FRUSTRATION AND HOPE: A REVIEW OF THE RESPONSE UNDERTAKEN IN NATIONAL EDUCATION PROGRAMMES TO LOWER-THAN-EXPECTED RATES OF FOREIGN LANGUAGE ACQUISITION. FRUSTRACIÓN Y ESPERANZA: UN ANÁLISIS DE LAS MEDIDAS ADOPTADAS EN LOS PROGRAMAS DE EDUCACIÓN NACIONAL ANTE LAS TASAS INFERIORES

This review and discussion focused on surveys and reports regarding the rate of foreign language acquisition and attainment within national compulsory education systems, academic analyses of the results obtained, and national programmes and initiatives designed to address the shortcomings revealed in them. The proposed reasons for lower-than-expected acquisition and attainment in foreign language learning within countries including the European Union, Colombia, Vietnam and Ecuador were reviewed in detail. These reasons, and the measures undertaken as a result to remedy these flaws, were evaluated through a comparison with recent academic research relating to these factors, and through comparison between the different experiences revealed in the surveys and reports. These comparisons found that some of the measures undertaken were fit for purpose and were likely to yield some improvements in acquisition rate, although to a lesser extent than those projected by some of the national programmes, while identifying some aspects which have been overlooked. The conclusion highlighted in which aspects of the teaching and learning of foreign languages requires greater focus is needed in order to effect the desired changes. Investigación y Desarrollo • Revista de Divulgación Científica y Cultural • Volumen 11 Diciembre 2016 • PP 5 -16 • Dirección de Investigación y Desarrollo • U.T.A. • Ambato Ecuador RESUMEN Este artículo de revisión y discusión se centró en los sondeos e informes sobre la tasa de adquisición y rendimiento en los idiomas extranjeros dentro de los sistemas de educación nacionales obligatorios, los análisis académicos de los resultados obtenidos, y los programas e iniciativas destinados a remediar las falencias identificadas. Las posibles causas del bajo rendimiento, menor al esperado, en la adquisición de idiomas extranjeros, observadas en algunos países de la Unión Europea, Colombia, Vietnam y Ecuador, se revisaron en detalle. Estos motivos y las medidas adoptadas como consecuencia para remediar las falencias fueron evaluados por medio de una comparación con investigaciones académicas recientes, así como entre las distintas experiencias recopiladas en los sondeos e informes. Como consecuencia de estas comparaciones, se concluyó que algunas de las medidas adoptadas sí son las adecuadas, y que darán lugar a una mejoría en la tasa de adquisición del idioma extranjero, aunque en menor grado que lo proyectado por algunos de los programas nacionales, y a la vez se identificarán otros aspectos que no han sido tomados en cuenta. La conclusión resaltó en cuales aspectos de la enseñanza y aprendizaje de idiomas extranjeros es necesario un mayor enfoque para poder efectuar los cambios proyectados.


INTRODUCTION
I n 2014, launching their second national programme for the reinforcement of English language teaching and learning, the Colombian Ministry of Education made this important reflection: Las horas asignadas a inglés en teoría son suficientes para alcanzar los niveles de la aspiración si fueran horas de la mejor calidad en un ambiente de aprendizaje con buenas condiciones [The hours assigned to English are, in theory, sufficient to reach the levels aspired to, only if they are the best of quality and hours, in a learning environment of a good standard].(Ministerio de Educación Nacional [MEN], 2014, p. 46) The aspirations referenced here were that nearly half of the student population reach level B2 by the end of high school education.The following reflection is a common one: how many hours of instruction are necessary to become fully independent in a foreign language, and which variables delay or even altogether impede the attainment of this level?
