Penny Ur: Vocabulary review: just reminding is not enough
ETAS Journal Volume 36, Number 2 (Spring 2019), pp. 28-30
I have selected this article because of its practical take on vocabulary revision. In addition to asking tough questions about what does and does not work when revising vocabulary, Ur also provides concrete examples of how to counter some of the challenges in a classroom setting.
Multiple review and vocabulary acquisition through reading
Both research and practice confirm that it is not enough to teach a new vocabulary item once: learners need to review it several times. however, it appears that reviewing by just reminding learners of the new item is not very effective: it is better to use tasks that require deliberate and effortful retrieval.
New vocabulary needs to be reviewed a number of times in order for its basic form and meaning to be mastered (Zahar, Cobb, & Spada, 2001; Webb, 2007). The actual number varies, but research indicates that in most cases it is at least ten, often more, depending on the difficulty of the item, the level, motivation, and ability of the student, and the effectiveness of the review technique used.
Given this assumption, how can we make sure that students get enough review of the vocabulary we teach them?
Those who support the Input hypothesis would claim that the provision of comprehensible input is a necessary and sufficient condition for language acquisition. In other words, give your students enough comprehensible spoken or written input contextualizing the language feature to the learnt, and it will be acquired. In the case of vocabulary, the emphasis is on input through reading (Krashen, 1989).
When applied to the learning of one’s first language, the hypothesis is convincing: probably most of us reached a high level of vocabulary knowledge in our mother tongue mainly through our reading. However, the sheer number of hours spent reading in the mother tongue in a mainly monolingual environment by a learner progressing through the school system probably reaches tens of thousands. Even in a multilingual country like Switzerland, a school student will spend thousands of hours reading texts in the language(s) which is/are the medium of instruction. This cannot be compared to the situation where a student is learning a foreign language, like English, through maybe three or four hours a week of instruction. In such a context, the number of hours spent by the student reading in the new language is only a fraction of the numbers quoted above.
If vocabulary acquisition derives mainly from encounter through reading – how much do you need to read in order to acquire how much vocabulary? Clearly, the more common the word, the less you need to read, since the number of times the more frequent words, like “go”, “big”, “man”, etc., are likely to occur in a text is relatively high. Once you progress even to medium-frequency words like “therefore”, “perfect”, “announce”, however, you will need to read a huge amount of text in order to re-encounter the words enough times to learn them permanently.
A study done by Zahar, Cobb and Spada (2001) looked at vocabulary acquisition through reading of French-speaking learners of English in Quebec. They selected a set of 30 words that occurred (most of them repeatedly) in a story, checked that students did not know them in advance, got them to read and listen to the story, and then tested how many of the words they had in fact learnt by the time they had finished.
They found that about one word was learnt satisfactorily for every 1 000 words read: a result broadly supported by those who have done similar research since then (Waring & Takaki, 2003; Cobb, 2007; Pigada & Schmitt, 2006). Zahar, Cobb and Spada’s conclusion is that in a situation where students spend about one English period a week reading in the target language, it would take them 29 years to acquire a vocabulary of 2 000 words (about A2 level) only through reading.
The point here is that acquisition of vocabulary through reading is effective, but it is not efficient. It works well if you have enough input. Learners acquiring their own language have enough input: those learning a new language only through a formal course of instruction do not.
Vocabulary acquisition through retrieval-based review
We need, therefore, to make sure that students re-encounter new vocabulary by deliberately providing opportunities for them to do so, in the form of focused vocabulary review activities. We cannot afford to wait until the vocabulary item comes up by chance in texts or classroom interactions.
Which of these activities do you most commonly use? (Note that the following suggestions refer to “words”, but could equally well refer to multi-word items such as “of course”).
1. The teacher reminds students of the words they have recently learnt, and asks them to repeat them.
2. Students write down the words in their notebooks.
3. Students take time to study and review the words recently taught.
4. The teacher asks students to read a text in which the words are encountered.
5. The teacher asks students to write down the words she dictates.
6. Students are given the word in English and a number of L1 possible equivalents: they choose the right translation.
7. The teacher says or writes the word in English, then asks the students to say or write down the mother-tongue equivalent.
8. Students do a gapfill exercise: fill in gaps in sentences from a “bank” of words, including the ones they were asked to learn.
