| Teaching Articles Home |

Volume 35
Professional Teaching Articles
April 2009
Article 2

Formats
PDF
SWF

Title
Young Language Learner Assessment:
A Case for Using Assessment Portfolios

Author
Gerry Lassche
Miyagi Gakuin Women's University,
Japan

Bio
Mr. Gerry Lassche holds an MA (TESOL), and holds a Certificate for Teaching English to Children from David English House. He has given Korean MOE sponsored seminars for elementary school teachers on assessment, materials design and teaching reading, and taught intensive “speaking upgrade” seminars, and was an invited consultant on the English Village project (Paju, Korea), which targeted upper elementary school students. He has lectured on behalf of the Ajou University Graduate School of Education (Suwon, Korea) on Teaching English through English and Teaching Young Learners, and has delivered core courses on TESOL Methodology and TESOL Practicum on the Ajou-Wisconsin TESOL Certificate Program.

Abstract
The introduction of English into the Japanese elementary school presents an important curriculum issue of suitability and accountability: how to ensure that teaching and learning is enjoyable, easy to understand, trustworthy, and can improve practices. The crucial consideration is that children differ in their rates of development and in their approaches to learning in many ways.  This paper, a description of important young learner (YL) differences, will provide a rationale for the use of assessment portfolios to answer that issue. Various examples of portfolio content will be provided which demonstrate individual tailoring of testing and feedback over the entire course of learning, in a way that shows progress through observed skill development for all stakeholders.

Key words:- approaches to learning, young learner (YL) differences, the use of assessment portfolios

Introduction
Elementary education in Japan will soon include the teaching of English to children from grade 5 and up. The Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) is tasked with realizing this vision for implementation in 2009 – 2010. While the administrative details have yet to be clearly specified, the broad goals of such a program were already laid out in a white paper published in 2003 on the MEXT website, called “Regarding the Establishment of an Action Plan to Cultivate Japanese with English Abilities.”
   It would appear that the government has provided ambiguous information on why it is proposing an early start to English teaching (Japan Politics and Policy, 2001). However, by carefully sifting through policy reports and media releases, Goto-Butler (2007) has offered a detailed and comprehensive rationale, and while it is beyond the scope of this paper to go into the specific complex of details, a summary is possible. The government was interested in internationalizing the curriculum (known as kokusaika), in an effort to make it more fully compete in the international business world. Eventually this process began to place more emphasis on the development of more communicative English skills, with sooner being thought better. The process culminated in a series of policy white papers from a panel of experts commissioned directly through the Prime Minister’s office, culminating in the plan mentioned above. With specific regard to elementary school education, MEXT (2003a) states:

It is important that experiential learning activities that are suitable for elementary school students are carried out, and that the motivation and attitude for children to communicate positively is fostered by providing children with exposure to foreign language conversation in an enjoyable manner, and by familiarizing them with foreign cultures and ways of living…The situation and content of English conversation activities at elementary schools will be surveyed and publicized through the Status Report on Improvements in English Education (SRIEE) mentioned previously. This will contribute to further approaches for improvement. (MEXT, 2003a; bold added)

Top priorities for educational reform then, as described by MEXT (2003b) include four main components: easy to understand classes; enjoyable classes that are free of worries; a process that is trusted by parents and communities; and finally, a system that can improve the provision for education. From the above, a summary of crucial points include:

1. the provision of developmentally-appropriate education (suitable, enjoyable, easy to understand)
2. accountability to stakeholders (SRIEE, parents, teachers, students)
3. feedback system for gauging curriculum development (improvement)

