Sustainable vocabulary acquisition in Japanese classroom with the help of Memrise

Abstract

This paper presents the methodology and results of creating Croatian language materials for e-learning and m-learning of Japanese language vocabulary through Memrise platform. Firstly, we have assessed previous research on the subject of vocabulary acquisition and e-learning, as well as the existing free platforms for language content creation. According to both the literature review and assessment of e-learning tools, Memrise was chosen as most fitting platform. Secondly, materials were creating using language textbooks Genki 1 and 2 as a reference, since they are main Japanese textbooks used in Croatia. Upon the completion of the e-courses, they were distributed to the students and their results recorded. After two semester usage with the group of 27 university students, those who have regularly used Memrise for learning have had 40% better grades at the end of both semesters. Although vocabulary is only a part of the cumulative language acquisition process, sustainable learning e-system can help students to improve all their language skills. We have found that students mostly enjoy using Memrise to study, and that it is beneficial to their results

Full text

Sustainable vocabulary acquisition in Japanese classroom with
the help of Memrise
Sara Librenjaka, Marijana Janjić b,
Kristina Kocijan c
Department of Information and Communication Sciences,
Faculty of Social Sciences and Humanities Zagreb
Ivana Lučića 3, 10000 Zagreb
a sara.librenjak@gmail.com
b marijanajanjic@yahoo.com
c krkocijan@ffzg.hr
Abstract
This paper presents the methodology and results of creating Croatian language materials for e-learning and m-learning of Japanese
language vocabulary through Memrise platform. Firstly, we have assessed previous research on the subject of vocabulary acquisition
and e-learning, as well as the existing free platforms for language content creation. According to both the literature review and
assessment of e-learning tools, Memrise was chosen as most fitting platform. Secondly, materials were creating using language
textbooks Genki 1 and 2 as a reference, since they are main Japanese textbooks used in Croatia. Upon the completion of the e-
courses, they were distributed to the students and their results recorded. After two semester usage with the group of 27 university
students, those who have regularly used Memrise for learning have had 40% better grades at the end of both semesters. Although
vocabulary is only a part of the cumulative language acquisition process, sustainable learning e-system can help students to improve
all their language skills. We have found that students mostly enjoy using Memrise to study, and that it is beneficial to their results.
Keywords: vocabulary acquisition, Japanese, e-learning, spaced repetition, Memrise
ON THE VOCABULARY ACQUISITION AND ITS ISSUES
Japanese study and vocabulary acquisition in Croatian learners
This paper deals with specific issue in language learning – the acquisition of vocabulary, and the development of
innovative e-learning and m-learning methods in order to overcome the issue in question. We will deal with Japanese
language, and Croatian learners in this paper. Japanese, as a non-Indo-European language, presents a challenge in the
vocabulary acquisition area, since a Croatian learner cannot memorize words with the aid of similarity (as is the case
with the Indo-European languages), or in most cases, is not exposed to native speakers to such extent that he or she can
acquire the language naturally. Since vocabulary and grammar are building blocks on which other language skills –
listening, reading, speaking and writing – can grow, it is important to ensure that learners develop their vocabulary well:
both in volume and with consistent speed.
In Croatia, most learners of Japanese use traditional methods of studying vocabulary. According to the survey across
Croatian cities where Japanese course is available (Janjić, Librenjak and Kocijan, 2016), only 25% of students have had
a chance to encounter e-learning and m-learning in their language classrooms, while they rate the relevance of such
study methods 4.5 out of 5, and the average interest of a student in using e-learning or mobile learning is 4.1 out of 5.
These figures state that digitalization of Japanese materials in Croatia is well wanted and necessary.
Students of Japanese in Croatia do not generally rate vocabulary acquisition as one of the biggest issues in learning
Japanese, but are stating the script (ideographic kanji characters, with around 2000 characters they need to memorize)
and speaking as the most difficult in their learning process, with 38% of students stating the script and 22% of students
stating speaking as most difficult, according to Janjić, Librenjak and Kocijan (2016). Still, one can argue that
vocabulary is one of the hidden areas that hinder one’s ability to be more expressive in their speech, and enable to
memorize more kanji if the word behind it is knows, as it is less of a difficulty for Japanese children who only need to

connect the shape of a character with the sound. We do recognize the importance of developing proper methods for
teaching kanji, and we have dealt with the issue in Librenjak, Kocijan and Dovedan Han (2012).