This same question has been raised by many other countries in recent years.Latin America and South-East Asia, in particular, are areas where national initiatives have been undertaken, and are currently in progress at analysing why the many hours given over to English instruction in school and university systems have not resulted in a higher level of attainment.The standard being sought in the long term is a B2 level of the CEFR (MEN, 2014;Ortega & Argudo, 2016;Nguyen, 2017), with a C1 level required for some university language programmes.These aspirations, and the reason why certain levels of attainment are important to the governments and economies of the countries concerned, are summarised in Figure 1.T his review discusses the objectives and interim results of national programmes in Vietnam (British Council Vietnam, 2018;Nguyen, 2017;Tuyet, 2015); Colombia (Bonilla & Tejado-Sánchez, 2016;British Council, 2015a;Portafolio, 2015;MEN, 2014); and Ecuador (Ortega & Argudo, 2016;British Council, 2015b; Ministerio de Educación del Ecuador [Mineduc], 2014; Ureña, 2014).Moreover, a number of countries within the European Union have experienced disappointing rates of progress in languages, as reported and analysed following the First European Survey of Languages Competences, (Costa & Almeida, 2015;Araujo & Costa, 2013;European Commission, 2012), and their results are reviewed here.
The actions undertaken, or suggested courses of action proposed by academics and authorities in these different contexts, are reviewed and compared to the findings and recommendations of a number of established historical studies and observations of factors affecting the rate of acquisition (Spolsky, 2014(Spolsky, , 1989;;Muñoz & Singleton, 2011;Muñoz, 2006;Zarker, 2000;Jackson & Kaplan, 1999).The discussion and conclusions highlight those aspects of the initiatives and analyses which are fit for purpose, while recommending changes to those aspects in which resources have been mis-directed and are unlikely to yield much progress in the rate of acquisition.

METHODOLOGY
Criteria used for the selection of data for the study L anguage acquisition is a vast topic: Spolsky (1989, 2014   identified 74 conditions which influence successful learning, while other authors have suggested more factors.Therefore, it is necessary to define the parameters of the present review and discussion and to set aside those factors which will have no bearing on the discussion here. Firstly, all of the contexts discussed refer to the length of the acquisition process of a foreign language, and therefore exclude any discussion of acquisition rates for students learning within an L2 community.The study contemplates chiefly the experiences of students who learn the L2 in a classroom and/or through self-study from within their own country.However, some consideration is given to the minority who receive an amount of additional exposure from within their own countries, sufficient to be regarded as a degree of immersion in the target language, and the opportunities for implicit learning that may result (Muñoz & Singleton, 201;DeKeyser, 2003).
Secondly, this discussion is based upon studies of large populations derived from national statistics.It, therefore, necessarily excludes the effects of individual variation in acquisition rate.Spolsky (1989, p. 15) postulated the formula: where K(f) is the future level of competence of the L2 being aimed for, and is to be achieved by taking K(p) (the current level of competence) and adding to it A, which is the individual's ability for languages; M, which are the individual's motivational factors; and O, which are the external opportunities for language learning and acquisition.For the purposes of this review and discussion, the considerations corresponding to factor 'A' are of no relevance.Furthermore, considerations pertaining to factor 'M' are not relevant, except where motivational factors may be identified as common to a student population as a whole.Therefore, the 'opportunity' factors are the variables bearing upon rate of progress and attainment to be considered.Here, this study follows the criteria adopted by several of the contributors to the Spolsky and Sung (2014) examination of conditions for English language teaching and learning in Asia (see for example, Moon, 2014, pp. 167-184).
Criteria used for the evaluation of national programmes T here are no universally accepted research results on how many hours of total study are required to reach a specified level of competence in a foreign language.One source of official data for which there are reliable details concerning the background and study habits of students was provided by The Foreign Service Institute (FSI).The FSI, a U.S. government body which trains diplomats in almost all world languages, produced an influential report (Jackson & Kaplan, 1999) which indicated average times for the acquisition of a B2/C1 borderline equivalent level of competence in different languages.For languages of the same linguistic branch, for example English and Spanish, they claimed around 600 classroom hours are required, based upon the results obtained by their students.This coincides with the recommendation of Cambridge English Assessment (2017) and the Alliance Française [French language-teaching organisation] (2017) for the same level of competence in their respective languages.These institutions both claim that level B2 may be achieved within 500 and 600 hours of guided instruction.These figures provide a guideline for comparisons with the situation in the different countries in this review.Figure 2 shows the current expected level of attainment, based on an average of all the data available from the countries and regions here reviewed, in comparison with the expectations of the FSI and Cambridge Assessment.In combination with the maximal expectations shown here, it is necessary to consider the circumstances in which students of the FSI (Jackson & Kaplan, 1999), or those who study for the higher-level Cambridge certificates (Cambridge English Assessment, 2017) learn the target language.These can then be used to evaluate the measures and policies adopted by the national programmes in this review.
