ArticlePDF Available

Do Longer School Days Have Enduring Educational, Occupational, or Income Effects? A Natural Experiment in Buenos Aires, Argentina

Authors:

Abstract

In 1971, longer school days were decreed for around half the public primary schools in the city of Buenos Aires, Argentina. Since participating schools were chosen roughly at random, an unusual opportunity for a natural experiment was created. In 2006 and 2007, we interviewed a sample of 380 alumni of the 1971 cohort, thirty years after their 1977 graduation in schools both with and without longer days. The main results are as follows. Students that attended double-shift (DS) primary schools had secondary school graduation rates 21 percent higher than those that attended single-shift primary schools. This result is mainly explained by what happened with the students with low socioeconomic status. Regarding tertiary and postgraduate educational levels, we have found both positive and negative impacts of DS. These last results, taken together with the absence of enduring effects of DS on income and employment and with the fact that DS students do not have a better knowledge of a second language, in spite of having had it as a subject in the school, suggest that the quality of the content and learning in DS schools was not good. These findings are very relevant when considering the extension of DS to other schools or to the whole educational system.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires,
Argentina
Juan J. Llach, Cecilia Adrogué and María Elina Gigaglia1
Abstract
In 1971, longer school days were decreed for around half the primary schools in the city of
Buenos Aires, Argentina. Since schools were chosen roughly at random, an unusual opportunity
for a natural experiment was created. In 2006 and 2007, we interviewed a sample of 380 alumni
of the 1971 cohort, thirty years after their 1977 graduation in schools both with and without longer
days. The main results are as follows. Students that attended double shift (DS) primary schools
had secondary school graduation rates 21% higher than those that attended single shift primary
schools. This result is mainly explained by what happened with the students with low
socioeconomic status. Regarding tertiary and postgraduate educational levels, we have found
both positive and negative impacts of DS. These last results, taken together with the absence of
enduring effects of DS on income and employment and with the fact that DS students do not
have a better knowledge of a second language of study in spite of having had it as a subject in
the school, suggest that the quality of the contents and of the learning in DS schools was not
good. This is something very relevant to take into account the time of considering the extension
of DS to other schools or to the whole educational system.
JEL: I2, J2
Resumen
En 1971, se decretó la extensión de la jornada escolar para aproximadamente la mitad de las
escuelas primarias de la ciudad de Buenos Aires, Argentina. Dado que las escuelas fueron
seleccionadas de una manera prácticamente aleatoria, se originó una oportunidad inusual para
un experimento natural. Entre 2006 y 2007, se entrevistó a una muestra de 380 alumnos de la
cohorte del año 1971, treinta años después de su graduación en 1977 tanto de escuelas de
jornada simple como de jornada completa. Los principales resultados son los siguientes: los
alumnos que asistieron a escuelas primarias de doble jornada (DJ) tuvieron una tasa de
graduación secundaria 21% más elevada que aquellos que asistieron a establecimientos de
modalidad simple. Esto se explica principalmente por lo ocurrido con los alumnos con bajo nivel
socioeconómico. Respecto al nivel terciario y a estudios post terciarios, encontramos impactos
negativos y positivos de la DJ. Estos últimos resultados, conjuntamente con la ausencia de
efectos perdurables de la DJ en los ingresos y en el empleo y con el hecho de que los alumnos
de DJ no tienen un mejor conocimiento de una segunda lengua a pesar de haber tenido dicha
materia en la escuela, sugieren que la calidad de los contenidos y de la enseñanza en las
escuelas de DJ no fue buena. Esto es algo muy relevante a tener en cuenta al momento de
considerar la extensión de la DJ a otras escuelas o a todo el sistema educativo.
JEL: I2, J2
1 Juan J. Llach is full professor and María E. Gigaglia is research assistant, both at IAE-Universidad
Austral, Argentina. Cecilia Adrogué - now at Universidad de San Andrés and CONICET (Argentina) -
was research assistant at IAE when most of this research was performed. The authors are very
grateful both to IAE-Universidad Austral and Fundación Ethos, whose support to this project was
decisive. The authors also thank the helpful comments of Ernesto Schargrodsky, Flavia Roldán,
Milagros Nores and the participants in seminars at IAE-Universidad Austral and the Argentine National
Academy of Education.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
2
1. Introduction
In 1971, longer school days were decreed for around half the primary schools in the city
of Buenos Aires, Argentina. The policy covered all the city neighborhoods and the schools
were chosen roughly at random. As a consequence, it was created an unusual opportunity
for a natural experiment. In 2006 and 2007, we interviewed a sample of 380 alumni of the
1971 cohort, thirty years after their 1977 graduation in schools both with and without longer
days. We tried to identify how the length of their school days affected their education,
occupation and income.
After this introduction we describe in section 2 the mains traits of the aforementioned
policy. Section 3, devoted to the review of the literature is longer than usual. We thought it
was important to review and to put in contact both, the older literature on the relationship
between the length of school schedules and academic results and, on the other hand, the
newer literature devoted to renew the educational production function approach using
random or natural experiments. Cross-references between different literatures are rare but,
from our point of view, they can help to a better understanding to the issues dealt with here.
Section 4 presents the design of the experiment and the characteristics of the data base.
Section 5 is devoted to show the main results of the experiment. We conclude in section 6
with a discussion of the results and some of their policy implications.
2. The policy and its context
2.1. The educational system in Argentina and Buenos Aires in 1970. Since the end of the
nineteenth century, the Argentine educational system has been traditionally governed by the
principles of free and universal access, laity in the public schools and, up to the late 1970s,
seven years of compulsory primary education.2 Although constitutionally in the hands of the
2 At the end of the 1970s, the first year of pre-school education was also decreed to be compulsory,
and in the early 1990s compulsory education was extended up to the tenth year.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
3
provinces, the federal government continued running some primary schools in most
provinces until the late 1970s and early 1980s. The private sector both religious and
secular was also authorized to run primary and secondary schools. The case of the city of
Buenos Aires was peculiar. As the capital of Argentina, until 1996 its administration was in
the hands of the federal government, and the same happened with its schools. Enrollment
rates in Argentina have been traditionally high when compared to other Latin American
countries, and much higher in the city of Buenos Aires than in the rest of the country. At the
federal level, enrollment rates for primary, secondary and tertiary education at the end of the
1970s were 93.9%, 63.3% and 18.9% respectively.
2.2. The policy of lengthening school days in the city of Buenos Aires. The policy
consisted in the introduction of a double shift (DS) or full time schooling3 in the primary
schools of the city of Buenos Aires. It began very gradually, in 1960, and was drastically
expanded to almost 50% of the schools in the late 1960s and early 1970s. The new
curriculum was put fully in place in 1971.4 For the purposes of school administration, the city
of Buenos Aires was and still is divided into 21 School Districts (SD).
Even where middle classes predominate in the city, there are important socio-economic
differences among their quarters and SD. The policy was originally conceived to achieve
both educational and social purposes (CNE, 1968 and 1971), and was evenly applied in all
the school districts, in such a way that in the early 1970s the proportion of DS primary
schools in every school district was around 50% of the total.
3 The traditional length of a primary schools schedule in Argentina has been between four and four
and a half hours, either in the mornings (more common) or in the afternoons. This system is known as
simple schedule or day (jornada simple). Accordingly, the new system was named full schedule or day
(jornada completa) or double shift (doble jornada). The length of the school day in the new system
was nearly eight and a half hours, including around two hours for lunch.
4 See Ministerio de Cultura y Educación (1970).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
4
From a social point of view, the idea was to provide a solution to the uneven
consequences of the increasing participation of women in the labor force. Whilst richer
households could pay for nurses or other domestic help, the poorer could not. For that
reason, the first DS schools were in poorer SDs. This policy was changed in the late 1960s,
when educational purposes began to dominate over social purposes. The change was
clearly shown in the parallel modification of emphasis of admission criteria. In 1968, DS
schools were ordered to give priority to: (1) the familiar, social and economic needs of the
candidates; (2) the proximity of the student‟s address to that of the school; and (3) the
students with sisters or brothers in the school.
In 1971, however, it was clearly established that the main and unavoidable condition to
be admitted was to live near the school. Only after that was fulfilled could the school consider
the following additional social criteria: (1) the family‟s unfavorable socioeconomic conditions;
(2) whether both parents were working, without domestic help; and (3) the number of
siblings.
The 1971 reform of the admission criteria was critical in reducing the selection bias of our
research. In addition, it is very well known that the address and the siblings-in-the-school
criteria have been predominant in the city of Buenos Aires.
Academic content was very precisely defined for the new DS schools. Extra time was
assigned to the following activities:
a) reinforcement of the academic content already in place, particularly language and
mathematics (35%);
(b) one on one teacher-assisted studying (25%);
(c) foreign language, typically English (12.5%);
(d) health, gym and saving habits (7.5%); and
(e) crafts and job training (20%).
Although originally minded to teach the children useful abilities for the labor market, much
of the craft and job training looked old-fashioned from the very beginning.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
5
3. Longer school days, enduring effects of education and natural experiments: a
review of the literature
As far as the authors are aware, no previous research has been done with the same
purposes and methods of this paper, that is, to assess the enduring educational,
occupational and income effects of longer school days. For that reason, we separately
review the literature on three different issues: the effects of instructional time on educational
outcomes; the enduring, lifetime effects of education and, finally, natural experiments
performed on education. It is a little surprising that the vast majority of the literature on the
effectiveness of instructional time was written between 1960 and 1990, as if the discussion
has disappeared since then. We also note at the end of the revision, however, a recent
revival of the interest in instructional time as an educational policy.