A case for Memrise as supported by research in learning strategies
Memrise is a vocabulary learning website which exists in the form of web application as well as an Android and
iPhone application free of charge. It provides the option of creating courses with desired vocabulary in almost any
language, or using any of the existing for individual learning . Its main qualities are ease of accessibility, option of
mobile learning at any place and time, employment of spaced repetition algorithms for better recall, providing
knowledge testing methods, and usage of gamification elements such as scoreboards. This section seeks to connect the
findings in learning strategies research with the functionalities of Memrise, thus proving it to be a feasible choice to use
in classroom setting, and not just individual and goal-less learning.
Learners sometimes lack the discipline to study outside of the classroom and rely too much on the teacher, but
according to Macaro (2001), the strategies for vocabulary acquisition are largely individual and learners need to rely on
themselves and their own practices more than completely relying on a teaching process. Digital methods that adapt to
the learner, i.e. spaced repetition methods which store individual memorized data and assess the knowledge according
to this data, would be beneficial in this context. Memrise uses spaced repetition in order to facilitate memorization and
recall, but does so on an individual level.
Still, self-regulated learning can be difficult to manually regulate at times. A learner would need to self-assess the
review intervals, which is not only difficult to predict, but also an unnecessary task which takes time that could be better
invested. Pyc, Agarwal and Roediger (2014) argue that it is important to review items at spaced intervals, but it is much
easier to do so using a computer regulated algorithm such as the one used in Memrise.
Lastly, testing is a method of vocabulary acquisition which, according to Brown, Roediger and McDaniel (2014) is
more efficient then studying without having to generate response, and especially if an effort is put into responding to a
question. Brown et al. state that students retain more information when they have to work harder to answer a question,
and when they just go over vocabulary lists, they tend to overstate their knowledge if they are not explicitly tested on it.
Along with this research, Huesler and Metcalfe (2012) have found that learners often avoid making the mistakes in
language learning, but are wrong in this belief, as error making is greatly correlated with better recall of word pairs.
Kornell, Hays and Bjork (2009) also state that feedback which comes after unsuccessful attempt leads to more learning,
then spending time to simply memorize the word pair. Since Memrise constantly uses testing methods, as opposed to
simply presenting the data, it should serve better in the task of helping students with vocabulary recall.
VOCABULARY E-LEARNING USING MEMRISE AND SIMILAR TOOLS
Comparison of Memrise and similar tools
Along with Memrise, there are few other online tools that provide similar functions. Since our goal was language
learning, with the vocabulary acquisition in the focus, not every flashcard and e-learning site was adequate. Table 1
presents similar tools and their characteristics.