The FSI points out that their data are average learning times under very specific conditions, which had been, in their view, optimised for their circumstances (Jackson & Kaplan, 1999).These may be considered the factors of 'opportunity':

RESULTS
Data from national surveys of attainment in foreign languages T he language survey carried out by the European Commission in 2012 revealed statistics regarding attainment among adolescents (average age 15) across 14 countries of the European Union (EU).Tomasi (2017, pp.123-124) reported that "the level of independent user (B1+B2) is achieved by only 42% of tested students in the first foreign language."This overall statistic masks considerable variation among the participant countries, and results from selected countries with differing levels of attainment can be seen in Table 1.The situation among language students in Colombia is considerably further from achieving expectations (see Figure 3 for a comparison of EU, Colombia and Vientnam): 59% of school students up to 11th Grade showed an effective knowledge of English at A0/A1, with only 6% achieving B1 level in 2013 (Bonilla & Tejado-Sánchez, 2016).For most university students in Colombia, continued English instruction is obligatory with at least 200 (and often considerably more) classroom hours being provided at this level.In 2012, 8% of these students achieved level B2 following these additional hours of instruction, with over 60% remaining at A1 level (MEN, 2014).
In Vietnam, the statistics for high school students are very similar.The average 16-year-old accumulates 600 hours of English language instruction, and is at the A0/A1 level.Admittedly, English is considered by many linguists harder to access for Vietnamese speakers than for Spanish speakers (Jackson & Kaplan, 1999), but the reported progress is slower than both the FSI's estimates, and the expectations of those involved in language education in Vietnam: 98% of Vietnamese students study English for 7 years but are unable to use it in basic communication.On average, only those who major in the language can score 5 out of 10 in the National English Entrance Exams to university.After entering university, most have to restart at beginner level.(Nhan, 2013, p. 147) • Recommended class size at the FSI was a maximum of 6 for English learners of Spanish or French.• The learners were all highly motivated, both intrinsically and extrinsically.• The teachers were all trained native speakers, or at least functionally bilingual.• A variety of approaches and learning formats were employed, including some level of immersion in the L2.
Furthermore, all students in the FSI, and most students studying for higher level Cambridge English assessments (Cambridge English Assessment, 2017) and higher level French assessments (DELF-DALF, 2018), are adults, or older adolescents.This is the first factor to be evaluated in the following discussion.
Frustration and hope: a review of the response undertaken in national education programmes to lower-than-expected rates of foreign language acquisition.Clearly, in these last three cases, each in themselves fairly representative of their wider regions (Education First, 2017a), (Education First, 2017b), as well as in some countries within the European Union (Araujo & Costa, 2013), rates of English language acquisition, both in secondary and in tertiary education, have been historically very low when compared to the potential for attainment highlighted in Figure 1.They remain relatively low despite some years of targeted intervention (Bonilla & Tejado-Sánchez, 2016;Nguyen, 2017;Ureña, 2014).
There are many inter-connected factors within the 'opportunity' conditions for foreign language acquisition (see for example Spolsky & Sung, 2014).Those which have been highlighted and explicitly addressed in the reports and initiatives commissioned by national governments and the European Commission are evaluated in the following section.

DISCUSSION
Factor 1: hours of study versus age of students A s was summarised in a thorough review of studies of the influence of learner age on rates of acquisition by Muñoz and Singleton (2011), adults and older adolescents experience a considerably higher rate of L2 acquisition than young adolescents, although the rates do begin to converge in most skill areas after several hundreds of hours of instruction, according to the Barcelona Age Factor Project (Muñoz, 2006).
Therefore, 600 hours of instruction for adults will equate to a higher level of attainment than 600 hours between late childhood and mid-adolescence.In all of the countries reviewed here, current compulsory provision of English language instruction amounts to at least 600 hours accrued by 15/16 years of age, with some areas providing up to 900 hours, (Araujo & Costa, 2013;Bonilla & Tejado-Sánchez, 2016;Ortega & Argudo, 2016).
