Logo Congreso

El Congreso Puertorriqueño de Investigación en la Educación es un evento que se celebra cada dos años en la Facultad de Educación de la Universidad de Puerto Rico, Recinto de Río Piedras. Desde sus inicios, su compromiso ha sido difundir la investigación en la Facultad de Educación, en el país y en la región del Caribe con el propósito de aportar a la educación puertorriqueña.

Centro de Investigaciones Educativas
Facultad de Educación
Universidad de Puerto Rico
Recinto de Río Piedras
P.O. Box 23304
San Juan, PR 00931-3304
email: cie.educacion@upr.edu
tel. (787) 764-0000, x. 4382, 4383, 4384, 4385
Fax (787) 764-2929


X Congreso Puertorriqueño de Investigación en la Educación

11, 12 y 13 de marzo de 2009


Dual Language Education: Lessons for Puerto Rico

Dr. Susana C. De Jesús
University of Puerto Rico, Río Piedras Campus

Palabras clave: palabras

Keywords: keywords

Cómo citar este artículo (estilo APA) / Citing this article (APA style):
De Jesús, S.C.. (2009). Dual Language Education: Lessons for Puerto Rico. Ponencia presentada en el X Congreso Puertorriqueño de Investigación en la Educación. Facultad de Educación, Universidad de Puerto Rico, Río Piedras. Recuperado de http://cie.uprrp.edu/congreso/2009/



Newspapers and media are often filled with dire reports of poor public schools and low functioning students, both in Puerto Rico and the United States. Yet, there are programs which run counter to the norm, and rarely receive publicity. Dual Language programs, sometimes called Dual immersion programs, or Two-way Dual Language programs, are “astoundingly” successful in the US and can be found flourishing in other countries in the world (Collier & Thomas, 1997a, 1997b, 2002, 2003, 2004; Cummins, 1999; De Jesús, 2008, 2009; Genevese, 1987, 1999; Howard et al. 2005; Lindholm, 1990, 1997; Lindholm-Leary, 2006; Torres-Guzmán, & Etxeberria, 2005).  This paper will summarize the program design and benefits of Dual language instruction, will briefly focus on one highly successful program which has been in operation for over 18 years, will then identify some important educational issues in Puerto Rico, and finally, will suggest possible options that might address and ameliorate some of those issues.

Dual language program design

Dual Language programs teach two languages effectively and simultaneously, by carefully following a specific program design, and systematically rotating instruction between the two languages. The method will work with any two languages: Spanish and Chinese, French and English, or Dutch and Papiamentu, for example. In the United States, at least, the most commonly needed languages are English and Spanish, as the majority of English Language Learners (ELL) in public schools are speakers of Spanish, and there is broad interest, for practical reasons, among English only students from the dominant culture, as well as Heritage language students from Latino families, to become bilingual and biliterate in Spanish.

Usually two teachers are paired together —each one teaching exclusively in one of the target languages. The students move from classroom to classroom, between the “two worlds”. Occasionally, because of class size requirements or the size of the eligible cohort, one bilingual teacher has a self-contained class, and s/he alternates instruction between the two languages.

To achieve a high functioning and successful Dual Language program, there are certain basic axioms —requirements of program design— that absolutely must be followed. Sometimes programs call themselves Dual Language, but do not follow the design that brings extraordinary success. Flawed methods can be somewhat successful, but when the program design is carried out rigorously and conscientiously, the results are “astounding” (Collier & Thomas, 2002, 2003, 2004; De Jesús, 2008, 2009).  The critical axioms are as follows:

When a single bilingual teacher works alone, and s/he rotates instruction between the two languages, this may seem easier, as it may appear to reduce some of the difficulties of joint planning between two teachers. However, the self-contained classroom usually does not function as well as the two classroom design. First, it is exhausting for the teacher —they burn out— and most teachers cannot do it for more than one or two years. Second, since most individuals are slightly dominant in one language over the other, behind closed doors there is often a tendency to revert to the teacher’s preferred language. This clearly destroys the 50:50 language distribution, the total separation of languages, equality of time, and the balance between subject matter and language. Further, if the teacher is an English speaker, and the children like her or him, they may unconsciously transfer this positive feeling, and “prefer” English. They may feel it is a better language and that they really do not need to learn Spanish.