Table 1. E-learning and m-learning websites and tools
Name of the
service
Website Tool type Creation
of own
materials
Mobile
apps
Pricing Spaced
repetition
algorithm
Testing types
Anki ankisrs.net downloadable app, simple
flashcards with powerful
algorithm
yes yes free,
24.99$
for iOS
yes self-checking
Cerego cerego.com elaborate memorization site
with many courses and nice
design
yes yes free,
10$/m
pro
yes multiple choice w/
none of the above
option, spelling,
audio questions
CoboCards cobocards.com/en flashcards web service with
slightly outdated design,
mainly for German market
limited yes free,
5.57$/m
pro
yes flashcards
viewing, no test
Cram www.cram.com simple flashcards with nice
and usable design
yes yes free no self-checking,
generated non-
dynamic test,
games
Duolingo www.duolingo.com non-customizable language no yes free no multiple choice,

courses spelling, audio
Memrise www.memrise.com elaborate memorization site
with many courses and nice
design
yes yes free,
9$/m pro
yes multiple choice w/
none of the above
option, spelling,
audio questions
Mnemosyne mnemosyne-proj.org flashcards downloadable app,
not web-based
yes yes free yes self-checking
Learn that
Word
www.learnthat.org English vocabulary and
spelling training site
no no free,
9.50$/m
pro
no quizzes, tests,
concentrated on
English voc and
spelling
OpenCards opencards.info flashcard app which can use
PowerPoint files
yes no free yes self-checking
Quizlet quizlet.com simple flashcards with nice
and usable design
yes yes free,
34,99$/y
pro
no self-checking,
generated non-
dynamic test,
games
Skritter www.skritter.com Japanese and Chinese
characters flashcards
no yes 14,99$/m
onth
yes writing, reading
SuperMemo supermemo.com flashcards downloadable app,
not web-based, a bit outdated
yes yes 66$ for
program
yes self-checking
Synap synap.ac web based flashcard quizzes
(MCQ)
yes no free yes multiple choice
quizzes with lots
of feedback
Memrise was chosen as one of the best e-learning tools, providing users – both teachers and learners – with most
functionalities, great application for mobile learning, and easily accessible and fun interface. Gamification element in
the form of the leaderboard was also expected to be an important advantage. Other tools also showed some advantages,
but were more applicable for some other area in language study, such as Anki for ideographic kanji characters, or
Quizlet or Cram for fast studying, using games or quality TTS (text to speech synthesis) of whole sentences for
pronunciation learning.
There is already some academic research on the usefulness of Memrise in classroom. Walker (2015), who did a case
study in using Memrise to teach Latin vocabulary, states that “the quantitative data strongly support the superiority of
Memrise as a way of learning Latin vocabulary, over the methods that students would otherwise self-select”. Her
students found using Memrise enjoyable and convenient, and she concludes that Memrise would be a useful tool for
Latin teachers to employ in teaching vocabulary.
Japanese courses already present on Memrise
As Memrise was chosen as the platform on which we would develop Croatian-Japanese materials for study, some
preliminary survey of already created materials was necessary. At the moment of creation of our courses, there was no
Croatian-Japanese materials online, so we reviewed English-Japanese materials, as English is used partially as a
language of instruction in Japanese language course at the University of Zagreb, and all used textbooks are in English.
There is already a significant number of Japanese language courses present at Memrise, most popular being the one
issued by Memrise itself (Japanese 1) with 164.000 users, and the one issued by user jlptbootcamp (Introduction to
Japanese), also a website dedicated to studying Japanese with the accent on passing the standardized test in Japanese
called JLPT (Japanese Language Proficiency Test). The latter is by far the most popular course, counting as much as
585.000 users. Other popular courses are following prominent Japanese textbooks, such as Minna no Nihongo or Genki,
or an online course and book called Tae Kim’s Guide to Japanese. There are some courses specifically designed for
various levels of JLPT test, N5 being the lowest and N1 being the highest level of the test. Lastly, there are some
courses specialized for the acquisition of some of the Japanese scripts (syllabic hiragana and katakana, or ideographic
kanji), and some very specialized vocabulary areas. Unfortunately, it may be difficult for users to be persistent in using
any of the courses without the proper context – a course or an opportunity to use what they have learned. According to
the survey conducted by Janjić, Librenjak and Kocijan (2016), only 20% of Japanese language students have previously
had used Memrise in their studies, but they have used it more than any other e-learning or m-learning tool.

CREATION OF CROATIAN MATERIALS FOR MEMRISE
In our research so far, we have recognized two main problems for Croatian students when it comes to using e-tools
and m-learning tools for the purpose of vocabulary acquisition. Firstly, there was no systematic course which follows
curriculum, so the students can make most use of its content. Secondly, there was no course in Croatian as the language
pair with Japanese, so even the learners with weaker command of English can make use of it. We believe that such
course is needed also for the purpose of strengthening the Croatian identity in the Japanese learning community, as
opposed to using English as a bridge between Japanese and Croatian learners. This is supported by the survey
conducted by Janjić, Librenjak and Kocijan (2016), where we found that 63% of learners would prefer the native
language materials, even though there is a disparity between young learners (high school students and young adults) and
older learners (past university age) in this area. Younger learners find English materials satisfactory in their pursuits,
while older learners would strongly prefer Croatian materials.
