Following the January and July 2023 issues, Emiko Izumi from the School of Education & Graduate School of Education at Kwansei Gakuin University discusses the content and results of her collaborative research on the teaching and assessment of reading and writing for early English as foreign language (EFL) learners in Japan
Reading and writing lessons should be conducted in small, incremental steps. Especially for Japanese learners of English, it is very difficult to learn the alphabet (upper- and lower-case letters) and to connect letters and sounds. The English language has a different phonological system and graphemes from Japanese, so it is important for learners to recognise phonemes and develop phonological awareness.
With this being said, we recommend a balanced approach that introduces bottom-up strategies, such as phonics instruction that teaches the rules for linking sounds and letters, and a top-down approach, such as understanding content from songs, stories, and picture books with a focus on meaning.
A bottom-up approach to teaching and assessing literacy: Correlations between self-rating and tests of alphabet and phonological awareness
Starting in 2020, when the latest national curriculum was introduced, students in the upper grades of Japanese public elementary schools are now able take twice-weekly foreign language classes. This article discusses a unique literacy education programme and the relationship between literacy and self-assessments of the students who participated in the programme.
In the district where the programme was implemented, about 15,000 students have received the systematic and explicit bottom-up instruction every year since 2017. This instruction involves about seven minutes of literacy instruction in a weekly class from the third grade.
Since 2016, Dr Mitsue Allen-Tamai, a member of Dr Izumi‘s research group, has been researching how students who received the literacy instruction evaluated their abilities. Among the many things we have learned, the Can-Do scale evaluation was very effective in allowing students to evaluate their ability accurately.
93 sixth graders participated in this study in 2022 and student self-appraisals were collected along with three criterion measures of reading ability: letter-name knowledge, sound-letter knowledge, and word knowledge.
The Can-Do scale self-assessment was used for letter-name knowledge and a three-point Likert scale assessment was used to evaluate their sound-letter knowledge (consonants, short vowels, consonant digraphs, and long vowels). We created experimental Can-Do scales for decoding.
This uses a five-level scale, asking at what level words can be read. Level 1 is the ability to read words with ‘consonant + short vowel + consonant,’ like cat and dog; level 2 includes words with consonant digraphs like shop and dish; level 3 includes words with consonant clusters like frog and clock; level 4 includes final-e words like cake and nice; and, finally, level 5 includes words with vowel digraphs like rain and boat.
As a result, 74% of the students were able to respond quickly to the uppercase letters and 67% to the lowercase letters, since the target was three or higher on a four-point scale. Based on the test results, the average percentage of correct responses for both uppercase and lowercase letters was close to 90%, so it can be said that the goal was achieved.
As for the relationship between sounds and letters, the children seemed to have the most difficulty with consonants, followed by two-letter consonants, short vowels, and finally, long vowels, in this order. The second level of the three-level scale- “fairly well understood” was most popular, which meant that many children understood those sound-letter associations.
However, one-quarter of the students answered, “I don’t understand at all” (level 1) for long vowels, so decoding words with long vowels was still perceived to be quite difficult for elementary school children.
Correlations between self-assessments showed that the strongest correlation was found between self-assessments of uppercase and lowercase letter knowledge (r = .914), followed by consonant and consonant digraph knowledge (r = .723), and finally between short and long vowel knowledge (r = .720).
Next, we examined the relationship between the skill measured in each test and its self-assessment. Interestingly, the highest correlation coefficient was found between each skill and the self-assessment of decoding, followed by that of consonants. It is interesting to note that the correlation coefficient between the test score of uppercase letters and the self-evaluation of decoding was higher than that between the test score and uppercase self-evaluation. The same was true for the lowercase, consonant, short vowel, and long vowel knowledge.
Teaching and assessing literacy through a top-down approach
As one sector of Dr Izumi’s research group, the top-down method for fostering elementary school students’ emergent literacy has been studied by Professor Mayumi Tabuchi and her colleagues. A series of effective English activities, followed by telling a story with the picture book, has been practiced and named the ‘Learning by Storytelling’ (LBS) method.
A qualitative analysis was conducted based on students’ responses to three prompts. There were three questions to match the goal of each lesson for students to choose from four scales and a space for their free description to assess their own performance.
The LBS method consists of four main parts:
- Interactive storytelling, that is, telling the story of a picture book while interacting with students;
- Reordering the plot of the story as a class and in groups;
- Retelling the story in a group; and
- An output activity depending on the age and cognitive level of the students.
At a private elementary school where students are taught English twice a week, sixth graders (N=30) had five lessons under this LBS method and the Can-Do scale self-assessments were conducted four times.
The results show the correlations between each unit quiz and its self-assessment increased (r = .492 to r = .691) as the lessons proceeded. It was also noted that those who assessed themselves as low as 1 (It is still very hard) or 2 (I could understand with the visuals in the book somehow) when asked “Do you understand the story?” had changed notably.
The qualitative analysis of the students’ comments showed they gradually gained confidence in understanding the contents of the stories in English, even with their limited knowledge, scaffolded by visuals and the teacher’s storytelling. Furthermore, they became more motivated to read the story on their own as they wanted to know what would happen next.
Teaching reading and writing to elementary school students in Japan started recently in 2020 and, due to the difficulty in understanding sound-letter relations and reading words, more research in this area is in urgent need. Besides the need to implement a balanced of both bottom-up and top-down approaches, we need to keep searching for more tools to assess Japanese people’s reading ability as it is such a complex area.
Our study suggests that using this Can-Do scale self-assessment as a formative tool should share the teachers’ goals with students, help students to realise for themselves what they can or cannot do, and lead to their increased self-efficacy.
