exploringtherelatedfactorsineducationqualitythrough...

Research ArticleExploring the Related Factors in Education Quality throughSpatial Autoregressive Modeling with Latent Variables A RuralCase Study

Anik Anekawati12 Bambang W Otok 2 Purhadi2 and Sutikno2

1Faculty of Teacher Training and Education Universitas Wiraraja Sumenep 69451 Indonesia2Department of Statistics Institut Teknologi Sepuluh Nopember Surabaya 60111 Indonesia

Correspondence should be addressed to Bambang W Otok bambang_wostatistikaitsacid

Received 30 June 2020 Accepted 18 September 2020 Published 7 October 2020

Academic Editor Enrique Palou

Copyright copy 2020 Anik Anekawati et al is is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

e principle of education for sustainable development (ESD) is that no child is left behind Hence the fourth sustainabledevelopment goal (SDG) of the United Nations (UN) emphasizes inclusion and equity in education by focusing on eliminatingdisparities among regions is study explores factors related to education quality through modeling in rural areas of SumenepRegency in East Java Indonesia Currently only a few kinds of research studies involve spatial data latent variables and at thesame time tests of their spillover effects e modeling herein is the spatial autoregressive model with latent variables (SAR-LVs)e latent variables were estimated using the weighted least square (WLS) method while the Lagrange multiplier (LM) test wasused for spatial dependence testing e parameters of the SAR-LVs were estimated using two-stage least square (2SLS) eresults show that the quality of education is directly influenced by the infrastructure of the schools but not by the socioeconomicconditions of the local communities e autoregressive spatial coefficient has a significant but negative effect which shows anegative spillover from districts with a lower quality of education to the ones with a high quality of education is is due to thestudentsrsquo competition to get registered for a favorite or good quality school in a particular district which stimulates the migrationof students from its neighboring districts is reveals the inequality of school quality since not all students can get access toschools with good quality rough this study some recommendations are given as a contribution to achieving the fourth SDGin Indonesia

1 Introduction

Indonesia is an archipelago with more than 16000 islandsthat are 5027 rural and 4973 urban by area It is alsoknown as a multiethnic country with over 1300 ethnicitiesdivided into 31 ethnic groups [1] e Indonesian Ministryof Education and Culture has identified 668 local languagesin Indonesia and its data show that 795 of the Indonesianpopulation aged above five years communicates daily usinglocal languages [1]

Based on its demographic and sociocultural character-istics Indonesia has the potential to have a high disparityparticularly in education access and quality erefore

Indonesia has tried to improve the educational attainment ofchildren from low-income households in the context ofequity in education for all In 1997 the Indonesian gov-ernment implemented a social safety net (SSN) program tobuffer the hardship resulting from the economic crisis eaim of the SSN program was to maintain the quality of thelearning process and school enrollment rates and to reducethe dropout rates e SSN program consists of scholarshipsfor the poor and subsidies for the schools [2 3] In 2005 theIndonesian government implemented the school opera-tional assistance (SOB) and an unconditional cash transferprogram (UCT) e SOB program provided subsidies forboth public and private schools with the aim of maintaining

HindawiEducation Research InternationalVolume 2020 Article ID 8823186 10 pageshttpsdoiorg10115520208823186

the quality of the educational service UCT program wasintended to anticipate the negative effects of the govern-mentrsquos plan to reduce a fuel subsidy

Kharisma [4] found that the provision of SOB had apositive effect on dropout rates for students aged between 16and 20 years but not for those aged between 7 and 15 yearsFurthermore the Indonesian government has constantlycommitted to not leaving any child behind in developmentas the core of the agenda 2030 on education for sustainabledevelopment (ESD) Since 2014 the Indonesian governmenthas provided smart cards to more than 24 million poorstudents and others who cannot attend school because offinancial issues Based on the report [5] Indonesia is one ofthe low-income or middle-income countries whose share ofthe state budget for education is likely to be high eamendment of the education budget was simultaneous withthe start of a decentralized government system

Since 1998 Indonesia has given greater autonomy tolocal governments due to the commencement of a refor-mation movement across Indonesia Education manage-ment was also delegated more to the local governments econstitution states that local governments are obliged toallocate a minimum of 20 of their budgets to the educationsector Furthermore local governments have a crucial role inproviding access to education

Muttaqin [6] found that local governments with moreresources can provide more scholarships to attract pupilsfrom poor families to return to school but this is not the casefor local governments with fewer resources such as localgovernments with greater rural areas At least there is adisparity in education quality between rural and urban areasand between western and eastern Indonesia Based onwelfare statistics [7] the percentage of the population whoare aged above 15 years that literate is 9749 in urban areasand 9313 in rural areas e school enrollment ratio of thepopulation aged between 16 and 18 years is 7521 in urbanareas and 6771 in rural areas e net enrollment ratio ofthe population for senior high school is 6433 in urbanareas and 5606 in rural areas Azzizah [8] point out theapparent gap of education among the eastern and westernprovinces in Indonesia using the dummy regression analysisand the most influencing factor was the poverty rate

e report [9] showed that the Indonesian children andyouth who do not have sufficient access to education are inremote and isolated areas and in poor communities isreport was based on data of the National SocioeconomicSurvey 2015 which stated that the composition of the groupof out-of-school children (OOSC) of elementary school agewas 81 from rural areas (based on location) 49 frompoor families (based on wealth) and 40 from Papua (basedon region)

e fourth sustainable development goal (SDG) oneducation is to ensure inclusive and equitable education andto promote lifelong learning opportunities for all One of thetargets is ldquoby 2030 to eliminate gender disparities in edu-cation and ensure equal access to all levels of education andvocational training for the vulnerable including personswith disabilities indigenous peoples and children in vul-nerable situationsrdquo One of the disparity indices is rural-

urban As reported in [10] youths aged between 20 and 24years living in rural areas in 101 low- and middle-incomecountries have on average an education period 26 yearsshorter than do those in urban areas Indonesia is one ofthese countries In Indonesia rural-urban disparities arewidely prevalent across indicators in each target of the fourthSDG since poverty is still concentrated in rural areas [9]e report [5] shows that children born in selected ruralareas in Indonesia whomigrated to an urban area as childrenwere found to have attained three more years of educationthan those who did not migrate is shows that migrationfrom rural to urban areas can increase educational attain-ment due to the low access to education in rural areas