The survey also showed us that most prevalent textbooks in Croatia are Genki 1 and Genki 2 by Japan Times, and
thus they were chosen for the basis of our courses. We have used Common European Framework of Reference for
Languages (CEFRL) to divide the courses into levels from A1 to B2. Textbooks Genki 1 and 2 were used as a reference
in constructing the A1 and A2 courses, respectively, with 768 words in A1 and 751 word in A2 level. For higher levels,
we referred to a number of textbooks, but have used Integrated Course of Intermediate Japanese and Tobira: Gateway to
Advanced Japanese as base textbooks for reference for B1 and B2, respectively. Each level counts about 700 new
words, and after A2 level they are represented both with syllabic kana script and ideographic kanji, so that a student can
review both ways. Words were paired with their pronunciations recorded by native speakers available in public domain
at Forvo website, and publicly created mnemonics were also provided. Bakken and Simpson (2011) argue that young
learners benefit from using mnemonics in their acquisition of knowledge, and as we are aware that young adults are
most prominent population amongst learners of Japanese in Croatia, we believed they would benefit from this approach.
Figure 1 shows Japanese courses word list and a demonstration of a study process.
Fig. 1. (a) Memrise word list; (b) An example of test while studying
However, mere vocabulary lists may not be sufficient for some students. The grammar of Japanese is significantly
different from Croatian, and some learners find it difficult to form a sentence even if they have mastered the necessary
vocabulary. In fact, grammar was listed as third most difficult feature of Japanese language by Croatian students in
survey by Janjić, Librenjak and Kocijan (2016). Having this in mind, we have also created accompanying levels (A1 –
B1, with B2 in making for upcoming higher level students) with grammar in context, i.e. Japanese sentences with their
Croatian equivalents. Each sentence contains only the vocabulary that was learned on the corresponding level of
vocabulary course, and has a new grammar expression in focus. Grammar courses also contain Croatian language
explanations for the forms that would appear in sentences on the next level. Sentences are also paired with their
pronunciation, so learners can naturally acquire the sound of the language as well. Figure 2 shows grammar courses on
Memrise: the explanation, sentence display and testing. Testing of the sentences is performed in three main manners:
multiple choice, typing the sentence, and arranging the scrambled pieces of a sentence. All modes of testing can be
conducted in Japanese, Croatian, or audio mode.

Fig. 2. (a) Grammar explanation in Croatian; (b, c) Examples of using the course
All Memrise courses constructed in this project were manually checked for errors, and updated frequently when an
error is reported. As they are tested in the classrooms, it is believed that they are reliable and correct for the large part,
so we can ensure that students do not learn an incorrect word or form. Higher levels are yet to be added audio, as they
were not yet tested on a larger scale, since there is not enough learners of that level. Table 2 summarizes Japanese
courses created on Memrise, as well as their main sources and number of users.
Table 2. A list of courses on Memrise created in this project
Number of units Main source Users Audio
Vocabulary A1 768 words Genki 1 85 yes
Grammar and sentences A1 134 sentences Genki 1 63 yes
Vocabulary A2 751 word Genki 2 28 yes
Grammar and sentences A2 180 sentences Genki 2 23 partially
Vocabulary B1 772 words Integrated approach to
intermediate Japanese
11 not yet
Grammar and sentences B1 266 sentences Integrated approach to
intermediate Japanese
10 not yet
Vocabulary B2 85 words
(currently in making)
Tobira: A gateway to
advanced Japanese
not published not yet
One can see that there are not so many learners of Japanese compared to those worldwide, but considering the fact
that there are only between one and two hundred active learners in Croatia all together (approximated according to the
results of the 2016 survey by Janjić, Librenjak and Kocijan), we can be satisfied with the current number of users after
an incomplete year of the courses’ existence.
IMPLEMENTATION IN THE JAPANESE CLASROOM
A method of implementation in the beginners Japanese course
In the academic year 2015/2016, these courses were implemented in Japanese language course at the University of
Zagreb. Japanese language 1 (in the winter semester) had 27 students, while Japanese 2 (in the summer semester) had
17 students who have passed Japanese 1. It was an elective course for students of various majors, and not a course for
Japanese majors. At these course, Memrise course Japanese A1 was used as an important part of the curriculum. The
textbook used in the course was Genki 1, which was completed over two semesters with 3 school hours a week. With
each of the 12 units in Genki 1, students were instructed to use Memrise consistently, as a preparation for the each unit.
