case study according to oxford dictionary

• Subscriber Services
• For Authors
• Publications
• Archaeology
• Art & Architecture
• Bilingual dictionaries
• Classical studies
• Encyclopedias
• English Dictionaries and Thesauri
• Language reference
• Linguistics
• Media studies
• Medicine and health
• Names studies
• Performing arts
• Science and technology
• Social sciences
• Society and culture
• Overview Pages
• Subject Reference
• English Dictionaries
• Bilingual Dictionaries

Recently viewed (0)

• Save Search
• Share This Facebook LinkedIn Twitter

Related Content

Related overviews.

case-history

thick description

participant observation

See all related overviews in Oxford Reference »

More Like This

Show all results sharing this subject:

Quick Reference

A research method that engages in the close, detailed examination of a single example or phenomenon. In some instances, it may be a version of ideographic rather than nomothetic investigation—seeking ...

From:   case study   in  Dictionary of the Social Sciences »

Subjects: Social sciences

Related content in Oxford Reference

Reference entries, case study n..

View all reference entries »

View all related items in Oxford Reference »

Search for: 'case study' in Oxford Reference »

• Oxford University Press

PRINTED FROM OXFORD REFERENCE (www.oxfordreference.com). (c) Copyright Oxford University Press, 2023. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single entry from a reference work in OR for personal use (for details see Privacy Policy and Legal Notice ).

date: 27 July 2024

• Cookie Policy
• Privacy Policy
• Legal Notice
• Accessibility
• [195.190.12.77]
• 195.190.12.77

Character limit 500 /500

Open Education Sociology Dictionary

Table of Contents

Definition of Case Study

( noun ) A detailed and in-depth study of a single case , involving an event,  group , individual, or organization .

Examples of Case Study

• An in-depth research project of male allies that try to prevent violence against women completed through interviews and surveys .
• Street Corner Society: The Social Structure of an Italian Slum (1943) by William Foote Whyte (1914–2000)
• Streetwise: Race, Class, and Change in an Urban Community (1990) by Elijah Anderson (born 1943)

Case Study Pronunciation

Pronunciation Usage Guide

Syllabification : case stud·y

Audio Pronunciation

Phonetic Spelling

• American English – /kAYs stUHd-ee/
• British English – /kAYs stUHd-ee/

International Phonetic Alphabet

• American English – /keɪs ˈstʌdi/
• British English – /keɪs ˈstʌdi/

Usage Notes

• Plural: case studies
• The term refers to both the subject and  method of study .
• Case studies are often based on ethnographic research and conducted through participant observation , instead of aggregate  data such as a sample of a population .
• Variant spelling: case-study
• Also called case report .

Additional Information

• Qualitative Research Resources – Books, Journals, and Helpful Links
• Quantitative Research Resources – Books, Journals, and Helpful Links
• Word origin of “case” and “study” – Online Etymology Dictionary: etymonline.com
• Feagin, Joe R., Anthony M. Orum, and Gideon Sjoberg, eds. 1991. A Case for the Case Study . Chapel Hill, NC: University of North Carolina Press.
• Gerring, John. 2009. Case Study Research: Principles and Practices . Cambridge: Cambridge University Press.
• Thomas, Gary. 2011. How to Do Your Case Study: A Guide for Students and Researchers . Thousand Oaks, CA: SAGE.

Related Terms

• cohort study
• generalizability
• longitudinal study
• qualitative research
• quantitative research
• reliability
• symbolic interactionism

Works Consulted

Bilton, Tony, Kevin Bonnett, Pip Jones, David Skinner, Michelle Stanworth, and Andrew Webster. 1996. Introductory Sociology . 3rd ed. London: Macmillan.

Crooks, Robert, and Karla Baur. 2014. Our Sexuality . 12th ed. Belmont, CA: Wadsworth Cengage Learning.

Ferrante, Joan. 2011. Seeing Sociology: An Introduction . Belmont, CA: Wadsworth.

Griffiths, Heather, Nathan Keirns, Eric Strayer, Susan Cody-Rydzewski, Gail Scaramuzzo, Tommy Sadler, Sally Vyain, Jeff Bry, Faye Jones. 2016. Introduction to Sociology 2e . Houston, TX: OpenStax.

Henslin, James M. 2012. Sociology: A Down-to-Earth Approach . 10th ed. Boston: Allyn & Bacon.