Consequently, are the countries surveyed here justified in their frustration or alarm at the slow rate of progress?According to tendencies reported by the Barcelona Age Factor Project (Muñoz, 2006), 600 hours of instruction between late childhood and mid-adolescence is sufficient to bring a majority of students to B1 level.If this analysis is applied to some areas of the European Union, then it would seem there is no justification for frustration, as adolescent learners are making the expected rate of progress for their age (see for example, statistics for Slovenia, Greece and the Netherlands, in European Commission, 2012).In other areas of Europe, the majority of students are at A2 level or below, following 600 hours of instruction, which is certainly less than expected.Araujo and Costa (2013) see the age of beginning training in the foreign language as an essential 'opportunity' for influencing attainment by the end of compulsory education: "In general, with respect to system-level policies, we can say that the most significant effect is the onset of language learning" (p.28).However, in Vietnam, Colombia and Ecuador (Nguyen, 2017), (Bonilla & Tejado-Sánchez, 2016), (Ureña, 2014), and in some European nations (see statistics, for example, for the U.K., France, Poland, Spain in European Commission, 2012), the rate of acquisition by mid-adolescence is so slow it seems that considerably reducing the age of onset would have little additional impact.Earlier onset has been a recommendation in both Colombia and Ecuador (see, for example MEN, 2014MEN, , 2015)), and yet if 600 hours between ages 10 and 16 has produced an attainment of level A1 at best, it seems unlikely that a further 300 hours between ages 6 and 10 will move students much further towards a level B1, or indeed B2, by age 16.
On the contrary, if rates of acquisition are far slower among children than among older adolescents, as a large number of studies have shown (Muñoz & Singleton, 2011), then those countries with low attainment by 16 should rather invest only in post-16 compulsory foreign language tuition.Mandatory university-level English instruction has been embraced by both Colombia and Ecuador, but at the same time as their having dedicated scarce resources to primary-level English teaching (British Council 2015a; British Council 2015b).Certain government policy may be mistakenly guided by the Critical Period Hypothesis (CPH) (Colombo, 1982) which suggests that native-like attainment may only result from some degree of L2 immersion during childhood.The CPH is itself a matter of considerable debate (Muñoz & Singleton, 2011), but, in any event, it has not been shown that failure to attain a B2 level is linked to a lack of early-childhood learning opportunities.
Factor 2: class size W ithin the education systems of the European Union, Vietnam, Colombia and Ecuador, class size is rarely less than 25, and may often be closer to 40 students (British Council, 2015a;British Council, 2015b;Araujo & Costa, 2013).There is no consensus on how class size correlates to attainment in language learning.Zarker's extensive survey of the impact of class size on both language learning and on other subjects suggested that the quantifiable benefits may be in the range of a few percentage points (Zarker, 2000).This is supported by the conclusions of Araujo and Costa (2013) who stated: "there is little effect of classroom size in foreign language learning" (p.28), and they even maintained that reduced class size coincided with lower attainment in some instances.
The indication that very large classes may be a threat to learning has been put forward by British Council reports (British Council, 2015b), but without suggesting any strong correlations.On the other hand, a number of small-scale studies do show considerable differences in learning progress (see for example, Yi, 2008) and certainly private academies and specialist language institutes around the world insist on reduced class sizes as a key benefit for learners.
None of the surveys and national programmes recommended a significant reduction in class size within their public education systems (MEN, 2014;Nguyet, 2017;Education First, 2017a, 2017b;Araujo & Costa, 2013).Certainly, class size in countries whose education system produces close-to-expected progress in the acquisition of English, such as the Netherlands or Sweden, class sizes are no different to those with poorer results (Araujo & Costa, 2013).Therefore, in the light of much contradictory evidence, and certainly with no clear indication of a positive correlation between class size and an improved rate of acquisition, all of the initiatives and recommendations surveyed are correct not to prioritise this factor by assigning scarce funding to smaller classes.
Factor 3: average levels of motivation T he average level of motivation of the students within the European survey was well below those described by the FSI (Araujo & Costa, 2013).Intrinsic motivation was particularly low compared with students who voluntarily attend classes at Frustration and hope: a review of the response undertaken in national education programmes to lower-than-expected rates of foreign language acquisition.
private language academies or as part of specialist programmes.However, Araujo and Costa found no positive correlation between intrinsic interest in language learning and attainment in European countries, and in some instances the opposite was true (2013, pp.36-37).