erefore some educational research studies have fo-cused on rural areas to provide more specific reviews Yueet al [11] researched policy and trends in rural educationfrom preschool to high school in China over 40 years from1978 to 2018 Rural schools continually improved in allaspects In 1980 only a small proportion of rural childrenattended preschool but in 2014 more than 90 of ruralchildren attended preschool Carrascal et al [12] focusedtheir research on rural primary schools in Mixco Guate-mala ey developed the ECO Kit as an effective teach-ingndashlearning tool that contributed to improving the visualliteracy and creativity of students In Finland digitalcommunication is being more understandable by ruralteachers than in the urban area and promoted parent-teacher partnership in rural better than did parents in theurban area [13] Cheng et al [14] documented the learningproblems of disadvantaged students in rural areas in Taiwane results showed that in rural elementary schools theeducation levels of most of the parents were below thecollege level students had low motivation and the ordinarystudents and disadvantaged students indeed showed dif-ferences Azzizah [8] found that poverty had a huge influ-ence on the rate of school enrollment particularly inIndonesiarsquos eastern province based on Joseph Nkurunzizarsquosanalysis (hindawi 2012) that students from very poor peoplehave lower odds of attendance in school Based on [9]poverty in Indonesia is still concentrated in rural areas

East Java is one of the provinces in western Indonesiaand it is relatively more advanced than others In East Javaprovince the OOSC rate in the primary school level is only7 while in junior secondary it is 12 and in seniorsecondary it is 14 ese numbers are relatively lowcompared to those in other provinces However one of theregencies in East Java has 100 underdeveloped regionsnamely Sumenep Regency [15] e percentage of poorpeople in Sumenep Regency is 1962 which is the fourthhighest in East Java and the human development index is6428 the fourth lowest in East Java [16]

e principle of ESD is that no child is left behind andemphasizes equity of education by focusing on eliminatingdisparities among regions and socioeconomic groupserefore in this study the factors related to educationquality in Sumenep Regency were explored through thespatial autoregressive model with latent variables (SAR-LVs) Sumenep is a regency with rural characteristics It isnecessary to study the modeling of education quality

2 Education Research International

specifically rural cases to get deeper and more specificinformation

For this modeling the SAR-LVs approach was usedsince it involves spatial data and latent variables Manystudies on education have involved spatial data Gille [17]tested the existence of education spillovers in a rural contextin India using fixed-effects vector decomposition consistingof the three-stage estimation e result was that one ad-ditional year of education of neighborsrsquo increases farmproductivity by 2 Gao et al [18] analyzed the disparity incompulsory education based on the perspective of the im-balanced spatial distribution using spatial autocorrelationmethods Xu et al [19] calculated the index of geographicaccessibility available opportunity and economic afford-ability to describe the social spatial accessibility In all thesestudies spatial data were used but the spillover effect wasnot tested erefore the aim of the study herein was toperform modeling of education quality that involves spatialdata and latent variables and to test the spillover effect ecase study takes Sumenep Regency since Sumenep has thecharacteristics of an archipelago and multiethnic as aminiature of Indonesia but represents rural areas

Sumenep Regency is one of the regencies in East JavaProvince precisely at the eastern end of Madura Island isregion is located between 113deg32prime54Prime and 116deg16Prime48 eastlongitude and 4deg55Prime00Prime and 7deg24Prime00Primesouth latitude eregion borders Madura Straits to the south the Java Sea tothe north Pamekasan Regency to the west and the Java Seaand Flores Sea to the east Geographically Sumenep Regencyconsists of land and islands with a total area of 209346 km2namely the mainland section of 114693 km2 (5479) andthe archipelago of 94653 km2 (4521) One hundred andtwenty-six islands consisting of 48 inhabited islands and 78uninhabited islands make up the archipelago as shown inFigure 1

Sumenep Regency is divided into 27 districts 18 districtsare on the mainland and 9 districts in the island Districts onthe mainland are Ambunten Batang Batang BatuanBatuputih Bluto Dasuk Dungkek Ganding Gapura GulukGuluk Kalianget Sumenep City Lenteng MandingPasongsongan Pragaan Rubaru and Saronggi Districts inthe island are Arjasa Gayam Giligenteng KangayanMasalembo Nonggunong Raas Sapeken and Talango epopulation consists of at least six ethnic groups namelyMadura Bajo Mandar Bugis Chinese Arabic andJavanese

In 2018 the number of senior high schools in SumenepRegency was 226 consisting of 13 public schools and 213private schools e number of private schools is far morethan that of public schools describing the communityrsquosindependent participation in education It is an interestingillustration if related to the level of poverty in the SumenepRegency Sumenep Regency has 334 villages 100 of whichwere classified as disadvantaged villages [15] Based on data[16] the percentage of poor people is 1962 (the fourthhighest in East Java) and the human development index is6428 (the fourth lowest in East Java)

2 Materials and Methods

21 Research Methods is study focuses on the modelingof education quality at senior high schools using the SAR-LVs model is modeling involves the latent variables andthe sample unit is locationspatial Bollen [20] defined SEMin two models namely the measurement model and thestructural model e measurement model represents therelationship between the manifest variable and exogenouslatent variables (1) or endogenous latent variables (2) whilethe structural model describes the relationship among thelatent variables (3)

x Λx ξ + δ (1)

y Λy η + εlowast (2)

η Blowastη + Γ ξ + ζ (3)

where η is a vector of an endogenous latent variable ξ is avector of an latent variable Blowast is a coefficient matrix thatshows the effect of an endogenous latent variable to anotherendogenous variables Γ is a coefficient matrix which showsthe effect of ξ to η ζ is a vector of random error y and x arethe vectors of observed variableΛy andΛx are the coefficientmatrices which show the relationship of y to η and x to ξ andεlowast and δ are the vectors of measurements of y and x re-spectively where εlowast sim NB(0Θεlowast) δ sim NA(0Θδ) and Θδand Θεlowast are the covariant-variant matrices