Research supports implementation of computer based flashcards into curriculum (Altiner, 2011), but it is important
to present it as an option and not a necessity, since not all students have the faculties to use e-learning and m-learning in
the same amount. Thus, usage of Memrise was optional, and students were given a list of words to study in a traditional

manner in case they did not chose e-learning or m-learning, but a large number of students gladly tried using Memrise
course. Before starting the new unit, they were given a vocabulary test, and were required to correctly answer 70%
words in order to pass it. Students who did not pass, had to retake the test at the next class. In addition to vocabulary
quizzes, teacher ensured that students use new vocabulary in speaking and writing exercises. Each week, leaderboard at
the Memrise site was inspected, and most avid users were noted.
Results
Out of the 27 students in Japanese 1 class, 15 were regulars users of Memrise course, and out of 17 students in
Japanese 2 course, 10 used Memrise regularly. This data was later compared to final grades of the course, as shown in
Table 2. Croatian grades range from 1 to 5, 5 being the highest possible grade, 2 lowest passing grade, and 1 meaning
the failure of the class.
Table 3. Results of implementation of Memrise A1 Japanese-Croatian course in the class
Japanese 1 Japanese 2 Cumulative
Regular Memrise users 55% 58% 56.5%
Average passing rate 85% 70% 77.5%
Average grade (if passed) 3.68 / 5 4 / 5 3,84 / 5
Average grade of Memrise users 4.63 / 5 4.5 / 5 4,56 / 5
Average grade of non-users 2.37 / 5 3 / 5 2.68 / 5
Difference between two groups + 48% + 33% + 40,5%
In both semesters, a bit over half of the students used Memrise on a regular basis, but almost all have tried it. Regular
users are considered those who have had score on a leaderboard each week in the given course. Generally, their average
score is significantly larger than non-users. Of course, they were more motivated in other areas of the course as well,
but the contribution of regular vocabulary learning cannot be ignored. We can conclude that using Memrise benefited
new students of Japanese, since they completed the course with better grades, even though quizzes and participation
formed only 1/3rd of the whole grade in the course. Grade is considered an apt measure of their knowledge, because they
were comprehensively tested in speaking, listening, writing, reading and participation. Students who have used
Memrise throughout the course have had 40% better grades in the course - and the knowledge behind it.
CONCLUSION
This article dealt with vocabulary acquisition of Japanese language for Croatian students, and usage of e-learning
and m-learning tools. Firstly, we presented the current state of Japanese learners in Croatia, and their issues in learning.
Even though they do not always recognize the importance of vocabulary acquisition, it is connected with every
language skill, such as speaking and writing which Croatian learners find most difficult in their study. Literature review
pointed us to seek a learning and teaching tool which enhances the recall by enabling students to make mistakes, and
test them often. Memrise, a web based application which also exists in the form of mobile application for m-learning,
was chosen as a platform which would be most usable in the classroom. Studies suggest that it was already successfully
employed in teaching vocabulary in the language classroom.
However, there was no Croatian language course with materials which follow Japanese language class curriculum, so
no sustainable learning was possible with the language pair Japanese-Croatian. We have thus created Japanese course
from levels A1 to B1, to be used in Croatian classrooms as an additional tool for teachers and learners. We have chosen
vocabulary items and sentences from most used Japanese textbook in Croatia, and added audio and visual mnemonics
for most of the content.
These courses were presented to students throughout Croatia, but one class of beginners using Memrise A1 course
was thoroughly monitored in their usage for two semesters. More than half of the students in the class opted to use
Memrise regularly every week. Even though that did not affect their grade directly, they did perform better on almost
every part of the class. Cumulatively, Memrise users group have had 40% better grades at the end of the class then the
group of non-regular users.