The extrinsic motivations resulting from having made English study obligatory for high school or even for university graduation (British Council, 2015a; Consejo de Educación Superior, 2013) are certainly strong, but a greater, albeit emerging, motivation may come from the phenomenon of the imaginary international English language community described by Canh (2014).In his study into language learning in Vietnam, Canh argues that globalised communication and interaction on economic, cultural and sporting themes have created an 'imagined community' to which most learners desire to have a connection.Therefore, traditional discussions of acculturation (Schumann, 1985) and integration to a specific group, previously more applicable to second language acquisition studies (Dornyei, 2001), may be substituted for a desire to belong to this 'imagined' global community and be equally felt by students of English as a foreign language in any context.
In the case of Colombia, the government has attempted to force a cultural shift towards an internationally-integrated population through the Colombia Bilingüe programme (Portafolio, 2015;British Council, 2015a).However there is no evidence as yet whether integration can successfully be made a national policy in this way.In any event, the permeation of this international English-speaking culture into society is universal and seemingly irresistible, and is therefore outside of the control of government initiatives.The impact of this motivational factor on rates of English acquisition remain to be seen and should be the subject of future study in national programmes.
Factor 4: teachers' linguistic competence T o return to the limitations highlighted above by the Colombian Ministry of Education (MEN, 2014) (see introduction), the quality of the hour's instruction is seen to be fundamental in determining the learners' rate of progress.The factor most frequently highlighted in the national programmes surveyed here (Tuyet, 2015;MEN, 2014;British Council, 2015b) is teachers' linguistic knowledge, that is, language skills, vocabulary and grammatical knowledge of the target language.National testing of English language teachers in 2010 revealed that 29% of all staff working in the Colombian public sector were at the B2 level.Following efforts, this percentage increased to 43% in 2013, leading to the current initiative, Colombia Bilingüe, which seeks to have all English teachers at B2 level or above by 2025.
A similar case is reported by the British Council in Vietnam, where they collaborate in teacher training with the Vietnamese Ministry of Education and Training [MOET].Starting from 2008, MOET raised the target for linguistic standards for English teachers to B2 for all English teachers, and C1 for high school teachers (British Council, 2018, para. 1).
This initiative came in response to the results of a nationwide test of English teachers' ability the B2 level.Only 3% of high school passed the test, 17% of primary school teachers being found to be at the A1 level, which led MOET to conclude that lack of student progress, despite receiving hundreds of hours of instruction, is due to professional inadequacy (Tuyet, 2015).
In Ecuador in 2012, less than 1% of teachers in the public sector were found to be at level B2 (Ecuador tiene falencias [Ecuador has flaws], n.d.).The Ecuadorian government then established, and currently maintains, the requirement that all specialist English teachers have at least a B2 level of competence.However, this remains far from being achieved.As in Colombia and Vietnam, the Ecuadorian government (Mineduc, 2014) places the greatest emphasis on the correlation between the linguistic competence of teachers and an adequate rate of foreign language acquisition among students (Ureña, 2014).
One criterion of minimum linguistic competence is provided by Cambridge English (2018) in the DELTA teacher training programme.A minimum of C1 is required for teachers to be considered proficient, or expert according to their assessment criteria.This level is generally achieved in Europe, where teachers' subject knowledge of the target language is ranged between B2 and C2 (European Commission 2012, pp. 206-207).Most language teachers in the public sector are required to have at least a bachelor's degree or equivalent in the relevant language, which in most European countries corresponds to a C1 level.In many cases, a large number of native-speaker teachers are employed.In Malta, 54% of public sector English teachers are native speakers.In Spain, France, Sweden and the Netherlands, the porcentage of native speakers is 20% (European Commission, 2012).There seems to be a strong correlation between teacher linguistic level and student attainment across all of the regions reviewed here.However, many European countries, including France, Spain and the UK, are far from achieving expected progress, despite the linguistic expertise of their teaching staff (see table 1).Vietnam, Colombia and Ecuador are right to place a strong emphasis on teacher subject knowledge, but they should not expect that remedying this factor alone will lead to better attainment, if this were the case, most European countries would already be achieving expected levels of attainment.