Anselin wrote the spatial autoregressive model (SAR)[21] as shown in equation (4)

ylowast λWylowast + Xlowastβ + ε (4)

where ylowast is a vector of a spatially lagged dependent variableXlowast is a matrix of an exogenous variable λ is a spatialautoregressive coefficient ρ is a coefficient of a spatial errorβ is a vector of parameters associated with an exogenousvariable W is a matrix of spatial weight with the maindiagonal elements being zero and ε is the disturbance withε sim N(0 σ2I)

e factor score is the result of estimating the latentvariables in the measurement model via equations (1) and(2) using the weighted least square (WLS) method efactor scores for an endogenous variable and an exogenousrandom variable are given in equations (5) and (6)respectively

1113954ξprime ΛxprimeΘminus1δ Λx1113872 1113873

minus 1ΛxprimeΘ

minus1δ1113872 1113873X (5)

1113954ηprime ΛyprimeΘminus1δ Λy1113872 1113873

minus 1ΛyprimeΘ

minus1δlowastY (6)

where X and Y are the random observation matricese SAR-LVs model is obtained by replacing the spa-

tially lagged dependent variable (ylowast) and exogenous variable(Xlowast) in the spatial model of equation (4) with factor scoresfrom equations (5) and (6) e dependent variable ylowast isreplaced by 1113954η and Xlowast is replaced by 1113954ξ us the SAR-LVsmodel is written as shown in equation (7)

Education Research International 3

1113954η λW1113954η + 1113954ξβ + ε (7)

where the error distribution of ε is

ε sim NT (I minus λW)e1113954ηt minus 1113954ξβΘ1113872 1113873 (8)

where Θ (I minus λW)(ΛyprimeΘminus1

δ Λy)minus 1(I minus λW) ande (1 1)prime

Anselin [22] developed diagnostics for spatial depen-dence using the Lagrange multiplier (LM) test e LMapproach seems reasonable and relatively easy based onestimation under the null hypothesis [23] ie in its mostsimple form e spatial dependence test of the SAR-LVsmodel as shown in equation (7) makes use of the LM testdeveloped by Anselin [22] but based on the error distri-bution as shown in equation (8)e value of the test statisticLM SAR-LVs is

LMλ minus p(W1113954ξβ)prime1113957ε1113872 1113873

2

p D (9)

where p (ΛyprimeΘminus1

δ Λy) D (e1113954ηt minus 1113954ξβ)primeWWprime(e1113954ηt minus 1113954ξβ)1113957ε

(1113954η minus 1113954ξβ) and e (1 1)primee parameters of SAR-LVs in equation (7) are esti-

mated using the using the two-stage least square (2SLS)method reviewed by [24 25]

e SAR-LVs model as shown in equation (7) can besimplified as follows 1113954η Zα + ε where Z (1113954ξ |W1113954η) andα (βprime | λ)prime e 2SLS method requires an instrumentvariable which is a joint of the 1113954ξ matrix and the W1113954ξ matrixor written as H (1113954ξ |W1113954ξ) e instrument variable H isvalid because it does not correlate with ε but correlates withregressor W1113954η e result of the estimation is

1113954α 1113954Zprime 1113954Z1113872 1113873minus 1 1113954Zprime1113954η (10)

where 1113954Z H(HprimeH)minus 1HprimeZ and 1113954α contains 1113954β and 1113954λ

e parameter significance test using the maximumlikelihood ratio test (MLRT) method and based on the errordistribution as in (8) is

c Λminus 2nminus 11113872 1113873(n minus 1) (11)

where Λminus2n 1 + (n(n + 1))(1113954ξ1113954β)prime(ΛyprimeΘminus1

δ Λy)IT(1113954ξ1113954β)prime ematerials and methods section should contain sufficientdetail so that all procedures can be repeated It may bedivided into headed subsections if several methods aredescribed

22 Indicator Latent Variable Model and Unit Sampleis study models the quality of education at senior highschools in Sumenep Regency Indonesia is modelinginvolves two exogenous latent variables and one endogenouslatent variable e exogenous latent variables are schoolinfrastructure and the socioeconomic condition whereas theendogenous latent variable is education quality

Indicators for the variables of education quality wereconstructed based on the study by Kemendikbud [26]Regulation in the study by Kemendiknas [27] was used todetermine the indicators for the variable of school infra-structure e indicators for each variable are shown inTable 1 Indicators are secondary data in Sumenep Fig-ure 2018 [16] and data from education office SumenepRegency 2018

e conceptual model was constructed of three latentvariables namely the education quality influenced by thesocioeconomic conditions of the community [26] and byinfrastructure facilities [28] e theoretical hypotheticalmodel was based on the conceptual model as shown inFigure 2

e observation units in this study are districts inSumenep Regency ere are 27 districts consisting of 18districts on the mainland and 9 districts on islands ereare 226 senior high schools consisting of 12 public schools1 public madrasa (Islamic school) school 72 privatenonmadrasa schools and 141 private madrasa schoolsTable 2 is a number and kind of senior high school in eachdistrict

Indonesia

Java island

Sumenep regency

Masalembo

Sapeken

KangayanArjasa

GiligentingGayam Raas

Nonggunung

Talango

Figure 1 A map of the districts in Sumenep Regency


23 9e Parameter Estimation Step and Significance Test

231 Evaluating the Measurement Model Evaluation of themeasurement model is conducted to ensure that indicatorsare constructing their latent variables properly is workused smart-PLS

232 Estimating the Latent Variable Factor scores wereobtained based on equations (5) and (6) and calculated usinga MATLAB program

233 Setting the Spatial Weights e spatial weighting usedthe Queen contiguity method In Queen contiguity Wij isdefined as 1 for an entity where the common side or thecommon vertex meets the region of concern and Wij isdefined as 0 for other regions [29]

234 Modeling the SAR-LVs e model was constructedbased on equation (7) with one endogenous variable twoexogenous variables and 27 observation units the SAR-LVsmodel can be rewritten as11139541113957ηi 1113954b0 + 1113954λ1113936

27j1inejWij1113954ηj + 1113954b1

1113954ξ1 + 1113954b21113954ξ2

235 Testing the Spatial Dependency e value of the teststatistic LM SAR-LVs is as given in equation (9) and wascalculated using a MATLAB program e LM statistic LMλfollows the asymptotic distribution of X2

