Mother Tongue Based - Multilingual Education : A Barrier to Cognitive Development?

Perhaps, children need to be taught to think first.
The desire to make classrooms enjoyable for children can work against education if the fact that learning requires work is ignored. Cognitive development is a process, and like other processes, requires both time and energy. Learning is rewarding and with children, the effort alone is integral to developing the brain. Without challenges, the brain will not master thinking processes. Making classrooms enjoyable means equipping the schools with the necessary materials and inputs for education, supporting adequately the teachers. It means creating an environment that is physically conducive to learning. It does not mean lowering the standards and expectations.

Having the mother tongue in school is important as it provides children a sense of familiarity, easing their transition from home to school. Learning to read and write in the mother tongue as a formal subject in the early years is, without doubt, important. Using the mother tongue as a medium of instruction is the questionable part. Claims have been made that children learn faster and more effectively when instruction is done in the mother tongue. Indeed, browsing through the internet, one will arrive at numerous articles that either support or dismiss this notion. For this reason, it is important to examine the evidence behind these claims. The lack of learning materials on science and math in the mother tongue requires a serious inspection of how the medium of instruction affects children in their cognitive development. Adopting mother tongue based - multilingual education requires substantial investments in money, time and energy.

To understand the role of the medium of instruction on learning, a deeper examination of what constitutes learning is required. While papers on how the medium of instruction affects learning abound, I would like to focus on a recent one published in the Journal of Education and Practice:
Specifically, the article of interest is
Role of Medium of Instruction on the Development of Cognitive ProcessesPDF
Pritimayee Senapati, Nirlipta Patnaik, Manaswini Dash60-68

The assessment of the students in this published study included the following tests (pp. 62-63):
Matching Numbers. This is a measure of planning. The test comprises rows of numbers. Each
row contains six numbers. The child’s task is to find and underline the two numbers that are the same in each row within a specified time limit. Each correct pair of matching numbers gets a score of ‘1’. The raw scores are transformed to a ratio score using the score conversion table.
Planned Connection. This is a measure of planning. In this test, the child’s task is to connect a
series of boxes containing numbers or letters in correct sequence. Some items involve a sequence of numbers only, while others involve a sequence of both numbers and letters (that is 1-a-2-b-3-c, etc.).Time taken to complete each item is recorded. The sum of the time taken for each item is the raw score. 
Non-verbal Matrices. This test of simultaneous processing requires the child to select one of the
options that best completes the matrix. The subtest uses the standard progressive matrix format and varies from completion of a simple pattern to completion of a 3×3 matrix of stimuli. A score of ‘1’ is given to each item passed. 
Figure Memory. This is a test of simultaneous processing which requires the child to identify a
geometric figure that is embedded within a more complex design. The stimulus figure is exposed for five seconds. The child is required to reproduce the same figure within a more complex design presented on a separate page. Each correct response obtains a score of ‘1’. 
Expressive Attention. This is a measure of attention. The child’s task is to read colour names
such as ‘blue’ and ‘yellow’ (item -1), to identify the colours of a series of rectangles(item -2) and then to identify the colour of the ink in which colour words are printed rather than to read the words (item 3). Basing on the correct number of responses in the item-3 and the time taken to complete it, the score is obtained using the conversion table. 
Receptive Attention. This test of attention requires the child to find and underline pairs of letters
that are the same from among rows of letters which contain both targets (pairs that match) and distracters (pairs that do not match). The child has to underline pairs of letters that are physically the same and have the same name. The raw score constitute the sum of ratio scores for both the physical as well as name match condition which are obtained from number of correct as well as the false detections and the time taken to complete each item using the ratio score conversion table. 
Word Series. This test measures successive processing. Here the child’s task is to repeat a series of words in the same order in which the examiner says them. A score of ‘1’ is given for correct recall of the words in their correct order. 
Sentence Repetition. This is a test of successive processing in which the child’s task is to repeat a series of sentences spoken by the examiner. The sentences contain colour names in place of content words. Perfect repetition of the sentence obtains a score of ‘1’.
And the authors present their results in the following table (p. 67): 
The paper also includes an analysis of variance (ANOVA). The authors share their discussion and conclusions from the study in the paper. One of their main conclusions is that "English Medium schooling has been found to facilitate the development of the cognitive processes studied." And their explanation (p.64):
"...Getting education through English involves differential curricular pressure and cognitive demands on the part of the students. The student has to understand the instructions presented in English which is not his mother tongue, to develop linguistic competence in it and simultaneously master the course content. For Odia medium children the task is much simpler as they have to master the course content only using a language already acquired earlier at home. However, this tougher task demand at an early age helps in faster development of cognitive processes in the children being educated through English medium...."
Statisticians could probably do additional analysis of the data presented in this paper but a cursory look at the numbers already reveals that the claim that instruction using the mother tongue is more effective may not be generally true. 

Closer to home, Rudy Coronel writes in his blog: 


"....The MTB-MLE is allegedly based on studies that showed that children who were taught in their dialect learned better than those who were taught in their second language, which is either Filipino or English.  I really do not know in which particular region the said study was conducted.  For its result  to be fair and conclusive, was there also a parallel study conducted in schools in the Tagalog region? 
I mean, one is made to understand that, according to the study, Bicolandia children who were taught via the Bicol dialect, for example, had learned better than when they were taught in English or Pilipino.  But how conclusive is the study and its findings?  Does it also apply in the Tagalog region, for example.  That is to say, the mother tongue is Filipino in this region, where the medium of instruction has always been Filipino in the lower grades. Has there also been a study showing that children in Bulacan, Laguna, Quezon and Batangas have generally been fast and better learners than their counterparts in the Bicol or Ilocos region?  I don't think and I refuse to believe so.  I think that people from the DEPED themselves will in good conscience admit the plain truth that, taking all other variables equal, our public school children in the Tagalog provinces are just generally in the same degree of learning capabilities and aptitudes as their counterparts in the non-Tagalog regions.  More clearly put, Kinder-to-Grade III children in the Tagalog regions are definitely not as better prepared for Grade IV and onwards than those from any other region.  To contradict this plain truth is not unlike contradicting human nature itself.   If there is ever a difference, I believe it is not because of the medium of instruction per se -- maybe the quality of teachers and the like...." 
Well, the following table provides the results of the National Achievement Test in 2005 and the Eastern Visayas region students scored the highest in Filipino:

