A Tailored Intensive
Vocabulary Trainer Using an Online Flashcard Site
Yoneoka, Judy (Kumamoto Gakuen University)
Key
Words： Learner
autonomy, Flashcards, Vocabulary learning, short-term memory, long-term memory
1. Introduction
Flashcards[i]
have been a standard foreign language vocabulary aid most likely from the
beginning of the invention of paper.
Based on developments in the past century in cognitive memory research,
however, flashcards have been developed into mechanized and systematic
vocabulary tools by B.F. Skinner and later by Sebastian Leitner (1972). With
the advent of computers, flashcards programs such as Lernkartei for German and
ALICE for Spanish appeared. Further developments in computer technology have
led to the flashcard system going online in various forms (freeware, shareware,
opensource, java) and on several sites, both commercial (MemoryLifter,
Flashcard Station, etc.) and non-commericial (Mentalcode, VocusPocus,
Studystack, etc.). Cellphone and PDA versions have also recently begun to
appear.
The
Leitner system (also called the “box system”), although in general use in
various flashcard systems on the Net, is relatively unfamiliar in ESL and
vocabulary circles. This is perhaps because flashcard systems in general are
often dismissed as ineffective in contrast to more sophisticated vocabulary
acquisition methods such as learning vocabulary in context or within semantic
fields. In the light of research on learner styles and autonomy, however,
flashcards offer a viable option for learners with the appropriate skills,
preferences, and motivation.
This
research will begin with an overview of the flashcard debate and discuss how
the Leitner system enhances the effectiveness of flashcard study. It will
proceed to explain a similar flashcard study system developed by the author and
document a two-month experiment in studying Korean online using this system.
Finally, it will be showed how online and offline flashcard resources can be
combined with the study system to provide a well-rounded vocabulary study
program for ESL university students.
1.1 The Flashcard Debate
An online discussion on
flashcards at how-to-learn-any-language.com/ reveals several differences in
opinion regarding their usefulness:
*Flash cards are easily the best vocabulary
tool out there... I'm a strict adherent to one-word-a-card … and it's worked so
far.
*No offence, but I disagree with you
completely. I don't think you can beat reading, reading and even more reading
for picking up vocabulary. …. I find them to be too much like a "phone
book" list of words and that's artificial for me.
*I'd agree with you that flash cards, per se,
don't necessarily have anything over general reading. However, when combined
with programmed spaced repetition via a Leitner-style box system (software or
hardware), they are an extremely effective method of learning and most
importantly retaining vocabulary.
*I've never been a big fan of flashcards … but
I am learning to appreciate them for their portability…Will the words be as
natural as if I'd encountered them in reading or listening? No. But it's better
than not reviewing or encountering them at all.
*For me I have had good luck with short term
retention with flashcards (for exams for example), but for long term retention
reading in my target language has really been the best. … I just find
flashcards very tedious and I don't think you can learn anything from something
you find tedious. Others find them very helpful and I think that's great,
whatever works!
The
discussion here brings up several strengths (portability, systematicity, effectiveness for
learning) and weaknesses (artificiality, lack of context, tediousness) of
flashcards in contrast to reading. To these could be added the advantages of
learner choice and autonomy and the disadvantage of having to spend time to
create cards (although perhaps can be included in the interpretation of “tedious”).
Another advantage for some learners is the quiz-like nature of the activity
itself. In general, we can say that flashcards lend themselves more to those
who enjoy puzzles and games than to those who would rather curl up with a good
book..
The
necessity of a spaced repetition component (also called “graduated interval
recall”) to ensure the effectiveness of flashcards for long-term retention is
also mentioned in the discussion. Spaced repetition means that a specific review
schedule is set and used by the learner to repeat stacks of cards at certain
intervals to ensure that the vocabulary learned is still in the long-term
memory. This component is often overlooked by occasional flashcard users, but
crucial to success with flashcards. It is an integral part of many online
flashcard systems and software, most of which are based on the Leitner box
system. Section 2 reviews the Leitner system briefly, and contrasts it with a
similar system developed by the author.
2. The
Leitner System: A Brief Review
The
Leitner box system (German psychologist Sebastian
Leitner) is a method for learning and retaining vocabulary in both short-term
and long-term memory. In its low-tech version, an actual box with several
compartments is used to organize flashcards in terms of relative retention in
the user’s memory. When cards from a certain box are reviewed, those that were
remembered are promoted to the next box, whereas those that were forgotten are
sent back to the first box. Thus, not only are cards reviewed systematically,
but those that are more problematic for the learner are reviewed more often. A
specific review schedule is set and adhered to, providing the spaced repetition
component described above.