Minor variations in these axioms of program design may seem trivial to teachers and administrators. To some it may appear inconsequential to slightly adjust certain aspects, especially if the changes may seem unimportant or logical. However, this is absolutely not the case, especially if teachers revert to their habits, attitudes or stereotypes without even realizing that they are making changes in the program design. But, small changes can be important and can negatively impact student achievement. Documentation from the national longitudinal study conducted by Collier and Thomas shows that a seemingly insignificant change in program design had a cumulative effect of reducing the rate of growth from 6 to 3.5 NCEs annually (Collier & Thomas, 2004, p. 13). In another study, teachers began grouping students within the classroom and teaching ESL to the Spanish dominant students, and English Language Arts to the English only dominant culture students. These teachers felt the change was “logical” because it was in tune with their old habits and attitudes, and did not realize they were changing the program design. However, this “small” change and “logical” adjustment negatively affected overall achievement. It reduced the rate of reading growth in those classrooms (De Jesús, 2008, p. 200). While students made progress, it was not the dramatic improvement made by students in other classrooms, or the progress expected in a Dual. In the rooms where this simple and “logical” change was made, the students outperformed their school peers, but not their district peers. Students in the other classrooms did.  The lesson is clear: little changes can have a huge impact. Teachers and supervisors need to understand the program design, not simply follow by rote. They must be in communication, must scrutinize and self-reflect, even when “good” results are not good enough.

The benefits of dual language instruction

As a benchmark of the extraordinary levels that can be reached in a Dual Language program, Figure 1 shows the pattern of achievement for English Language Learners in NCEs on standardized tests (in English reading) aggregated from 4-8 year longitudinal studies (Collier & Thomas, 1997).  It is important to underscore that the students represented in this data are English language learners, not native speakers of English. They began school in the U.S. with no proficiency in English. In Reading, on tests administered in English, these students achieved at the 61 NCE level and outperformed their nativeEnglish speaking peers by 11 NCEs. Figure 1 indicates their achievement by the highest arching line on the graph. The dotted line at the 50 NCE level indicates the average achievement of native-English speaking, dominant culture students from mainstream general education programs, excluding Special Education students. 

Figure 1

Figure 1

Students in a Dual language program routinely outperform their school and district peers. They normally reach the highest state levels of expected proficiency in Reading and Math. This is extremely important, since the demographic profile of Dual language programs indicates that the ELL students serviced by Duals are typically low SES, from poor and immigrant communities (Robledo Montecel et al, 2002). This outstanding achievement is not predicted by the literature regarding expected academic outcomes and graduation rates for low SES or immigrant students.  Further, in their study of second generation immigrant students, Portes and Rumbaut showed that ELLs who are not fluent bilinguals are among the lowest academic achievers in the US, among those most likely to drop out of school, and most likely to become marginalized and trapped in a permanent underclass. In their study, fluent bilingualism —an automatic consequence of Dual language programs— was shown to be the marker for high school graduation, future educational attainment, social adjustment and economic success. (Portes & Rumbaut, 2001).

Collier and Thomas published their national, longitudinal study in 2002, which examined the school records of over one million students, from five school districts, in different areas of the United States —urban, rural, south, east, west— and found that the Dual language program was the only bilingual option, including the highly popular ESL, that could raise the level of academic performance of English Language Learners, and bring them up to par with, or to a  level higher than, the academic achievement of dominant culture mainstream students. In other words, the Dual program, when carried out rigorously and conscientiously, is the only programcapable of eliminating or vastly reducing the Achievement Gap that exists in the US between students who are poor, urban and English language learners, and their dominant culture peers.

Furthermore, and perhaps most surprisingly, Dual language programs benefit all students in the treated cohort, both language learners and dominant culture students alike. Students from the dominant culture, who participate in Dual language programs, outperform their dominant culture peers in Reading and Math, on standardized achievement tests. Thus, all Dual language students can achieve a level of academic performance unattained by students from other programs. This aspect of Dual language instruction has important implications for Puerto Rico, as discussed below.

