See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257658048

Mother tongue instruction in Lubuagan: A case study from the Philippines

Article  in  International Review of Education · December 2011

DOI: 10.1007/s11159-011-9246-4

CITATIONS READS

44 5,530

2 authors:

Stephen L. Walter Diane Dekker

Graduate Institute of Applied Linguistics LEAD Asia
15 PUBLICATIONS   116 CITATIONS    10 PUBLICATIONS   96 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Good answers to tough questions in mother tongue-based multilingual education View project

Listening to Filipino Teachers’ Voices as They Negotiate Their Identities within the Context of Language Policy Change to Mother-Tongue-Based Multilingual Education
View project

All content following this page was uploaded by Diane Dekker on 12 August 2021.

The user has requested enhancement of the downloaded file.

Int Rev Educ (2011) 57:667–683
DOI 10.1007/s11159-011-9246-4

Mother tongue instruction in Lubuagan: A case study

from the Philippines

Stephen L. Walter • Diane E. Dekker

Published online: 13 November 2011

Ó Springer Science+Business Media B.V. 2011

Abstract In the modern era, the prevailing model of public education has been
that of ‘‘one size fits all’’, with private schooling being a small but notable
exception. Language (of instruction) was generally viewed as a minor variable
readily overcome by standard classroom instruction. As researchers have sharpened
their focus on the reasons for educational failure, language has begun to emerge as a
significant variable in producing gains in educational efficiency. This paper reports
the intermediate result of a controlled study in a very rural area of a developing
country designed to examine the effect of language of instruction on educational
outcomes. In the experimental schools, children are taught to read first in the local
language (via the local language) and are taught other key subjects via the local
language as well. English is taught as a subject. Teachers in the control or standard
schools continue the standard national practice of teaching all subjects in either
English or Filipino, neither of which is spoken by children when they begin school.
Year-end standardised testing was done in all subjects throughout grades one to
three as a means of comparing the two programme methodologies.

Keywords Philippines
Language policies
Lubuagan mother tongue-based

multilingual education programme
Lilubuagen
Filipino
English

Evaluation of educational outcomes primary 1 to 3

Résumé L’enseignement en langue maternelle à Lubuagan : une étude de cas des

Philippines – À notre époque moderne, le modèle prédominant de l’enseignement
public est celui de la « taille unique », l’enseignement privé étant une exception

S. L. Walter (&)
Graduate Institute of Applied Linguistics (GIAL), Dallas, TX, USA
e-mail: [email protected]

D. E. Dekker
SIL International, Manila, Philippines
e-mail: [email protected]

123
668 S. L. Walter, D. E. Dekker

restreinte mais non négligeable. La langue (d’instruction) était traditionnellement

considérée comme une variable mineure facilement maı̂trisée par l’enseignement
standard en classe. Lorsque les chercheurs se penchèrent sur les raisons des échecs
scolaires, la langue est alors apparue comme une variable significative susceptible
d’accroı̂tre l’efficacité éducative. Les auteurs de cet article présentent les résultats
intermédiaires d’une étude contrôlée réalisée dans une région très rurale d’un pays
en développement, qui vise à examiner l’impact de la langue d’instruction sur les
résultats éducatifs. Dans les écoles expérimentales, les élèves apprennent à lire en
premier lieu en langue locale (au moyen de la langue locale) et sont instruits dans
les autres disciplines principales également en langue locale. L’anglais y est
enseigné en tant que matière. Les enseignants des écoles témoins ou classiques
poursuivent la pratique nationale générale d’enseigner toutes les matières soit en
anglais soit en filipino, langues que les enfants ne parlent pas au moment de leur
scolarisation. Des tests standard ont été administrés en fin d’année dans toutes les
matières de la première à la troisième classe, en vue de comparer les méthodologies
des deux programmes.

Zusammenfassung Muttersprachlicher Unterricht in Lubuagan : eine Fallstudie

aus den Philippinen – ,,Allen das Gleiche‘‘ ist heutzutage das vorherrschende
Modell in der öffentlichen Bildung, wobei der private Unterricht eine kleine, aber
bemerkenswerte Ausnahme darstellt. Die (Unterrichts-)Sprache galt bislang allge-
mein als nebensächliche Variable, die sich im normalen Unterricht ohne Weiteres
bewältigen lasse. Erst seit die Wissenschaft sich genauer mit den Gründen für das
Bildungsversagen beschäftigt, stellt sich heraus, dass die Sprache ein wichtiger
Faktor ist, wenn es darum geht, Bildung effizienter zu machen. In diesem Artikel
wird das Zwischenergebnis einer kontrollierten Studie vorgestellt, mit der die
Auswirkungen der Unterrichtssprache auf den Bildungserfolg in einer stark ländlich
geprägten Region eines Entwicklungslandes untersucht werden sollen. In den
Versuchsschulen lernen die Kinder das Lesen zuerst in der lokalen Sprache (durch
die lokale Sprache) und auch andere wichtige Fächer werden in der lokalen Sprache
unterrichtet. Englisch wird als Fach unterrichtet. Die Lehrkräfte in den Kontroll-
oder Normalschulen unterrichten gemäß der landesüblichen Praxis alle Fächer
entweder auf Englisch oder auf Filipino, wobei die Kinder bei Schuleintritt keine
dieser beiden Sprachen beherrschen. Der standardisierte Test am Jahresende wurde
in allen Fächern mit den Klassen eins bis drei durchgeführt und diente dem Ver-
gleich der beiden Methodiken des Programms.