9. Students are given a word and asked to compose a sentence contextualizing it.
10. Students are asked to give a definition of a word they were taught.
11. The teacher says or writes the word in the students’ mother tongue, then asks the students to say or write down the English equivalent.
12. Students are asked to provide the word appropriate to a definition or context.
Let’s look at these more closely: what sort of learning does each imply?
The first five of the examples above get students to look at, write down, and/or hear, or repeat the words again, but does not require them to make any effort to retrieve from memory either form or meaning. In all the others, they do need to make such an effort.
In numbers 6-10, they are provided with the written or spoken form of the word and asked to do something that demonstrates their understanding of its meaning.
In numbers 11 and 12, they are given the meaning and asked to recall the word.
Note that only the last two options ask students for productive knowledge, namely, the ability to produce the target item themselves to express a meaning. All the others only require receptive knowledge (the ability to recognize and understand the item when they see or hear it). If you look through the vocabulary exercises in your textbook, you will find that the vast majority – if not all – require receptive knowledge only.
It appears from research that the first five options – the kinds of exercises that require attention, but no active effort by the learner to retrieve the item from memory – have only a relatively weak effect in reinforcing learning. The retrieval-based ones are significantly more effective.
The first time I really appreciated this point was when I came across an article by Karpicke and Roediger (2008), which compared the vocabulary-learning outcomes of students who reviewed the items through what they defined as “study”, as compared to those who reviewed through being “tested”.
Study meant providing opportunities for students to look at and review the target items, while tested involved requiring them to retrieve the item by, for example, providing a translation of a target item into the mother tongue.
The bottom line was that those students who performed the tested procedures as the main component of their vocabulary review scored significantly higher on a summative test at the end of the experiment than did those in the study group. A study by Kang, Gollan, and Pashler (2013), which contrasted vocabulary retrieval with imitation or repetition, achieved similar results.
Laufer and Rozovski-Roitblat (2011) defined the question slightly differently: What difference does the type of task make to the number of times you need to review? In other words, if you give students retrieval tasks – ones that challenge them to make the effort to remember the word’s meaning – does this mean that they will need fewer encounters in order to master it permanently?
The answer is ‘yes’.
In a delayed test, students who did the retrieval tasks but encountered the target items in all only three to four times got better results than those who only studied them (encountered in text, and looked up meanings wherever they needed to) but encountered them six or seven times.
The conclusion is that having students make the effort to retrieve a target item from memory leads to better learning, as compared to either just reminding them, or having them repeat, copy, or look it up in a dictionary. It seems that the act of successful, if effortful, retrieval oils the wheels of the remembering process, so that next time recalling requires less effort.
There are, however, two practical problems with all this. First, doing retrieval exercises takes more time than just repeating, hearing, writing, or re-reading the items. One must ask: is the improvement in learning worth the extra time invested? Second, what happens if we ask learners to retrieve a word, and they cannot do so?
Time is an absolutely crucial issue for those of us teaching a limited number of hours a week to students who are exposed to relatively little English outside the classroom. It is important, therefore, to design retrieval-based tasks so that they are cost-effective relative to the time invested. One strategy is to (temporarily) abandon the principle of contextualization, and review single items on their own: ask our students to recall, for example, words they learnt recently, or translate them.
Even contextualizing tasks can be kept relatively brief: ask them to think of a short phrase using the word (e.g. “What sorts of things could you describe using this adjective?”), or a single sentence (e.g. “Can you think of a sentence about yourself, using this word?”). It is also probably a good idea for in-class work to give the cues and ask students to produce their responses orally (much faster than writing them out). The written exercises can always be done for homework.
Now for the second issue: failure to successfully retrieve an item. I would like to make two points here. First, we need to design the task so as to maximize the chances that they will remember; in other words, to provide plenty of “scaffolding” and support.
We might, for example, give them the first letter of the target item, or provide a set of items to choose from, hoping that the correct one will jog their memory. It is also vital to administer such tasks fairly soon after the students have been taught the item, or have successfully done a previous retrieval, so that they have not had time to forget it.
This is where the idea of expanding rehearsal or spaced repetition is so useful. The principle being that if you want someone to remember something, get them to recall it immediately after they have been taught it, then a short time later, then wait a bit longer before the next recall … and so on.