This extraordinary pattern of achievement is likely the result of the cognitive impact of learning in an immersion environment, the cognitive stretch (De Jesús, 2008, pp. 208-209; De Jesús, 2009) for all students, on both sides of the Dual – ELL  and ED alike. There is ample literature, going back 40 years or more, to document the consistent high achievement of all students in Dual Language programs (Collier & Thomas, 1997a, 1997b, 2002, 2003, 2004; Cummins, 1989, 1999; De Jesús, 2008, 2009; Genevese, 1987, 1999; Hakuta & Diaz, 1984; Howard et al, 2005; Lindholm, 1990, 1991, 1997; Lindholm-Leary, 2005, 2007; Ramirez et al, 1991; Torres-Guzmán et al, 2005; Torres-Guzman & Etxeberria, 2005). For further information, see the Center for Applied Linguistics at http://www.cal.org, and the National Clearinghouse of English Language Acquisition at the George Washington University, Washington, DC, http://www.ncela.gwu.edu.

A case in point

A study in progress at this time is measuring the treatment effect of participation in a Dual Language program. This venue, a small urban district of about 3,000 students located near a large Northeastern city in the US, had some features that make statistical treatment of data possible. The district is fortunate to have a mature, well-run Dual Language program, currently in its 18th year of implementation. It has always required standardized testing in Reading and Math, but under No Child Left Behind (NCLB) exams are mandated in English for 4th graders in Language Arts/Literacy (Reading) and Math. Data was available for several years and is currently being analyzed. A sample of student outcomes, presented below, was disaggregated according to participation in one of the three programs offered by the district: general education, the Dual language program, and the Department of Education transitional bilingual program.

In this district, approximately 60% of the students are African American, about 35% are Latino, and 5% come from Caucasian, Asian, Native American or Middle Eastern backgrounds. Regarding the Socio-Economic-Status, SES, 64% of students qualify for Free or Reduced Lunch, which is commonly used as a marker for poverty/low income, as the district is not permitted to ask questions regarding family income or the parents’ educational level.

Figure 2 presents this demographic information. Noteworthy is that the students in the Dual language program were identified as poorer than their district peers – 77% of them qualified for free or reduced lunch, as compared to 64% in the district as a whole. Furthermore, district personnel estimated that at least 20% of the students in the Dual language program were thought to be from undocumented immigrant families. Teachers and staff working with these students and families may have an accurate estimate. If true, the figure suggests an overall high poverty level, and indicates that many students in the Dual language program were among those social and economic groups least likely to succeed (Portes & Rumbaut, 2001).

Figure 2. District and Program Demographics: Racial or Ethnic Background and SES

Background District
Combined Data
Dual Language
Caucasian, Middle Eastern
Asian, Native American 
Eligible for Free or Reduced
Price Lunch

This study consists of a cross section of 4th graders, with about 360 observations each year, for a 6 year period. Data on student outcomes for Language Arts/Literacy (Reading) and Math on the NCLB mandated achievement tests were obtained from the district for 2002-2007. Since the state changed the testing instrument, only some of the data can be compared. Some of the variables are: 1) Achievement Scores: Math and LAL exams were the dependent variables, 2) Gender, 3) Program: General Ed program, Dual Language, and Transitional Bilingual Program, 4) Language Dominance: Spanish or English, 5) SES: free lunch eligibility, and 6) Ethnicity. Ordinary Least Square (OLS) regressions were run, as well as Fixed Effect (FE) regressions, controlling for unobservable differences between schools, and statistical inferences are being drawn (DeJesús, 2009). A sample of the data is presented below, in the form required by NCLB, comparing Dual language program outcomes, with district and state achievement levels in percentage of students achieving proficiency levels, or higher scores.  

Figure 3. Percentage of students who achieved scores of Proficiency or more on mandated DOE achievement exams (in English) in the areas of Language Arts/Literacy and Math, as shown by Dual Language program, District and State achievement, 2005 to 2007.


This region is one of the higher achieving districts in the state, as typically between 65 and 78% of 4th grade students achieve proficiency or above in Language Arts and Math on the state instrument. The state levels are higher, with over 80 to about 87% achieving proficiency.

