ANÁLISIS DE LA RETENCIÓN ESCOLAR A TRAVÉS DE MODELOS MULTINIVEL: UN ESTUDIO EN PORTUGAL

Este artículo investiga la retención escolar en Portugal utilizando microdatos. Con la aplicación de modelos multinivel, se analiza el número de retenciones en la educación obligatoria -variable explicada-, considerando como posibles variables explicativas las características individuales del estudiante, su contexto familiar en términos sociodemográficos y económicos y sus condiciones de vida. También se investiga el impacto de la propia escuela por sí misma en la variable endógena. Los resultados obtenidos confirman la importancia de las características individuales y familiares y de las condiciones de vida en la retención escolar. El análisis empírico muestra que la probabilidad de fracaso de un estudiante varía de una escuela a otra, lo que demuestra la importancia del impacto de la escuela en la retención escolar.

REPETICIÓN ESCOLAR; PARTICIPACIÓN SOCIAL; MODELO MULTINIVEL

Autoría

Amélia Bastos

Universidade de Lisboa (ULisboa); Instituto Superior de Economia e Gestão (ISEG); Centro de Matemática Aplicada à Previsão e Decisão Económica (CEMAPRE), Lisboa, Portugal; abastos@iseg.ulisboa.ptUniversidade de LisboaPortugalLisboa, Portugal Universidade de Lisboa (ULisboa); Instituto Superior de Economia e Gestão (ISEG); Centro de Matemática Aplicada à Previsão e Decisão Económica (CEMAPRE), Lisboa, Portugal; abastos@iseg.ulisboa.pt

https://orcid.org/0000-0003-0211-7299

Maria Eugénia Ferrão

Universidade da Beira Interior, Covilhã; Centro de Matemática Aplicada à Previsão e Decisão Económica (CEMAPRE), Lisboa, Portugal; meferrao@ubi.ptUniversidade da Beira InteriorPortugalCovilhã, Portugal Universidade da Beira Interior, Covilhã; Centro de Matemática Aplicada à Previsão e Decisão Económica (CEMAPRE), Lisboa, Portugal; meferrao@ubi.pt

Centro de Matemática Aplicada à Previsão e Decisão Económica, Lisboa, PortugalCentro de Matemática Aplicada à Previsão e Decisão EconómicaPortugalLisboa, PortugalCentro de Matemática Aplicada à Previsão e Decisão Económica, Lisboa, Portugal

https://orcid.org/0000-0002-1317-0629

NOTA:

Amélia Bastos has contributed to the revised versions of the manuscript, and approved the final version. She has made substantial contributions to conception and design of the manuscript, has drafted the first version of the literature review and revised the methodology section. In addition, she has contributed to the conclusion. Maria Eugénia Ferrão has contributed to the revised versions of the manuscript, and approved the final version. She has made substantial contributions to conception and design of the manuscript, has drafted the first version of the methods and empirical findings sections, and revised the literature review section. In addition, she has contributed to the conclusion.

SCIMAGO INSTITUTIONS RANKINGS

Figuras | Tablas

Figuras (2)
Tablas (3)

Thumbnail

FIGURE 1
MARGINAL EFFECT OF SOCIAL PARTICIPATION FOR STUDENTS ATTENDING THE 9^TH GRADE

Thumbnail

FIGURE 2
MARGINAL EFFECT OF SOCIAL PARTICIPATION VARYING ACROSS SCHOOLS

Thumbnail

TABLE 1
VARIABLE DEFINITIONS AND SUMMARY STATISTICS

Thumbnail

TABLE 2
ESTIMATES OF THE THREE LEVEL POISSON MODEL FOR NUMBER OF GRADE REPETITIONS

Thumbnail

TABLE 3
ESTIMATES OF THE THREE LEVEL POISSON MODEL FOR NUMBER OF GRADE REPETITIONS. RANDOM COEFFICIENT OF SOCIAL PARTICIPATION

FIGURE 1 MARGINAL EFFECT OF SOCIAL PARTICIPATION FOR STUDENTS ATTENDING THE 9^TH GRADE

Source: Authors’ elaboration.

FIGURE 2 MARGINAL EFFECT OF SOCIAL PARTICIPATION VARYING ACROSS SCHOOLS

Source: Authors’ elaboration.

TABLE 1 VARIABLE DEFINITIONS AND SUMMARY STATISTICS

VARIABLE	MIN	MAX	MEAN	SD	PROPORTION	N MISSING
Male	---	---	---	---	0.46	7
Mother’s education ≤ ISCED 2	---	---	---	---	0.076	11
ISCED 2 < Mother’s education ≤ ISCED 3	---	---	---	---	0.623	11
ISCED 3 < Mother’s education ≤ ISCED 8	---	---	---	---	0.301	11
Father’s education ≤ ISCED 2	---	---	---	---	0.089	56
ISCED 2 < Father’s education ≤ ISCED 3	---	---	---	---	0.722	56
ISCED 3 < Father’s education ≤ ISCED 8	---	---	---	---	0.189	56
Poverty status	---	---	---	---	0.286	6
Housing index	-12.502	1.511	0.012	0.937	---	13
Health index	-5.941	1.016	0.000	1.000	---	27
Social participation index	-2.649	1.940	-0.013	1.000	---	7

Source: Own calculation (2019).

