1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.
-
Upload
keila-lass -
Category
Documents
-
view
240 -
download
3
Transcript of 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.
![Page 1: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/1.jpg)
1
Size and Depth of Vocabulary: What Does the Research Show?
Norbert Schmitt
![Page 2: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/2.jpg)
2
History
• Two year review of vocabulary studies pertinent to size vs. depth
• Results reported at AAAL 2012• Manuscript submitted to Language
Learning• 3 reviewers gave plenty of feedback• Latest version: focus on conclusions
![Page 3: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/3.jpg)
3
Size and Depth
• Vocabulary has often been characterized in terms of size vs. depth of knowledge
• The distinction is widespread, but one depth is not easy to pin down
• One reviewer states that depth is “the wooliest, least definable, and least operationalisable construct in the entirety of cognitive science past or present”
![Page 4: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/4.jpg)
4
Size and Depth
• It is time to start thinking about this distinction more rigorously
• The various conceptualizations and measurements of depth make it difficult to start from a theoretical framework
• So start from an empirical perspective to inform the debate:
– Review all studies that have a measurement of size and at least one measurement of depth
![Page 5: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/5.jpg)
5
Vocabulary Size
• Size = the number of lexical items ‘known’
![Page 6: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/6.jpg)
6
Vocabulary Size
• Size = the number of lexical items ‘known’
(to some criterion of mastery, i.e. depth)
![Page 7: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/7.jpg)
7
Vocabulary Size
• Size = the number of lexical items ‘known’
(to some criterion of mastery, i.e. depth)
• Every size test is also a depth test in the
sense that a certain criterion of mastery
must be met
![Page 8: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/8.jpg)
8
Vocabulary Depth
• Depth / Quality = How well do you know those items?
• What can you do with those items?
• Very broad: can be conceptualized and operationalized in a variety of ways
![Page 9: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/9.jpg)
9
Size and Depth
• The relationship between size and depth depends on:
• How both are conceptualized
• How both are measured
![Page 10: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/10.jpg)
10
Conceptualizing Depth
• Receptive vs. Productive Knowledge
• Usually connected with the 4 skills
![Page 11: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/11.jpg)
11
Reading, Writing, Speaking, Listening
• Depth could be seen as how well word can be employed in the four skills
• Vocabulary size correlates with all four skills
• But little research which shows how well individual lexical items are employed in the skills
![Page 12: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/12.jpg)
12
What is Involved in Knowing a Word
Nation (2001) Form spoken
receptive/productive
written word parts
Meaning form and meaning
concept and referents
associations
Use grammatical functions
collocations
constraints on use (register, frequency …)
![Page 13: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/13.jpg)
13
Size and Depth
• Knowledge of all of the word knowledge aspects taken together can be conceptualized as a relatively comprehensive depth of knowledge
• But each aspect can be known to various degrees of mastery
![Page 14: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/14.jpg)
14
Degree of Knowledge
Schmitt, 2010a: 38
![Page 15: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/15.jpg)
15
Vocabulary Depth
• So, improving knowledge of any individual word knowledge aspect can be considered as adding to depth
• Not an all-or-nothing concept
• Anything that improves mastery can be considered additional depth
![Page 16: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/16.jpg)
16
Vocabulary Depth
Depth =
•Degree of mastery of the form-meaning link•Polysemous word meanings•Derivations (word family members)•Collocations
•Other word knowledge aspects but were not found in research in conjunction with a size measure
![Page 17: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/17.jpg)
17
Vocabulary Depth
• Depth in this conceptualization concerns individual lexical items
• Only a small number of items can ever be measured, so unclear how generalizable the depth measures can be
![Page 18: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/18.jpg)
18
Conceptualizing Size and Depth
Meara and Wolter, 2004: 89
![Page 19: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/19.jpg)
19
Lexical Organization
• Concerns lexicon as a whole rather than individual lexical items
• Depth could be seen as any of the word knowledge connections between items
• But how to measure it?• Word associations
– Difficult to interpret– Idiosyncratic to individuals– Good measure of organization?– WAF main measure, but ?