Resumen Enseñanza en la lengua materna, en Lubuagan: estudio de caso de

Filipinas – En la era moderna, el modelo prevaleciente en la educación pública era
el de ‘‘un patrón de aplicación general’’, donde las escuelas privadas eran una
pequeña, pero notable excepción. La lengua (de enseñanza) generalmente era
considerada como variable de menor importancia que se podı́a superar fácilmente
mediante una instrucción estándar en las aulas. Cuando los investigadores centraron
la atención en las razones del fracaso educativo, la lengua comenzó a surgir como
variable importante para la obtención de ventajas en la eficiencia educativa. Este
trabajo informa sobre los resultados, obtenidos a mediano plazo, de un estudio

123
Mother tongue instruction in Lubuagan 669

controlado en un área muy rural de un paı́s en vı́as de desarrollo, designado para

examinar el efecto de la lengua de instrucción sobre los resultados de la educación.
En las escuelas experimentales, primero se les enseña a los niños a leer en la lengua
local (mediante la lengua local); asimismo, también se les enseñan otras asignaturas
clave mediante la lengua local. El inglés se enseña como asignatura. Los docentes
en las escuelas estándar o de control mantienen la práctica nacional estándar de
enseñar todas las asignaturas en inglés o filipino, siendo que ninguna de estas
lenguas son habladas por los niños cuando empiezan la escuela. Para comparar
ambas metodologı́as del programa, se realizaron pruebas estandarizadas que abar-
caban los tres primeros grados, en todas las asignaturas.

Introduction

The role of language of instruction (LOI) in the educational process has been long
debated (Baker 1996; Fishman et al. 1996; Dutcher 2004; Skutnabb-Kangas 2000)
with no clear consensus in sight – at least not in terms of practice. Even the massive
and compelling evidence supplied by the work of Wayne P. Thomas and Virginia
Collier (1997, 2002) has failed to deter those convinced that language of instruction
is not a salient issue in education – or at least not in terms of the quality of
educational outcomes. In the most extreme contrast, their research showed non-
English-speaking children who received no instructional support in their first
language finishing school, on average, at the 11th percentile while those who
received a full six years of instructional support in their first language finished as a

123
670 S. L. Walter, D. E. Dekker

group at the 70th percentile. In career terms, this is the difference between being a
day labourer and having access to a broad range of professional opportunities in
industry, science and education.
The research done by Thomas and Collier was carried out entirely in the United
States, leading many to question whether the differentials they observed might
extend to education systems and processes in developing countries. This paper
reports preliminary results of a longitudinal study being carried out in one such
country – the Philippines – designed to test whether the ‘‘language of instruction’’
effect would hold there as well.

Philippine language policy

During the Spanish colonial period, Spanish was the primary language of instruction
and the public use of vernaculars in any domain was forbidden. In 1898, when the
Americans became colonisers, English became the language of education (Act No.
74, 21 January 1901, Philippine Commission 1901) and Philippine languages were
not permitted in the schools (Sibayan 1985). While President McKinley recom-
mended the use of local languages in the classroom, a massive influx of American
teachers effectively resulted in a monolingual English-based education system
(Gonzalez 2001, p. 4).
The 1935 Philippine Constitution (article 13, section 2; Philippine Constitution
1935) referred to plans for ‘‘the development and adoption of a common language
based on one of the existing native languages’’. Commonwealth Act No. 570
(Commonwealth 1940) declared Tagalog as the basis of the national language,
along with English. In 1959, Education Secretary José Romero issued a Department
Order stating that the national language would be called Pilipino to distinguish it
from its Tagalog base and to give it a national identity. The 1973 Constitution
(Philippine Constitution 1973) designated Pilipino as the new national language and
as an official language, along with Spanish and English. The 1987, post-People
Power I Constitution (Philippine Constitution 1987) declared Filipino (now spelled
with an F) as the national language as well as one of the official languages along
with English. Spanish was dropped as an official language. The 1987 Constitution
(in force as of 1994) also stipulated the creation of a new language body, Komisyon
ng Wikang Filipino (Commission of the Filipino Language) for the development
and intellectualisation of Filipino. Three Constitutions (1935, 1973 and 1987) have
therefore decreed that the national language is Filipino; however, there seems a
clear intent that English should remain as an official language.

Language-in-education policy in the Philippines

The revised language policy of 1987 (Quisumbing 1987) prescribes the use of
English for teaching maths, science and English while Filipino is prescribed for
teaching all other subjects. However, observation shows that teachers typically
begin teaching in the required language (either English or Filipino) and repeat the