3.1. Longer school days (allocated/scheduled time) and educational outcomes. This
approach has a long tradition in educational sciences. Up to the middle of the last century it
was mostly inspired in the behaviorist and even the Taylorist schools of thought (CIPPEC,
2006). Carroll (1963) was perhaps the first to adopt a more pedagogically oriented approach.
Most of the literature since then has agreed that increasing the allocation of instructional time
has positive but small impacts on educational achievements, and that these impacts tend to
be higher, the lower the countries‟ GDP and the students‟ socio-economic status (SES).
One of the oldest and most comprehensive revisions of this literature is Cotton (1989).
She distinguishes different varieties of allocated time: school time (number of school days
and hours per day); classroom time (hours spent at classrooms); instructional time (the
portion of classroom time spent teaching students particular knowledge, concepts or skills);
engaged time or time-on-task (portions of time during which students are paying attention to
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
6
a learning task and attempting to learn); and academic learning time (ALT).5 Cotton
emphasizes the importance of keeping in mind this taxonomy, because comparisons such as
the ratio of school time to instructional time and the ratio of classroom time to time-on-task
have produced shocking findings. Only about half of the typical school day is actually used
for instruction and the students are engaged in learning activities for only about half of their
in-class time.6 Cotton concludes7 that: (a) there is a small, positive relationship between
allocated time and students' achievements; (b) there is a stronger, but still small, relationship
between time-on-task and achievements; (c) there is a strong and positive association
between ALT and students' achievements and attitudes; (d) there are greater achievements
and enhanced attitudes when time-on-task is interactive with the teacher; (e) lower-ability
students benefit more from increases in allocated or engaged time, while higher-ability
students only benefit very slightly, if at all; (f) the benefits are greater in highly structured
fields of study, such as mathematics and foreign languages.
Only one year after Cotton, Berliner (1990) also begins his revision by reminding of the
multi-dimensionality of instructional time. He emphasizes that the popularity of research on
5 ALT refers to that portion of engaged time that students spend working on tasks at an appropriate
level of difficulty for them and experiencing high levels of success.
6 The rest of the time was typically expended in classroom procedural matters, transitions, disciplinary
matters, dead time or off-task activities. After reviewing McMeekin (1993), Thrupp (1998) and Martinic
(2002), CIPPEC arrives to the same conclusions regarding the scarcity of effective classroom time,
adding that the problem is more serious in regions like Latin America and in poor socio-economic
environments.
7 Cotton (1989) revised 57 research studies - mainly from developed countries - concerned with the
relationship between the educational time factors cited above and the student outcomes of
achievement and attitudes. Twenty-nine are primary sources and 28 secondary (reviews, syntheses,
and meta-analyses).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
7
allocated or scheduled time is just a consequence of the easiness to measure it.8 Contrary to
the majority of the literature,9 Berliner concludes that: (a) despite the difficulties of working
with a complex variable like raw instructional time in developed countries, the effect of
quantity of instruction on achievement is clear and of great relevance to policy debates about
education; and (b) the effects of quantity and quality of schooling are much clearer in lesser
developed countries.
He adds that Hyman, Wright and Reed (1975) is one of the few studies on the enduring
effects of quantity of schooling on the overall quality of the life one leads. Sixteen years after
Berliner, Belleï (2006) concludes from his revision10 that most of the American studies on the
subject agree: (a) on the existence of a positive and statistically significant relationship
between instructional time and academic achievements of the students; (b) on the modest
size that relationship; (c) that it is stronger for students with initially low academic
achievements; and (d) that it tends to be curvilinear, showing diminishing returns to scale to
the increase in instructional time. Belleï argues that the methodological limitations of most of
the reviewed studies are huge and come from the small and non-randomly selected samples,
the limited range of the independent variables, the cross-sectional nature of most of the
studies, the short periods of time involved (generally less than one year) and, finally and
most importantly, that it is not clear to what extent the reviewed studies were controlled by
other factors with the potential to affect the findings.
8 He also points out that most of the contemporary arguments on the role of allocated instructional
time on students' achievements were an outgrowth of the Coleman report (Coleman et al., 1966), with
its skeptical view regarding the impact on educational outcomes of increased school resources of
almost any kind, including time.
9 In a shorter revision, Pittman et al. (1986) coincide with Berliner's first conclusion.
10 In addition to other already quoted papers, he revises Jencks et al. (1972); Bloom (1976); Wiley
(1976); Borg (1980); Fisher et al. (1980); Frederick and Walberg (1980); Karweit and Slavin (1981);
Brown and Saks (1986, 1987) and Link and Mulligan (1986).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
8
Finally, Fuller and Clarke (1994), quoted by CIPPEC (2006), analyze literature
specifically referred to developing countries and conclude that the effect of instructional time
on educational outcomes is stronger there. CIPPEC (2006) does not find many studies on
Latin America, but all of them coincide with the results obtained by Fuller and Clarke
(1994).11
School term’s length. A close family of studies has analyzed the educational,
occupational and income effects of the length of the school year. In his own study, Belleï
(2006) evaluates, with a natural experimental methodology, the Chilean "Full School Day-
Program", designed to increase the yearly high school instructional time from 955 to 1,216
hours. Every year, an additional group of high schools has been integrated into the program,
thereby "potentially establishing a natural experiment. The selection of the schools was not
done randomly, but decided by the government according to certain criteria. Belleï uses a
differences-in-differences approach and argues that it provides an unbiased estimate of the
causal effect of the program on students' academic achievement, as measured by
standardized tests. His main findings indicate that the program had positive effects on
students' achievements, both in language (between 0.05 and 0.07 standard deviations) and
mathematics (around 0.07 sd12), and stronger in rural and municipal schools than in urban
and private schools.
Marcotte (2005) also performed a natural experiment on the effects of instructional time
on Maryland's primary school test scores. He found that natural variation in snowfall over
time, which influenced the number of effective school days, has a small but significant effect
on students performance in their exams.13 According to Pischke (2007), most of the studies
11 Cardoso (Uruguay, n/d); Cervini (Argentina, 2001); Ministerio de Educación de Chile (2003); ANEP
(Uruguay, 2003) and Beleï (Chile, 2006).
12 Standard deviation.
13 Only 1 to 2% fewer students tested in harsh winters performed satisfactorily in math than did
students examined after mild winters.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
9
on the effects of the length of the school term, including his own, find that they are positive
and significant only regarding educational outcomes like avoiding repetition, but not as
regards test scores, future earnings or employment.
3.2. Enduring effects of education. The classical reference here is Hyman, Wright and
Reed (1975), who analyze responses to general knowledge questions in public opinion
surveys between 1947 and 1974.14 Based on the fact that the higher the respondents' level
of educational accomplishment, the more often were the correct responses given, the
authors conclude that "education produces large, pervasive, and enduring effects on
knowledge and receptivity to knowledge" (Hyman et al., 1975: 109). This effect of education
is successfully controlled by gender, religion, ethnicity, geographical origin, age,
socioeconomic background and current occupational status.
Wolfle (1980) emphasizes, however, that the study is weak because the authors could
not control for early intelligence or propensity to learn, in contrast to the vast literature
demonstrating that intelligence has a strong effect upon socioeconomic achievements, and
that it is likely that all analyses of educational effects which do not include IQ variables suffer
severe, although unknown, specification errors.15 Using a causal model of the enduring
effects of education, including the estimated effects of intelligence measures, he concludes
14 In these surveys, people of different ages and educational attainment were polled on their
knowledge of a wide variety of issues, from identifying prominent public figures to responding to
questions on vocabulary.
15 His point is very relevant because it is very uncommon nowadays to include intelligence measures
in studies of the determinants of educational outcomes. Meghir and Palme (2003, 2004) are some of
the exceptions.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
10
that previous studies have seriously overestimated the enduring effects of education.16 A
very important point added by Wolfle is that education does increase general intelligence, in
such a way that its indirect effect on vocabulary through adult IQ is five times the size of the
direct effect.17
More recently, in the age of the methodologies of instrumental variables and natural
experiments, Duflo (2001) studies the educational and labor market outcomes of the
construction of 61,000 primary schools in Indonesia in a very short period (1974-1978).
Measuring the effects twenty years after the program, in 1995, she finds that: (a) it increased
0.25 to 0.40 the average years of education; (b) it improved by 12% the probability that an
affected child would complete primary school; and (c) it raised wages ranging from 3% to
5.4%.18 Combining both effects, she estimates economic returns to education ranging from
6.8% to 10.6%.19 She also warns about the risks of generalizing her results to other contexts
because a number of factors, such as the strong emphasis on education in Indonesia at that
16 He recognizes, however, that his results are conditional until confirmed by longitudinal studies in
which intelligence scores are obtained for a representative sample of children, and their subsequent
levels of educational, intellectual, and verbal achievements are measured.
17 This is the interpretation I would like to draw-that education's primary effect is a generalized
development of adult cognitive skills, not necessarily the retention of specific bits of knowledge
(Wolfle, 1980, p.113).
18 She thinks that the increase in wages she finds proves that there is a combined effect of quality and
quantity changes in education leading to an increase in human capital.
19 She also argues that her 2SLS estimates are similar to OLS estimates and also similar to most
estimates reported for developed countries, but smaller than estimates reported in Psacharopoulos
(1994) for developing economies.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
11
time, the possibility of general equilibrium effects of the program on the returns to education20
and the fact that the program induced variations only at the primary school level, while
returns to secondary education might have been different. Additionally, individuals whose
education level changed because of the program may experience returns to education that
differ from the population average because, for instance, only individuals with high expected
returns responded quickly to the enrollment opportunities open with the program. Duflo
recognizes that the program increased the quantity of education and that it is sometimes
feared that deterioration in the quality of education might result from this type of program,
offsetting any gain in quantity.
