood potential optimally so it can form the basic ability and behavior according to the chil- dren’s development stage. In response to the government policy, the public has shown their concern for the problems of education, protection, and care of early-aged children with a variety of ser- vices in accordance with their conditions and capabilities. Public awareness of the impor- tance of early childhood education in the op- timal development of children's potential has been shown with various active participation in the implementation and improvement of services. Although various policies have been issued by the government, in fact, there are still many problems that exist in the imple- mentation of early childhood services in- cluding in Badung Regency, Province of Bali. There are still many children who have not gained early childhood services. This fact ac- knowledged by the General Director of Early Childhood Education, Non-Formal and In- formal that although the policies have been established and socialized related to early childhood development, in fact, of the 28, eight million children aged 0-6 years in late 2009, who gained an early childhood educa- tion services is just around 53.7% (Ministry of National Education of Republic of Indonesia, 2011, p. iii). Research by Hiryanto (2007) about the mapping of the quality achievement level of early childhood programs in Yogyakarta re- https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 - 3 ISSN: 2460-6995 (Online) veals that by the views of the implementation guidelines to the suitability of early childhood education with the real conditions of the pro- gram implementation based on the ten bench- marks of national education, in the implemen- tation of early childhood education in the Yogyakarta Province, some problems can be found as follows: (1) The variation in the im- plementation of education; (2) The existence of the age groupings that do not fit the guide- lines because of limited infrastructure and educators; (3) There are still some educators who have not received training; (4) The ratio of the teachers and students number is not ideal. The research by Hermawati (2007) in a children daycare in Beringharjo, Yogyakarta, found two drawbacks of the input variables, namely, the teacher’s educational background and caregiver qualifications that are not rele- vant to the tasks. In the process variables, the problem is the immeasurability of mentoring by a caregiver. It is associated with the major- ity of low caregiver’s education. In addition, the assistance has not been done regularly by the organizers. The public access to the chil- dren's daycare Beringharjo is limited because of limited capacity. Based on the preliminary study conducted at the early childhood insti- tution in Badung Regency, Province of Bali, it was also found many problems related to the implementation of early childhood education (ECE), such as the quality and quantity of early childhood teachers are still relatively low and the number of teachers is on the average of three to four people. In terms of process, the kindergarten student has been taught reading, writing, and arithmetic skills because, according to the teacher, if it is not done, then no parents want to enroll their children to the institution. In order to give good quality of early childhood education (ECE), in accordance with the existing standards, early childhood education services need to be evaluated regu- larly. According to Nugraha (2010, p. 3), good quality of early childhood education service is regularly evaluated and the results are acted upon appropriately. The same opinions are also expressed by Mardapi (2012, p. 12), that the improvement of the education quality can be achieved through improving the quality of learning by the improvement of the quality of the assessment system. Therefore, the avail- ability of quality evaluation instruments is very important, in which it can be used by the government to evaluate early childhood edu- cation services continuously. By the results of evaluation activities, we will be able to know the things that have been achieved, whether a program can meet the established criteria or not. Currently, the evaluation of early child- hood services internally has not done thor- oughly. Likewise, in Bali Province, even in Badung Regency, based on the preliminary study which has been conducted, the Depart- ment of Education has never done an evalu- ation of the existed early childhood services. The quality determination of early childhood institutions is often based on the frequency of competition participation and the number of early childhood learners. This is caused by the absence of evaluation instruments of early childhood services that have been tried and tested, both in terms of validity and reliability. The evaluation results will provide accurate information when obtained through evalu- ation using reliable instruments. Until now, the government, especially the Badung De- partment of Education, has not had a stan- dard instrument for evaluation that can