   What this means, therefore, is that there must be instruments for assessing the quality of teaching, and its impact on learning. This assessment, as part of the learning environment, should also be characterized as stress-free (“free of worries”), enjoyable, associative, easy, and one that aims at improvements in performance, as well as engaging stakeholders in such a way as to inspire trust and confidence in what is taking place in local elementary classrooms.
             Maley lamented that “when it comes to assessing the progress of young language learners, we often find ourselves driven back on testing materials which are more appropriate for use with older learners” (Ioannou-Georgio and Pavlou, 2003, iii). This is a concern that has been echoed in other countries, and MEXT is well-aware of the dangers of age-inappropriate assessment: “the simple introduction of junior high school English education at an earlier stage as well as teacher-centered methods for cramming knowledge should be avoided.” (MEXT, 2003a)
   MEXT states that “teaching methods relating to English education at elementary schools will continue to be developed.” It is hoped that the reinvention of the education wheel will not be carried, that a review of current international best practice elsewhere for elementary education will suffice. In this paper, then, I will review the characteristics that differentiate elementary school learners from older, high school-level learners. From there, an enunciation of a set of educational guidelines that specify clearly how assessment should be carried out will follow. An assessment procedure that best exemplifies and fulfills the priorities and characteristics that MEXT has in principle already agreed to will be described and discussed to spotlight the use of assessment portfolios. With the help of this, I will launch into a description of portfolio contents, with pertinent examples drawn from the existing literature, and evaluate each example in terms of its age-appropriacy as well as its applicability to the local EFL setting.

Review of Relevant Developmental Literature
McKay (2006) suggests that young learners differ from older learners in 3 broad areas, and these will be discussed in relation to testing:
1. growth factors (which includes cognitive, socio-emotional, and physical issues)
2. literacy factors
3. vulnerability issues.

Growth factors
First, growth factors are especially important, because children are developing, day by day. Children differ greatly in their individual rates of development, as well as their general development at particular ages within particular skill areas. For example, one child might be socially very competent, and yet demonstrate more linguistic errors than another child, as can be the case in a bilingual home (Riordan, 2005). A test that focused only on a child’s linguistic development and not social development might cause undue alarm for a teacher or parents.
   Cognitive issues at young ages include shorter attention spans, understanding the connection between a cause and its effect, understanding how parts can relate to a whole, organizing information in their minds for short-term and long-term memory recall. With a shorter attention span, a testing environment which required a child to pay close attention for more than 15 or 20 minutes would elicit boredom or fatigue. Following detailed instructions to perform some test task, rather than through play or experimentation, would also prove difficult for most children. Because children might not understand the importance of the rubric, or perhaps that the rubric failed to engage them, short-term recall would probably suffer. Children tend to learn best through direct experience, where they can see and relate an object within its environment. Hypothesizing about some situation, imagining possible effects, requires an ability to abstract that is developmentally unavailable at younger ages. This becomes almost impossible when the imagined object is beyond the child’s range of experience (ie. answering questions on a story involving playing outside in winter, when the child has never seen snow). Using a meta-language (ie grammatical terms) to identify parts of a sentence, (a test item often employed for vocabulary or grammar tests in middle school), would not be appropriate for young language learners (YLLs).
   Socio-emotional issues arise due to the fact that the child is still learning how cope with increasing detachment from the family unit (for example, the mother as primary care-giver), and how to relate to others who are not family, or even not familiar. This can create some anxiety and dependency in children. In a testing situation that involved cooperating with other children, the degree of familiarity with the social setting would need to be considered, and the child’s needs for recognition and secure affirmation attended to. As well, the child would probably be very sensitive to negative feedback. Montessori (1912) has talked about the fact that children generally experience failure for the first time in the classroom.
   With physical issues, developmental variation is quite common. Children’s bodies are still growing, and they are still developing both fine and large motor skills. Tests that require students to write their answers or draw pictures might be simply measuring fine motor skills rather than underlying linguistic abilities. Brain researchers have found that physical movement is correlated with neuronal connections (Center on the Developing Child at Harvard University, 2007). Allowing a student some opportunities for movement while testing, rather than sitting still, would be a more natural accommodation. Classes can sometimes reach up to 40 students, but hopefully MEXT policies will start to take effect on reducing class sizes at less than 35 students (MEXT, 2001). Also, because group activities tend to be more cooperative and encourage social interaction, as well as lessening the pressure to perform as indiviudals (Paul, 2003, 41ff), group-oriented assessment might be a more effective way to assess a child’s communicative abilities, and this could divide the load with a large class.