Factor 5: other aspects of instructional quality T he quality of instruction involves many factors, such as the teacher's classroom practice, their knowledge of methodologies, and their classroom management skills, as well as resources such as the quality, or indeed availability of textbooks.
The nature of the learning curriculum and of teacher training in generally accepted pedagogical methods has largely been standardised in most educational contexts, especially following the adoption of the Common European Framework of Reference for Languages by most countries (MEN, 2014), which demands an emphasis on communicative competence, and due to collaborations with expert international organisations such as the British Council (British Council, 2018;British Council 2015a, 2015b), as well as the use of internationally published textbooks which conform to accepted standards.However, there is evidence of continued inefficient practices in the language classroom, such as avoidance of the L2 for instruction and a perpetuation of grammar-translation techniques (Espinosa, 2015), which may offset some of the advantages of teacher linguistic competence.
Nonetheless, within Europe, Araujo and Costa (2013) found no evidence that differing approaches and methodologies had any significant impact on language acquisition (pp.33-34), and Spolsky finds no variation in approach to be directly correlated with acquisition: "Any intelligent and disinterested observer knows that there are many ways to learn languages and many ways to teach them, and that some ways work with some students in some circumstances and fail with others" (Spolsky, 1989, p. 383).
A qualitative factor found to be significant by Araujo and Costa (2013, p. 35) is described as how good language lessons are perceived to be by students.For example, where a teacher prefers to engage their students with low-level cognitive tasks, such as copying texts, or appears to take little interest in student learning, it is likely that learning will be impeded.This general quality of lesson may be one of the most significant factors influencing the rate of learning, and this is certainly manifested by the national school inspectorate in the UK (Ofsted, 2015).Their report into modern foreign language learning [MFL] from 11-14 years old observed that "too often, inspectors found teaching that failed to challenge and engage pupils" and concluded that "achievement was not good enough in just under half of the MFL classes observed" (Ofsted, 2015 p. 5).
Figure 4 illustrates the different levels of competence in the different countries and regions in this review, and compares them to standards at the FSI.
Frustration and hope: a review of the response undertaken in national education programmes to lower-than-expected rates of foreign language acquisition.
The UK was in last place in the European Survey of Languages, although with reference to the teaching of French and Spanish as foreign languages, rather than English.Results were similar to those seen in Latin America and Vietnam, that is, 80% at level A1 or below (European Commission, 2012, p. 232).There may be parallels to be drawn between this situation and those of Colombia, Ecuador or Vietnam, yet there are no published inspectorate reports pertaining to the quality of lessons in these countries.Whereas, in Colombia, respondents in surveys manifested a positive opinion of the general quality of language lessons at school and university (British Council, 2015a, pp. 55-56), (British Council 2015b).Greater rigour and transparency in inspections and lesson observations may lead to much needed reform in this area.On the other hand, in the UK, MFL attainment has plateaued for many years despite this transparency (Tinsley & Board, 2017).
Factor 6: exposure to the foreign language outside of the classroom I n some countries, such as Sweden, the Netherlands and Malta, a majority of students do reach level B2 by the end of high school (Costa & Almeida, 2015).In these European countries, the target language enjoys a very strong presence in the cultural and working environment, allowing students hundreds, or even thousands, of additional hours of passive exposure.The level of impact on language acquisition of such passive exposure outside of the target language country, as part of foreign, rather than second language learning, has not been reliably quantified.
However there is certainly a consensus that the presence of the target language in the environment is a strong complement to language acquisition (Costa & Almeida, 2014;British Council, 2015a, 2015b).
The opportunities for passive exposure include English language television and film, is readily available in several countries (Costa & Almeida, 2014)  Ecuadorians who consider themselves fluent in English certainly place great importance on this aspect of acquisition: The largest shares of those that felt confident in their English skills (advanced and fluent) said that it was due to social and cultural factors such as watching English language films and television (29%), listening to music with English lyrics (22%) and speaking English with friends (15%) and at work (15%) (British Council, 2015b, p. 40).
A similar situation was reported in Colombia (British Council, 2015a).However, this survey refers to those who already see themselves as very competent in the language.The general situation in Colombia and Ecuador is a cultural and professional environment with a negligible presence of the English language.Both British Council reports consider a failure to engage with the English language outside of the classroom as a key factor in poor English acquisition among the population as a whole.