However, the Dual outcomes are extraordinary. Virtually 100% of all Dual language students achieve proficiency or higher each year, surpassing not only their district peers, but also the high rate of proficiency on the state level. This is particularly impressive given the district and program demographics: low SES students, a “minority” district, where more than 65% are on a poverty level, and where the vast majority —95% of the student population— comes from Latino or African-American families. No other program in this district, or in the state, including the affluent districts, is able to produce high scoring students consistently, year in and year out. Thus, when a Dual language program design is carried out rigorously and conscientiously, these are the results that show the “astounding effectiveness of dual language education” (Collier & Thomas, 2004), which eliminates the achievement gap between second language learners and their dominant culture peers, and benefits all students in the treated cohort.

Some observations regarding achievement in the public schools of Puerto Rico

According to archives of the US government, NCLB has achieved the following goals:

School Report Cards and district assessment information is supposed to be easily available to parents through a link on the Puerto Rico Department of Education website. But, the link marked NCLB leads to information on the National Assessment of Educational Process (NAEP) exam, not achievement under NCLB, and the data refers back to 2003 and 2005. The link for School Report Cards leads to another page, which gives samples of achievement data – not the School Report Cards – for individual schools in Puerto Rico, but not the District of Puerto Rico. Another link leads to a non-governmental organization giving data from districts and states, but Puerto Rico is not among them. At this point, the average parent might give up in frustration.

Another difficulty in assessing the achievement levels in Puerto Rico is that each state or territory selects its own assessment instrument, and establishes its own level of proficiency. In other words, what is considered proficient in Mississippi might not be considered so in New York. Comparisons among states and territories cannot be made because the measures are all different. The only national measure is the National Assessment of Educational Progress (NAEP). This test is considered difficult, and it is voluntary. Puerto Rico does not participate in the Spanish and English measures. In 2003, Puerto Rico began to administer its own achievement test, Pruebas Puertorriqueñas de Aprovechamento Académico (PPAA) to public school students. About 25% of the K-12th grade students attend private school. This test might not accurately reflect achievement of students in Puerto Rico, but it might give insight into the effectiveness of the public school program. In this test there are 3 levels 1) basic, 2) proficient, and 3) advanced proficient. It is important to understand two key points: students who score on the basic level are not proficient —they are one step below; students who fail to reach this basic level are even further below proficient —they are at least two steps below.

On the PPAA, in 2002-3 and 2003-4, more than 50% of the public school students who were tested failed to reach proficiency in Math, Spanish and English as a Second Language (Puerto Rico, like other venues, sets its own standards of proficiency, yet more than 50% of the students who were tested failed to reach this minimum standard). Thus, if the students had taken the NAEP test, “it is likely… that the performance of public school students in Puerto Rico would have appeared even less satisfactory” (Rivera-Batiz & Ladd, 2006, p. 199).

NCLB’s mandate for annual testing gives an appearance of scientific rigor and quantitative measurement that may not always pertain. Under NCLB, testing is highly political. The consequences of failing to reach proficiency are dire, and may involve school takeovers and other undesirable consequences. Thus, decisions in establishing the proficiency levels have the appearance of self-interest. Proficiency levels are sometimes changed, and sometimes the state or territory changes their testing instruments. This politicalization of the “standardized” testing process exists in all jurisdictions under the US Department of Education, including Puerto Rico.   

“[T]he proportion of students performing below a standard of proficiency is closely related to the chosen standard. The higher is the standard; the lower will be the proportion of students who meet it. Given that under NCLB each state sets its own standards….most, if not all states, have adopted proficiency benchmarks well below, (and sometimes substantially below) NAEP standards (National Assessment of Educational Progress)…[which is] the only comparable measure across states”(Rivera-Batiz & Ladd, 2006, p. 199). Another observer agreed: “the law lets each state write its own tests and define proficiency as it sees fit. In all but a few states, the standards for proficiency have been set far lower than the national assessment test [the NAEP]…” (Dillon, 2007).