be used by the Department of Education or by the head of the kindergarten institution as an internal evaluation. The education service is a system which consists of interlinked components and they mutually determine each other. These compo- nents are the input, process, and product. The component of inputs includes infrastructure, students, teachers, curriculum, and also sub- ject matter. The component of processes in- cludes lesson planning, implementation, and evaluation. The component of products on early childhood education services includes the achievement level of early childhood de- velopment, such as moral religious, social- emotional, language, cognitive, and physical- motor development. In evaluating early child- hood education services, these components should be evaluated continuously. Therefore, it is necessary to develop an instrument to https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran 4 - Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 ISSN: 2460-6995 (Online) evaluate the inputs, processes, and products of early childhood services. Based on that description, the instruments developed in this study is only an instrument for measuring the products, that is, an instrument to evaluate the achievement level of the early childhood development, which includes: (a) moral-reli- gious, (b) social-emotional, (c) language, (d) cognitive, and (e) physical-motor. This is caused by the limited costs, energy, and time available. Based on those aforementioned back- grounds, the problems in this study are as follows. How is the instrument for measuring the level of childhood development? How is the quality of the developed instruments, both in terms of validity and reliability? Based on the problems, the purpose of this research is to develop evaluation instruments that can be used to evaluate the level of early childhood development, particularly for kindergarten group B so as to provide complete and ac- curate information for program managers and to assess the quality of the evaluation instru- ments developed. The products of this study are a set of an evaluation instrument for early childhood education services, particularly for kindergar- ten group B. The evaluation instrument of early childhood education services limited to instruments tends to measure the achieve- ment level of early childhood development, which includes moral-religious, social-emo- tional, language, cognitive, and physical-motor development. The development of an evalu- ation model for early childhood services pro- gram is very beneficial, both theoretically and practically. Theoretically, this study is useful as a contribution to developing the existed evaluation methodology to generate new concepts in the field of evaluation science. Practically, the results of this study are useful for teachers, principals of early childhood/ kindergarten, as well as the Department of Education. For teachers in early childhood education (ECE), a kindergarten teacher, in particular, this instrument can be used to measure the effectiveness of the performed services and the results can be useful as a basis to make corrections to educational ser- vices. Method Development Model This study is a research and develop- ment (R & D), which aims to produce a prod- uct in the form of a set of instruments in order to evaluate the level of early-aged chil- dren’s development (specifically kindergarten group B). The development research adopts the model which was proposed by Borg and Gall (1983, p. 775). The ten steps of devel- opment by Borg and Gall were then sim- plified into four steps, namely: (1) preliminary investigation, (2) design phase, (3) testing, evaluation, and revision, and also (4) imple- mentation. In the early stages, we conducted sev- eral activities, including a preliminary study, review the theory of instrument evaluation models, early childhood education, as well as review the results of research that has been done. In the design phase, the draft instru- ment was designed in order to measure the level of early childhood development, which consists of instruments for measuring prod- ucts of services and the test design. In the pilot, evaluation, and revision phase, expert validation and testing of the instruments that have been designed in kindergarten were conducted. The data of test results were then analyzed. If the results of the analysis show that the instrument is not yet good, then it can be revised and tested again until a final prototype eligible fit model (good prototype). Tests were conducted in three phases. In the implementation phase, the instruments that have been well and subsequently tested were implemented. Development Procedure Several steps were taken in developing the instrument for measuring the level of early childhood development. Each step is elabo- rated as follows. Drafting of the Design At this stage, evaluation instruments to evaluate the product of early childhood ser- vices were structured, which consist of instru- ments for measuring religious morality, social- emotional, language, cognitive, and