(1)

236 Estimating the Parameters e parameters of theSAR-LVs model were obtained based on equation (10) andcalculated using a MATLAB program

237 Testing the Statistical Significance e value of theparameter significance test is as given in equation (11) andwas calculated using a MATLAB program e statistic of cfollows the asymptotic distribution of F1(nminus1) If cgt F1(nminus1)then it has a spatial effect

24 9eoretical Hypothesis A hypothetical model wasproposed as shown in Figure 2 that shows the path hy-potheses in which socioeconomic conditions and schoolinfrastructure directly affect the quality of education espatial effect is to be tested since observation units are lo-cations Furthermore the following hypotheses wereproposed

Hypothesis H1 there is a spatial effectHypothesis H2 school infrastructure has direct effectson the quality of educationHypothesis H3 socioeconomic conditions have directeffects on the quality of education

3 Results and Discussion

31 Results e review of the measurement model wasbased on convergent validity composite reliability and thesignificance test Convergent validity was measured by theloading factors that are the coefficients of correlation be-tween the indicators and their latent constructs Indicatorsthat have high loading factors contribute to explaining latentconstructs more Table 3 shows the loading factor value ofeach indicator

Table 1 Latent variables and their indicators

Latent variable Indicator Brief description

Education quality

Y11e ratio of the gross enrolled number of senior high school students to the number of children aged

between 15 and 18 years in each district

Y12e ratio of the number of accredited senior high schools with at least B level to the total number of

senior high schools in each districtY13 e average of national exam scores of senior high school students in each district

School infrastructure

X11e proportion of the number of schools having minimum classroom space according to the

regulations of the national education ministry

X12e proportion of the number of schools having laboratories according to the regulations of the

national education ministry

X13e proportion of the number of schools having libraries according to the regulations of the national

education ministry

Socioeconomicconditions

X21e ratio of the number of households running a home industry or having a shop at home to the total

number of households in each district

X22e ratio of the number of households using clean water to the total number of households in each

district

X11 X12 X13

Y11

Y12

Y13

X22X21

Quality ofeducation

Schoolinfrastructure

Socioeconomiccondition

Figure 2 eoretical hypothetical model of quality of education atsenior high schools


Some references stated that a factor loading of 050 wasconsidered to have sufficient validation to explain the latentvariable [30] Sharma [31] mentioned that researchers hadused cutoff values of factor loading as low as 040 All in-dicators in Table 3 are more than 040 so it is said to besufficient to explain all the latent variables of educationquality school infrastructure and socioeconomic condition

Reliability testing was carried out to prove the accuracyand consistency of instruments in measuring latent vari-ables In this study the reliability was testedassessed usingcomposite reliability Generally the composite reliabilityvalue of greater than 06 is acceptable [32] Table 4 sum-marizes the results of the composite reliability test of threelatent variables

As shown in Table 4 all the composite reliability valuesof each latent variable are more than 06 It indicates that alllatent variables are reliable or indicators are consistent inmeasuring each of the latent variables

In the SEM PLS method the significance of the modelcannot be tested because the data distribution is unknownerefore the bootstrapping resamplingmethod was used to

conduct the significance test Table 5 shows a summary ofthe significance test results for the measurement model withα 10

As can be seen from the previous table the only invalidindicator is X21 representing the ratio of the number ofhouseholds running a home industry or having a shop athome to the total number of households However ingeneral these indicators can still be used to construct latentvariables

e result of the identification of the spatial effect on themodel of the education led to the SAR-LVs model at thesignificance level α 5 e results of the parameters es-timation and significance test are summarized in Table 6

In general the SAR-LVs model for the education qualityof the senior high school is11139541113957ηi 96604 minus 00021113936

27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2

where 1113954ηi is the quality of education in ith district 1113954ξ1 is theinfrastructure and 1113954ξ2 is the socioeconomic condition

Models of the education quality of the senior highschools in the SAR-LVs model for several districts are asfollows

Table 2 Number and kind of school in each districtCode Districts Number of school Explanation010 Pragaan 26 4 private nonmadrasa and 22 private madrasa020 Bluto 14 1 public 1 private nonmadrasa and 12 private madrasa030 Saronggi 3 3 private madrasa040 Giligenteng 6 1 private nonmadrasa and 5 private madrasa050 Talango 1 1 private nonmadrasa060 Kalianget 1 1 public070 Kotasumenep 14 2 publics 7 private nonmadrasa 1 public madrasa and 4 private madrasa071 Batuan 2 1 public and 1 private nonmadrasa080 Lenteng 19 1 public 4 private nonmadrasa and 14 private madrasa090 Ganding 18 6 private nonmadrasa and 12 private madrasa100 Guluk Guluk 21 8 private nonmadrasa and 13 private madrasa110 Pasongsongan 6 1 private nonmadrasa and 5 private madrasa120 Ambunten 6 1 public 2 private nonmadrasa and 3 private madrasa130 Rubaru 9 2 private nonmadrasa and 7 private madrasa140 Dasuk 5 1 private nonmadrasa and 4 private madrasa150 Manding 4 2 private nonmadrasa and 2 private madrasa160 Batuputih 2 1 private nonmadrasa and 1 private madrasa170 Gapura 8 1 public 2 private nonmadrasa and 5 private madrasa180 Batang Batang 8 4 private nonmadrasa and 4 private madrasa190 Dungkek 5 3 private nonmadrasa and 2 private madrasa200 Nonggunong 2 2 private nonmadrasa210 Gayam 3 1 public 1 private nonmadrasa and 1 private madrasa220 Raas 5 3 private nonmadrasa and 2 private madrasa230 Sapeken 18 1 public 5 private nonmadrasa and 12 private madrasa240 Arjasa 6 1 public 4 private nonmadrasa and 1 private madrasa241 Kangayan 8 4 private nonmadrasa and 4 private madrasa250 Masalembu 6 1 public 2 private nonmadrasa and 3 private madrasa

Table 3 e value of the loading factor

Quality of education School infrastructure Socioeconomic conditionIndicator Loading factor Indicator Loading factor Indicator Loading factorY11 0725 X11 0890 X21 0444Y12 0844 X12 0702 X22 0910Y13 0574 X13 0685