TABLE 2 ESTIMATES OF THE THREE LEVEL POISSON MODEL FOR NUMBER OF GRADE REPETITIONS

	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR
	ALL FIXED PARAMETERS			FIXED PARAMETERS WITH α=5%
Constant	-3.604	0.027	0.405	-3.076	0.046	0.353
Boy (1) vs. Girl (0)	0.364	1.439	0.174	0.299	1.349	0.149
Mother’s education ≤ ISCED 2 (1) vs Mother’s education > ISCED 2 (0)	0.662	1.939	0.214	0.942	2.565	0.175
Father’s education ≤ ISCED 2 (1) vs Father’s education > ISCED 2 (0)	0.447	1.564	0.230	---	---	---
Poverty status (poor=1)	0.415	1.514	0.183	0.404	1.498	0.166
Housing index	-0.084	0.919	0.052	---	---	---
Health index	-0.240	0.787	0.077	-0.261	0.77	0.081
Social participation index	-0.278	0.757	0.088	-0.192	0.825	0.068
Grade or School year	0.088	1.092	0.041	0.065	1.067	0.030
Random effect at level 2 (class)	0.228		0.139	0.194		0.117
Random effect at level 3 (school)	0.234		0.164	0.199		0.138
Number of cases	784			838

Source: Authors’ elaboration.

TABLE 3 ESTIMATES OF THE THREE LEVEL POISSON MODEL FOR NUMBER OF GRADE REPETITIONS. RANDOM COEFFICIENT OF SOCIAL PARTICIPATION

	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR
	ALL FIXED PARAMETERS			FIXED PARAMETERS WITH α=5%
Constant	-3.652	0.026	0.410	-3.162	0.042	0.357
Boy vs. Girl	0.362	1.436	0.174	0.301	1.351	0.149
Mother’s education ≤ ISCED 2 (1) vs Mother’s education > ISCED 2 (0)	0.686	1.986	0.214	0.945	2.573	0.176
Father’s education ≤ ISCED 2 (1) vs Father’s education > ISCED 2 (0)	0.445	1.576	0.230	---	---	---
	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR	PARAMETER ESTIMATE	EXPECTED COUNT	STANDARD ERROR
	ALL FIXED PARAMETERS			FIXED PARAMETERS WITH α=5%
Poverty status (poor=1)	0,409	1.505	0.183	0.404	1.498	0.166
Housing index	-0,084	0.919	0.052	---	---	---
Health index	-0,242	0.785	0.077	-0.340	0.712	0.104
Social participation index	-0,313	0.731	0.107	-0.193	0.824	0.069
Grade or School year	0,091	1.095	0.041	0.070	1.073	0.030
Random effect at level 2 (class)	0,208		0.135	0.182		0.115
Random effect at level 3 (school)	$[\begin{matrix} 0,256 & - 0,017 \\ - 0,017 & 0,036 \end{matrix}]$		$[\begin{matrix} 0,169 & 0,005 \\ 0,005 & 0,006 \end{matrix}]$	$[\begin{matrix} 0,266 & - 0,078 \\ - 0,078 & 0,046 \end{matrix}]$		$[\begin{matrix} 0,171 & 0,076 \\ 0,076 & 0,006 \end{matrix}]$
Number of cases		784				838

Source: Authors’ elaboration.

Como citar

copiar

Fundação Carlos Chagas Av. Prof. Francisco Morato, 1565, 05513-900 São Paulo SP Brasil, Tel.: +55 11 3723-3000 - São Paulo - SP - Brazil
E-mail: cadpesq@fcc.org.br

Acompanhe os números deste periódico no seu leitor de RSS

Versão para download de PDF

PDF

Inglês
Português

Versões e tradução automática

Versão original do texto

English
Português

Tradução automática

Google Translator
Microsoft Translator

Como citar

RIS

BIBTEX

Outros formatos de citação e exportação:

SciELO - Scientific Electronic Library Online
Rua Dr. Diogo de Faria, 1087 – 9º andar – Vila Clementino 04037-003 São Paulo/SP - Brasil
E-mail: scielo@scielo.org

Leia a Declaração de Acesso Aberto

Métricas

ANÁLISIS DE LA RETENCIÓN ESCOLAR A TRAVÉS DE MODELOS MULTINIVEL: UN ESTUDIO EN PORTUGAL

PlumX

Mensagem

[1] NOTA:

Amélia Bastos has contributed to the revised versions of the manuscript, and approved the final version. She has made substantial contributions to conception and design of the manuscript, has drafted the first version of the literature review and revised the methodology section. In addition, she has contributed to the conclusion. Maria Eugénia Ferrão has contributed to the revised versions of the manuscript, and approved the final version. She has made substantial contributions to conception and design of the manuscript, has drafted the first version of the methods and empirical findings sections, and revised the literature review section. In addition, she has contributed to the conclusion.