123
Mother tongue instruction in Lubuagan 671

same content in the local language to ensure student comprehension of the

curriculum content. Or teachers may codeswitch within the same statement
(Gonzalez 1998; Young 2002). In practice, this means that local languages are used
to explain the curriculum content to students rather than using those languages
specifically as the media of instruction and teaching English and Filipino as
subjects. In 2004, President Gloria Macapagal-Arroyo instituted a return to English
as the primary medium of instruction in an effort to enhance the competitive edge of
Filipinos in the international labour market (DepEd 2003).
With over 160 living languages, multilingualism is common in the Philippines
and multilingual education has often been a controversial topic of conversation.
Over the last several decades multiple studies, including PCSPE (1970), World
Bank (1988), EDCOM (1991), PESS (1998), PCER (2000) and BESRA (2006),
have recommended the use of the vernacular in the early years of education.1
In an attempt to implement a mother tongue-based national educational
programme bridging learners from their first language to the languages of
education, Filipino and English, Andrew Gonzalez, Secretary of the Department
of Education, Culture and Sports (DECS) instituted DECS Memo No. 144 s. 1999
(DECS 1999) to develop foundational literacy skills (Cruz 2004). This programme
was expanded with the inclusion of more schools and more languages by DECS
Undersecretary for Programs and Projects, Isagani Cruz (DECS Memo No. 243 s.
2000, DECS 2000). This resulted in the expansion of the 1974 Bilingual Education
Policy to a ‘‘still-unnamed and unacknowledged Multilingual Education Policy’’
(Cruz 2004). The Basic Education Curriculum (DECS, changed to DepEd in 2001,
Order No. 43s 2002, DepEd 2002) as implemented by Secretary Raul Roco
maintained a focus on the central role of language in education and retains the
multilingual policy begun in the expansion of the Regional Lingua Franca Program.
The most recent and probably strongest statement is DepEd Order No. 74 2009
(DepEd 2009), Institutionalizing Mother Tongue-Based Multilingual Education
(MLE). At the time of writing it is the intent of the Department of Education to
implement fully a strong MLE programme beginning in each learner’s first language
and bridging them to Filipino and English.

The Lubuagan multilingual education programme

The context

The Lubuagan Kalinga Mother Tongue-Based Multilingual Education (MLE)

Program is a response to the late Secretary Andrew Gonzalez’ call for piloting
innovative approaches to literacy and education in minority language communities.
The municipality of Lubuagan lies in the province of Kalinga in the Cordillera
Mountains of the northern Philippines with a population of around 12,000. Lubuagan
and its barrios host one school district with 13 elementary schools. There are two
private high schools and one public high school. Lubuagan is a monolingual, almost

1
Brigham and Castillo (1999).

123
672 S. L. Walter, D. E. Dekker

homogenous community in which those who move to the area for business purposes
or through marriage learn and use the local language, Lilubuagen.2 Ilocano, the
regional language of wider communication, is primarily used when one travels
outside Lubuagan. Consequently, the children in Lubuagan typically begin school
speaking Lilubuagen but no other language (Dekker and Young 2005).
Lilubuagen is being used in all domains except in the church. Children growing
up in Lubuagan speak Lilubuagen solely until they attend school and begin learning
the languages of education, Filipino and English. Those Lubuagens who have
completed their education are quite proud of their ability to speak English. Most
Lubuagens desire that their children learn English in order go out of the area to find
work. English is seen as providing the best access to a better life, while Filipino
enables communication with the prominent group but provides less opportunity for
work and is therefore seen as being less important. Knowledge of Ilocano is
expected because of the necessity of doing business in the provincial capital.

The rationale

The Lubuagan MLE programme teaches the Filipino and English languages through
the mother tongue rather than through immersion in these two languages. Using the
learners’ first language to teach them Filipino and English provides the compre-
hensible input necessary for learning second languages (Krashen 1991, 2000). After
oral proficiency is developed in Filipino and English, literacy in these second
languages is introduced, bridging from literacy in the first language.
Additionally the programme does not change any of the standard governmental
competencies for all subjects. Rather the focus is on using the first language as the
medium of instruction to ensure comprehension of curriculum content for mastery.
Reading and writing are taught first in the learners’ mother tongue. Basic fluency in
reading is developed in the first language first, while Filipino and English language
lessons develop oral communication skills before reading in those languages is
introduced. Teachers had to learn to read and write their own language first before
they were competent to teach first language literacy. This was done in a series of
writer’s workshops where teachers were bridged from reading and writing the
national language to their own language. Because their own language has only
recently been standardised to written form, spelling is often difficult. This will
change as more and more print literature is made available in the language and
people become accustomed to seeing their language in written form.
In the traditional classroom, curriculum content is taught through Filipino for
social studies and Filipino language and English for maths, science and English.
This creates a scenario in which young learners must learn a language at the same
time they are learning new concepts taught in that language.
The Lubuagan MLE programme incorporates cultural content in order to
optimise learning of the curriculum content (Dekker and Dumatog 2004). This is
done by beginning each lesson with what is familiar to the learner, related to his or

2
Lubuagan is the spelling of the town, Lubuagen are the people and Lilubuagen is the name of the
language.

123
Mother tongue instruction in Lubuagan 673

her everyday experiences, and building on that to introduce the new content. All
beginners’ reading material consists of stories familiar in content to the learner so
that the focus is on the process of reading rather than distracting him or her with
new situations and unfamiliar people and places. Teachers incorporate oral
literature, local history, local arts, craft and music in the curriculum as well to
support the learners’ home culture. This strategy provides a link between what the
learner knows, his or her first language and culture, and the unknown, or what they
need to learn – Filipino, English and the curriculum content.
Beyond the more local concerns about educational efficiency and culture
preservation lies a broader national debate about educational policy and strategy.
Will the country achieve its Millennium Development Goals (MDG) in education?
Does the country have in place appropriate educational strategies for the 21st
century? What language or languages should be used for instructional purposes to
achieve national goals for educational development? Dalisay Maligalig and José
Ramon Albert (2008, p. 23) report data suggesting that the country has actually
regressed slightly during the last 15 years on several key indicators of the MDG in
education. In a national case study prepared for the Education for All (EFA) Global
Monitoring Report, Rhona Caoli-Rodriguez (2007) similarly reports that national
initiatives (such as decentralisation) undertaken as part of the EFA initiative show
little evidence of having raised key educational indicators in the Philippines.
Ricardo Nolasco (2008) suggests that national educational policy with respect to
language(s) of instruction additionally constrains educational effectiveness for
many Philippine children, especially those living in rural areas and in regions of the
country where neither English nor Filipino – the official languages of instruction – is
widely known.