![Page 20: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/20.jpg)
20
Lexical Fluency
• Daller, Milton, and
Treffers-Daller (2007)
see fluency as a
separate dimension
• Can see fluency as depth (i.e. depth does not have to be knowledge, but can be seen as employability (skills, automaticity)
![Page 21: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/21.jpg)
21
Recognition and Recall of the Form-Meaning Link
• Laufer and Paribakht (1998) • VLT (form recognition)• PVLT (form recall)• Recall-recognition r=.89 (EFL: Israeli high school)
.72 (ESL: Canadian university)
PVLT ÷ VLT ratio EFL% ESL%
Combined 77 62
2,000 94 84 3,000 76 58 5,000 62 6310,000 46 44
![Page 22: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/22.jpg)
22
Depth of Form-Meaning Link
• As vocabulary size increases (and frequency level decreases), the recall/recognition gap increases
• Learners are more likely to have both form recognition and form recall mastery at the higher frequencies (i.e. smaller gap)
• Less likely to have form recall mastery at the lower frequency levels (i.e. form recognition mastery only)
• Form recall lags both form recognition and meaning recall
![Page 23: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/23.jpg)
23
Knowledge of Written vs. Spoken Word Forms
• van Zeeland (2013) used a meaning recall interview to measure the written and spoken vocabulary knowledge of advanced L2 learners
• Results showed a stronger correlation between written and spoken word knowledge than found by Milton and Hopkins (2006) (r = .85 vs. .68).
• The relationship between learners’ knowledge of written and spoken vocabulary furthermore remained constant as overall scores increased
• These results suggest that knowledge of vocabulary in the two modes may be more closely related than suggested by checklist test results (Milton and colleagues)
![Page 24: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/24.jpg)
24
Depth of Spoken-Written Mastery
• Some evidence that very small vocabularies might be mainly known phonologically
• Somewhat larger vocabulary sizes shift to being known predominately orthographically
? Advanced learners tend to have relatively
balanced spoken/written vocabularies, while
lower-level students are prone to the type of
imbalanced vocabularies found by Milton and
colleagues?
![Page 25: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/25.jpg)
25
Knowledge of Derivatives
• Correlations between size and derivation/suffix knowledge • Schmitt and Meara (1997) r
– recall of suffix derivation .27 and .35– Recognition of suffix derivation .37 and .41
• Kieffer and Lesaux (2008)– recall of derivation .53 and .46
• Kieffer and Lesaux (2012) – recall of derivation .50 -.57
• Noro (2002) – recall function of affix .42, .54, .69
• Mochizuki and Aizawa (2000)– recall function of affix .54 - .65
![Page 26: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/26.jpg)
26
Knowledge of Derivatives
• Milton (2009) reviews Mochizuki and Aizawa’s results and concludes:
• A vocabulary size of 3,000-5,000 families is necessary for affixes to be mastered
• But even at 5,000 families, some affixes may not be known well
![Page 27: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/27.jpg)
27
Depth of Derivative Knowledge
• Size is only modestly related to knowledge of affixes and derivatives
(system learned before items?)
• Milton suggests that a threshold might exist
(3,000-5,000 families?)
![Page 28: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/28.jpg)
28
Knowledge of Collocation
• Gyllstad (2007) high-proficiency Swedish ESL • VLT• 2 collocation tests
– COLLEX 5: a 3-option form recognition test– COLLMATCH 3: a yes/no collocation judgement
• Size - collocation (r=.90)
• 10,000 families >90% on both collocation tests • 5,000 families ≈ 85%• 3,000 families ≈ 70%
• With larger vocabulary sizes, it is possible to recognize collocations well
![Page 29: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/29.jpg)
29
Knowledge of Collocation
• Laufer and Waldman (2011) review the literature and conclude:
– receptive knowledge is related to general vocabulary knowledge
– productive knowledge of collocation lags behind knowledge of individual words
– the problem with collocations is not recognition, but in using them properly, i.e. productive mastery
![Page 30: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/30.jpg)
30
Lexical Fluency
• Laufer and Nation (2001) Israeli university
• VORST (computerized, timed, modified VLT)
• There was an increase in lexical access speed at a vocabulary size of around 5,000 word families
• The larger the vocabulary size, the faster the access speed
2,000 – speed (r = -.38) 3,000 – speed (r = -.40) 5,000 – speed (r = -.50)10,000 – speed (r = -.67)
![Page 31: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/31.jpg)
31
Depth of Automaticity
• The larger the vocabulary size, the faster the access speed