What might be the achievement levels in Puerto Rico if NAEP standards applied, instead of the PPAA standards? In 2007, the NAEP released data from Puerto Rico regarding Math achievement results which were obtained in 2003 and 2005.  “In 2003, at the eighth grade, 4% of Puerto Rico students scored at or above Basic as compared to 67% nationally” [in the US]. “In 2005, at the eighth grade, 6% of Puerto Rico students scored at or above Basic as compared to 68% nationally” [in the US]. “The ‘at or above Basic’ category includes Basic, Proficient, and Advanced  levels”. In “2005 Results by Content Area”…for eighth grade, Puerto Rico students scored below the national average in all five areas…. The score gaps ranged from 49 points [below the national average] in geometry to 76 points [below the national average] in measurement” (http://nces.ed.gov/whatsnew/commissioner/remarks2007/3_39_2007.asp

Thus, fewer than 4% in 2003, and 6% in 2005 actually reached proficiency in the NAEP test. The website goes on to explain that the 2003 and 2005 scores cannot be compared to each other because of changes in the testing instrument. Therefore, the 2005 score of 6% cannot be seen as an “improvement” over the 2003 score of 4%. In any case, basic is still less than proficient. That means in 2003 practically all, 96% of 8th grade students, and in 2005 practically all, 94% of 8th grade students tested in Puerto Rico failed to reach even a basic score in Mathematics, let alone to reach a proficiency level.  Let the data speak for itself.

History of NAEP Participation and Performance
Scale Score
Achievement Level
Percent at or Above
State Avg.
[Nat. Avg.]*
Mathematics (scale: 0-500)
  • Scale Scores
  • Achievement Levels
  • Coss-State Comparion Maps:
    • Scale scores
    • Percent at or Above Proficient
* Includes public schools only
  •  Significantly different (p < .05) from students in Puerto Rico.
  • SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center
  • for Education Statistics, National Assessment of Educational Progress (NAEP), 2003
  • Mathematics Assessment.
  • _ Puerto Rico _ Nation

Lessons for Puerto Rico

While the benefits of high stakes testing are questionable (Berliner, 2009), and especially the benefits of highly politicized high stakes testing under NCLB, there is nevertheless a broad pattern evident in these data. Something drastic needs to be done when virtually all students fail to reach the minimum proficiency levels.

In contrast, almost 100% of the students in the well structured Dual language program in the US were able to achieve proficiency, for several consecutive years. More than ¾ of these students were low SES and immigrant families. In contrast, more than 50% of the public school students failed to reach proficiency in Puerto Rico on the PPAA in Math, Spanish and English, and almost 96% (or 94%) failed to reach proficiency on the NAEP in Math. Regardless of the specifics of the PPAA as compared with the NAEP, there is clearly a gap that is statistically significant and educationally disturbing. The data is so lopsided that the point has to be clear.

If in Puerto Rico, a Dual language program were instituted in the venues that supported a classic Dual —where there are native speakers of both languages— and if the Dual axioms of program design were followed, then it is extremely likely that a large percentage of students would be able to reach proficiency levels, or higher, in Spanish, Math and English. Depending on how well the program is run, the academic achievement of students would increase.

If instead of a classic Dual, a program with Dual features were instituted in the venues that did not support the classic model —those that do not have native speakers of both languages— and if the Dual axioms of program design were again followed, it is also likely that a large percentage of students would be able to achieve proficiency levels or higher. Although the dramatic achievement that the classic Dual produces might not necessarily occur in a program with Dual features, the improvement over the current state of achievement would be likely. With the devastating test scores and data described above, this program option would seem to be of interest to the Department of Education as one measure to be explored for school improvement. Without quibbling over tests or levels of proficiency, the lopsided data makes it abundantly clear that the majority of students in Puerto Rico are in need of a more effective instructional program.

Final observations and conclusions

The high success of Duals depends on a strict adherence to the axiomatic elements of the program design. This in turn depends on committed and diligent teachers and supervisors who understand the model, and who follow it, not by rote, but because they realize the potential benefits. Dual teachers need to be willing to work and plan together to prepare lessons for their students, and conscientious supervisors, who also understand the program design, need to be able to support teachers and staff in accomplishing program goals, and help maintain the integrity of implementation. The beauty of Dual language instruction is that success does not depend on the children —they can be poor, or wealthy, immigrant or from the dominant culture. Success depends on the quality of implementation of the program.  