Online
versions of the Leitner system provide both the flashcards and the
compartments, and promote or demote those cards automatically. The commercial
cite MemoryLifter, for example, explains the algorithm used in their system as
follows (2004, online):
- Initially, all cards are derived from the Pool,
which contains all cards available.
- Check whether any of the boxes are full. If so, ask
for a card from that box. If none of the boxes are full, place a new card
from the pool in the first box and ask it.
- If the card is answered correctly, promote it to the
next box; otherwise, demote it to box 1.
- When placing a card in a box, put it at the end.
When asking a card from the box, ask in the order it was placed in the box
(first in - first out principle).
- Continue with step 2 - 4 until done.
On
some Leitner-based online systems, the spaced repetition intervals are preset.[ii]
On Flashcard Exchange (2004, online) for example, “an e-mail is sent when it is time to review a cardfile again. The
default review periods are based on the number of times you have successfully
completed a cardfile.” After the first completion of a set, the next review is
four days later, after the second completion it is seven days later, etc. Another characteristic of the Leiter-based system is that it
concentrates on receptive retention of the target word, and does not
specifically require production. Cards can be flipped to test production, of
course, but this is generally treated only as an optional side step in the
learning system. This means that practice of writing, a crucial part of
learning a language with a different writing system such as Chinese, Japanese
or Korean for English speakers, may tend to be overlooked.
This
difficulty is solved in a similar system developed independently by the author
in the 1970s to study Japanese (hereafter referred to as the JY method) using
homemade kanji cards, and adapted to the paid system available on www.flashcardexchange.com to study
Korean in 2006. In the JY method, the study schedule is set by the day, fixing
an intuitively easy-to-follow review plan. Cards are promoted weekly from the
Pool, first into a reading stack and then into a writing stack. Missed cards
are not demoted, but stay in the same stack. Promoted cards are replaced by new
cards from the Pool so that the reading stack retains a set number of cards
(20-25 were used when the author studied Japanese, 50-100 in the online Korean
version). Each stack is labeled with a day of the week, on which they are to be
reviewed.
Figure
1 demonstrates the first three weeks of the JY weekly repetition flashcard
method. In the first week, only reading stacks are studied, but both reading
and writing stacks are ready for study from the second week. In the figure,
after testing the daily reading cards, a total of 45 cards were recalled
correctly after one week, and graduated from the reading files to newly-formed
writing files. New cards were added to keep the number in each reading stack at
20.
|
|
Monday |
Tuesday |
Wednesday |
Thursday |
Friday |
Saturday |
Sunday |
|
WEEK １ |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
|
140 cards studied |
|||||||
|
WEEK 2 |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
|
140 cards |
↓ 5 |
↓7 |
↓８ |
↓３ |
↓６ |
↓１２ |
↓４ |
|
45 cards |
WRITE (5) |
WRITE (7) |
WRITE (8) |
WRITE (3) |
WRITE (6) |
WRITE (12) |
WRITE (4) |
|
195 cards studied (45 new cards introduced
from pool) |
|||||||
|
WEEK 3- |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
READ (20) |
|
140 cards |
↓ 4 |
↓ 5 |
↓ 3 |
↓ 9 |
↓ 6 |
↓ 8 |
↓ 3 |
|
＋83cards |
WRITE (9) |
WRITE (12) |
WRITE (11) |
WRITE (12) |
WRITE (12) |
WRITE (20) |
WRITE (7) |
|
223 cards |
↓ -2 |
↓-4 |
↓-3 |
↓-3 |
↓ -1 |
↓-7 |
↓-1 |
|
- 20 cards |
DONE (2) |
DONE (4) |
DONE (3) |
DONE (3) |
DONE (1) |
DONE (7) |
DONE (1) |
|
203 cards studied (83 new cards introduced
from the pool, 20 cards completed) |
|||||||
Figure 1. The JY Weekly Repetition Flashcard Method
Although the actual number of cards reviewed and remembered each
day will vary, personal experience has shown that one to two practice sessions
each day results in approximately 1/4 of the cards being recalled the next
week. This means that of the 140 cards studied weekly in Figure 1,
approximately 35 cards will “graduate” into the writing and done files
respectively. That is, from the third week approximately 35 new words can be
consistently expected to be acquired each week. Each card takes at least 3
weeks to progress to the done stack, but can take as long as needed to be
acquired. Especially difficult-to-remember cards have been known to linger in a
stack for several months or more!
The
choice of whether and when to review the “done” cards is up to the learner, but
should probably take place whenever a done stack reaches a certain number (say
20 or more) of cards. Cards that were not remembered should go back into the
pool to restart the process.