• The larger the vocabulary size (and the lower the frequency level) the stronger the relationship between size and fluency
• Hint of a threshold: there is an increase in lexical access speed at a vocabulary size of around 5,000 word families
![Page 32: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/32.jpg)
32
Lexical Organization
• Henriksen (2008) Danish high school
Grades 7 10 13
• VLT – association recognition .72 n.s. n.s. • VLT – association recall .85 .69 .55
• The relationship between size and association is stronger at lower grades than more advanced ones
• Since the students had increasing vocabulary sizes at all three grades, we can also interpret the results to show a stronger size-association relationship for smaller vocabulary sizes than larger ones
![Page 33: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/33.jpg)
33
Lexical Organization
• Greidanus, et al. (2004) • Advanced Dutch learners of French • Form recognition and form recall • New WAF (paradigmatic, syntagmatic, and analytic)
• Form recognition – association (r= .70)
• Form recall – association (r= .81)
![Page 34: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/34.jpg)
34
Lexical Organization
• Size – association correlations
.18 .36 .39 .43 .49 .51 .55 .61 .62 .69 .69
.70 .70 .72 .78 .81 .81 .82 .85 .85 .86 .89
• Overall, fairly strong correlations
![Page 35: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/35.jpg)
35
Lexical Organization
• Size – association correlations
.18 .36 .39 .43 .49 .51 .55 .61 .62 .69 .69
.70 .70 .72 .78 .81 .81 .82 .85 .85 .86 .89
Association recall
Association recognition (WAF)
![Page 36: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/36.jpg)
36
Lexical Organization
• Size – association correlations
.18 .36 .39 .43 .49 .51 .55 .61 .62 .69 .69
.70 .70 .72 .78 .81 .81 .82 .85 .85 .86 .89
Lower vocabulary size
Larger vocabulary size
![Page 37: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/37.jpg)
37
Lexical Organization
• 3 studies showed a trend for larger vocabulary sizes having stronger size – association correlations
• 2 studies showed a trend for larger vocabulary sizes having weaker size – association correlations
![Page 38: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/38.jpg)
38
Lexical Organization
• 3 studies showed a trend for larger vocabulary sizes having stronger size – association correlations
• 2 studies showed a trend for larger vocabulary sizes having weaker size – association correlations
• Does larger size relate to better lexical organization?
• Evidence seems mixed at this point
(Problems with measuring organization?)
![Page 39: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/39.jpg)
39
Tentative Conclusions
• Does greater vocabulary size relate to greater depth of vocabulary knowledge?
• Yes, generally
• But how strongly depends on what ‘depth’ is
![Page 40: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/40.jpg)
40
Tentative Conclusions
• How one views the size-depth relationship should depend on one’s purpose of use
• If one wishes to discuss the nature of vocabulary in general, particularly with practitioners, then the distinction is useful
![Page 41: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/41.jpg)
41
Tentative Conclusions
• If one’s purpose is to characterize vocabulary knowledge in more precise terms:– theorizing– designing and interpreting research– assessment
• Depth is probably too vague a term to be useful
• Need to state lexical aspect addressed and focus on that
![Page 42: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/42.jpg)
42
Tentative Conclusions
• Virtually all aspects of vocabulary knowledge seem interrelated
• This makes it difficult to discuss any particular conceptualization of depth in isolation
• This makes it difficult to conceptualize overall depth as anything but the combined interrelationships between word knowledge aspects
![Page 43: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/43.jpg)
43
Tentative Conclusions
• The most widely-used vocabulary tests are size tests, and they typically describe their results as the number of words ‘known’
• But they do not define what this actually entails
• Test developers need to explicitly state what correct answers on their tests entail, and what degree of depth they represent
![Page 44: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/44.jpg)
44
Tentative Conclusions
• There can be no clear conceptual distinction between size and depth
• Size by definition is the number of lexical items ‘known’ to some criterion level of mastery
• But the criterion will always be some measure of depth, and so the two will always be confounded
![Page 45: 1 Size and Depth of Vocabulary: What Does the Research Show? Norbert Schmitt.](https://reader035.fdocuments.net/reader035/viewer/2022062221/56649c995503460f94955fc3/html5/thumbnails/45.jpg)
45
Questions / Comments
Comments to help me understand
size/depth better?
?
Size? Depth?
www.norbertschmitt.co.uk