We can conclude that creating a systematic, sustainable courses which follow curriculum in the classroom benefit
students’ vocabulary acquisition as well as their overall progress. It is important that teachers and instructors include
that course in their class through quizzes and exercises which would provide the students the opportunity to use and
show their knowledge. Memrise also provides the function of keeping score each week and month, and displays it in

leaderboard. This gamification element cannot be ignored in the education of young adults, since it served beneficially
in their motivation. Overall, after two semesters of use, we would recommend similar courses for vocabulary
acquisition of foreign language, both on individual level and in the classroom.
Acknowledgements
This research was supported by European Union, European Social Funds.
We would also like to thank our dedicated students and collaborators Marija Bilić, Iva Borovec, Ana Horvatin, Tea
Lukačević, Veronika Kovač, Tena Omerović, Fran Široki and Laura Tandarić the hard work they put into inputting data
into courses and checking the content.
References
M. Janjić, S. Librenjak, K. Kocijan. (2016) Croatian Students’ Attitudes towards Technology Usage in Teaching Asian languages - a Field Research.
Proceedings of the 39th International Convention MIPRO, Rijeka. pp.1051-1055.
Librenjak S., Vučković K., Dovedan Han, Z. (2012) Multimedia assisted learning of Japanese kanji characters. Proceedings of the 35th International
Convention MIPRO, Rijeka. pp-1284-1289.
Macaro, E. (2001). Language strategies in foreign and second language classrooms. London: Continuum.
Brown, P.C., Roediger, H.L., & McDaniel, M.A. (2014). The science of successful learning. In Make it stick. Cambridge, MA: Harvard University
Press.
Huelser, B.J. & Metcalfe, J. (2012). Making related errors facilitates learning, but learners do not know it. Memory & Cognition 40, pp. 514-27.
Pyc, M.A, Agarwal, P.K., & Roediger, H.L. (2014). Test-enhanced Learning. In Benassi, V.A, Overson, C.E. & Hakala, C.M. (Eds.). Applying the
science of learning in education: Infusing psychological science into the curriculum. pp. 78-90. Retrieved March 21, 2015. from the Society for the
Teaching of Psychology web site: http:// teachpsych.org/ebooks/asle2014/index.php
Kornell, N., Hays, M.J. & Bjork, R.A. (2009). Unsuccessful Retrieval Attempts Enhance Subsequent Learning. Journal of Experimental Psychology:
Learning, Memory, and Cognition 35 (4), pp. 989-998.
Walker, L. (2015). The impact of using Memrise on student perceptions of learning Latin vocabulary and on long-term memory of words. Journal of
Classics Teaching, 16, pp 14-20
Banno, E., Ohno, Y., Sakane, Y., Shinagawa, C., Takashiki, K., Japan Times, ltd., & JP Trading, Inc. (1999). Genki: An integrated course in
elementary Japanese = shokyu Nihongo "genki". ̄Tokyo: Japan Times.̄ ̄
Banno, E., Ohno, Y., Sakane, Y., Shinagawa, C., Takashiki, K., Japan Times, ltd., & JP Trading, Inc. (2011). Genki II: An integrated course in
elementary Japanese = shokyu Nihongo "genki". ̄Tokyo: Japan Times.̄ ̄
Council of Europe (2011). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Council of Europe. website:
http://www.coe.int/t/dg4/linguistic/Cadre1_en.asp
Miura, A., & McGloin, N. H. (2008). An integrated approach to intermediate Japanese: Chukyu no Nihongō. [revised edition] Tokyo: Japan Times.
Oka, M., Tsutsui, M., & Kondo, J. (2009).̄ Jokyu e no tobira: Kontentsu to maruchimedia de manabu nihongo = Tobira : gateway to advanced ̄ ̄
Japanese learning through content and multimedia. Tokyo: Kuroshio Shuppan.̄ ̄
Forvo, the pronunciation dictionary. web site: http://forvo.com/
Bakken, J. & Simpson, C. (2011) Mnemonic Strategies: Success for the Young-Adult Learner. Journal of Human Resources & Adult Learning, Vol. 7
Issue 2, 79
Altiner, C. (2011) Integrating a computer-based flashcard program into academic vocabulary learning. Graduate Theses and Dissertations. Paper
10160.