Literacy factors
Second, literacy experiences vary greatly from child to child. Some children enter the class having been read to often by the parents, where other children may not yet have developed the association from sound to symbol (Dyson and Genishi, 1993, 127). In an increasingly audio-visual wired world, the TV can play a large role in determining and limiting exposure to written materials (Puckett and Black, 2002, 481). First language children have the advantage of more or less developed oracy. When approaching a second language, however, developing L2 oracy parallels L2 literacy. Yet, at the same time, the assessment tradition often relies heavily on a written format as noted by Maley (Ioannou-Georgio and Pavlou, 2003, iii). Children from an EFL environment have the added difficulty of using language that has no connection to their worlds of experience. Asian EFL contexts almost without exception require students to learn a completely different alphabet and script, which is not the case in many European countries (France, Germany, Italy, etc while using different languages use the a similar romantic alphabet), giving a distinct advantage to European children versus Japanese, Korean or Chinese children when taking age-appropriate tests, like Council of Europe (COE) tests or the Cambridge tests. This can explain the emphasis most Japanese elementary school programs place on speaking and listening alone; yet, Paul (2003, 83) suggests that such obstacles are not necessarily insurmountable, if the assessment material is pitched at a simpler level, and in a way that engages the child.

Vulnerability factors
Third, apart from handicaps, all things being equal, children generally learn a first language fluently. It is only when learning an unfamiliar topic in the school setting that children first become vulnerable to a sense of their own inadequacy (ie EFL). An extended quote from Montessori (1912, p. 237) can help illustrate the point:

A widespread prejudice [is]… the belief that the child left to himself gives absolute repose to his mind. If this were so he would remain a stranger to the world, and, instead, we see him, little by little, spontaneously conquer various ideas and words. He is a traveler through life, who observes the new things among which he journeys, and who tries to understand the unknown tongues spoken by those around him. Indeed, he makes a great and voluntary effort to understand and to imitate.

From these exchanges, the child develops a sense of worth and value, which is affirmed constantly in the home setting, and this quality of family interactions and communication patterns has profound downstream effects upon later achievement in the school setting (Amatea, Smith-Adcock, and Villares, 2006). In the classroom setting, suddenly the child may be thrust into an experience of receiving negative feedback (for the first time) from her new primary caregiver (for the first time). This makes the child especially vulnerable to testing situations which provide feedback and achievement scores, and may slant a teacher’s perspective toward that child’s achievement and progress. This kind of testing can become important way stations for making crucial administrative and pedagogical decisions for later schooling, even though standardized tests for young children can often be hampered by validity and reliability problems (such as in the USA, as described by Goodwin and Goodwin, 1993, 456). They found, for example, that many content areas that should be measured in young children (for example, motivational competence) are ignored; what measures are used often don’t correspond to actual performance (for example, language readiness); and finally, the end-users of such testing are not trained or competent in their interpretation or application. A very important point they raise is the issue of what children should be tested for: what they can do, rather than what they cannot, and this fits in well with the MEXT’s stated policy for language education to be stress-free and to engender positive attitudes and a sense of success.

Testing guidelines
By combining ideas from Goodwin and Goodwin (1993), Hasselgreen (2005), McKay (2006), and the National Association for the Education of Young Children (1997), several suggestions for testing YLLs can be derived. Careful introduction of the activities and materials should be given individually tailored to each child to respect her developmental level. During the assessment process, support should be continuously provided, and be related directly and concretely to immediate performance of the child to make it more salient.
   Testing scenarios that set up a child for testing, and then leave them to independently “sink or swim” with the process is not advisable with younger children is. Given the child’s vulnerabilities and need for security, assessment should be conducted in familiar settings. The place where the child usually learns is the best place to elicit optimal performance. The people whom the child trusts (the homeroom teacher) can provide the most appropriate feedback and support. The process the child undergoes should be continuous, seamless extensions of activities the child already has engaged during regular lessons. Assessment should elicit optimal performance, focusing on what the child can do rather than on deficits, to give the child a sense of success with the materials, and a clear sense that she is progressing and achieving.
   One of the most important outcomes of assessment is to create positive attitudes to language and learning – far from the “exam hell” characteristic of later learning stages. Exam hell refers to the period of time in which Japanese students prepare for taking entrance examinations at top-ranking universities. The process is grueling, mind-numbingly intense, and miserable, so much so that many students arrive at university burnt out, or keep re-taking the exam so that they can go to the school of their choice. Newspapers often signal the advent of exam season with an article decrying the practice, such as Gordenker (2002) in the Japan Times.
   Finally, because children are at such variance in development, assessment should: a) directly observe performance, not extrapolate it from paper-and-pencil proxies; b) be flexibly applied to give adequate coverage of a spectrum of skill sets; and c) be given in series, at many times, since a one-shot application will not be indicative of the progression the child is making from day-to-day.