Also based on a natural experiment, Meghir and Palme (2003 and 2004) evaluate,
around forty years later, the impact on educational attainment and earnings of a major school
reform that took place in the 1950s in Sweden. The reform had many common elements with
those in other European countries at that time, and included an increase in the years of
compulsory schooling, a new national curriculum and the abolition of selection by ability into
academic and non-academic streams at the age of twelve. The authors find that the reform
increased both the educational attainment and the earnings of those whose parents only had
compulsory education. However, the earnings of those with more educated parents declined
- possibly because of a dilution of quality at the top end of the education levels. Although this
study is a benchmark in the research of long lasting effects of education, it was not possible
to separate in it the effects of the increase in the quantity of education - years of schooling -
from the qualitative ones - such as the new curriculum or the elimination of selection by
ability at the age of twelve. The effects found were also small (see Table 1).21
20 Because of the measuring of the returns twenty years after the program, in an environment where
the education levels were higher than when the program began, individuals‟ returns may be lower than
they would be in other developing countries.
21 As a benchmark for the magnitude of these effects, Anders Björklund (2000) estimates the wage
premium per additional year of education to be 4.6% for Sweden (Meghir and Palme, 2004).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
12
Finally, but perhaps most importantly, Schweinhart et al. (2005) report the results of the
High/Scope Perry Preschool Study Through Age 40. They identify “both the short- and long-
term effects of a high-quality preschool program on young children living in poverty.They
come from a randomized experiment in which a sample of 123 low-income African American
children - who were assessed to be at high risk of school failure - was split into a treatment
group of 53 and a control group of 68. They measure a variety of outcomes at ages 3, 11, 14,
15, 19, 27 and 40. They find “evidence of positive effects on program-group children‟s
intellectual performance, school experiences, lifetime earnings and crime rates: Their school
achievement was at a higher level, they were more committed to school, and more of them
graduated from high school than members of the no-program group. In their adult lives
program participants have achieved higher earnings and committed fewer crimes than
members of the no-program group (pp. xv and xvi).
3.3. Natural and randomized experiments in education. Fortunately, during this century,
there has been a blossoming of a new family of natural and random experiments on the
educational and labor market effects of different educational policies. It has renewed the
hopes of a better understanding of this very relevant question, after the disappointing results
of the vast educational production function research program that followed the challenge
posed by Coleman et al. (1966).22 Although more accurate than the previous research
program, the most important common trait of this new vintage is that most of the effects of
the measured educational policies on educational outcomes are positive but modest, as can
22 Two relevant contributions to get a balance of the educational production function research program
are Glewwe (2002) and Akerlof and Kranton (2002).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
13
be observed in Table 1.23 Regrettably, only a few experiments have studied the enduring
effects of educational policies.
Insert table 1 near here
A brief summary of the results shown in Table 1 is as follows. The educational policy
(treatment) with more intense and widespread effects on income, educational and
employment outcomes is high-quality preschool education (Schweinhart et al., 2005). Also
worth mentioning are class size at the primary level, with strong effects on test scores‟ gaps
(Piketty and Valdenaire, 2006); conditional cash transfer effects on access to tertiary
education (Barrera-Osorio et al., 2008); training program impacts on wages and employment
(Attanasio et al., 2008); effects of preschool meals on attendance (Kremer and Vermeersch,
2004); and school construction on graduation rates and school attendance (Duflo, 2001;
Berlinsky et al., 2006). Surprisingly, and perhaps due to identification problems, vast school
reforms appear as having more modest effects (Meghir and Palme, 2003 and 2004; Hoxby
and Rockoff, 2006). Regarding the lengthening of the school year, the only effects analyzed
up to now are test scores, and the three studies reviewed (Marcotte, 2005; Belleï, 2006 and
Sims, 2006) show smaller impacts on them than other treatments shown in Table 1. Finally,
some of the studies, as Banerjee et al. (2005) and Hoxby and Rockoff (2005), emphasize the
importance of quality over “quantity” of education. This, as well as the critical importance of
early childhood education, appears to be clearly proved by Schweinhart et al. as can be seen
in its impact on a rich set of dependent variables, from income to intellectual development.
23 Piketty argues that the class size is perhaps the clearest case at the time of assessing the
superiority of natural experiments.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
14
3.4. Recent revival of instructional time as an educational policy. The already quoted,
recent works of Meghir and Palme (2003 and 2004), Banerjee et al. (2005), Belleï (2006) and
CIPPEC (2006)24 reveal a sort of revival of the instructional time as an educational policy,
perhaps due to the fact that it is a strategy relatively simple to implement. However, impacts
of additional instructional time give the impression of being very modest up to now, validating
the point made by Cotton (1989): "Significant increases in the quantity of schooling would be
required to bring about even modest increases in achievement."
4. The design of the experiment and the database
4.1. General approach. We used the kind of natural experiment methodology normally
applied to observational studies. As it is well known, in this methodological framework, the
estimation of the effects of the treatment may be biased because of the existence of
confounding factors, and the comparison of sample means among groups is not optimal. To
avoid this problem, we used the propensity score matching25 to reduce the bias in the
estimation of the outcomes, that is, comparing the outcomes using treated and control
subjects who are as similar as possible (Becker and Ichino, 2002). Since this method is not
sufficient to estimate the Average Effect of Treatment on the Treated (ATT), we used the
Kernel matching approach, in which all units of the treatment group are matched with a
weighted average of all units of the control group, with weights that are inversely proportional
to the distance between the propensity scores of treated and controls.
24 CIPPEC (2006) reviews some of the policy-oriented papers that analyze increased instructional time
from different (mainly positive) points of view, including Husti (1992); Pereyra (1992 a and b); Slavin
(1996); Martinic (1998 and 2002); Aguerrondo (1998), and more recently Feldfeber et al. (2003);
Boissiere (2004) and Llach et al. (2006). They also refer to some critical studies like Kurweit (1985)
and National Education Commission (1994).
25 Defined by Rosenbaum and Rubin (1983) as the conditional probability of receiving a treatment,
given the pre-treatment characteristics.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
15
4.2. Data base and sample. The database comes from a randomly sampled survey
applied to the 1971 cohort. The selection of this cohort is methodologically very relevant,
because it was the first to attend the primary schools of the city of Buenos Aires after the
generalization of the DS policy. Although we cannot say that this device eliminates the
selection bias problems typical of these studies, it very probably helps at least to reduce
them. The survey included items such as educational attainment at all levels, information
about subjects‟ parents, current SES and labor status. The questionnaire included both
closed and open questions. To design the sample (Table 2), two variables were taken into
account: the UBN (unsatisfied basic needs)26 and the 1980 primary enrollment of the
schools‟ districts.
Insert table 2 near here
We interviewed people who finished the primary school in 1977 in: (a) double shift (DS)
schools, where the policy was implemented in 1971 (treatment group), and (b) simple shift
(SS) schools with similar characteristics in pre-treatment variables such as socioeconomic
status (SES) and geographic proximity (control group).
5. Results: more education does not imply better education
5.1. Mean differences. Mean differences between treatment and control groups are
presented in Table 3.27
26 Unsatisfied basic needs (UBN) is a compound index of social indicators like housing quality,
employment status and educational attainment.
27 Descriptive statistics of pre-treatment and treatment variables are shown in Appendix 1.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
16
Insert table 3 near here
Some of the relevant variables that cannot be considered to have the same mean in both
groups are the following: (a) conclusion of high school is higher in the treatment group (TG)
and changes of jobs are higher in the control group (CG), both significant at 1%; (b)
knowledge of a foreign language, a job related to the career and the educational level of the
spouse are higher in TG; postgraduate studies are higher in the CG and their members
abandoned the second tertiary study later; all of them significant at 5%; and (c) conclusion of
the second tertiary study, the quality of postgraduate studies and the current SES are higher
in the TG, and unemployment frequency is higher in the CG, all of them significant at 10%.
Tables 4 to 6 show the main and statistically significant mean differences in outcomes‟
variables of students originally coming from low, medium and high SES, respectively. In the
case of low SES students, conclusion of high school remains higher for TG and significant at
1%; the same happens with knowledge of a foreign language, but now significant at 5%; and
primary school grade repetition is lower in TG, significant at 5%, while high school repetition
is higher for the TG, but only significant at 10%.
Insert table 4 near here
In the case of medium SES students, both the conclusion of a second tertiary study and
the quality of postgraduate studies are higher in the TG, but the members of the CG
abandoned the second tertiary study later; all these results are significant at 1%. Contrary to
what happened with low SES students, high school graduation rates are higher in the CG,
and the same happens with the frequency of unemployment, being all these results
significant at 5%. The originally middle SES group is the only one in which a significant (at
10%) mean difference appears with their current SES, that is higher in the CG.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
17
Finally, in the case of the high SES group, repetition in high school is higher in the TG
and the quality of postgraduate studies is higher in CG, both of them significant at 10%.
Insert tables 5 and 6 near here
5.2. Educational, occupational and income effects. With the exception of students‟
nationality and gender, pre-treatment variables cannot be considered to have the same
mean (Appendix 1). It follows that mean differences between TG and CG in the outcome
variables cannot be considered as the result of the treatment per se. In order to control for
these pre-existing differences, we adopt the Kernel‟s propensity score matching. In the
following tables we present the estimation of the average effect of the policy on the treatment
group (Attk), as reflected in the mean differences attributable to the treatment after
controlling them for the differences in pre-treatment variables. On the other hand, to estimate
the impact of the double shift, we eliminate the outliers in the control group. In other words,
we do not include in the sample individuals whose propensity score is lower than the
minimum probability observed in the treatment group.