Brief history of the Lubuagan experimental project

Before the Lubuagan MLE programme was implemented, a Steering Committee

was established by the community to discuss language and education issues. After
meeting monthly for 18 months, the steering committee recommended implement-
ing the programme as a pilot project for evaluation. Originally the Lubuagan
programme was called the First Language Component-Bridging Program (FLC-BP)
and based on the Tuwali FLC-BP developed by Lou Hohulin of the Summer
Institute of Linguistics (SIL). Dr. Gloria Baguingan of Nueva Vizcaya State
University (NVSU) provided the first two training events in Lubuagan in 1998, each
covering 3 days. Several Lubuagan teachers also took courses at NVSU during three
summers. Subsequently the author (Diane Dekker) took primary responsibility for
the development of the programme.
Experimental implementation in grade one began at St. Teresita’s School during
the 1999–2000 school year. The programme was not continued at St. Teresita’s after
2000. In 2000 several teachers in the public school system began implementing the
programme in grade one. Each summer teacher training workshops focused on
curriculum adaptation for incorporating local culture and materials production. A
variety of materials have been developed, including both monolingual and trilingual
story books, big books, charted stories, a first language primer and translated

123
674 S. L. Walter, D. E. Dekker

textbooks for maths and social studies, grade one. Other teaching and learning
materials include a cultural calendar and multilingual maths calendar.
Initially, the teaching of all subjects except Filipino and English was done in the
mother tongue. At a later point teaching Filipino and English through the mother
tongue as language of instruction was implemented. This method has proved
effective for strengthening the acquisition of those languages among Lubuagan
students.
By 2005, when the longitudinal study was launched, there were nine teachers in
three schools using the MLE methodology in the primary grades. In 2006 the
Division Superintendent commended the Lubuagan teachers for raising overall
achievement throughout the Lubuagan district.

The research questions

The Lubuagan MLE programme was launched on the general premise that use of the
mother tongue would result in improved educational outcomes. But not all
stakeholders valued equally the various outcomes which could be emphasised and
measured. Accordingly, the design which guided the research being reported here
has sought to attend to those outcomes deemed important to the broadest range of
stakeholders.
Parents tend to measure educational effectiveness in terms of whether or not their
children are/will be able to get a ‘‘good job’’ in the ‘‘outside world’’. Most typically
this means to them that their children must gain adequate proficiency in the
‘‘language of jobs’’. The fear that their children will be stymied in their mastery of
the ‘‘language of jobs’’ is the primary reason parents give for insisting that their
children be educated only in this language. In the case of the Philippines, the
language(s) of jobs are English and Filipino. Therefore, our first research question
was:
Does participation in a programme in which there is heavy usage of the first
language (L1) as a language of instruction compromise the development of
proficiency in L2 (the language of jobs)?
Educators want to see their children do well on tests comparing their children to
those from other schools. The usual assumption is that since such testing is normally
done in L2, children will better master curricular content (and thus do better on
standardised tests) if instructed entirely by means of this language. Hence our
second research question was:
Does participation in a programme in which there is heavy usage of the first
language as a language of instruction compromise mastery of overall
curricular content?
National policy-makers think in terms of a skilled workforce and competitive status
relative to neighbouring countries. Measures of particular concern at this level
include tests of ability in maths and science. This suggests an obvious additional
research question:

123
Mother tongue instruction in Lubuagan 675

Does participation in a programme in which there is heavy usage of the first

language as a language of instruction compromise the development of
proficiency in such key areas as maths, science and reading?

Methodology

The major schools in the Lubuagan area agreed to be divided into two groups – one
in which the local language is used as a language of instruction (the experimental
group) and one in which the prevailing national model of providing instruction in
English and Filipino is used (the control group). The largest school participates in
both models with one section of each grade in the experimental group and one
section of each grade in the control group. Assignment to one of the two sections is
random with the exception of occasional specific requests from parents. The result
of this arrangement is three ‘‘schools’’ or classes in the control group and three in
the experimental group.
For assessment purposes, tests were developed in the following content areas:
reading, maths, Filipino, English and social studies (Makabayan3). The language of
the test followed the scheme used in instruction (Table 1).
The tests were constructed by the research team (the authors plus two highly
experienced local teachers) using the national curriculum statement of learning
outcomes for each grade. The team first extracted lists of learning outcomes in each
content area and then constructed test items designed to measure mastery of each
learning outcome. The tests were first prepared in English (reading, maths and
English) or Filipino (Filipino and social studies). Each test item on each test was
individually reviewed by the entire team together to ensure clarity, aptness and
appropriateness. Then, as needed, the tests were rendered in Lilubuagen and again
checked for clarity and aptness as expressed in Lilubuagen.
Most test items used a multiple choice format to reduce ambiguity in scoring.
The set of tests included multiple examples of how to use the multiple choice format
in test-taking. This format was additionally explained by the test administrators
several times as they administered the tests to ensure that children understood the
format.