3. A pilot
experiment in Korean vocabulary acquisition using the JY method with www.flashcardexchange.com
Flashcard
exchange (hereafter referred to FE) is an online flashcard database with over
4,000,000 user-created flashcards available for anyone to study. The ability to
create, use and review flashcards is free for all users, but the compartment
function, which allows for subdivisions or compartments within flashcard files,
is only available to paying members (a one-time $19.95 fee). The addition of
compartments allows for a Leitner-type review system, which is built into the
fee-based program and available for paying users.
In
order to adapt the JY weekly repetition method for use with this site, the
built-in review schedule was not used.
Rather, files were named by the day of the week (e.g. Monread, Wedwrite,
etc.). Each file was created with three compartments. The first compartment
holds the cards to be tested; the third compartment houses cards that were
previously promoted. The second compartment serves two functions: (1) when
testing cards, it serves as a holding place for cards that have been remembered
until they can be copied and pasted into the corresponding writefile and
promoted up to the third compartment, and (2) when studying cards, it is a
temporary holder for those cards that have successfully been processed in the
learner’s short term memory. These are then demoted to the first compartment
after the day’s study is complete.
This
modified system was used to study Korean vocabulary over a period of 2 months,
from 2/28 to 4/30/2006. Two reading files (50-100 words each) and one writing
file were studied daily. The words to be studied were arranged in 50-word
subfiles[iii]
from two sources on FE: (1) a file of 739 low intermediate level words created
by bair787, and (2) an advanced file of 2316 words recreated by the author (the
original file by another author has since been deleted). Thus approximately
3055 words in 62 50-word sets were prepared for study.
The
first week began with 700 words, i.e. 100 words in 2 50-word read files per
day. From the second week, when a file fell below 50 words, a new file of 50
more words was added.[iv]
Study sessions lasted approximately an hour, and were repeated during the day
if desired (usually no more than once). By the end of the two months, another
1439 words had been introduced into the system, at a rate of approximately 24
new words a day (=1 new file every two days) Of the total 2139 words in the
system, 1124 had advanced to the writing file at the end of two months, and 672
of these had graduated out of the system entirely. This indicates that over
half of the words had been retained in long-term memory for a week at least
once, averaging out to approximately 19 words per day.
It
must be noted that the author had previous experience studying Korean and was
already familiar with approximately 250 of the words on the low intermediate
list, although had never written them. This number is approximately 37% of the
672 graduated words. Assuming that these previously known words were all among
the “graduates”, however, still leaves 672-250= 422 new words acquired in
writing and 1124-250=874 in reading. Of the 19 words per day mentioned above, 5
of these would have corresponded to words previously known.
To
test retention, approximately half of the 672 graduated words were tested again
at random 45 days later. Retention rate was found to be approximately 80%.
Adjusting this to account for the previously-known words mentioned above (assuming
again that all previously known words had already been graduated and were being
retained at 100% rate), the retention rate for newly acquired words worked out
to approximately 64% after 1 1/2 months.
4. Adapting and supplementing Flashcard Exchange for ESL students
As mentioned in the
introduction, appropriate skills, preferences and motivation are prerequisite
for success with flashcards. Thus it may not be suitable for in-class work with
a variety of students. However, if introduced and offered as a study tool, it
could well be used by students who have time, facilities (i.e. a computer
available) and self-motivation. It is recommended that the student make use of
the compartments by paying the one-time fee, but a suitable substitute for
compartments can be created on the site by making and renaming files combined
with much copying and pasting.
It
is suggested that students begin with the 2235-word General Service List, which
is available on the FE site with Japanese translations. For university
students, however, most of this list should be already known. More advanced
students can try the 570-word Academic Word List (AWL, Coxhead 2001) available
on the site.
One
disadvantage of using FE—which ironically is perhaps the main advantage of its
low-tech counterpart—is its lack of portability and accessibility. Many
students may not have access to Internet at home, and may not have sufficient
daily chunks of time to use the system at school. However,:this disadvantage
can be overcome by shrewd combination with other downloadable online resources.
Pauker
(http://sourceforge.net/projects/pauker/) is a freeware downloadable Java
application into which FE files can be directly imported. It uses Unicode and
thereby avoids special character issues common on other sites. A set of cards
is reviewed in approximately 15 minutes, with each card passing through
ultra-short term memory (18 seconds), short-term memory (12 minutes) and a
final general review. When Internet access is impossible, this freeware provides
a viable offline alternative for reading files. However, due to the timed
nature of the program, it is not suitable for practicing writing files.
Figure 2．Pauker Screenshots
Although
Pauker can be used by students who have computers at home but no Internet
access, this still does not help those students who do not have computers
available. A cell phone version of Pauker has recently become available, but
whether it is compatible with any or all of the Japanese cellphone platforms
has not yet been determined. Other cellphone and PDA based flashcard software
is also easily found on the Net, but again compatibility both with Japanese
language and Japanese celllphones may prove to be problematic. This is a
subject for further study.