physical- https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 - 5 ISSN: 2460-6995 (Online) motor development. All of those instruments were in the form of a questionnaire on a Likert scale with five points. These instru- ments are the first draft. Expert Judgment In order to check the content validity and refine the instrument draft, it was valid- ated by experts, namely, academicians or lec- turers and practitioners (kindergarten teacher), and also the user of the instrument (head/ deputy head of the kindergarten). The expert validation process used FGD (focus group discussions) model. The implementation of the FGD was conducted in two stages. The first FGD was conducted by ten academicians (lecturers) from the post-graduate program of Universitas Negeri Yogyakarta. When the in- strument was revised in accordance with aca- demicians' suggestions (lecturers), it was fol- lowed by another FGD and readability test by three kindergarten heads and also 17 kinder- garten teachers. After the test was carried out, it was continued by the assessment of the instrument. Tests The draft of the instrument that has been revised based on the advice obtained in the FGD was piloted in kindergarten to determine the fit model of the measurement, construct validity, and reliability. The instru- ment test was conducted in three stages, namely, the first, second, and third with the increasing number of test subjects. The num- bers of kindergarten were: 10, 13, and 18 and 160, 260, and 360 kindergarten children as the subject. Data Analysis The data about the comprehensiveness and also clarity of the instrument which were obtained from the experts were then analyzed descriptively. The data which were taken from the results of the field test were then analyzed using Confirmatory Factor Analysis (CFA) in order to find out the goodness of fit (GoF) as well as determine the validity and reliability, with the 8.8 Lisrel program. In determining the goodness of fit, several indicators were employed, including: (a) the value of chi- square p-value ≥ 0.05, (b) root mean square error of approximation (RMSEA) ≤ 0.08, and goodness of fit index (GFI) ≥ 0.9 (Ghozali & Fuad, 2008, pp. 29–31; Latan, 2012, p. 53). The construct reliability was calculated based on lambda (λ) for each indicator, and the error variance (δ) indicator. In the descriptive-qualitative analysis, the average score of the quantitative data that were obtained through an assessment instru- ment was calculated, then were converted into qualitative data with scale 5, and then finally were interpreted qualitatively. The results of the qualitative analysis were used as the basis for determining whether the developed instru- ment was good or not. In converting the quantitative data into qualitative data with scale 5, a modification of rules which were developed by Sudijono (2011, p. 329) was employed. The criteria of the instrument as- sessment which were used are presented in Table 1. Implementation After the last product of the instrument had been analyzed, a good prototype was im- plemented in 18 kindergartens. When it is depicted in the chart, the whole developing process of the instrument model of the early childhood development is clearly illustrated in Figure 1. Table 1. Criteria of Instrument Assessment Average Score Qualification Conclusion > 4.2 Very good Can be an example > 3.4 – 4.2 Good Can be used without any revision > 2.6 – 3.4 Quite good Can be used with a little revision > 1.8 – 2.6 Less good Can be used with some revision ≤ 1.8 Bad Cannot be an example https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran 6 - Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 ISSN: 2460-6995 (Online) Annotation: : activity process : next step : activity result : : review for revision : analysis result Figure 1. Flowchart of Instrument Model Development Procedure Table 2. Instrument of Development Result Instrument for Evaluating Early Childhood Services Instrument Evaluated Item Instrument Form Instrument for measuring the level of early childhood development Achievement Level of Development: a. Moral-Religious Questionnaire b. Social-Emotional Questionnaire c. Cognitive Questionnaire d. Language Questionnaire e. Physical-Motor Questionnaire Findings and Discussion The instrument for measuring the level of early childhood development consists of five components, namely the instruments for measuring the development of moral-reli- gious, social-emotional, language, cognitive, as well as physical-motor components. The type of the product instrument of the early child- hood services developed is clearly presented in Table 2. Validation Result of Experts and Practitioners The instrument assessment by experts and practitioners was directed into four main aspects, namely: (a) the clarity of instrument guidance, (b) the completeness of instrument indicators, (c) the suitability of the indicators with the point, and (d) the effectiveness of Indonesian. The assessment used a scale of 5 with the lowest score was 1 and 5 was the highest. Anal isis