DS is found to have positive and significant effects in the following educational outcomes.
(a) The conclusion of high school: secondary school graduation rate is 21% higher in the TG,
and this is one of the most relevant results of our research. (b) The access to, and the
conclusion in time of, a second tertiary study. (c) The probability to have a job related to the
career during the studies. (d) The educational level of the spouse and of the household head.
On the other hand, the educational variable where the treatment appears as having
negative and significant effects are the timely conclusion of the first tertiary study and the
conclusion of postgraduate studies. Another positive and significant effect of the DS is to
change jobs less frequently (Table 7).
Insert table 7 near here
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
18
We also calculated the effects of the treatment for each one of the three levels of
students‟ household‟s SES. In Tables 8 to 10 we only show statistically significant effects. In
the group of low SES students, TG‟s high school graduation rate is 22% higher again, a
very relevant result; a second tertiary study once finished the first one is 25% more probable
and, on the other side of the coin, repetition in high school is 13% higher. In the group of
medium SES, a big variety of results appears. On the one hand, DS shows a positive and
significant effect in the timely conclusion of the second tertiary study -once the first is
finished- as well as in the quality of the postgraduate studies and in the probability of having
a job related to the career while studying. On the other hand, DS has a small, but negative
and significant impact on the conclusion of high school and the access to tertiary studies.
Another negative, but stronger effect is on the access to postgraduate studies. The middle
SES one is the only group in which DS has a positive impact on income but, surprisingly, a
negative impact on the current students‟ SES.28 In the case of the high SES group, DS
appears associated to higher repetition in high school, lower graduation rates in
postgraduate studies and lower income.
Insert tables 8, 9 and 10 near here
6. Discussion and policy implications
We have shown that the introduction of longer school days in half of the primary schools
of the city of Buenos Aires in 1971 has significantly improved only one, but very relevant,
educational outcome. Students that attended DS primary schools had a secondary school
graduation rate 21% higher than those that attended single shift primary schools. Moreover,
28 This is not incompatible, because SES is measured only by educational and employment quality
variables without including income.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
19
this result is mainly explained by what happened with the low SES students. Regarding the
tertiary and postgraduate educational levels, we have found both positive and negative
impacts of DS. These last results, taken together with the absence of enduring effects of DS
on income and employment and with the fact that DS students do not have a better
knowledge of a second language of study in spite of having had it as a subject in the school,
suggest that the contents‟ and learning‟s quality in DS schools was not satisfactory. This is
something very relevant to keep in mind at the time of considering the extension of DS to
other schools or to the whole educational system. Our paper helps to emphasize that even
more important than having longer school days are the contents of the additional hours. Just
to give an example, academic results‟ improvement could be very different if those hours are
just an extension of the current curriculum or, instead, if they allow the disadvantaged
students to develop their skills and abilities through the teaching and learning of a second
language, sports, arts and technologies, i.e., the same subjects that their advantaged mates
can normally learn and practice.
The general meaning of our results coincides with most of the literature reviewed in this
paper, even the one performed with the more demanding methodology of natural
experiments. Impacts of a wide variety of educational policies can be big and relevant
regarding the “quantity” of education, but very seldom as regards the quality of education or
other lifelong effects. The main exceptions we found are a high-quality preschool
(Schweinhart et al., 2005) and, potentially, class size in primary schools of low SES children
(Piketty and Valdenaire, 2006). In the first case, the effects are not only important, but also
widespread to personality, labor, income and citizenship outcomes. In the second case, the
impact is very strong on test scores. Taken all these results (including this paper‟s) together,
it sounds true the claim made by Piketty and Valdenaire (2006), who argue that a targeted
allocation of resources to poorer - or the poorest - schools and students could have a
significant impact in reducing educational inequalities, and that this effectiveness will be
much greater if it concentrates from early childhood onwards. However, it also seems clear
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
20
that we still need more and better research to understand what specific policies are needed
to improve the quality of education for the poor.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
21
Insert Appendix 1 near here
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
22
References
Aguerrondo, I. (1998). América Latina y el desafío del tercer milenio. Educación de mejor
calidad con menores costos, PREAL, Documentos de trabajo, 10.
Akerlof, G. A. and R. E. Kranton (2002). “Identity and schooling: some lessons for the
economics of education”, Journal of Economic Literature, XL, 4 (1167-1201).
Anderson, L. (1983). "Policy implications of research on school time", The School
Administrator, 40 (25-28).
ANEP (Administración Nacional de Educación Pública, 2003). Resultados en escuelas
de tiempo completo y escuelas de áreas integradas. Evaluación Nacional de
Aprendizajes en lenguaje y matemática, 6to año enseñanza primaria, 2002, Segundo
Informe, Montevideo: ANEP.
Angrist, J. and V. Lavy (1999). "Using Maimonides‟ rule to estimate the effect of class
size on scholastic achievement", Quarterly Journal of Economics, 114, 2 (533-574).
Attanasio, O.; A. Kugler and C. Meghir (2008). Training disadvantaged youth in Latin
America: evidence from a randomized trial, NBER WP 13931.
Banerjee, A.; S. Cole; E. Duflo and L. Linden (2005). Remedying education: evidence
from two randomized experiments in India, NBER WP 11904.
Barrera-Osorio, F.; M. Bertrand, L. L. Linden and F. Perez-Calle (2008). Conditional cash
transfers in education: design features, peer and sibling effects: evidence from
randomized experiment in Colombia, NBER WP 13890.
Becker S. and A. Ichino (2002). “Estimation of average treatment effects based on
propensity scores”, The Stata Journal 2, Number 4, (358-377).
Belleï, C. (2006). Does lengthening the school day increase students' academic
achievement? Results from a natural experiment in Chile, Harvard Graduate School of
Education, Qualifying Paper.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
23
Berliner, D. C. (1990). "What's all the fuss about instructional time?" in The Nature of
Time in Schools. Theoretical Concepts, Practitioner Perceptions, New York and London:
Teachers College Press, Teachers College, Columbia University.
Berlinsky, S.; S. Galiani and P. Gertler (2006). The effect of pre-primary education on
primary school performance, University College, Universidad de San Andrés and World
Bank. London and Washington, DC.
Betts, J. R and E. Johnson (1998). A test of diminishing returns to school spending,
mimeo, University of California at San Diego.
Björklund, A. (2000). “Educational policy and returns to education”, Swedish Economic
Policy Review, spring, 2000, 7(1), (71-105).
Bloom, B. S. (1976). Human Characteristics and School Learning, New York: McGraw-
Hill.
Bloom, B. S; G. F. Madaus and J. T. Hastings (1981). Evaluation to Improve Learning,
New York: McGraw-Hill.
Boissere, M. (2004). Determinants of primary education outcomes in developing
countries, OED, Washington DC: The World Bank.
Borg, W. (1980). "Time and school learning", in C. Denham and A. Lieberman (editors),
Time to Learn, Washington D.C.: Department of Education, National Institute of
Education (33-72).
Bressoux, P; F. Kramarz and C. Prost (2004). Caractéristiques des enseignants, taille
des classes et réussite des élèves : une estimation sur des classes de CE2 en 1991,
mimeo, Université de Grenoble et Crest.
Brown B. and D. Saks (1986). "Measuring the effects of instructional time on student
learning: evidence from the beginning teacher evaluation study", American Journal of
Education, 94, 4 (480-500).
Brown B. and D. Saks (1987). "The microeconomics of the allocation of teachers' time
and student learning", Economics of Education Review, 6, 4, (319-337).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
24
Browning, M. and E. Heinesen (2003). Class size, teachers hours and educational
attainment, Working Paper 2003-15, University of Copenhagen.
Card, D. and A. Krueger (1992). "Does school quality matter? Returns to education and
the characteristics of public schools in the United States", Journal of Political Economy,
100 (1-40).
Cardoso, M. (n/d). Calidad y equidad en las escuelas de tiempo completo: un análisis de
sus resultados en las evaluaciones estandarizadas 1996 y 1999.
Carroll, J. B. (1963). "A model of school learning", Teachers College Records, 64 (723-
733).
Cervini, R. (2001). "Efecto de la oportunidad de aprender sobre el logro en matemáticas
en la educación básica argentina", Revista Electrónica de Investigación educativa, 3, 2,
http://redie.uabc.mx/vol3no2/contenido-cervini.html.
CIPPEC (Centro de Implementación de Políticas Públicas para la Equidad y el
Crecimiento, 2006). Estudio para la Implementación de una Política Nacional de
Extensión de la Jornada Escolar. Buenos Aires: CIPPEC.
CNE (Consejo Nacional de Educación) (1968). Escuela de jornada completa.
Organización y funcionamiento, Buenos Aires.
----------------------------------------------------- (1971). Reglamento orgánico para escuelas de
jornada completa.
Coleman, J.; E. Q. Campbell, C. J. Hobson, J. McPartland, A. M. Mood, F. D. Weinfeld
and R. L. York (1966). Equality of Educational Opportunity, Washington D.C: U.S.
Government Printing Office.
Cotton, K. (1989). "Educational time factors", School Improvement Research Series
(SIRS): http://www.nwrel.org/scpd/sirs/4/cu8.html.
Cotton, K. and K. Wikelund, (1990). Educational time factors, Northwest Regional
Educational Laboratory: http://www.nwrel.org/scpd/sirs/4/cu8.html.