(1) Masalembo with no one neighboring district11139541113957η(250) 96604 + 231211113954ξ1 + 012861113954ξ2

(2) Nonggunong with one neighboring district11139541113957η(200) 96604 minus 00021113954η(210) + 231211113954ξ1 + 012861113954ξ2

(3) Saronggi with five neighboring districts 11139541113957η(030)

96604 minus 000041113954η(020) minus 000041113954η(080) minus 0000041113954η(060)

minus0000041113954η(070) minus0000041113954η(071) +231211113954ξ1+ 012861113954ξ2(4) Sumenep City with six neighboring districts

11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)

32 Discussion ere are three items to be discussednamely the spatial effect and the influence of the schoolinfrastructure and socioeconomic conditions on the qualityof education e autoregressive spatial coefficient signifi-cantly effects at the 5 significance level is means thatthere is a correlation between the education quality of thesenior high schools in one district and the one in othercontiguous districts However the autoregressive spatialcoefficient is negative It indicates the opposite of thecommon spillover effect ie the districts with a highereducation quality are supported by or gains a spillover effectfrom the neighboring districts with lower education qualitye opposite of the spillover effect can be illustrated as thephenomenon ldquothere is sugar so there are antsrdquo ie there aremigrations of good students to districts with a highernumber of high-quality schools

e following social phenomenon can explain this sit-uation ere is a tendency among Indonesian people tosend their children to their favorite schools to ensure a goodquality education even though the schools are located farfrom their home In urban areas there are many favoriteschoolsmdashnot only public schools but even many privateschools that are preferred by the community because of theirgood quality In rural areas particularly Sumenep Regencythe favorite schools are mostly public schools Districts thathave many favorite schools tend to have many applicants orprospective students and they select the best students in-cluding the ones from neighboring districts Consequentlyschools of good quality tend to be in demand by thecommunity even in other districts especially the directlyneighboring districts On the contrary schools in the directlyneighboring districts lose good students from their owndistrict or gain students of poorer quality from other districtssince the quality of new students is one of the determinants

of education quality As a result of this phenomenon dis-tricts are classified into two categories namely receiving andlosing districts e receiving districts are districts whosenumber of new students of senior high school is more thanthe number of junior high school graduates in the same yearOn the opposite the losing districts are the ones that losejunior high school graduates Figure 3 shows both districtcategories including the directions of the student migrationis figure is created using the number of new students ofsenior high school and the number of junior high schoolgraduates in Sumenep Regency in 2018 according to theCentral Bureau of Statistics (BPS) [16]

Examples for receiving and losing districts are SumenepCity (Kotasumenep) and Saronggi respectively (Figure 3)Based on the BPS data [16] the number of junior high schoolgraduates in Sumenep City was 1582 and the number ofnew students of senior high school was 2041 It means thatSumenep City received students for the high school levelfrom its neighboring districts Referring to (12) there is anegative spillover in terms of good students input toSumenep City from its neighboring districts namely Sar-onggi Rubaru Kalianget Manding Batuan and GapuraSumenep City has three public senior high schools that arestrongly favored by the community or by students while itsthree neighboring districts Saronggi Rubaru and Mandingdo not have any public schools e other neighboringdistricts Kalianget Batuan and Gapura have only onepublic school is condition stimulates the migration ofstudents who have graduated from junior high school in theneighboring districts to register for senior high schools inSumenep City In other words the high schools in SumenepCity always have the privilege of choosing the best studentsIn general the favorite public schools in Sumenep regencyare dominated by people of a good economic level who couldsupport improvements in the quality of schools

As an example of the losing districts Saronggi has nopublic school and has five neighboring districts Based ondata [16] Saronggi had 305 junior high school graduates and71 new senior high school students in 2018 It lost seniorhigh school students due to migration to the neighboringdistricts all of which have favorite public schools (Figure 3)e remaining 71 new senior high school students registeredfor three private madrasa schools in Saronggi

e studentsrsquo migration in Sumenep City and Saronggidistricts show that students compete to get registered forfavorite or good quality schools It also reveals the inequalityof the school quality since not all students can have access toschools with a good quality e urbanization correlatedwith gender equality in education because the urban pop-ulation is generally also more receptive to modern views onfemale education for 57 developing countries in 1970ndash2010[33] To eliminate school favoritism the Indonesian gov-ernment issued a zoning system policy through the Ministerof Education and Culture Regulation No 17 of 2017 No 14of 2018 and No 51 of 2018eMinistry argues that studentadmission via the zoning system will improve access toeducation services in public schools for all and removefavorite schools as a basis for school selection However thezoning system policy is only for public schools

Table 4 Composite reliability

Latent variable Composite reliabilityQuality of education 0762School infrastructure 0806Socioeconomic condition 0653


Sumenep Regency implemented a zoning system policystarting in the middle of 2018 Sumenep has only 13 publicsenior high schools out of a total of 226 senior high schoolswhich are spread across 11 districts ese 13 public seniorhigh schools are not enough to accommodate all senior highschool students in Sumenep As a result many privateschools have emerged as community initiations usprivate schools play an important role in bridging the gapand providing access to education in rural areas Howeverprivate schools are generally inferior in terms of educationquality and facilities [34] As many as 677 of privateschools in Sumenep Regency has a C accreditation or noteven accredited e standardization of private and publicschools is very important in order to obtain uniform qualityamong schools especially when this zoning system is ap-plied If a public school is unable to accommodate students

in a district then a private school with almost the samequality will be able to accommodate them

e result of this modeling describes the influence of theschool infrastructure variable as follows each level of schoolinfrastructure improvement is followed by an increase of23121 in the quality of education in a district is meansthat indicators of the school infrastructure variable based onregulations by the Ministry of National Education arepositively correlated to indicators of the education qualityvariable based on the strategic planning of education for2015ndash2019 is finding is supported by the research by Bediand Garg [35] where the effectiveness of private versuspublic schools in Indonesia based on labor market earningswas studied Schools were classified into four different typesnamely public private nonreligious private Islamic andprivate Christian schools One of the results revealed that

RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students

The number of public schools

The possibility direction ofstudents migrations

Figure 3 e receiving and losing districts and the directions of the studentsrsquo migration