Table 1 Map of test instruments administered in the Lubuagan district in 2008

Experimental group Control group

Instructional language Test language Instructional language Test language

Reading Lilubuagen Lilubuagen English English

Maths Lilubuagen Lilubuagen English English
English Lilubuagen and English English English English
Social studies Lilubuagen and Filipino Lilubuagen English and Filipino Filipino
Filipino Lilubuagen and Filipino Filipino Filipino Filipino

3
Makabayan is the Tagalog term for social studies.

123
676 S. L. Walter, D. E. Dekker

Test administration and scoring

Instructions on how to take the test were given to all children in both groups in
Lilubuagen so that all children heard the same explanations and had the same
training in test-taking. Two test administrators – both native speakers of Lilubuagen
and speakers of both English and Filipino as second languages – administered all the
tests to ensure parity of administration.
All tests were presented in written form with each child having a personal copy
of each test. In grades one and two, each test item in each test was read to the
children by the test administrator (for both controls and experimentals) to
compensate for limited or unpredictable levels of reading skill among this group.
No further explanation or hint was given about the test item. The test administrators
did not read the list of possible answers leaving this task to the test-taker. Grade
three children were responsible for the entire test with the test administrator simply
proctoring the test.
After the tests were administered, they were scored locally, with the scoring spot-
checked by one of the authors who speaks Lilubuagen. This spot-checking identified
two items on the maths test for one grade which were inconsistently scored, so these
two items were discarded from all tests for that grade for both control and
experimental groups before analysis.

Findings

Table 2 summarises the overall results of the testing by grade and experimental
grouping. The score reported in the ‘‘Mean’’ column is the average raw score for all
students tested on all tests for that grade. The score in the ‘‘Percentage’’ column is
simply the mean score converted to a percentage based on the total number of test
items in all tests combined.
Several points stand out in Table 2. First, the (percentage) scores across grades
are quite uniform suggesting appropriate scaling of the test items in the test
instruments. Secondly, the Experimental group consistently scored 21 to 22
percentage points higher than the Control group. And third, the level of statistical
significance (of the difference) is high.

Table 2 Combined test scores (all five tests administered) for all three grades
Controls Experimentals Statistics

N Mean Percentage N Mean Percentage T-test P

Grade 1 73 40.10 53.5 69 56.90 75.9 7.89 0.000

Grade 2 94* 55.20 56.9 42 75.48 77.8 10.98 0.000
Grade 3 60 44.23 53.9 56 61.64 75.1 7.08 0.000

* The unbalanced number of subjects in the two groupings in Grade 2 is due to the fact that one school
did not adhere to its assigned model for two consecutive years so that the cohort for Grade 2 this year in
that school had to be treated as being in a different model

123
Mother tongue instruction in Lubuagan 677

Since the test instruments were designed as criterion-referenced tests, the data
suggest that children in the control schools are consistently mastering only a little
over 50 per cent of the content while those in the experimental schools are
consistently mastering 75 to 80 per cent of the curricular content.

Comparisons by major content areas

Maths and reading

Reading and maths are normally considered the most fundamental of the basic skills
to be taught and mastered in early basic education. Therefore, performance in these
areas is of critical interest to all stakeholders in basic education as students will not
be capable of working successfully at the higher grades if they do not master these
skills. Table 3 presents the results of testing done in maths and reading.
The results for reading and maths mimic the overall scores quite closely. Apart
from grade two, maths is the subject for which there is the greatest difference
between control and experimental groups with a 27–33 per cent differential. The
differential in reading is relatively constant at 23–24 percentage points. Looking at
the experimental schools, we note a very slight tendency for the reading scores to be
improving through the three grades along with a slight tendency for the maths scores
to be decreasing. It will take more years of testing to determine whether this is a
genuine trend or a statistical accident.

Second language acquisition

One of the most persistent objections to the use of mother-tongue instruction is the
assertion – or the assumption – that when children are being taught by means of

Table 3 Comparison of test results in maths and reading

Controls Experimentals Statistics

N Mean Percentage N Mean Percentage T-test P

Grade 1
Reading 73 8.97 52.8 69 12.83 75.5 6.86 0.000
Maths 73 8.32 48.9 70 13.76 82.1 9.19 0.000
Grade 2
Reading 94 12.63 54.9 42 18.00 78.3 9.22 0.000
Maths 94 13.00 61.9** 42 16.86 80.3 7.79 0.000
Grade 3
Reading 60 7.47 53.4 56 11.09 79.2 7.16 0.000
Maths 60 7.42 49.5 56 11.43 76.2 7.25 0.000

** One school which has been designated as an experimental school but in which the teacher has for the
last two years chosen to use English rather than mother tongue in her classroom did much better in maths
than in any other subject, thus influencing performance on maths for the control schools for Grade 2

123
678 S. L. Walter, D. E. Dekker

Table 4 Comparison of performance of the control and experimental groups on tests of knowledge in
Filipino and English, the two second languages being taught in the Lubuagan schools
Controls Experimentals Statistics

N Mean Percentage N Mean Percentage T-test P

Grade 1
Filipino 73 6.85 57.1 70 8.21 68.4 3.42 0.001
English 73 8.97 52.8 70 12.31 72.4 5.88 0.000
Grade 2
Filipino 94 7.79 51.9 42 12.21 81.4 14.88*** 0.000
English 94 12.63 54.9 42 14.29 62.1 4.89 0.000
Grade 3
Filipino 60 12.39 62.9 56 14.12 70.6 2.32 0.022
English 60 7.47 53.4 56 10.80 77.1 6.92 0.000