Dee, T. and M. West (2008). The non-cognitive returns to class size, NBER WP 13994.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
25
Duflo, E. (2001). "Schooling and labor market consequences of school construction in
Indonesia", American Economic Review, 91, 4 (795-813).
Duflo, E.; P. Dupas, M. Kremer, and S. Sinei (2006). Education and HIV/AIDS
Prevention: Evidence from a Randomized Evaluation in Western Kenya, mimeo.
Eide, E. and M. H. Showalter (1998). "The effect of school quality on student
performance: a quantile regression approach", Economic Letters, 58 (345-50).
Evans, D.; M. Kremer, and M. Ngatia (2008), The Impact of Distributing School Uniforms
on Children’s Education in Kenya, mimeo.
Feldfeber, M.; N. Gluz and M. Gómez (2003). La jornada completa en la Ciudad de
Buenos Aires, Serie Estudios de Base, Vol. 3, Buenos Aires: Gobierno de la Ciudad
Autónoma de Buenos Aires, Secretaria de Educación, Dirección de Investigación.
Fisher, C.; D. Berliner; N. Filby; R. Marliave; L. Cahen and M. Dishaw (1980). "Teaching
behaviors, academic learning time and student achievement: an overview", in C. Denham
and A. Lieberman (editors), Time to Learn, Washington D.C.: Department of Education,
National Institute of Education, (7-32).
Frederick, W. and H. Walberg (1980). "Learning as a function of time", Journal of
Educational Research, 73, 4, (183-94).
Fuller, B., Clarke, P. (1994). "Raising school effects while ignoring culture?. Local
conditions and the influence of classroom tools, rules and pedagogy", Review of
Educational Research, 64, 1, (119-157).
Furstenberg, F. F. and D. Neumark (2005). School-to-career and post-secondary
education: evidence from the Philadelphia Educational Longitudinal Study, NBER
Working Paper 11260.
Glewwe, P. (2002). “Schools and skills in developing countries: education policies and
socioeconomic outcomes”, Journal of Economic Literature, XL, 2, (436-482).
Glewwe, P.; M. Kremer and S. Moulin (2007). Many children left behind? Textbooks and
test scores in Kenya, NBER Working Paper 13300.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
26
Goux, D. and E. Maurin (2007). "Close neighbors matter: neighborhood effects on early
performance at school", Economic Journal, 117 (1193-1215).
Grogger, J. (1996). "Does school quality explain the recent black/white wage trend?",
Journal of Labor Economics, 14 (231-253).
Heckman, J., A. Lane-Farrar and P. Todd (1996). "Does measured school quality really
matter? An examination of the earnings quality relationship" in (G. Burtless, ed.) Does
Money Matter? The Effect of School Resources on Student Achievement and Adult
Success (192-289), Washington, DC: Brookings Institution Press.
Heyneman, S. and W. Loxley (1983). "The effect of primary school quality on academic
achievement across twenty-nine high- and low-income countries", American Journal of
Sociology, 88, 6, (1162-94).
Heyns, B. (1978). Summer Learning and the Effects of Schooling, New York: Academic
Press.
Hyman, H. H.; C. R. Wright and J. S. Reed (1975). The Enduring Effects of Education,
Chicago: Chicago University Press.
Honzay, A. (1986-87). "More is not necessarily better", Educational Research Quarterly,
11 (2-6).
Hoxby, C. M. and J. E. Rockoff (2005). The Impact of Charter Schools on Student
Achievement, mimeo.
Husti, A. (1992). "Del tiempo escolar uniforme al tiempo escolar móvil", Revista de
Educación, 298, Madrid: Ministerio de Educación y Ciencia.
Jencks, C.; M. Smith; H. Acland; M. Bane; D. Cohen; H. Gintis; B. Heyns and S.
Michelson (1972). Inequality, New York: Basic Books.
Karweit, N. L. (1976). "A reanalysis of the effect of quantity of schooling on achievement,
Sociology of Education, 49, (235-46).
Karweit, N. and R. E. Slavin (1981). "Measurement and modeling choices in studies of
time and learning", American Educational Research Journal, 18, 2, (157-71).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
27
Karweit, N. (1985). "Should we lengthen the school term?", Educational Researcher, 14,
6, June-July.
Kremer, M.; S. Moulin and R. Namunyu (2003). Decentralization: a cautionary tale,
mimeo.
Kremer, M. and C. Vermeersch (2004). School meals, educational Attainment, and
school competition: evidence from a randomized evaluation, World Bank Policy Research
Working Paper, WPS3523.
Krueger, A. B. (1999). “Experimental estimates of education production functions”,
Quarterly Journal of Economics, 114, 2 (497-532).
Lee, J. W. and R. Barro (2001). "School quality in a cross-section of countries",
Economica, 68 (465-88).
Levin, H. and M. Tsang (1987). "The economics of student time", Economics of
Education Review, 6, 4, (357-364).
Link, C. and J. Mulligan (1986). "The merits of a longer school day", Economics of
Education Review, 5, 4, (373-81).
Llach, J. J. et al. (2006). El desafío de la equidad educativa. Diagnóstico y propuestas,
Buenos Aires: Granica.
Marcotte, D. E. (2005). "Schooling and test Scores: a mother-natural experiment",
Economics of Education Review, 26 (629-640).
Martinic, S. (1998). Tiempo y aprendizaje, LCSHD Paper Series, 26, The World Bank,
Human Development Department.
Martinic, S. (2002). El tiempo y el aprendizaje en América Latina, Serie Políticas, PREAL,
Santiago.
McMeekin, R. W. (1993). "La investigación al servicio de la educación: tiempo y
aprendizaje", Proyecto Principal de Educación en América Latina y el Caribe, Boletín 30,
UNESCO-OREALC.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
28
Meghir, C. and M. Palme (2003). Ability, Parental Background and Education Policy:
Empirical Evidence from a Social Experiment, The Institute for Fiscal Studies, WP 03/05.
-------------------------------------- (2004). Educational Reform, Ability and Family Background,
The Institute for Fiscal Studies, WP 04/10.
Ministerio de Cultura y Educación (Argentina, 1970). Boletín de Comunicaciones, XV, 21,
31 de agosto.
Ministerio de Educación de Chile (2003). Factores que inciden en el rendimiento de los
alumnos. Prueba SIMCE Básico 2002. Nota Técnica, Santiago de Chile:
Departamento de Estudios y Estadísticas.
National Education Commission on Time and Learning (1994). U.S. Department of
Education (1994), Prisoners Of Time. Report of the National Education Commission on
Time and Learning, USA Department of Education.
http://www.ed.gov/pubs/PrisonersOfTime/
Pastorino, H. O. (2000). La Escuela de Jornada Completa. Crónica Histórica y Aportes
para su Organización, Buenos Aires: Ediciones Caminos.
Pereyra, M. A. (s/f). "La reforma de la no reforma del tiempo escolar en España.
¿Escuela con jornada partida o continuada? Itinerario reflexivo sobre un tema recurrente
para el futuro de la enseñanza pública española", en Propuesta Educativa, Buenos
Aires: FLACSO.
Pereyra, M. A. (1992a). "La construcción social del tiempo escolar", Cuadernos de
pedagogía, 206.
Pereyra, M. A. (1992b), "La jornada escolar en Europa", Cuadernos de Pedagogía, 206.
Piketty, T. (2004). L’impact de la taille des classes et de la ségrégation sociale sur la
réussite scolaire dans les écoles françaises : une estimation à partir du panel primaire
1997, http://jourdan.ens.fr/piketty/_mpublic/ipublic.php.
Piketty, T. and M. Valdenaire (2006). L'impact de la taille des classes sur la réussite
scolaire dans les écoles collèges et lycées français. Estimations à partir du panel
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
29
primaire 1997 et du panel secondaire 1995, Les Dossiers, Ministère de l‟Éducation
nationale, de l‟Enseignement supérieur et de la Recherche. Direction de l‟évaluation et
de la prospective, 173.
Pischke, J. (2007). "The impact of the length of the school year on student performance
and earnings: evidence from the German short school years", Economic Journal, 117,
October (1216-1242).
Pittman, R.: R. Cox and G. Burchfiel (1986). "The extended school year: implications for
student achievement", Journal of Experimental Education, 54.
Psacharopoulos, G. (1994). “Returns to investments in education: a global update.”
World Development, September, 22, 9 (132543).
Rizzuto, R. and P. Wachtel (1980). "Further evidence on the returns to school quality",
Journal of Human Resources, 15 (240-254).
Rosenbaum P. R. and D. B. Rubin (1983). “The central role of propensity score in
observational studies for causal effects.Biometrika, 70 (1), (41-55).
Sims, D. P. (2006). “Strategic responses to school accountability measures: It‟s all on the
timing”, Economics of Education Review, 27 (58-68).
Schweinhart, L. J., J. Montie, Z. Xiang, W. S. Barnett, C. R. Belfield and M. Nores (2005).
Lifetime Effects. The High/Scope Perry Preschool Study Through Age 40, Ypsilanti,
Michigan: High Scope Press.
Schultz, T. P. (2004). “School subsidies for the poor: evaluating the Mexican
PROGRESA Poverty Program, Journal of Development Economics, 74, 1 (199-250).
Slavin, R. E. (1996). Salas de clases efectivas, escuelas efectivas: plataforma de
investigación para una reforma educativa en América Latina, Santiago de Chile: PREAL,
Documento de Trabajo.
Thrupp, M. (1998). “The art of the possible: organizing and managing high and low socio-
economic schools”, Journal of Education Policy, 13, 2, (197-219).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
30
Walberg, H. (1986). "Syntheses of research on teaching", in M. Wittrock (editor),
Handbook of Research on Teaching, Third Edition, New York: Macmillan (214-229).