Table 5 e significance test results

Quality of education School infrastructure Socioeconomic conditionIndicator T value Interpretation Indicator T value Interpretation Indicator T value InterpretationY11 2849 Valid X11 9427 Valid X21 1322 NonvalidY12 4061 Valid X12 4183 Valid X22 3616 ValidY13 2045 Valid X13 1904 Valid

Table 6 e estimation results of the parameters

Variable Coefficient Significance test HypothesisSpatial autoregressive coefficient (λ) minus0002 Significant H1School infrastructure (b1) 23121 Significant H2Socioeconomic condition (b2) 01286 Not significant H3Constant (b0) 96604 Significant mdash


students of schools with poor infrastructure earned between25 and 57 (for public and private nonreligious schools)less than those of schools with adequate infrastructure eschool infrastructure in Sumenep Regency directly influ-ences the quality of education It becomes an important partto be prepared carefully and continuously by the localgovernment so that the ESD can be achieved

Socioeconomic condition was found to not have a sig-nificant impact on the quality of education Azziza [8]revealed that socioeconomic inequality caused educationaldisparities in eastern and western Indonesia but the researchdid not explain the actual situation in the regions InSumenep Regency the number of private schools hasreached 9425 of the total (213 of 226 schools) of allschools there 531 is public 044 is public madrasaschools 3186 is private nonmadrasa schools and 6239is private madrasa schools Private schools as a represen-tation of the communityrsquos independent participation ineducational services reflect the communityrsquos socioeconomiccondition Meanwhile based on statistical data [36] eco-nomic growth and the gross regional domestic product(GRDP) per capita in Sumenep are below the averageeconomic growth and GRDP of regencies in East Java Sari[34] argued that in general private schools are inferior interms of quality and facilities A large number of privateschools in the Sumenep regency do not reflect yet the so-cioeconomic condition of the local community ey stillrely on the SOB fund Nurkholis [37] suggested that thequality of madrasa schools was still low when compared withother schools for the same education level In Sumenep themajority of senior high schools are private madrasa schoolsese schools were built by the community to poor qualitystandards since religious idealism was prioritizedereforein this model the socioeconomic variable has no effect to thequality of education

4 Conclusions

emain conclusions of this study are as follows (1) there isa negative spillover effect due to the migration of goodstudents from districts to the others with higher educationquality (2) the school infrastructure directly influences thequality of education and (3) socioeconomic conditions donot have a significant impact on the quality of educationis study recommends that it is necessary to standardizethe quality of all schools especially private schools inSumenep Regency so that the zoning system can functionwell In the standardization process the role of the centralgovernment is crucial e central government shouldprovide guidance or assistance to the local governments inachieving the desired outcomes when implementing thezoning system such as by accelerating the provision ofteachers facilities and infrastructure Furthermore theeducation quality of rural private schools needs to be im-proved since this also means providing better education forchildren from low-income families to achieve equality ofaccess and quality of education for all children In that wayESD can be carried out properly Research for specific re-gions (eg in this case Sumenep Regency) should be

conducted to obtain more specific information about theproblems that are present and certainly distinct from thosein other regions e study of specific cases can be used as abenchmark for making policy Each policy thus madeprovides solutions that are relevant to regional problems andcapabilities

Data Availability

e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

e authors declare that there are no conflicts of interestregarding the publication of this paper

References

[1] B P S Kewarganegaraan Suku Bangsa Agama Dan BahasaSehari-Hari Penduduk Indonesia Hasil Sensus Penduduk2010 Badan Pusat Statistik Jakarta Indonesia 2011

[2] S Sumarto A Suryahadi and W Widyanti ldquoDesigns andimplementation of Indonesian social safety net programsrdquo9e Developing Economies vol 40 no 1 pp 3ndash31 2002

[3] R Sparrow ldquoProtecting education for the poor in times ofcrisis an evaluation of a scholarship programme in Indo-nesiardquo Oxford Bulletin of Economics and Statistics vol 69no 1 pp 99ndash122 2007

[4] B Kharisma ldquoDampak program bantuan operasional sekolah(BOS) terhadap tingkat putus sekolah di Indonesia analisisDIDrdquo Jurnal Ekonomi Kuantitatif Terapan vol 6 p 10 2013

[5] UNESCO Global Education Monitoring Report Summary2019 Migration Displacement and EducationndashBuildingBridges NotWalls United Nations Educational Scientific andCultural Organization Paris France 2018

[6] T Muttaqin ldquoDeterminants of unequal access to and qualityof education in Indonesiardquo Jurnal Perencanaan Pemban-gunan 9e Indonesian Journal of Development Planningvol 2 no 1 2018

[7] BPS BPS Welfare Statistics 2019 BPS-Statistics IndonesiaJakarta Indonesia 2019

[8] Y Azzizah ldquoSocio-economic factors on Indonesia educationdisparityrdquo International Education Studies vol 8 no 12p 218 2015

[9] KEMDIKBUD UNICEF SDG4 Baseline Report for IndonesiaIndonesia Ministry of Education and Culture and the UnitedNations Childrenrsquos Fund (UNICEF) Jakarta Indonesia 2017

[10] UNESCO Global Education Monitoring Report Summary2016 Education for People and Planet Creating SustainableFutures for All United Nations Educational Scientific andCultural Organization Paris France 2016

[11] A Yue B Tang Y Shi et al ldquoRural education across Chinarsquos40 years of reform past successes and future challengesrdquoChina Agricultural Economic Review vol 10 no 1 pp 93ndash118 2018

[12] S Carrascal M Magro J Anguita and M Espada ldquoAcqui-sition of competences for sustainable development throughvisual thinking A study in rural schools in Mixco Guate-malardquo Sustainability vol 11 no 8 p 2317 2019

[13] A-M Kuusimaki L Uusitalo-Malmivaara and K TirrildquoParentsrsquo and teachersrsquo views on digital communication in


Finlandrdquo Education Research International vol 2019 ArticleID 8236786 7 pages 2019

[14] C-H Cheng Y-C Wang and W-X Liu ldquoExploring therelated factors in studentsrsquo academic achievement for thesustainable education of rural areasrdquo Sustainability vol 11no 21 p 5974 2019