Assessment Portfolios: Evaluation and Application
Assessment portfolios meet every one of the guidelines outlined above. Because they are conducted in the classroom by the homeroom teacher, they can be given in ways sensitive to individual children’s needs, and in a familiar place using familiar procedures. The teacher is best placed to observe her young learners continuously, and the placement of portfolios in classroom bookshelves for easy access by the children allows them to be continuously updated, reflected over, and negotiated upon, so that they represent the best samples of the student’s performance.
   Assessment portfolios can contain checklists with instructional objectives on regular activities; rating scales, that could handle skills with several components (eg. following directions in different situations); screening tests; anecdotal records, which are restricted to factual observations and non-judgmental records on the child’s performance; the child’s own notes, self-evaluation efforts, journal entries, and in general examples of the child’s best work culled from writing assignments, drawings and craftwork, and audio performances.

Also important to bear in mind is that this material is derived from an ESL source (Oxford University Press), and so many of the skills shown here might be characteristic of high schools rather than elementary schools in Japan. More appropriate content for an EFL context might be “can identify letters”, “can follow the main story by drawing a cartoon panel”, etc. Graded level readers even pose some difficulty. When dealing with absolute beginners, “stories” might only consist of pictures with single word or phrase descriptors.
   While MEXT does not currently envision including reading in its elementary curriculum, Nikolova (2008) offers a compelling argument for its inclusion. According to the Ministry of Education’s own survey, allowing reading in the class is something the kids say that they want. Nikolova also suggests that reading activities give students more choice over their learning content. Finally, much content in middle school and high school concentrates on reading materials, so having students start early at acquiring literacy skills in a communicative content should only accelerate their learning capacity in later years.
   Rating scales. The child herself would use this scale to indicate the depth and breadth of her knowledge on a particular topic. In Figure 3 below (McKay, 2006, 192), the child would write or draw anything she knew about the topic “insects.” Again, in Japan, “insect” might be too abstract a concept for young children to know, or even more importantly to use, so a more common descriptor like “bugs”, or even a picture of a generic-looking bug, might be more appropriate. This is the kind of activity that can be given to the child without any notion that “testing” or “assessment” is being done. This can be woven seamlessly into a lesson.

When a child has gone through several lessons, and has completed rating scales for several, she can choose for herself what scale she thinks represents her best work, and this can be placed in her portfolio. The portfolio contents can be added to or removed as time goes on, so if a better rating scale appears, it can replace the older. Furthermore, if a child’s knowledge of bugs increases, this scale can easily accommodate that expanding knowledge domain. Again one should remember not to use score or grade, but only a happy face to indicate the teacher’s acknowledgement of effort (Paul, 2003, 115ff).
   Screening tests. These are given to children when they embark upon their course of learning, to give the teacher some indication of their entry-level point. These represent a little more of a sticky issue. In ESL settings, the use of Council of Europe or Cambridge University (COE) based tests is perhaps more common (see figure 4 a & b for rating level examples). In Japan, MEXT is planning on implementing an English program from grades 5. Yet the COE guidelines expect children to be ready to “give a short prepared talk” in which they “give their opinions”. Such a guideline would of course be inappropriate to expect in a curriculum which only has one hour of English content per week. COE breakthrough level, the beginner level, does not give sound-letter association any recognition even at a grade 1 level. Vocabulary items are recognized only when embedded within a communicative context (“tell what the weather is like”).