Another
previously discussed disadvantage of flashcards is that of lack of context.
When using FE to study established vocabulary lists such as the GSL and AWL,
however, supplementary online materials that allow a greater range of study
choices are relatively easy to find. With respect to the AWL, especially, the
following sites offer some excellent self-study supplementary material:
(1)
The AWL Highlighter and Gapmaker at
Nottingham University. On the Academic Vocabulary site at Nottingham University
http://www.nottingham.ac.uk/%7Ealzsh3/acvocab/index.htm , the AWL highlighter
can be used to produce text with the AWL vocabulary highlighted. This allows
students to produce their own readings in which they can review the vocabulary
they have studied in context. The Gapmaker on the same site can be used to
produce similar cloze style exercises for testing and reviewing the vocabulary.
Figure 3.
Results screenshots of the AWL highlighter (left) and Gapmaker (right)
(2)
Activities for ESL Students (http://a4esl.org/) a project of the Internet TESL
Journal (http://iteslj.org/) with
contributions by teachers around Japan and the world. It boasts a series of multiple
choice quizzes for both the GSL and AWL contributed by Kelly Quinn. Figure 4
shows a screenshot of one such quiz.
Figure 4.
Multiple choice quiz for AWL from Activities for ESL Students (http://a4esl.org/ )
(3) Vocabulary Exercises for the Academic Word List (http://web.uvic.ca/~gluton/awl/ ) is
part of the Gerry’s Vocabulary Teacher site. There are several Hot-Potatoes
style matching exercises here for practice with the AWL.
Figure 5．Matching
quiz for AWL from Gerry’s Vocabulary Teacher (http://web.uvic.ca/~gluton/awl/ )
5. Conclusion and Directions
for Further Research
The present study describes a
variation of the Leitner method (the JY method) and documents a self-conducted
experiment by the author using this method to study Korean vocabulary on
Flashcard Exchange. It also describes how a similar method can be used with ESL
students to study the General Service List and Academic Word List, and presents
several complementary sites freely available.
This
method enhances the original flashcard benefits of effectiveness of learning
and retention by providing a review schedule that is intuitive and easy to
follow, as well as requiring production of vocabulary in writing. Moving the
system online and availing oneself of ready-made cards removes the tedious
requirement of having to create or input cards.
The
biggest drawback to using flashcards online is the lack of portability, one of
the greatest benefits of the paper version of the method. However, several
sites and software programs are available for cell phones (although they may
not be compatible with those in Japan) as well as PDAs. A stand-alone flashcard
drill system is also on the market in Japan, but unfortunately cannot import
the ready-made cards from FE. Even so, portability of the JY method should be
possible in the near future with appropriate software.
Aside
from portability, another difficulty in implementation would be how to make a
program such as this available to motivated students. The best solution would
be to structure an elective vocabulary course around it, which would require
daily effort from all registered students. Another would be to offer a
semi-self-study website linking all of these materials, but it is doubtful as
to how much the site would actually be used. In today’s world, where
opportunities abound at every corner of the Internet and “ESL vocabulary study”
produces 16,500 results on Google, students cannot be expected to choose from
the smorgasbord of study tools available what will actually work best for them.
6. References
Coxhead, Averil (2001) The Academic Word
List <http://www.vuw.ac.nz/lals/research/awl>
“Learning Theory - How MemoryLifter Works as a
Memorization Tool” (2004) on Memory Lifter Website
<http://www.memorylifter.com/learning/flash-cards.html>
Leitner, Sebastian (1972) So Lernt Man Lernen
(in German)., Germany: Herder Verlag
“Leitner Cardfile System” on Flashcard Exchange
Website, 2005 Tuolumne Technology Group, Inc
<http://www.flashcardexchange.com/docs/leitner>
Micheloud, Francois, ed. (2006) Flashcard Forum
on How-to-learn-any-language.com
<http://how-to-learn-any-language.com/forum/keyword.asp?kw=87>
Saruwatari, Asuka, Yoshihara, Shota and Suzuki,
Chizuko (2006) “Development of a Computer Assisted Flashcard System”, LET Kyushu-Okinawa Bulletin No 6 p.
13-22.
Schaefer, Jorgen (1997) “So Lernt Man Lernen”
(Book Review, In German), <http://www.forcix.cx/books/leitner72.html>
“System, Apparatus and Method for Maximizing
Effectiveness and Efficiency of Learning, Retaining and Retrieving Knowledge
and Skills” WO/2001/050439 Patent description,
<http://www.wipo.int/pctdb/en/wo.jsp?IA=US2000035381&DISPLAY=DESC>