[15] P P M D Dirjen Keputusan Direktur Jenderal Pemban-gunan Dan Pemberdayaan Masyarakat Desa Nomor 201Tahun 2019 Tentang Perubahan Kedua Atas KeputusanDirjen Pembangunan Dan Pemberdayaan Masyarakat DesaNomor 30 Tahun 2016 Tentang Status Kemajuan DanKemandirian Desa 2019

[16] BPS Sumenep in Figures 2018 BPS-Statistics of SumenepRegency Sumenep 2018

[17] V Gille ldquoEducation spillovers empirical evidence in ruralIndiardquo Indian Growth and Development Review vol 5 no 1pp 4ndash24 2012

[18] Y Gao Q He Y Liu L Zhang H Wang and E CaildquoImbalance in spatial accessibility to primary and secondaryschools in China guidance for education sustainabilityrdquoSustainability vol 8 no 12 p 1236 2016

[19] Y Xu W Song and C Liu ldquoSocial-spatial accessibility tourban educational resources under the school district systema case study of public primary schools in Nanjing ChinardquoSustainability vol 10 no 7 p 2305 2018

[20] K A Bollen Structural Equations with Latent Variables JohnWiley amp Sons New York NY USA 1989

[21] L Anselin Spatial Econometrics Methods and Models Vol 4Kluwer Academic Publisher Dordrecht Netherlands 1988

[22] L Anselin ldquoLagrange multiplier test diagnostics for spatialdependence and spatial heterogeneityrdquo Geographical Analy-sis vol 20 no 1 pp 1ndash17 1988

[23] T S Breusch and A R Pagan ldquoe Lagrange multiplier testand its applications to model specification in econometricsrdquo9e Review of Economic Studies vol 47 no 1 pp 239ndash2531980

[24] H H Kelejian and I R Prucha ldquoA generalized spatial two-stage least squares procedure for estimating a spatial autor-egressive model with autoregressive disturbancesrdquo 9eJournal of Real Estate Finance and Economics vol 17 no 1pp 99ndash121 1998

[25] H H Kelejian and I R Prucha ldquoA generalized momentsestimator for the autoregressive parameter in a spatial modelrdquoInternational Economic Review vol 40 no 2 pp 509ndash5331999

[26] Kemendikbud Rencana Strategis Kementerian PendidikanDan Kebudayaan Tahun 2015mdash2019 2015

[27] Kemendiknas Peraturan Menteri Pendidikan NasionalNomor 24 Tahun 2007 2007

[28] P Barrett A Treves T Shmis D Ambasz and M Ustinova9e Impact of School Infrastructure on Learning A Synthesis ofthe Evidence e World Bank Washington DC USA 2019

[29] J P LeSage 9e 9eory and Practice of Spatial EconometricsDepartment of Economics University of Toledo Toledo OHUSA 1999

[30] J F Hair W C Black and B J Babin Multivariate DataAnalysis Prentice-Hall Upper Saddle River NJ USA 7thedition 2009

[31] S Sharma Applied Multivariate Techniques John Wiley ampSon Inc New York NY USA 1st edition 1996

[32] K S Sujit and B K Rajesh ldquoDeterminants of discretionaryinvestmentsrdquo SAGE Open vol 6 no 1 2016

[33] G Oslashstby H Urdal and I Rudolfsen ldquoWhat is driving genderequality in secondary education Evidence from 57

developing countries 1970ndash2010rdquo Education Research In-ternational vol 2016 Article ID 4587194 18 pages 2016

[34] V A Sari ldquoEducational assistance and education quality inIndonesia the role of decentralizationrdquo Population and De-velopment Review vol 45 no S1 pp 123ndash154 2019

[35] A S Bedi and A Garg ldquoe effectiveness of private versuspublic schools the case of Indonesiardquo Journal of DevelopmentEconomics vol 61 no 2 pp 463ndash494 2000

[36] BPS Produk Domestik Regional Bruto KabupatenKota DiProvinsi Jawa Timur Menurut Lapangan Usaha 2014ndash2018BPS Provinsi Jawa Timur Surabaya Indonesia 2019

[37] N Nurkolis ldquoEducational improvement towards effectivemadrasahrdquo in Proceedings of the 1st Yogyakarta InternationalConference on Educational ManagementAdministration andPedagogy (YICEMAP 2017) Atlantis Press YogyakartaIndonesia 2017


the quality of the educational service UCT program wasintended to anticipate the negative effects of the govern-mentrsquos plan to reduce a fuel subsidy

Kharisma [4] found that the provision of SOB had apositive effect on dropout rates for students aged between 16and 20 years but not for those aged between 7 and 15 yearsFurthermore the Indonesian government has constantlycommitted to not leaving any child behind in developmentas the core of the agenda 2030 on education for sustainabledevelopment (ESD) Since 2014 the Indonesian governmenthas provided smart cards to more than 24 million poorstudents and others who cannot attend school because offinancial issues Based on the report [5] Indonesia is one ofthe low-income or middle-income countries whose share ofthe state budget for education is likely to be high eamendment of the education budget was simultaneous withthe start of a decentralized government system

Since 1998 Indonesia has given greater autonomy tolocal governments due to the commencement of a refor-mation movement across Indonesia Education manage-ment was also delegated more to the local governments econstitution states that local governments are obliged toallocate a minimum of 20 of their budgets to the educationsector Furthermore local governments have a crucial role inproviding access to education