While the argument of this paper has highlighted the dramatic testing data, that is not the real argument for Dual language education. Testing data is only the vehicle for showing Dual language effectiveness. The real benefits are not the number of children who reach proficiency on a test, but the number of children who learn to read and write in two languages, to do math and science in two languages, and to think, and to reason and to understand the world in two languages. This program is good education. Children do well on tests when they are learning, and thinking and doing.

Initially, there are start up costs: books and materials need to be obtained in both languages, and most importantly, professional development is needed. Teachers and administrators cannot just read a handout or listen to a short explanation, and then do the program. There needs to be in-depth professional training, preferably on-going, because the program asks teachers and administrators to function in a new way. But, once set up, and when the teachers and supervisors can comfortably do what is asked of them, then Dual language programs require relatively low maintenance costs. What is key to the success of the program is not just money, but the quality of implementation: the professionalism of teachers, supervisors and administrators, their knowledge base, and their commitment to the program and to its success. In this respect, the question of teacher education and professional development is very important. Teachers and supervisors need to understand the program design, or they will revert to their old habits and their old ways, perhaps without realizing it.

The benefits of implementing Dual language programs or programs with Dual features in Puerto Rico are enormous, and far outweigh the costs. Whether in the classic design or the model with Dual features, these programs prepare children to be bilingual and biliterate —learning a second language without losing the mother tongue— raise the level of performance, and develop children academically and cognitively. Essentially, they work.


Christian, D., Montone, C., Carranza, I., & Lindholm, K. (1996). Profiles in two-way immersion education. Language in Education: Theory and Practice: 89. Washington, DC: Center for Applied Linguistics.

Collier, V. P. (1992). A synthesis of studies examining long term language minority student data on academic achievement. Bilingual Research Journal, 16, 187-222

Collier, V., & Thomas,W. (1997a). School effectiveness for language minority students, NCBE Resource Collection Series, No. 9, Washington, DC: National Clearinghouse for Bilingual Education.

Collier, V., & Thomas, W. (1997b). Two languages are better than one. Educational Leadership, 55(4), 23-26. Available at http://www.ascd.org/cms/objectlib/ascdframeset/index.cfm?publication= and http://www.ascd.org/publications/ed_lead/199712/toc.html

Collier, V., & Thomas, W. (2002). A national study of school effectiveness for language minority students' long-term academic achievement. Santa Cruz, CA: Center for Research on Education, Diversity and Excellence (CREDE), University of California-Santa Cruz, CA and Washington, DC: Center for Research on Education, Diversity &Excellence. Available at http://repositories.cdlib.org/crede/finalrpts/1_1_final or http://www.crede.ucsc.edu/research/llaa/1.1_final.html

Collier, V., & Thomas, W. (2003, October). The multiple benefits of dual language. Educational Leadership, 61(2). Retrieved from http://www.ascd.org/publications/ed_lead/200310/toc.html

Collier, V., & Thomas, W. (2004). The astounding effectiveness of dual language education for all. NABE Journal of Research and Practice, 2(1), 1-20. Retrieved from  http://njrp.tamu.edu/2004/.htm or http://njrp.tamu.edu/2004/PDFs/Collier.pdf

Cummins, J. (1992). Empowerment through biliteracy. In J. V. Tinajero & A. F. Ada, The power of two languages: Literacy and biliteracy for Spanish speaking students (pp. 9-25). New York, NY: Macmillian/McGraw-Hill. 

Cummins, J. (1999, October). Alternative paradigms in bilingual education research: Does theory have a place? Educational Researcher, 28(7), 26-32 and 41.

De Jesús, S. (1995). The attitudes and opinions of Latino parents regarding language learning in English and Spanish. Doctoral Dissertation. Teachers College, Columbia University, NY.

De Jesús, S. (2008, Fall). An astounding treasure: Dual language education in a public school setting.  El CENTRO: Journal of the Center for Puerto Rican Studies, 20(2), 192-217.

De Jesús, S. (2009). A Dual language Spanish and English program, PreK to 5th grade: 18 years of success. Paper presented at the X Congreso Puertorriqueño de Investigación en la Educación, Universidad de Puerto Rico, Río Piedras, March 12, 2009.   