Therefore, in the majority of the countries surveyed in this review, there is a general lack of opportunity for the passive exposure as a complement to the study of English considered of great importance by numerous experts (Costa & Almeida, 2014;Council of Europe, 2017;European Commission, 2011).
Moreover, exposure to reading texts in English was also shown to be a strong element among fluent English users in Ecuador and Colombia.In both countries, nearly 50% of those who considered themselves at an advanced level of English stated they read frequently, either for pleasure, or as a requirement for work or study (British Council, 2015a, 2015b).Conversely, those qualified as weak in English stated they seldom or never read in the L2.
None of the countries surveyed here have undertaken an explicit programme, or invested resources in allowing their citizens to receive greater exposure to written English.Colombia has undertaken a nationwide reading initiative for both L1 and L2 (MEN, Source: compiled by author based on data in European Commission (2012).
However making subtitling rather than dubbing mandatory on national television or cinema has so far not been implemented by any country to date, despite recommendations of the potential impact of such a measure (European Commission, 2011).

CONCLUSION
C ompulsory national education systems around the world face the challenge of equipping citizens with a level of L2 competence sufficient for them to participate in the global labour market, and allow them to join the international "imagined community" for social and cultural interaction.Many of these systems have so far failed in that goal, but are dedicating increasing resources in an attempt at remedying the situation.600 or 700 hundred hours of tuition at secondary or tertiary level is not likely to bring more than a small proportion of students to B2 level, unless the specific conditions present at the FSI, for example, could be practically implemented in national compulsory education.In the few countries where this goal is achieved, this review has shown that there is a culture of students' being exposed to, and voluntarily engaging with, hundreds of hours of L2 input beyond the classroom.Implementing the same culture in other countries would require a paradigm shift on the part of both governments and citizens.
Certain modest improvements, for example bringing most students to B1 level, may be attainable if the initiatives of the kind reviewed here for Colombia, Vietnam and Ecuador continue.All of these countries are right to insist on a far higher level of linguistic competence for their teaching staff.Furthermore, are also right to adopt international standards of teaching and assessment.
A major area of concern is that the general quality of lessons has either not been measured and explicitly addressed, or at least few governments have disclosed any perceived threat to learning of this kind.Institutional inspections and future academic studies of lesson quality in terms of efficient use of time, the use of higher cognitive activities, and teacher interest and motivation for the progress of their students, if undertaken, may reveal great deficiencies with the potential to undermine progress in other areas.
Finally, increasing the time students spend learning the L2 within the compulsory education system is unlikely to give any benefits unless other aspects of instructional quality are addressed first.In any event, moving toward earlier onset of language education in primary school or before will not be as effective as continuing compulsory L2 instruction into tertiary education, at least until the age of 18.In this way, making B2 competence a mandatory requirement for completing either academic or vocational tertiary education may prove to be the most time-efficient means to reaching the required level of attainment, as long as the other reforms discussed here are implemented at all levels of the compulsory system.

Figure 1 .
Figure 1.Schematic of the levels of the CEFR used as references in the review.

Figure 2 .
Figure 2. Expected level of attainment after 600 hours of instruction.

Figure 3 .
Figure 3. Level of attainment at the end of compulsory education.

Finally,
Ecuador reports a similar experience to that seen in Colombia and Vietnam.In the case of Ecuador, alarm at the very low level of attainment of English students led to the CRADLE initiative, followed by the current programme of Reinforcement of English Language Teaching (Ortega & Argudo, 2016; Ministerio de Educación de Ecuador [Mineduc], 2014).Slight improvement was afterwards recorded in the level of English in the population as a whole, but Ecuador remains below Colombia in standardised international proficiency indices (Education First, 2017a, 2017b).

Table 1 .
Countries representing lower, middle and upper ranges of attainment in languages.
. Its impact is stronger in listening skills, as may be seen in the data compiled by the First European Survey on Language Competences (European Commission, 2012) and summarised in Table 2 below.Countries whose television and cinema are dominated by original version English language content higher score in listening skills:

Table 2 .
Top EU countries by listening skill