Preliminary Research Instrument Design Expert Validation First Draft Revision Test Revision Draft Analysis Good Bad Prototype Implementation Development Result https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 - 7 ISSN: 2460-6995 (Online) Based on the average score given by the experts, the mean obtained is 4.1 in total. In line with the conversion guidelines, the mean is at intervals of 3.4 to 4.2 and is classified as good. Based on the assessment conducted by teachers and heads of kindergarten, it is ob- tained a mean score of 4.29 in total. The mean score according to those criteria is also quite good. The total mean score of the two asses- sors groups is 4.2 It means that the instru- ment has been well conducted and can be used without any revision, shown in Table 3. Instruments Measurement Model Based on the analysis, all items on all instruments of the three pilot phases are signi- ficant (t> 1.96), meaning that all items can be used to measure the construct well. In the third test, there are some items of achieve- ment level instruments for language develop- ment that have smaller factor loading than 0.5, i.e. 0.49 and 0.48. Since it is approaching 0.5, then it is rounded to 0.5. Thus, all instru- ments have good construct validity. By look- ing at the model fit, on the third test, all re- quirements of model fit are met, both the p- value (≥ 0.05), RMSEA (≤ 0.08), and GFI (≥ 0.9). The construct reliability (CR) of all in- struments are above 0.7 in all three stages of the test. Thus, based on the three stages of the test, all of the instruments have good con- struct validity, reliability, and goodness of fit. Those three phases’ analyses are presented in Table 4. In this study, five instruments for meas- uring the level of early childhood develop- ment were developed, namely: instruments for measuring moral-religious, social, lan- guage, cognitive, and physical-motor develop- ment. The instrument developed is in the form of a questionnaire. Instrument indica- tors are based on indicators of the level of achievement of early childhood development contained in the Regulation of the Minister of National Education No. 58 of 2009 on the Standard for Early Childhood Education, spe- cifically the standard level of achievement of Table 3. Recapitulation of Experts and Practitioners Validation Validator Validator Number Average of Score Qualification Experts 10 4.10 Good Practitioners 20 4.29 Good Total 30 8.40 - Mean 4.2 Good Table 4. Summary of Analysis Result for Instrument Measurement Model of Product and Outcome Instrument Number of Point Test No. Chi-Square RMSEA GFI λ < 0.5 CR Score p-value Moral-Religious Development 25 1 308.30 0.07 0.029 0.87 2 0.89 2 311.77 0.058 0.023 0.91 - 0.91 3 307.31 0.075 0.019 0.94 - 0.91 Social-Emotional Development 26 1 330.54 0.081 0.027 0.86 - 0.92 2 333.69 0.07 0.022 0.91 - 0.91 3 331.38 0.089 0.018 0.93 - 0.92 Language Development 24 1 282.39 0.060 0.030 0.87 - 0.70 2 276.32 0.089 0.022 0.90 - 0.75 3 286.48 0.051 0.02 0.94 2 0.82 Cognitive Deevelopment 26 1 331.27 0.066 0.028 0.86 - 0.87 2 326.05 0.089 0.021 0.91 2 0.80 3 330.72 0.075 0.018 0.93 1 0.76 Physical-Motor Development 27 1 356.76 0.077 0.027 0.86 4 0.72 2 351.72 0.094 0.02 0.91 3 0.85 3 355.86 0.076 0.018 0.93 - 0.82 Life Skills 30 1 439.09 0.081 0.025 0.84 1 0.72 2 437.05 0.092 0.019 0.90 - 0.76 3 447.32 0.055 0.018 0.92 2 0.74 https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran 8 - Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 ISSN: 2460-6995 (Online) development. The procedure for developing this instrument follows five steps, namely: (1) the design preparation phase, (2) the expert validation phase, (3) the testing phase, (4) the data analysis phase, and (5) the implemen- tation phase. The draft instruments that have been compiled are then validated by experts to see indicator depth, formulation of ques- tions or statements, language effectiveness, and others. The experts who validated the in- strument consisted of ten peoples, who came from several fields of science, namely: two measurement experts, three evaluation ex- perts, one education management expert, two primary education experts, and two childhood education experts. The goal is that the instru- ment can be assessed in various aspects, so as to produce a quality instrument. After being revised based on the FGD input, the instrument was tested to determine the construct validity and reliability. The trial was conducted in three stages, with the num- ber of trial subjects increased. Two assump- tions underlie the thinking of why the test was conducted in three stages, namely: (1) increas- ing variety and the number of trial subjects three times expected to reach all kinds of characteristics, both kindergarten and existing students, and (2) with the representation of all kindergarten characteristics and students, then a good instrument will be obtained, which is an instrument that can be applied to all ex- isting kindergarten. Based on the results of the test data analysis that conducted from the first stage to the