Dillon, S. (2007, September 25). Scores show mixed results for Bush’s Education Law. New York Times. Retrieved from http://www.NYTimes.com

Genesee, F. (1987). Learning in two languages: Studies of immersion in bilingual education. Cambridge, MA: Newbury House.

Genesee, F. (1999). Program alternatives for linguistically diverse students. Santa Cruz, CA and Washington, DC: Center for Research on Education, Diversity & Excellence. Retrieved from http://www.cal.org/crede/pubs/edpractice/EPR1.pdf

Hakuta, K., & Díaz, R. (1984). The relationship between bilingualism and cognitive ability. A critical discussion and some new longitudinal data. In K.E. Nelson (Ed.), Children’s language (Vol.5, pp 319-344). Hillsdale, NJ: Lawrence Erlbaum.

Howard, E. R., & Sugarman, J. (2007). Realizing the vision of two-way immersion: Fostering effective programs and classrooms. Washington, DC and McHenry, IL: The Center for Applied Linguistics and Delta Systems, Co.

Howard, E., Sugarman, J., Christian, D., Lindholm-Leary, K, & Rogers, D. (2005). Guiding principles for Dual language education. Center for Applied Linguistics, Washington, D.C. Retrieved from http://www.cal.org/twi/guidingprincipals.htm

Krashen, S., Tse, L., & McQuillan, J. (1998). Heritage language development. California: Language Education Associates. Retrieved from http:://www.sdkrashen.com/main.php3
Lindolm, K. (1990). Bilingual immersion education: Criteria for program development. In A. M. Padilla, H. H. Fairchild & C. M. Valadez (Eds.), Bilingual education: issues and strategies (pp 91-105). Newbury Park, CA: Sage.

Lindholm, K. (1991). Theoretical assumptions and empirical evidence for academic achievement in two languages. Hispanic Journal of Behavioral Science, 13, 3-7.

Lindholm, K., & Aclan, Z. (1991). Bilingual proficiency as a bridge to academic achievement: Results from bilingual/immersion programs. Journal of Education, 173, 71-80.

Lindholm-Leary, K. (2005). Review of research and best practices on effective features of Dual Language educational programs. Draft Report, Center for Applied Linguistics, Washington, D.C. Retrieved from http://www.cal.org   

National Assessment of Educational Progress. (2007). Results, Puerto Rico, 2007, Mathematics, Commissioner’s Report, Executive Summary. Retrieved from  http://nces.ed.gov/whatsnew/commissioner/remarks2007/3_29_2007.asp

Pérez, B. (2004). Becoming biliterate: A study of two-way bilingual immersion education. Mahwah, NJ: Lawrence Erlbaum Associates.

Ramírez, J. D., Yuen, S. D., & Ramey, D. R. (1991). Executive summary, Final report: Longitudinal study of structured English immersion strategy, early-exit and late-exit transitional bilingual education programs for language minority children. (U.S. Department of Education Contract No. 300-87-0156). San Mateo, CA: Aguire International.

Rivera-Batiz, F., & Ladd, H. (2006). Education and economic development. In S.Collins, B.Bosworth, & M. Soto-Class (Eds.) The economy of Puerto Rico: Restoring growth (pp.189-254). Washington D.C. and San Juan, PR: The Brookings Institution Press & Center for the New Economy. Retrieved from http://www.brookings.edu.

Robledo Montecel, M, & Danini Cortez, J. (2002, Spring). Successful bilingual education programs: Development and the dissemination of criteria to identify promising and exemplary practices in bilingual education, the national level. Bilingual Research Journal, 26(1).

Torres-Guzmán, M. E. (2002, Spring). Dual language programs: Key features and results. Directions in Language and Education, 14.

Torres-Guzmán, M. E., & Etxeberria, F. (2005). Modelo B/dual language programmes in the Basque Country and the USA. International Journal of Bilingual Education and Bilingualism, 8(6), 506-528.

Torres-Guzmán, M. E., Kleyn, T., Morales-Rodriguez, S., & Han, A. (2005). Self-designated dual-language programs: Is there a gap between labeling and implementation? Bilingual Research Journal, 29(2), 453-474.