*** The difference in Grade 2 is exaggerated by the fact that the weakest of the experimental schools was
treated as a control school in Grade 2 because that is the way it had functioned for this and the previous
year

their first language, valuable time is being lost in learning the second language
which is commonly the language of higher education, commercial activity and the
political world. In most Philippine schools, two second languages are being taught –
Filipino and English – making ‘‘the language issue’’ even more critical in terms of
policy. Table 4 compares the performance of control and experimental groups of
children on tests of their ability in these two languages.
Table 4 provides clear evidence that the use of the primary language as a language
of instruction is not compromising children in learning the second language.
Contrary to popular belief, the children who are receiving most or all of their
instruction via English did consistently worse on a test of knowledge of English than
did the experimental children. Furthermore, the difference was even greater in grade
three than in grade one though there is not sufficient evidence to claim that the
differential has been increasing through the grades. The same advantage exists for
learning Filipino, though the difference is much less pronounced between the two
programmes (the result in grade two is probably a chance anomaly).

Performance by programme and school

If/when an educational intervention has systematic impact and the intervention has
been introduced at the level of full classes or schools, we can expect to see evidence
of this impact in a comparison of experimental and non-experimental (or control)
schools. Table 5 presents the relevant data for the schools tested as part of the
Lubuagan assessment.
The data in Table 5 are clearly suggestive but not totally unambiguous. Two of
the three experimental schools consistently ranked at the top in all three grades.
Similarly, the bottom two positions were consistently occupied by control schools

123
Mother tongue instruction in Lubuagan 679

Table 5 Ranking of schools participating in the Lubuagan assessment

Grade 1 Grade 2 Grade 3

School Mean School Mean School Mean

LCS (X) 63.62 Ag-agama (X) 78.22 Ag-agama (X) 75.52

Ag-agama (X) 61.75 LCS (X) 73.42 LCS (X) 62.18
LCS (C) 58.29 Mabilong (C) 60.05 LCS (C) 55.11
Mabilong (X) 47.08 LCS (C) 57.48 Mabilong (X) 44.94
Uma (C) 33.21 Dangoy (C) 54.58 Uma (C) 40.64
Dangoy (C) 32.39 Uma (C) 46.19 Dangoy (C) 39.04

Note X stands for experimental group, C stands for control group

whose level of performance tended to be approximately half that of the top

experimental schools.
Are the results in Table 5 due to the effects of the intervention or the product of
significant differences in quality of instruction in the schools involved? From the
researchers’ first-hand knowledge of the schools, we know that there is significant
variation in the quality of instruction and adherence to the intervention in at least
one of the schools. At the same time, data from similar research in Cameroon
(Walter 2011) provide strong evidence that the intervention (use of L1 as a medium
of instruction) accounts for much of the increased performance. In the Cameroonian
study all 12 experimental schools outscored all 12 control schools although the
differential was minimal between the weakest of the experimental schools and the
strongest of the control schools.

Discussion of findings

The test results for 2008 show a consistent advantage for children in the
experimental programme (MLE) across all three grades and all subjects in the
curriculum, although the advantage varies considerably from grade to grade and
subject to subject. How compelling is the advantage? (How strong the effect?)
Statistically, the advantage is large (t = 7.08; P \ 0.000 for Grade 3). In
distributional terms, the advantage is 1.1 standard deviation, a very large differential
for an educational innovation. In terms of relative gains in educational efficiency
(measured solely by means of test scores), the gain is approximately 40 per cent
with the largest gains being seen in the most core subjects (48.3 per cent for reading
in grade three; 53.9 per cent in grade three maths; 67.8 per cent in grade one maths).
These findings thus provide strong initial evidence that the use of local languages
for instructional purposes, instead of compromising, actually enhances mastery of
curricular content including in the more critical areas of maths and science (second
and third research hypotheses).
Even in English, where the expectation is that children in schools receiving all of
their instruction in English would do well, children in the experimental programme
registered a 44.4 per cent advantage in performance. In fact, children in Grade 3

123
680 S. L. Walter, D. E. Dekker

appear to be ‘‘pulling away’’ from their peers in the English-medium schools since
the gain in Grade 1 was 37.2 per cent and in Grade 2 it was 13.1 per cent. Again, we
find that the early use of L1 as a language of instruction has not compromised the
development of proficiency in English, but rather appears to provide positive
support for the development of such proficiency (first research hypothesis).
The MLE programme currently provides only three years of instructional support
in the first language. While the data from the recent testing certainly provide strong
evidence of the ability of MLE to produce significant gains in educational efficiency,
students will need to be tracked for several more years to determine whether the
effect is lasting and, if so, how strong it might be. The research (and theorising) done
by Thomas and Collier (1997) suggests that the impact will persist but will be less
than if mother tongue support were being provided for at least six years.

Possible sources of confounding

For a number of reasons, we also need to exercise some caution in our interpretation
of these results. First, the sample size is small in this programme. At present, the
programme in Lubuagan includes three control schools and three experimental
schools. In a programme of this size, a good year or a bad year by a single school
can significantly affect outcomes.
Second, we must note that variation in teacher quality can significantly affect
findings. Other research (Walter and Davis 2006) has shown that it is quite common
to encounter very large variations in teacher quality from one school to another. In
combination with a limited sample, this situation can produce considerable
confounding in experimental results.
Third, informal inquiry has established that teachers vary in their use of language
in the classroom. If one teacher does a lot of code-switching (between English and
Lilubuagen) in an MLE classroom while another carefully follows prescribed
practice (for language use), then the impact of this experimental variable becomes
more difficult to assess.