Wiley, D. E. (1976). "Another hour, another day: quantity of schooling, a potent path for
policy", in W. Sewell, R. Hauser and D. Featherman (editors), Schooling and
Achievement in American Society, New York: Academic Press.
Wiley, D. and A. Harnischfeger (1974). "Explosion of a myth: quantity of schooling and
exposure to instruction: major educational vehicles", Educational Researcher, 3, (7-12).
Woessmann, L. (2003). "Schooling resources, educational institutions and student
performance": the international evidence", Oxford Bulletin of Economics and Statistics, 86
(497-513).
Wolfle, L. M. (1980). "The enduring effects of education on verbal skills", Sociology of
Education, 53, April, (104-114).
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
31
Tables
Table 1. Compared results of natural (N) and random (R) experiments on outcomes of
educational policies. Primary (P) and Secondary (S) education
Authors (date)
Results
A. Preschool improvements
1. Schweinhart et
al. (2005, R,
preschool).
Graduation. High school graduation rates of 77% in the
treatment group and 60% in the control group.
Income. 60% of the treatment group vs. 40% of the
control group earning ≥ US$20.000.
Quality of life. The treatment group at 40 also had lower
crime rates, higher employment rates, more fathers
assuming child-rearing responsibilities and higher scores
in various intellectual and language tests at very different
ages.
2. Kremer and
Vermeersch (2004,
R, preschool)
Attendance. In spite of the increased fees in treatment
schools, attendance to them improved by 8.5 pp (31%).
Attendance gains were both for current students and
students who had never attended before.
3. Berlinsky et al.
(2006, N,
preschool).
Attendance. One year of preprimary school increases
average third grade test scores by 8% of a mean or by
0.23 of the sd of the distribution of test scores.
Non cognitive skills. Preprimary school attendance
positively affected student‟s self-control in the third grade
as measured by behaviors such as attention, effort, class
participation, and discipline.
B. Increase in school’s resources
4. Duflo (2001, N, P
and S).
Graduation. 12% increase in the probability of primary
school completion.
Years of education. Increase of 0.25 to 0.4 of a year.
Wages. 3% to 7% increase in wages.
Rates of return. 6.8% to 10.6% for primary education.
5. Glewwe et al.
(2007, R, P).
Test scores. No increase in test scores, contrary to the
results of the previous literature. Textbooks increased
scores for students with high initial academic
achievement. Students with weaker academic
backgrounds did not benefit from the textbooks. Many of
them could not read the textbooks, which were written in
English, most students' third language.
6. Duflo et al.
(2006) and Evans et
al. (2008), R, P.
Attendance. For younger pupils, 6 pp increase (7%) in
school attendance and 13 pp (15%) increase for student
without a uniform prior to program. For older pupils,
13.5% decline in absence.
Years of education. Years of enrollment increased by 0.5
year (13%).
29 „Duration‟ refers to the time span between the treatment and the measurement of their effects.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
32
C. Class size
7. Krueger (1999,
R, P).
Test scores. Performance of students in smaller classes
increased by 4 percentile points the first year and by 1
percentile point per year in subsequent years. Test
scores in smaller classes rose by about 0.22 sd. Class
size had a larger effect for minority students.
Future earnings. A 0.22 sd improvement in test scores
resulting from smaller class sizes implied an
improvement of 1.7% and 2.4% average male and
female earnings, respectively.
8. Piketty and
Valdenaire (2006,
N, P and S).
Test scores and test scores’ gaps. The reduction of one
pupil per primary class allowed an increase in the range
of 0.3-0.4 points in math's test scores (0.7 in less socio-
economically endowed contexts). These results implied
that 5 pupils less per classroom in the poorer zones
could lead to close 46% of the tests results‟ gap between
them and the non-poor zones. With the same policies,
the gap would be closed 22% at the college level and
only 4% at the lycée.
9. Dee and West
(2008, R, early P).
Non cognitive skills. Early-grade class-size reductions
did improve subsequent student initiative, but these
effects did not persist into the 8th grade. Smaller classes
in the 8th grade led to 0.05 to 0.09 improvements in
measures of student engagement, persisting two years
later. Internal rate of return was 4.6% overall and 7.9 in
urban schools.
D. Conditional cash transfers
10. Schultz (2004,
R, P and S)
Attendance. Between 3.4 and 3.6 pp increase in
attendance for all children in grades 1 to 8. A 11.1 pp
increase (19%) in attendance for students who have
completed 6th grade and 14.5 pp increase for girls who
have completed 6th grade. Spillovers to ineligibles in
treatment villages of 5 pp (7%) in secondary enrollment.
11. Barrera-Osorio
et al. (2008, R, P
and S)
1. Direct cash transfers.
Attendance, permanence. Increases of 2.8 pp in school
attendance; 2.6 pp in school permanence.
Years of education. Increases of 2.8 pp in the following
year‟s enrollment and 23 pp in the probability to
matriculate in tertiary studies.
Graduation. 4.0 pp increase in graduation rates
2. Part of the cash transfer postponed.
Enrollment in both secondary and tertiary institutions
increased over the basic treatment by 3.6 and 3.3 pp,
respectively, without reducing the current attendance.
Spillovers. Negative spillovers on the non-selected
siblings and some positive peer effects on educational
outcomes.
D. Additional teaching support
12. Banerjee et al.
(2005, R, P).
Test scores. 0.14 / 0.28 sd increase in test scores
compared to non-treated peers. One year after the
program, initial gains faded to about 0.10 sd.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
33
13. Banerjee et al.
(2005, R, P).
Test scores. 0.35 and 0.47 sd increase in math scores in
the first and second year of the program respectively.
After that, the increase tended to fall.
E. Peers effect
14. Goux and
Maurin (2007, N, S).
Repetition. The probability of repetition was found to be
0.20 sd higher for adolescents (15-16 years) living in
neighborhoods with a higher proportion of mates that
have already been held back a grade at age 15.
F. Training programs
15. Attanasio et al.
(2008, R, Training).
Earnings and employment. In the case of women,
training increased wage and salaried earnings, the
probability of being employed, the amount of days and
hours worked, and the probability of having a formal job
with a written contract. Smaller effects on men: training
only increased wage and salaried earnings and the
probability of having a formal job and with a written
contract, but not of having employment. Salaried
earnings increased 18% for women and 8% for men.
Cost-Benefit. The rate of return of the program emerging
from cost-benefit analysis was 13.5% for women and
4.5% for men. On-the job training intensity increased the
returns of the program.
G. School reforms
16. Meghir and
Palme (2003, 2004,
N, S).
Years of education. Increase by 0.298 of a year, entirely
due to the increase in the educational attainment of
those with unskilled fathers.
Earnings. Overall, 1.42% increase in earnings. For those
with unskilled fathers, the reform increased earnings by
3.4%.
Rates of return. If all the changes in earnings were due
to changes in the quantity of education, the results
implied returns of 6.0% for low ability individuals, 11.6%
for those with high abilities and 8.4% overall. If other
variables played a role those returns were upper
bounds.
17. Hoxby and
Rockoff (2005, N,
P).
Compared to their lotteried-out fellow applicants,
students who applied, and attended to, charter schools
an average of two years starting in the elementary
grades, scored about 6 national percentile rank points
higher, both math and reading.
G. Length of the school year
18. Marcotte (2005,
N, P and S).
Test scores. Only between 1 and 2 percent fewer
students performed satisfactorily after harsh winters than
did students examined after mild winters.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
34
19. Belleï (2006, N,
S).
Test scores. The program had a positive effect on
students' achievement, both in language (between 0.05
and 0.07 sd) and math (around 0.06 sd). Stronger
effects were attained in rural and municipal schools than
in urban and private schools.
20. Sims (2006, N,
P)
Test scores. Clear and positive relationship between
math scores of 4th graders and days of preparation. The
implied effect was small in both in an absolute sense
and relative to other educational reforms.
Table 2. Characteristics of the sample
Total
Shift
Parents‟ SES
Students‟ SES
Simple
Double
High
Medium
Low
High
Medium
Low
Gender
Male
Obs.
185
86
99
22
73
90
61
59
65
%
48.7%
50.0%
47.6%
53.7%
50.0%
46.6%
56.0%
41.8%
50.0%
Female
Obs.
195
86
109
19
73
103
48
82
65
%
51.3%
50.0%
52.4%
46.3%
50.0%
53.4%
44.0%
58.2%
50.0%
Total
Observations
380
172
208
41
146
193
109
141
130
%
100%
100%
100%
100%
100%
100%
100%
100%
100%
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
35
Table 3. Outcomes’ variables
Outcomes
Obs.
Mean
Std.
Dev.