third stage, the following results were obtained. The results of the first phase of the trial show that the five instruments developed were still lacking. After the items points were revised, the second trial was conducted. The results of the second phase of the trial (main trial) show that the fit model of instrument had become better. Only some items of in- struments still have deficiency. The items of instrument was revised again and the third trial was conducted. In this study, all the poor instruments have been revised in two stages, the results of the third stage of the test show that all instruments have good fit model, val- idity, and reliability. Therefore, all instruments developed have a good measurement model, because: (a) all values of χ2 are low (p ≥ 0.05), (b) all RMSEA ≤ 0.08, and (c) all GFI values ≥ 0.9. The coefficients of construct reliability (CR) are all above 0.7. Thus, all instruments developed have good quality. Conclusion Based on the research findings, two points of conclusion can be drawn. Each point is elaborated as follows. (1) The instru- ments for measuring the achievement level of early childhood development developed in this research consist of five components: instrument to evaluate the achievement level of moral-religious, social-emotional, cognitive, language, and physical-motor development. (2) According to the assessment of experts and practitioners, the instruments developed have good quality and can be used without any revision. All developed instruments have good validity, reliability, and goodness of fit. References Ashiabi, G. S. (2007). Play in the preschool classroom: Its socioemotional significance and the teacher’s role in play. Early Childhood Education Journal, 35(2), 199–207. https://doi.org/ 10.1007/s10643-007-0165-8 Beard, K. Y., & Sugai, G. (2004). First step to success: An early intervention for elementary children at risk for antisocial behavior. Behavioral Disorders, 29(4), 396–409. https://doi.org/10.1177/ 019874290402900407 Berk, L. E. (2007). Development through the lifespan (4th ed.). Boston, MA: Pearson Education. Borg, W. R., & Gall, M. D. (1983). Educational research: An introduction (4th ed.). New York, NY: Longman. Ghozali, I., & Fuad, F. (2008). Structural Equation Modeling: Teori, konsep, dan aplikasi dengan program Lisrel 8.80. Semarang: Badan Penerbit Universitas Diponogoro. Hermawati, I. (2007). Evaluasi program Pendidikan Anak Usia Dini (PAUD) bagi https://doi.org/10.21831/reid.v6i1.21996 https://doi.org/10.21831/reid.v6i1.21996 I Wayan Gunartha, Tajularipin Sulaiman, Siti Partini Suardiman, & Badrun Kartowagiran Copyright © 2020, REiD (Research and Evaluation in Education), 6(1), 2020 - 9 ISSN: 2460-6995 (Online) anak dari keluarga miskin di tempat penitipan anak (TPA) Beringharjo, Yogyakarta. Yogyakarta: Departemen Sosial RI, Badan Pendidikan dan Penelitian Kesejahteraan Sosial, Balai Besar Penelitian dan Pengembangan, Pelayanan Kesejahteraan Sosial. Hiryanto, H. (2007). Pemetaan tingkat pencapaian mutu program pendidikan anak usia dini (PAUD) di Provinsi DIY. (Laporan penelitian, tidak diterbitkan). Yogyakarta: Lembaga penelitian UNY. Diklus: Jurnal Pendidikan Luar Sekolah, 6(11), 127–149. Retrieved from https://journal.uny.ac.id/index.php/dik lus/article/view/5787 Latan, H. (2012). Structural Equation Modeling: Konsep dan aplikasi menggunakan program Lisrel 8.80. Bandung: Alfabeta. Mann, E. A., & Reynolds, A. J. (2006). Early intervention and juvenile delinquency prevention: Evidence from the Chicago longitudinal study. Social Work Research, 30(3), 153–167. https://doi.org/ 10.1093/swr/30.3.153 Mardapi, D. (2012). Pengukuran, penilaian, dan evaluasi pendidikan. Yogyakarta: Nuha Medika. Ministry of National Education of Republic of Indonesia. (2011). Petunjuk teknis penyaluran bantuan alat permainan edukatif. Jakarta: Directorate of Early Childhood Education Development, Ministry of National Education of Republic of Indonesia. Nugraha, A. (2010). Evaluasi pembelajaran untuk anak usia dini. Bandung: Universitas Pendidikan Indonesia. Regulation of the Minister of National Education No. 58 of 2009 on the Standard for Early Childhood Education. , (2009). Rushton, S. (2011). Neuroscience, early childhood education and play: We are doing it right! Early Childhood Education Journal, 39(2), 89–94. https://doi.org/ 10.1007/s10643-011-0447-z Samuelsson, I. P. (2011). Why we should begin early with ESD: The role of early childhood education. International Journal of Early Childhood, 43(2), 103–118. https://doi.org/10.1007/s13158-011- 0034-x Sudijono, A. (2011). Pengantar evaluasi pendidikan. Jakarta: Raja Grafindo Persada. Valentine, K., Thomson, C., & Antcliff, G. (2009). Early childhood services and support for vulnerable families: Lessons from the Benevolent Society’s Partnerships in Early Childhood program. Australian Journal of Social Issues, 44(2), 195–213. https://doi.org/ 10.1002/j.1839-4655.2009.tb00140.x Woolfolk, A. (2007). Educational psychology (10th ed.). Boston, MA: Allyn & Bacon. https://doi.org/10.21831/reid.v6i1.21996