Conclusion

The testing done in 2008 found an approximate gain of 40 per cent in educational
outcomes (test scores). Several major questions present themselves in response to
this finding. How significant is this level of improvement? How does this finding
compare to gains (or losses) in similar programmes in similar countries? Will this
gain persist? Are further gains possible with additional tweaks to the existing
educational system?
In a similar study being carried out in Cameroon, Walter and Trammell (2008)
found average gains of over 200 per cent for grade one children being instructed in
the local language. In 2009 the reported gains were approximately 125 per cent for
grade one and 60 per cent for grade two. Gains were the greatest in reading and
maths and less in mastery of the second language (the language of instruction in the
control schools).

123
Mother tongue instruction in Lubuagan 681

In Jacob Cohen’s classic work on power and effect size in statistical analysis, an
effect size of 0.8 is considered to be large (Cohen 1988). The effect sizes observed
in the Lubuagan programme range from 1.31 to 1.61, indicating very large effect
sizes for the mother tongue innovation.
A striking feature of the Lubuagan research data is the variation in teacher
quality. The students of high-performing teachers outscored those of low-
performing teachers by as much as 70 per cent (within each of the instructional
models). In the Cameroonian research (Walter and Trammell 2008), the differences
were even greater – as much as 100 per cent – again with similar levels of variation
observed within instructional models. Comparable results were found in research
done in Eritrea (Walter and Davis 2005).
A tentative emerging hypothesis is that the greatest effect size for L1 instruction
in the early grades is to be realised in those contexts in which the existing quality of
educational delivery is the weakest. Only further and broader research will confirm
or disconfirm this hypothesis and the larger ones which motivated the Lubuagan
experiment in education.

References

Baker, C. (1996). Foundations of bilingual education and bilingualism (2nd ed.). Clevedon, UK:
Multilingual Matters.
BESRA (Basic Education Sector Reform Agenda). (2006). Retrieved October 10, 2011 from
http://www.fwwpp.org/BESRA.html.
Brigham, S., & Castillo, E. (1999). Language policy for education in the Philippines. Technical
background paper, no. 6, Philippines Education for the 21st Century—the 1998 Philippines
Education Sector Study. Manila: Asia Development Bank and World Bank.
Caoli-Rodriguez, R. (2007). The Philippines country case study (Country profile commissioned for the
Education for All Global Monitoring Report 2008. Education for All by 2015: Will we make it?).
Paris: UNESCO.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale: Lawrence Erlbaum
Associates.
Commonwealth. (1940). Commonwealth Act No. 570, 1940. Retrieved October 10, 2011 from
http://www.librarylink.org.ph/featarticle.asp?articleid=87.
Cruz, I. (2004). Reengineering Filipino, English and the lingua franca in basic education. Philippine
Journal of Linguistics 34(2)–35(1), 61–69.
Dekker, D. E., & Dumatog, R. (2004). First language education in Lubuagan, Kalinga, northern
Philippines. Paper presented at the conference on language development, language revitalization
and multilingual education in minority communities in Asia, Bangkok, Thailand. Retrieved October
10, 2011 from http://www.sil.org/asia/ldc/parallel_papers/dumatog_and_dekker.pdf.
Dekker, D., & Young, C. (2005). Bridging the gap: The development of appropriate educational strategies
for minority language communities in the Philippines. Current Issues in Language Planning, 6(2),
182–199.
DECS (Department of Education, Culture and Sports). (1999). Memo No. 144 s. 1999. Lingua franca
education project. Retrieved October 10, 2011 from http://www.deped.gov.ph/cpanel/uploads/
issuanceImg/DM%20No%20144%20s%201999.pdf.
DECS (Department of Education, Culture and Sports). (2000). Memo No. 243 s. 2000. Training workshop
batch 2 on the lingua franca education project. Retrieved October 10, 2011 from http://www.deped.
gov.ph/cpanel/uploads/issuanceImg/DM%20No%20243%20s%202000.pdf.
DepEd (Department of Education). (2002). Order No. 43s 2002. The 2002 basic education curriculum.
Retrieved October 10, 2011 from http://www.deped.gov.ph/cpanel/uploads/issuanceImg/DO%2043_
08-29-02_00001.pdf.