Figure 4a. Screening test levels, grades 1 – 3

Figure 4b. Screening test levels, grades 4 – 6

Again, in an EFL context, a screening test would need to be broken down into smaller steps. At the same time, Japan is increasingly becoming a more heterogeneous society. The luxury of assuming and therefore treating every student as the same is not warranted. Some children come from bilingual homes, or have had extensive experience living abroad, and screening tests need to take those factors into account. If levels are over-simplified, and that plateau at a level far below a bilingual child’s level, that child may well be forced to take boring and tedious classes, with teachers who speak less English than they do (cf. Riordan, 2005). Being vulnerable, a child may feel compelled to submerge their linguistic advantage, or be shushed or punished by the teacher if they demonstrate it. Goto-Butler (2007) also points to growing presence of foreign residents living in Japan, comprising 1.5% of the total population, which is a 50% increase over the last ten years, and currently these children are not cared for in the current system.
   Anecdotal records. In figure 5 (below), a portfolio review is provided (Ioannou-Georgio & Pavlou, 2003, p. 28). In this example, the teacher assesses the child’s development across a range of language modalities. This is an important consideration, as children may show differential progress, with some better at fine motor control (writing) than other areas (talking less because of shyness). I wonder if these types of reviews might fall by default to the native speaker teacher or teaching assistant, and thus if the use of English in the review might take away the benefit. The review, of course, is not done just to satisfy some administrative prerogative, but more importantly is done in an effort to communicate with both the parents (who may not speak English themselves) and the child, who will not understand many of the terms or their implications (terms like improve, context, unknown, handwriting, combining, paragraphs, etc). Thus it could be given in Japanese as well.
   The use of this kind of review would also be restricted due to classroom size (hard to do this with 30 or 40 kids in a class, based on one hour of observation), and the teacher’s already overburdened schedule (yet one more report!). Thus, in Japan, the review form would have to be simplified in several ways: with MEXT prioritizing speaking, and having no reading or writing components in the elementary curriculum, this cuts the content in half.

Figure 5. Anecdotal – Portfolio review

More details about the kind of syllabus content, and a check on the child’s grasp of associated content, would be helpful (ie. the review says “her handwriting has improved”, but does not indicate in what way? Neater? More content? More artistic in combination with graphics?). It cannot be emphasized enough, that children follow altogether normal and different paths of development, and to compare one child with another is tantamount to saying that there is one path to development that all children should follow in tandem. Montessori (1912, p. 107) referred to these types of activities as collective lessons, because it places limits on children to develop and experiment in their own way. Paul (2003) agrees that such child-centered approaches to teaching, and by extension assessment, should be tailored to individuals, and not as a collective, norming exercise.

At the same time, if MEXT does decide to include basic reading and writing skills in the curriculum, I would suggest that MEXT avoid the traditional method for teaching kanji, as described by Reid (1998, pp. 141ff) and the experience his daughter had in elementary school. He describes a curriculum in which his daughter, along with other elementary school students, learned 80 kanji characters in 1st grade, 160 in 2nd grade. This increased to 200 in 3rd grade, etc, until they graduated from high school with about 2000 characters learned. Even though some children may already have known the content, students were led to practice drawing the character hundreds of times, at which point, “everyone moved on the next character… [and] the same characters in the same way in precise lockstep with everybody else” (Reid, 1998, pp. 142–143).
   The repetitive process might be understandable for learning a system in which there are so many different characters and so little time in which to learn them; on the other hand, the 24-letter English alphabet is much simpler to learn, and perhaps could be taught through flashcard games and artistically-driven projects (posters, pictures, etc) that were more inclined to a child’s individual nature (Paul, 2003, p. 88ff; Montessori, 1912, p. 246ff). Repetition may indeed occur, but is determined by the child, not the teacher, to a need that she perceives, and may continue to experiment with the content in different contexts, and at a time which correspond to the child’s inner schedule, not outward mandates (Montessori, 1912, 346ff).

Given the de-emphasis on writing in the Japanese curriculum, the teacher would probably need to exchange a written text with a purely pictorial representation. Perhaps an alternative could involve children make journal entries via an iPod. Even just a sentence or two, or some words sounded into the device, could make for an interesting online record for parents to listen in on. Notice that the teacher only responds to the content in a positive way. There is no evaluation of effort, and achievement is viewed simply towards recognizing the effort that was made, and affirming the experience that the child had as valuable.