Muttaqin [6] found that local governments with moreresources can provide more scholarships to attract pupilsfrom poor families to return to school but this is not the casefor local governments with fewer resources such as localgovernments with greater rural areas At least there is adisparity in education quality between rural and urban areasand between western and eastern Indonesia Based onwelfare statistics [7] the percentage of the population whoare aged above 15 years that literate is 9749 in urban areasand 9313 in rural areas e school enrollment ratio of thepopulation aged between 16 and 18 years is 7521 in urbanareas and 6771 in rural areas e net enrollment ratio ofthe population for senior high school is 6433 in urbanareas and 5606 in rural areas Azzizah [8] point out theapparent gap of education among the eastern and westernprovinces in Indonesia using the dummy regression analysisand the most influencing factor was the poverty rate

e report [9] showed that the Indonesian children andyouth who do not have sufficient access to education are inremote and isolated areas and in poor communities isreport was based on data of the National SocioeconomicSurvey 2015 which stated that the composition of the groupof out-of-school children (OOSC) of elementary school agewas 81 from rural areas (based on location) 49 frompoor families (based on wealth) and 40 from Papua (basedon region)

e fourth sustainable development goal (SDG) oneducation is to ensure inclusive and equitable education andto promote lifelong learning opportunities for all One of thetargets is ldquoby 2030 to eliminate gender disparities in edu-cation and ensure equal access to all levels of education andvocational training for the vulnerable including personswith disabilities indigenous peoples and children in vul-nerable situationsrdquo One of the disparity indices is rural-

urban As reported in [10] youths aged between 20 and 24years living in rural areas in 101 low- and middle-incomecountries have on average an education period 26 yearsshorter than do those in urban areas Indonesia is one ofthese countries In Indonesia rural-urban disparities arewidely prevalent across indicators in each target of the fourthSDG since poverty is still concentrated in rural areas [9]e report [5] shows that children born in selected ruralareas in Indonesia whomigrated to an urban area as childrenwere found to have attained three more years of educationthan those who did not migrate is shows that migrationfrom rural to urban areas can increase educational attain-ment due to the low access to education in rural areas

erefore some educational research studies have fo-cused on rural areas to provide more specific reviews Yueet al [11] researched policy and trends in rural educationfrom preschool to high school in China over 40 years from1978 to 2018 Rural schools continually improved in allaspects In 1980 only a small proportion of rural childrenattended preschool but in 2014 more than 90 of ruralchildren attended preschool Carrascal et al [12] focusedtheir research on rural primary schools in Mixco Guate-mala ey developed the ECO Kit as an effective teach-ingndashlearning tool that contributed to improving the visualliteracy and creativity of students In Finland digitalcommunication is being more understandable by ruralteachers than in the urban area and promoted parent-teacher partnership in rural better than did parents in theurban area [13] Cheng et al [14] documented the learningproblems of disadvantaged students in rural areas in Taiwane results showed that in rural elementary schools theeducation levels of most of the parents were below thecollege level students had low motivation and the ordinarystudents and disadvantaged students indeed showed dif-ferences Azzizah [8] found that poverty had a huge influ-ence on the rate of school enrollment particularly inIndonesiarsquos eastern province based on Joseph Nkurunzizarsquosanalysis (hindawi 2012) that students from very poor peoplehave lower odds of attendance in school Based on [9]poverty in Indonesia is still concentrated in rural areas

East Java is one of the provinces in western Indonesiaand it is relatively more advanced than others In East Javaprovince the OOSC rate in the primary school level is only7 while in junior secondary it is 12 and in seniorsecondary it is 14 ese numbers are relatively lowcompared to those in other provinces However one of theregencies in East Java has 100 underdeveloped regionsnamely Sumenep Regency [15] e percentage of poorpeople in Sumenep Regency is 1962 which is the fourthhighest in East Java and the human development index is6428 the fourth lowest in East Java [16]

e principle of ESD is that no child is left behind andemphasizes equity of education by focusing on eliminatingdisparities among regions and socioeconomic groupserefore in this study the factors related to educationquality in Sumenep Regency were explored through thespatial autoregressive model with latent variables (SAR-LVs) Sumenep is a regency with rural characteristics It isnecessary to study the modeling of education quality









x Λx ξ + δ (1)









minus 1ΛxprimeΘ

minus1δ1113872 1113873X (5)


minus 1ΛyprimeΘ

minus1δlowastY (6)




1113954η λW1113954η + 1113954ξβ + ε (7)







2

p D (9)
















Indonesia

Java island

Sumenep regency

Masalembo

Sapeken

KangayanArjasa


Nonggunung

Talango








27j1inejWij1113954ηj + 1113954b1

1113954ξ1 + 1113954b21113954ξ2


(1)









Education quality











education ministry





district

X11 X12 X13

Y11

Y12

Y13

X22X21

Quality ofeducation













27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2












11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References

















































x Λx ξ + δ (1)









minus 1ΛxprimeΘ

minus1δ1113872 1113873X (5)


minus 1ΛyprimeΘ

minus1δlowastY (6)




1113954η λW1113954η + 1113954ξβ + ε (7)







2

p D (9)
















Indonesia

Java island

Sumenep regency

Masalembo

Sapeken

KangayanArjasa


Nonggunung

Talango








27j1inejWij1113954ηj + 1113954b1

1113954ξ1 + 1113954b21113954ξ2


(1)









Education quality











education ministry





district

X11 X12 X13

Y11

Y12

Y13

X22X21

Quality ofeducation













27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2












11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References










































1113954η λW1113954η + 1113954ξβ + ε (7)







2

p D (9)
















Indonesia

Java island

Sumenep regency

Masalembo

Sapeken

KangayanArjasa


Nonggunung

Talango








27j1inejWij1113954ηj + 1113954b1

1113954ξ1 + 1113954b21113954ξ2


(1)









Education quality











education ministry





district

X11 X12 X13

Y11

Y12

Y13

X22X21

Quality ofeducation













27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2












11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References















































27j1inejWij1113954ηj + 1113954b1

1113954ξ1 + 1113954b21113954ξ2


(1)









Education quality











education ministry





district

X11 X12 X13

Y11

Y12

Y13

X22X21

Quality ofeducation













27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2












11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References


















































27j1inejWij1113954ηj + 231211113954ξ1 + 012861113954ξ2












11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References















































11139541113957η(070) 96604 minus 000031113954η(030)

minus 000031113954η(130) minus 000031113954η(060)

minus 000031113954η(150) minus 000031113954η(071)

+ 231211113954ξ1 + 012861113954ξ2

(12)













RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References













































RUBARU MANDING

BATUPUTIH

BATANG BATANG

DUNGKEK

GANDING LENTENG

BLUTO SARONGGI

BATUAN

KALIANGET

KOTASUMENEPGAPURA

0

0

0

0

0

0

0

1

1

3

1

1

1

Receiving students

Losing students











4 Conclusions



Data Availability




References












































4 Conclusions



Data Availability




References










































exploringtherelatedfactorsineducationqualitythrough...

Documents