Second, even with all this support, there will still be situations where a student cannot recall the item. In such a case, give them a bit of time to think, and if even then they cannot remember, re-teach it. (Then ask them to retrieve it again shortly after.)
Summary and conclusions
The main points of this article are the following:
Learners need to review a new vocabulary item several times in order to master its basic form and meaning.
They will not normally get enough of such review through incidental encounter of reading or listening.
It is important to provide planned and focused review activities of the new vocabulary we teach them.
These review activities are most effective if they require the student to retrieve from memory either the meaning (in response to the written or spoken form) or the form (in response to a cue giving the meaning).
Textbook activities tend to focus on receptive tasks (where learners demonstrate understanding of the meaning of a given item) rather than productive (where they actually produce the item themselves to express a meaning).
Factors that can contribute to the effectiveness of retrieval-based activities are: economical use of time for maximum review opportunities; support, scaffolding, and encouragement by the teacher; and the principle of expanding rehearsal.
What are the practical implications for teachers?
First, we need to look critically at our course materials: do they provide enough retrieval-based vocabulary review? Many provide only one or two reminders or tasks reviewing new vocabulary, which are not enough, and some do not provide any at all. In either case, we will need to supplement with our own ideas, or look for them on the Internet or in teacher handbooks (Ur, 2012).
Second, course materials normally provide students with few opportunities to practise the new items productively. This is mainly because materials writers (and many teachers) prefer exercise questions that have one right answer: they are much easier to administer and check. As soon as we give a task that asks students to produce their own language, we allow them choice. If, for example, we give a definition, or a context, or a mother-tongue equivalent, and ask students to produce a vocabulary item – they may come up with the one we are looking for, or they may come up with something else that fits the cue. Nevertheless, it is important to do such activities, and here again we will probably need to supplement the textbook.
Third, note that while the processes described here will help your students master the basic form and most common meaning(s) of an item, vocabulary learning does not stop there. They will need also to learn where it is appropriate or inappropriate to use it, with what other words it collocates, in what ways it is different from the mother-tongue equivalent, its connotations, and so on. These are outside the scope of the present article, but you will find ideas in teacher handbooks (including my own, listed below) or on the Internet.
Cobb, T. (2007). Computing the vocabulary demands of l2 Reading. language learning and Technology, 11(3), 38–63.
Kang, S. H., Gollan, T. H., & Pashler, H. (2013). Don’t just repeat after me: Retrieval practice is better than imitation for foreign vocabulary learning. Psychonomic Bulletin & Review, 20(6), 1259–1265.
Karpicke, J., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319, 966–968.
Krashen, S. (1989). We acquire vocabulary and spelling by reading: Additional evidence for the input hypothesis. The Modern Language Journal, 73(4), 440–464.
Laufer, B., & Rozovski-Roitblat, B. (2011). Incidental vocabulary acquisition: The effects of task type, word occurrence and their combination. Language Teaching Research, 15(4), 391–412.
Pigada, M., & Schmitt, N. (2006). Vocabulary acquisition from extensive reading: A case study. Reading in a Foreign Language, 18(1), 1–28.
Ur, P. (2012). Vocabulary activities. Cambridge: Cambridge University Press.
Waring, R. & Takaki, M. (2003). At what rate do learners learn and retain new vocabulary from reading a graded reader? Reading in a Foreign Language, 15(2), 130–163.
Webb, S. (2007). The effects of repetition on vocabulary knowledge. Applied Linguistics 28, 1: 46–65.
Zahar, R, T. Cobb, T. & Spada, N. (2001). Acquiring vocabulary through reading: Effects of frequency and contextual richness. Canadian Modern Language Review, 57(4), 544–572.
Penny Ur was born in the UK, studied at Oxford and Cambridge Universities, and moved to Israel in 1967. She has thirty-five years’ experience as an English teacher in elementary, middle and high schools, and has also taught BA and MA courses at Oranim Academic College of Education and Haifa University. Her primary interest is practical aspects of effective English language teaching, on which she has presented lectures and workshops at TESOL, IATEFL, and other English teachers’ conferences worldwide. In 2013 she was awarded the OBE for services to English language teaching. Her books include Grammar Practice Activities(2nd Edition) (2009), Vocabulary Activities(2012), A Course in English Language Teaching(2012), Discussions and More(2014), Penny Ur’s 100 Teaching Tips(2016), all published by Cambridge University Press.