Conclusion
Children differ in their rates of development and in their approaches to learning in many ways. These differences should be celebrated and nurtured. Now that MEXT is about to introduce English content into the elementary level curriculum, teaching and learning can be held accountable in a way that can be of service to all stakeholders (especially to parents and their children), in a way that is enjoyable, easy to understand, trustworthy, and one that can improve upon traditional practices that may not have given enough attention to YL differences in the past. To do this, the use of assessment portfolios is fundamental. This paper has shown with various examples that portfolios can be individually tailored to each child’s learning experience, through encouraging teacher feedback which does not attempt to norm performances, and is negotiated with the young learner over the entire course of learning, in a way that shows progress through observed skill development.

References

Amatea, E., Smith-Adcock, S., Villares, E. (2006). From family deficit to family strength: Viewing families' contributions to children's learning from a family resilience perspective. Professional School Counseling, 9(3), 177 – 189.

Center on the Developing Child at Harvard University (2007). A science-based framework for early childhood policy: Using evidence to improve outcomes in learning, behavior, and health for vulnerable children. Available: http://www.developingchild.harvard.edu

Dyson, A. and Genishi, C. (1993). Visions of children as language users: Language and language education in early childhood. In B. Spodek (Ed.), Handbook of research on the education of young children (pp. 122-136). New York: Macmillan.

Goodwin, W. and Goodwin, L. (1993). Young children and measurement: Standardized and non-standardized instruments in early childhood education. In B. Spodek (Ed.), Handbook of research on the education of young children (pp. 441–463). New York: Macmillan.

Gordenker, A. (2002). 'EXAM HELL': Getting into the rat race in middle school. Japan Times, January 11. Available: http://search.japantimes.co.jp/cgi-bin/ek20020111ag.html

Goto Butler, Y. (2007). Foreign language education at elementary schools in Japan: searching for solutions amidst growing diversification. Current issues in language planning, 8(2), 129-147.

Hasselgreen, A. (2005). Assessing the language of young learners. Language Testing, 22(3), 337–354.

Ioannou-Georgio, S. & Pavlou, P. (2003). Assessing young learners. Oxford: Oxford University Press.

Japan Policy & Politics. (2001). Panel to propose English in elementary schools. January 22. Available: http://findarticles.com/p/articles/mi_m0XPQ/is_/ai_70201696?tag=artBody;col1

Montessori, M. (1912). The Montessori method. (Tr. Anne George). Stokes: New York. Available: http://digital.library.upenn.edu/women/montessori/method/method.html

McKay, P. (2006). Assessing young language learners. Cambridge: Cambridge University Press.

MEXT (2001). Bringing out talent and bringing up creativity. Ch 3 in Japanese Government Policies in Education, Culture, Sports, Science and Technology.
http://www.mext.go.jp/b_menu/hakusho/html/hpac200101/hpac200101_2_035.html

MEXT. (2003a). Action Plan to Cultivate “Japanese with English Abilities”. Retrieved March 12, 2008 from: http://www.mext.go.jp/english/topics/03072801.htm

MEXT. (2003b). The education reform plan for the 21st century: The rainbow plan – The seven priority strategies. Retrieved March 12, 2008 from:
http://unpan1.un.org/intradoc/groups/public/documents/APCITY/UNPAN008147.htm

National Association for the Education of Young Children (1997). Guidelines for decisions about developmentally appropriate practice. Retrieved October 15, 2007 from: http://www.naeyc.org/about/positions/dap4.asp

National Center for Languages (2006). My languages portfolio. Retrieved October 1, 2007 from: http://www.nacell.org.uk/

Nikolova, D. (2008) English-teaching in elementary schools in Japan: A review of a current government survey. Asian EFL Journal, 10(1). Available:
http://www.asian-efl-journal.com/March_08_dn.php

Paul, D. (2003). Teaching English to children in Asia. Hong Kong: Pearson-Longman.

Pucket, M. & Black, J. (2002). The young child. 3rd Ed. New York: Macmillan.

Reid, T. (1998). Confucius lives next door: What living in the east teaches us about living in the west. New York: Vintage

Riordan, B. (2005). Language policy for linguistic minority students in Japanese public schools. IULC Working Papers Online, 5. Retrieved May 4, 2008 from:
https://www.indiana.edu/~iulcwp/pdfs/05-riordan.pdf