Treated
group’s
mean (I)
Control
group’s
mean (II)
(I-II)
Educational
Repetition in primary school (1: yes, 0:no)
380
0.06
0.23
0.05
0.07
-0.02
º
Conclusion of high school (1: yes, 0: no)
380
0.90
0.30
0.94
0.86
0.08
***
Repetition in high school (1: yes, 0:no)
375
0.21
0.40
0.22
0.19
0.03
º
Tertiary (post-secondary) studies (1: yes, 0:no)
343
0.83
0.38
0.83
0.83
0.00
º
Conclusion of the first tertiary study (1: yes, 0: no)
285
0.76
0.43
0.77
0.75
0.02
º
Timely conclusion of the first tertiary study (1: yes,
0:no)
216
0.67
0.47
0.65
0.70
-0.05
º
Year in which the first tertiary study was interrupted
64
2.44
1.28
2.50
2.36
0.14
º
Second tertiary study, if the first was finished (1: yes,
0:no)
216
0.26
0.44
0.28
0.24
0.04
º
Conclusion of the second tertiary study if the first one
was finished (1: yes, 0: no)
57
0.61
0.49
0.69
0.50
0.19
*
Timely conclusion of the second tertiary, if the first
one was finished (1: yes, 0:no)
35
0.66
0.48
0.67
0.64
0.03
º
Year in which the second tertiary study was
interrupted, if the first one was finished
22
1.68
0.89
1.27
2.09
-0.82
**
Postgraduate studies (1: yes, 0:no)
67
0.96
0.21
0.92
1
-0.08
**
Type of postgraduate studies (1: post degree, 2:
master, 3: PhD)
64
1.41
0.61
1.51
1.28
0.03
*
Conclusion of the postgraduate studies (1: yes, 0:
no)
64
0.80
0.41
0.74
0.86
-0.12
º
Knowledge of a foreign language (1: yes, 0:no)
380
0.87
0.33
0.90
0.84
0.06
**
Labor
Work during the studies (1: yes, 0:no)
380
0.65
0.48
0.68
0.62
0.06
º
Job related to the career during the studies (1: yes,
0:no)
247
0.61
0.49
0.66
0.55
0.11
**
Number of hours worked during the studies
244
6.80
2.17
6.80
6.80
0.00
º
Quality of the present job (1: stable, 2: occasional)
331
1.02
0.14
1.02
1.02
0.00
º
Unemployed status of the person surveyed (% rate)
380
0.04
0.19
0.05
0.03
0.02
º
Changes of job since the person began working
378
3.63
3.98
3.17
4.18
-1.01
***
Number of times unemployed since the person
began working
374
0.66
1.32
0.56
0.77
-0.21
*
Quality of the present job of the household head (1:
stable, 2: occasional)
362
1.01
0.12
1.02
1.01
0.01
º
Unemployed status of the household head (% rate)
380
0.02
0.15
0.03
0.02
0.01
º
Income
Net current monthly income (*)
295
3.23
2.05
3.19
3.27
-0.08
º
Others
Students' SES at the present (1: low, 2: medium, 3:
high)
380
1.94
0.79
2
1.88
0.12
*
Presence of children in the current household (1:yes,
0:no)
380
0.78
0.42
0.78
0.77
0.01
º
Number of children in the current household
380
1.55
1.11
1.50
1.60
-0.10
º
Educational level of the spouse (2: lowest, 10:
highest)
261
6.80
1.98
7.04
6.50
0.54
**
Educational level of the household head (2: lowest,
10: highest)
380
7.24
2.07
7.33
7.15
0.18
º
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
36
* Significant at 10%. ** Significant at 5%. ***Significant at 1%. ºNot significant.
(*) 1: less than $323, 2: $324 -$484; 3: $485-$635, 4:$636-$806, 5: $807-$968, 6:$969-$1290, 7: >$1290.
Table 4. Outcomes of students from low SES households
Outcomes
Obs.
Mean
Std.
Dev.
Treated
group’s
mean (I)
Control
group’s
mean (II)
(I-II)
Educational
Repetition in primary school (1: yes, 0:no)
193
0.08
0.27
0.04
0.11
-0.07
**
Conclusion of high school (1: yes, 0: no)
193
0.83
0.37
0.91
0.76
0.15
***
Repetition in high school (1: yes, 0:no)
189
0.28
0.45
0.33
0.22
0.11
*
Knowledge of a foreign language (1: yes, 0:no)
193
0.83
0.37
0.89
0.78
0.11
**
Labor
Number of hours worked during the studies
103
7.25
2.11
7.53
6.96
0.57
*
Changes of job since the person began working
192
3.69
3.62
3.18
4.16
-0.98
**
Others
Number of children in the current household
193
1.67
1.10
1.52
1.81
-0.29
**
* Significant at 10%. ** Significant at 5%. ***Significant at 1%.
Table 5. Outcomes of students from medium SES households
Outcomes
Obs.
Mean
Std.
Dev.
Treated
group’s
mean (I)
Control
group’s
mean (II)
(I-II)
Educational
Conclusion of high school (1: yes, 0: no)
146
0.97
0.18
0.94
1.00
-0.06
**
Conclusion of the second tertiary study if the first one
was finished (1: yes, 0: no)
26
0.50
0.51
0.69
0.20
0.49
***
Timely conclusion of the second tertiary study if the
first one was finished (1: yes, 0:no)
13
0.62
0.51
0.72
0.00
0.72
**
Year in which the second tertiary study was
interrupted if the first one was finished
13
1.92
0.95
1.00
2.50
-1.50
***
Type of postgraduate studies (1: post degree, 2:
master, 3: PhD)
35
1.40
0.65
1.63
1.13
0.51
***
Labor
Changes of job since the person began working
145
3.76
4.74
3.16
4.63
-1.47
**
Number of times unemployed since the person
began working
146
0.58
1.55
0.41
0.84
-0.43
**
Others
Students' SES at the present (1: low, 2: medium, 3:
high)
146
2.26
0.69
2.20
2.36
-0.16
*
* Significant at 10%. ** Significant at 5%. ***Significant at 1%.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
37
Table 6. Outcomes of students from high SES households
Outcomes
Obs.
Mean
Std.
Dev.
Treated
group’s
mean (I)
Control
group’s
mean (II)
(I-II)
Educational
Repetition in high school (1: yes, 0:no)
41
0.10
0.30
0.14
0.00
0.14
*
Type of postgraduate studies (1: post degree, 2:
master, 3: PhD)
13
1.62
0.65
1.44
2
-0.56
*
Labor
Job related to the career during the studies (1: yes,
0:no)
34
0.88
0.33
1
0.64
0.36
***
* Significant at 10%. ** Significant at 5%. ***Significant at 1%.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
38
Table 7.Average effects of the double shift and standard errors
Outcomes
ATTK
Boot se
Educational
Repetition in primary school
-0.10
0.09
Conclusion of high school
0.21*
0.09
Repetition in high school
0.03
0.07
Tertiary (post-secondary) studies
-0.04
0.06
Conclusion of the first tertiary study
0.06
0.14
Timely conclusion of the first tertiary study
-0.16*
0.06
Year in which the first tertiary study was interrupted
-0.16
0.30
Second tertiary study if the person finished the first
one
0.15*
0.10
Conclusion of the second tertiary study if the first one
was finished
0.06
0.23
Timely conclusion of the second tertiary study if the
first one was finished
0.51*
0.24
Year in which the second tertiary study was
interrupted if the first one was finished the first one
-1.18*
0.56
Postgraduate studies
-0.07
0.14
Type of postgraduate studies
0.29
0.20
Conclusion of the postgraduate studies
-0.21*
0.11
Knowledge of a foreign language
0.09
0.09
Labor
Work during the studies
0.13
0.10
Work related to the career during the studies
0.21*
0.09
Number of hours worked during the studies
-0.06
0.50
Quality of work of the person surveyed at the present
-0.03
0.06
Unemployed status of the person being surveyed at
the present
0.01
0.04
Changes of job since the person began working
-1.43*
0.59
Unemployed status since the person began working
-0.29
0.23
Quality of work of the household head at the present
-0.03
0.06
Unemployed status of the household head at the
present
-0.01
0.04
Income
Net monthly income of the person being surveyed at
the present
0.02
0.63
Others
Students' SES at the present
0.20
0.18
Presence of children in the current household
0.02
0.08
Number of children in the current household
0.03
0.17
Educational level of the spouse
0.93*
0.33
Educational level of the household head
0.73*
0.42
*Significant at 5%
Notes: a) Attk: Average effect of treatment with Kernel matching. b) Boot se: Bootstrapping of
standard errors.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
39
Table 8.Students from low SES households
Average effects of the double shift and standard errors
Outcomes
ATTK
Boot se
Educational
Conclusion of high school
0.22*
0.09
Repetition in high school
0.13*
0.07
Second tertiary study if the person finished the first
one
0.25*
0.16
Labor
Changes of job since the person began working
-1.55*
0.07
*Significant at 5%
Notes: see Table 7.
Table 9.Students from medium SES households
Average effects of the double shift and standard errors
ATTK
Boot se
Educational Outcomes’ variables
Conclusion of high school
-0.06*
0.03
Tertiary studies
-0.05*
0.03
Timely conclusion of the second tertiary study if the
first one was finished
0.73*
0.15
Year in which the second tertiary study was
interrupted if the first one was finished
-1.77*
0.41
Postgraduate studies
-0.44*
0.10
Type of postgraduate studies
0.66*
0.19
Conclusion of the postgraduate studies
-0.21*
0.11
Labor Outcomes’ variables
Work during the studies
-0.17*
0.06
Job related to the career during the studies
0.34*
0.18
Income Outcomes’ variables
Net monthly income of the person being surveyed at
the present
1.04*
0.59
Other Outcomes’ variables
Students' SES at the present
-0.50*
0.12
*Significant at 5%
Notes: see Table 7.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
40
Table 10.Students from high SES households
Average effects of the double shift and standard errors
ATTK
Boot
se
Educational Outcomes’ variables
Repetition in high school
0.14*
0.08
Conclusion of the postgraduate studies
-0.33*
0.16
Labor Outcomes’ variables
Work during the studies
-0.13*
0.08
Changes of job since the person began working
0.39*
0.22
Income Outcomes’ variables
Net monthly income of the person being surveyed at
the present
-1.57*
0.85
Other Outcomes’ variables
Number of children in the current household
1.22*
0.71
*Significant at 5%
Notes: see Table 7.