123
682 S. L. Walter, D. E. Dekker

DepEd (Department of Education). (2003). Memorandum #189s. 2003. Establishing the policy to
strengthen the use of the English language as a medium of instruction in the educational system.
Retrieved October 14, 2011 from http://www.deped.gov.ph/cpanel/uploads/issuanceImg/DM%
20189_06-19-03_00001.pdf.
DepEd (Department of Education). (2009). Order No. 74 2009. Institutionalizing mother tongue-based
multilingual education (MLE). Retrieved October 10, 2011 from http://www.deped.gov.ph/cpanel/
uploads/issuanceImg/DO%20No.%2074,%20s.%202009.pdf.
Dutcher, N. (2004). Expanding educational opportunity in linguistically diverse societies. Washington,
DC: Center for Applied Linguistics.
EDCOM Congressional Commission on Education. (1991). Making education work: An agenda for
reform. Manila, Quezon City: Congress of the Philippines.
Fishman, J. A., Conrad, A. W., & Lopez, A.-R. (Eds.). (1996). Post-imperial English: Status change in
former British and American colonies, 1940–1990. Berlin: Mouton de Gruyter.
Gonzalez, A. (1998). The language planning situation in the Philippines. Journal of Multilingual and
Multicultural Development, 19(5 & 6), 487–525.
Gonzalez, A. (2001). Looking at some of DECS’ problems from a linguist’s point of view. Bonifacio P.
Sibayan Professorial Chair Lecture, De La Salle University, Manila
Krashen, S. (1991). Bilingual education: A focus on current research. Occasional Papers in Bilingual
Education 3. Washington DC: National Clearinghouse for Bilingual Education.
Krashen, S. (2000). Bilingual education, the acquisition of English, and the retention and loss of Spanish. In
A. Roca (Ed.), Research on Spanish in the US: Linguistic issues and challenges. Somerville, MA:
Cascadilla. Retrieved October 10, 2011 from http://www.languagepolicy.net/archives/Krashen7.htm.
Maligalig, D. S., & Albert, J. R. (2008). Measures for assessing basic education in the Philippines. In
A. Bernardo (Ed.), The Paradox of Philippine education and education reform: Social science
perspectives. Manila: Philippine Social Science Council.
Nolasco, R. M. (2008). The prospects of multilingual education and literacy in the Philippines. In
A. Bernardo (Ed.), The paradox of Philippine education and education reform: Social science
perspectives. Manila: Philippine Social Science Council.
PCER (Philippine Commission on Educational Reform). (2000). Philippine agenda for educational
reform: The PCER report. Manila, Philippines.
PCSPE (Presidential Commission to Survey Philippine Education). (1970). Education for national
development; new patterns, new directions. Rizal, Philippines.
PESS (Philippines Education Sector Study). (1998). Philippine education for the 21st century. The 1998
Philippines education sector study. Manila: Asian Development Bank and World Bank.
Philippine Commission. (1901). Act No. 74 of the Philippine Commission, 21 January 1901. Retrieved
October 10, 2011 from http://philippinelaw.info/statutes/act74-education-act-of-1901.html.
Philippine Constitution. (1935). Retrieved October 10, 2011 from http://www.chanrobles.com/
1935constitutionofthephilippines.htm.
Philippine Constitution. (1973). Retrieved October 10, 2011 from http://www.chanrobles.com/1973
constitutionofthephilippines.htm.
Philippine Constitution. (1987). Retrieved October 10, 2011 from http://www.chanrobles.com/
philsupremelaw1.htm.
Quisumbing, L. R. (1987). The 1987 policy on bilingual education. Department of Education, Culture and
Sports Order No. 52. Manila: DECS.
Sibayan, B. P. (1985). The Filipino people and English. In K. R. Janowsky (Ed.), Scientific and
humanistic dimensions of language: Festschrift for Robert Lado (pp. 581–593). Amsterdam: John
Benjamins.
Skutnabb-Kangas, T. (2000). Linguistic genocide in education—of worldwide diversity and human rights.
Philadelphia, PA: Lawrence Erlbaum.
Thomas, W. P., & Collier, V. (1997). School effectiveness for language minority student. Retrieved
October 10, 2011 from http://www.ncela.gwu.edu/files/rcd/BE020890/School_effectiveness_for_
langu.pdf.
Thomas, W. P., & Collier, V. (2002). A national study of school effectiveness for language minority
students’ long-term academic achievement. Santa Cruz, CA, and Washington DC: Center for
Research on Education, Diversity and Excellence. Retrieved October 10, 2011 from http://crede.
berkeley.edu/research/llaa/1.1_final.html.

123
Mother tongue instruction in Lubuagan 683

Walter, S. L. (2011). Mother tongue-based education in developing countries: Some emerging insights.
Retrieved October 13, 2011 from http://www.educationfasttrack.org/media/library/blog/Steve-
Walter-Mother-TongueInsights.pdf.
Walter, S. L., & Davis, P. (2005). Eritrea national reading survey—September 2002. Dallas, TX: SIL
International.
Walter, S. L., & Trammell, K. R. (2008). Mother tongue education in Kom (Cameroon)—a first report.
Research report presented to the Ministry of Education in Cameroon.
World Bank. (1988). Education in Sub-Saharan Africa: Policies for adjustment, revitalization and
expansion. A World Bank Policy Study. Retrieved October 10, 2011 from http://www.eric.ed.gov/
ERICWebPortal/search/detailmini.jsp?_nfpb=true&_&ERICExtSearch_SearchValue_0=ED312205
&ERICExtSearch_SearchType_0=no&accno=ED312205.
Young, C. (2002). First language first: Literacy education for the future in a multilingual Philippine
society. International Journal of Bilingual Education and Bilingualism, 5(4), 221–232.

The authors

Stephen L. Walter is Associate Professor and chair of the Language Development Department of the
Graduate Institute of Applied Linguistics in Dallas, TX. In addition, he is an International Literacy and
Education consultant for SIL International. His current research interests are primarily focused on
multilingual education in developing countries. His most recent publication is: The language of
instruction issue: framing an empirical perspective, in Spolsky and Hult, (eds). The Handbook of
Educational Linguistics.

Diane Dekker is currently a MTBMLE Consultant for SIL International in the Philippines. Her interest in
language and education issues led her to begin research in 1988 on mother tongue-based multilingual
education among the Lubuagan people in the Philippines. Her most recent publication is Current Issues in
Language Planning 6(2) (2005). Her collaborative research with Steve Walter was presented to the
Philippine Congressional Hearing on Education in 2008.

123
View publication stats