And from "First Things First: A Commentary on DepEd's K +12";
...It is unfortunate that amidst the lack of sound evidence, although this paucity in data has been emphasized and repeated so many times in published reviews and articles, various components have been incorporated in the K+12 plan with “panacea” stamped on them.  The following paragraphs highlight specific examples.
The mother tongue based multiple language education (MTBMLE) is one example. In 2009, the US Supreme Court issued an opinion (Horne vs.Flores) that Structured English Immersion (SEI) works better than bilingual education.  It was a narrow decision (5 against 4) so it is not a clear judgment against MTBMLE, but it sure is a clear sign that MTBMLE is not “panacea”.  Recent news from the state ofCaliforniaalso indicates that multilingual education is likewise not working well (see “English-Learning Students Far Behind Under English-Only Methods”). 
The world experts in MTBMLE are careful in promoting MTBMLE.  To make a strong case in favor of MTBMLE, data must show that high dropout rates are unquestionably due to using a second language as medium of instruction (Smits et al., 2008). I strongly recommend taking a closer look at Table A.1 of this study by Smits et al. because thiscontains data pertinent to the Philippines. Specifically, the paper states: “The figures presented in columns 4 and 8 of the table give an indication of the part of the attendance differences that is due to differences in the background characteristics. For both age groups the reduction is 25 percent or more in 13 of the 22 countries. So in the majority of countries the background characteristics play a role of importance. This result provides support for hypothesis H1.” Hypothesis H1 of this paper is “The differences in educational outcomes among linguistic groups are (partly) due to socioeconomic differences and/or differences in urbanization of the place of living among the groups.” The Philippines lists 45 and 48% in columns 4 and 8, respectively. In this light, the Philippines is among the three odd countries listed that show very strong correlation between school retention and socioeconomic factors, the others are Ghana and Peru.  In Table B1, page 41 of the paper, data from the Philippines clearly suggest that the various language groups in the country do not differ from each other in a significant manner in terms of dropout rates....


  1. i dont understand the last bullet. if half of the dropout rate is due to socio-economic factors, then the rest might be due to language. further, the fact that the different language groups arent too different is evidence for mother tongue -- language of instruction is either english or filipino, neither of which is a mother tongue.

  2. The table in the study presents the difference in dropout rates among various linguistic groups. It is not the total dropout rate. Thus, the 45% that remains is not 45% of the dropout rate, but 45% of the difference in dropout rates between language groups. The differences in dropout rates among language groups is smaller than the actual dropout rates. In addition, Tagalogs who are taught using Tagalog (Flipino) as medium are not seeing a significantly lower dropout rate than the other groups.

  3. it doesnt matter. i can re-write the objection as " if half of the (difference in) dropout rate(s) is due to socio-economic factors, then the rest might be due to language"

    "Tagalogs who are taught using Tagalog (Flipino) as medium are not seeing a significantly lower dropout rate than the other groups."

    is that in the paper? thats important to highlight it if were.

  4. You can rewrite your question, but bear in mind that "half of the (difference in) dropout rate(s)" is a much smaller number than "half of the dropout rates". You can find the raw data for the various language groups in Tables 3 and 4 in page 72 of the Smits et al. paper. The odds that a Tagalog pupil will dropout is 1.25 in primary school, while for a Cebuano, it is only 0.79, the average odds for all the other language groups is 1.02. The odds are much higher for Tagalogs to drop out of school.

  5. To illustrate a case that is visibly a language oppression, look at the case of Nepal. The odds that a Nepali will dropout from primary is 0.27, contrast this with the other language groups Maithili and Bhojpuri, which have odds at 2.16 and 2.26, respectively.

  6. this should point to the weirdness of this paper. in Table A1 they stress 7-11 and 12-16 analysis, but its not clear how A1 is derived from B4. Even in B4, the figures do NOT correspond to the primary/secondary education columns, when they should! its weird; dont know what to make of it.

    the average(d) estimates in table A1 show significant variation potentially attritutable to language, unexplained by their soco-econ factors -- approximately half. thats huge.

    the second objection is, even for Tagalogs, english is not their mother tongue. the most important subjects -- the one that are internationally tested and compared, are taught in english. Arguably, the bilingual advantage that tagalogs supposedly possess are muted.

  7. It is a UNESCO paper - reporting on EFA. Your second objection does not make sense since the report is not dealing with international tests or subjects, it is dealing strictly with school dropouts. The whole thing need to be read so that you would understand what the paper is about. School dropouts in the Philippines are dominated by socio-economic factors (child labor, poverty, and low educational attainment of mother - language is not the major factor).

  8. actually, its the opposite. they explicitly say soci-econ is not a major factor. i can pull out the exact quote if u want.

  9. That is true for some language groups, like in Nepal. It probably holds true as well for some IP in the Philippines, but not for the major languages in the Philippines. That is why I specifically quoted the paragraph that shows that in the Philippines, socio-economic factors are dominant. Other studies on school dropouts in the Philippines indicate this as well. Child labor is high in the Philippines - this is a major factor in school dropouts. Poverty and malnutrition are high - these are important factors.


Post a Comment