Do longer school days have enduring educational, occupational or income effects?
A natural experiment on the effects of lengthening primary school days in Buenos Aires, Argentina
J.J.Llach, C.Adrogué and
M.E.Gigaglia
41
Appendix 1 Descriptive statistic on pre treatment and treatment variables30
Obs.
Mean
Std.
Dev.
Treatment
group‟s
mean
(I)
Control
group‟s
mean
(II)
(I-II)
Pre-Treatment Variables
Nationality (1: argentine, 0: foreigner)
380
1.00
0.07
1.00
1.00
0.00
º
Age
372
41.52
0.70
41.44
41.62
-0.17
***
Gender (1: men, 0: women)
380
0.49
0.50
0.48
0.50
-0.02
º
School SES (1: low, 2: medium, 3: high)
380
2.00
0.75
2.09
1.90
0.19
***
Parent's SES (1: low, 2: medium, 3: high)
380
1.60
0.68
1.69
1.49
0.19
***
Type of School (1: men and women, 2: women,
3:men)
380
1.40
0.71
1.62
1.13
0.49
***
Number of students
375
25.56
5.73
27.83
22.79
5.05
***
Number of sections
380
1.67
1.16
1.90
1.40
0.13
***
Father's educational level (1: lowest, 10: highest)
380
4.47
2.10
4.68
4.20
0.51
**
Mother's educational level (1: lowest, 10: highest)
370
5.08
2.37
5.17
4.67
0.48
**
Treatment Variables
Foreign language as a subject (1: yes, 0:no)
380
0.52
0.50
0.93
0.02
0.90
***
Cultural activities (1: yes, 0: no)
380
0.67
0.47
0.67
0.67
0.00
º
Place where the cultural activities take place (0:
nowhere, 1: at school, 2: at home, 3: at home and
at school)
380
1.31
0.97
1.32
1.31
0.01
º
Presence of lunch service (1: yes, 0: no)
378
0.59
0.49
1.00
0.08
0.92
***
Assistance to the lunch service (1: yes, 0:no)
221
0.66
0.48
0.70
0.07
0.62
***
*significant at 10%. ** significant at 5%. *** significant at 1%. ºnot significant
30 Some other characteristics of both groups are: (a) more than half of the students changed their
primary school; (b) most of those who attended a double shift primary school changed to a simple shift
high school; (c) 76% of the students who started tertiary studies concluded them, and 70% of them did
so in time; (d) many surveyed people worked during their studies and that work was related to what
they were studying.
... For instance, Ghana has a 98% net primary school enrolment, but more than two-thirds of its students perform at basic and below basic levels in national standardized exams. To improve student performance, developing countries such as Peru, Argentina, Colombia and the Dominican Republic are increasingly identifying the expansion of the school calendar year as an important educational policy to achieve the best results in their students in terms of academic performance (Llach, Adrogue, & Gigaglia, 2009;Behrman, Parker, & Todd, 2012). ...
Article
Full-text available
This study investigated leadership opinion on extended school duration and the academic performance. The purpose of the study was to determine whether the SHS students performed better when the duration was extended to four years. The qualitative method was employed for the study. A population of 149 teachers who taught both the 3 and 4-year SHS students and 3 headmasters were drawn from three schools in Central Region via purposive and convenient sampling techniques respectively. The unstructured interview guide was used to collect the data. Nine(9) school leaders were interviewed and the interview data was analysed qualitatively. Two (2) research questions were formulated to guide the study. The major finding of the study was that the 4-year cohorts of students performed better than the 3-year cohorts. The MOE and GES, Ghana should consider extending the SHS education duration to four (4) years as a better option to improve students’ academic performance.
... The extension of the school day has been widely discussed as a mechanism to improve the quality of education (Holland, Alfaro, and Evans 2015;Bellei 2009;Link and Mulligan 1986;Llach, Adrogué, and Gigaglia 2009). Increasing the time students spend in school is an important decision for policy makers and school administrators, especially in face of the argument that a longer school day could produce better academic, social and labor market outcomes. ...
Article
This paper evaluates the impact of extension of the school day in the context of Brazil’s Mais Educação Extended School Day Program. Using school level longitudinal data, we find that the program reduces the dropout rates of students in all grade levels, raises the enrollment of students in grades 6–9, but reduces the enrollment of students in grades 10–12. Moreover, the estimates indicate that the impact on grade promotion is positive for students in grades 6–9, but negative for students in lower grades. Finally, the program seems to increase repetition rates for students in all grade levels.
Chapter
Full-text available
En este capítulo se presenta el procedimiento llevado a cabo para desarrollar una investigación cualitativa con enfoque hermenéutico que tiene como objetivo central fundamentar un principio didáctico de transversalidad para posibilitar la formación integral en la Educación Superior. Esta es una metodología que trabaja desde el interior del ser que investiga, desde su vivencia, para que desarrolle sus propios procesos de comprensión que le permitan interpretar lo real y formarse.
Article
This article addresses teacher perception regarding how excessive amounts of extracurricular activities could interfere with regular lessons and, ultimately, affect student academic performance. A survey probing personal opinions regarding this phenomenon was completed by thirty teachers from public and private primary schools and high schools in the Costa Rican province of Puntarenas. The main objective of this study was not only to determine teacher perception, but also to compare collected data with already existing theory on this subject. Results show that even though some of those activities are relevant in training students to become good citizens and future professionals, an overload of extracurricular activities will negatively impact their achievements and knowledge acquisition.
Article
Full-text available
Recibido 10 de agosto de 2012 • Corregido 05 de noviembre de 2012 • Aceptado 07 de noviembre de 2012Hace más de dos décadas que asistimos en América Latina y en Argentina, en particular, al desarrollo de diferentes políticas orientadas a ampliar las jornadas escolares. Entendemos que la implementación de estas políticas es una oportunidad para analizar el comportamiento de la organización escolar ante los intentos de modificar uno de sus componentes duros, como lo es el tiempo; y que se constituye en laboratorios naturales que permiten analizar cuánto resiste la organización clásica del tiempo escolar, en qué cambia y cómo esos cambios (de ocurrir) afectan los restantes componentes de lo escolar (espacios, agrupamientos, etc.). Este artículo presenta el estado de la cuestión de los estudios más relevantes en dos campos de indagación en el ámbito de la educación primaria o elemental que intervienen en esta problemática: por un lado, los que refieren al tiempo escolar, su organización y ampliación y; por otro lado, las investigaciones que abordan los componentes estructurales y estructurantes de lo escolar. La revisión bibliográfica realizada nos indica que las investigaciones que han abordado el tiempo escolar y las distintas políticas y programas de extensión de la jornada escolar como objeto de indagación no dan cuenta de las vicisitudes que estas encuentran a la hora de intentar modificar los componentes duros del dispositivo escolar y, por otro lado, que las investigaciones que toman como objeto de estudio el ‘dispositivo escolar’ no han atendido las distintas experiencias de ampliación del tiempo escolar, aun cuando el tiempo es uno de los elementos constitutivos de dicho dispositivo.
Article
Full-text available
Este documento sintetiza los resultados de las investigaciones disponibles que estudian el efecto de la ampliación de la jornada escolar sobre el logro académico, medido en pruebas estandarizadas en matemáticas y lenguaje. Para esto, se emplea un metanálisis con efectos aleatorios de 14 referencias que cumplieron los criterios de selección y fueron publicados entre 2007 y 2017. Estas publicaciones incluyen resultados para 8 países. Los documentos del metanálisis surgen de una búsqueda en 6 bases de datos electrónicas y 2916 referencias iniciales. El efecto promedio global estimado para logro académico de matemáticas en primaria es de 0,04 desviaciones estándar (SD) y secundaria 0,10SD. A su vez, el efecto promedio global para lenguaje de 0,04SD y 0,08SD para primaria y secundaria, respectivamente. Los tamaños de efecto promedio son mayores en secundaria que en primaria. Para todos los resultados y niveles de escolaridad existe una considerable heterogeneidad en los tamaños del efecto. This document seeks to consolidate the results of the available research who studies the effect of extending the school day on achievement, measured by standardized tests in mathematics and language. A meta-analysis was used with random effects in 14 references that satisfy the selection criteria and were published between 2007 and 2017. These publications include results for 8 countries. The meta-analysis documents arise from a search in 6 electronic databases and 2916 initial references. The size of the overall average effect estimated by means of meta-analysis for academic achievement of mathematics in primary school is 0.04 standard deviations (SD) and for secondary is 0.10SD. An average global effect is estimated for language of 0.04SD and 0.08SD for primary and secondary, respectively. The average effect sizes are higher in secondary than in primary. For all results and levels of schooling there is considerable heterogeneity in effect sizes.
Test scores. 0.35 and 0.47 sd increase in math scores in the first and second year of the program respectively. After that, the increase tended to fall
  • Banerjee
Banerjee et al. (2005, R, P). Test scores. 0.35 and 0.47 sd increase in math scores in the first and second year of the program respectively. After that, the increase tended to fall. E. Peers effect
R, preschool) Free school meals. Kenya Duration: 1 year. Effects of a vast preprimary classrooms construction program
  • Kremer
  • Vermeersch
Kremer and Vermeersch (2004, R, preschool) Free school meals. Kenya. Duration: 1 year. Effects of a vast preprimary classrooms construction program. Argentina. Duration: 4/5 years.
Dev. Treatment group " s mean (I) Control group " s mean (II) (I-II)
  • Mean Obs
  • Std
Obs. Mean Std. Dev. Treatment group " s mean (I) Control group " s mean (II) (I-II)