Marsh, Ian, and Mike Keating, eds. 2006.  Sociology: Making Sense of Society . 3rd ed. Harlow, England: Pearson Education.

Oxford University Press. (N.d.) Oxford Dictionaries . ( https://www.oxforddictionaries.com/ ).

Shepard, Jon M. 2010. Sociology . 11th ed. Belmont, CA: Wadsworth.

Shepard, Jon M., and Robert W. Greene. 2003.  Sociology and You . New York: Glencoe.

Thompson, William E., and Joseph V. Hickey. 2012. Society in Focus: An Introduction to Sociology . 7th ed. Boston: Allyn & Bacon.

Wikipedia contributors. (N.d.) Wiktionary, The Free Dictionary . Wikimedia Foundation. ( http://en.wiktionary.org ).

Cite the Definition of Case Study

ASA – American Sociological Association (5th edition)

Bell, Kenton, ed. 2013. “case study.” In Open Education Sociology Dictionary . Retrieved July 27, 2024 ( https://sociologydictionary.org/case-study/ ).

APA – American Psychological Association (6th edition)

case study. (2013). In K. Bell (Ed.), Open education sociology dictionary . Retrieved from https://sociologydictionary.org/case-study/

Chicago/Turabian: Author-Date – Chicago Manual of Style (16th edition)

Bell, Kenton, ed. 2013. “case study.” In Open Education Sociology Dictionary . Accessed July 27, 2024. https://sociologydictionary.org/case-study/ .

MLA – Modern Language Association (7th edition)

“case study.” Open Education Sociology Dictionary . Ed. Kenton Bell. 2013. Web. 27 Jul. 2024. < https://sociologydictionary.org/case-study/ >.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

A Critical Study of the Word meanings in Dictionaries: a case of Oxford Advanced Learners Dictionary, seventh edition

2015, Journal of Literature, Languages and Linguistics

One of the main functions of a dictionary is to give the definition of meanings of various words. This view is supported by various writers who have attempted to explain the meaning of a dictionary. Sally (2005) for instance, views a dictionary as a book that gives a list of words of a language in alphabetical order and explains what they mean or gives a word for them in a foreign language. This paper therefore, seeks to have a critical assessment on the way this function is implemented and wonders whether the goal is fully attained or not. The paper focuses on examining the definition of meaning of various words in dictionaries based on the Oxford Advanced Learner's Dictionary (OALD), 7 th edition. The strengths and weaknesses of word meanings in dictionaries form the base of this study. The paper concludes that, despite the weaknesses that OALD demonstrates in defining the meanings of various words, this dictionary seems to have more strengths than weaknesses, something which gives hope that dictionaries have to a good extent attained the goals for their existence. OALD is one of the outstanding dictionaries in the world recently. It has a remarkable contribution to the Linguists as well as other learners of the English language in the entire world. The publication of the OALD seventh edition comes 250 years after the appearance of the first comprehensive dictionary of the English language compiled by Samuel Johnson. Much has changed since then. The English that Johnson described in 1755 was relatively well defined, still essentially the national property of the British. Since then, it has been dispersed, diversified, adopted and adapted as an international means of communication by communities all over the globe. (Sally, 2005) This dictionary, just like other dictionaries, deals with word meanings in two ways: by explanation and by exemplification. However, whatever the way that a dictionary may use in defining the word meanings, a crucial question remains on the suitability or quality of the meaning provided. It is this question then which motivated me to study critically various definitions provided by OALD to the word meanings by assessing their strengths and weaknesses. 2. Different views on how the dictionary is regarded. Different writers have tried to conceptualize the term 'dictionary'. Landau (2001) defines a dictionary as a text that describes the meanings of words, often illustrates how they are used in the context and usually indicates how they are pronounced. Hartmann (1983: 3) in Kiango 2000 defines a dictionary as a reference book or list of words (usually in alphabetical order) together with a guide to their meaning, pronunciation, spelling or equivalencies in other languages. Again, a dictionary may be regarded as a lexicographical product that is characterized by three significant features: it is prepared for one or more functions, it contains data that have been selected for the purpose of fulfilling the function, its lexicographic structure links, and establishes relationship between the data so that they can meet the needs of users and fulfill the functions of a dictionary (Nielson, 2008). A comprehensive dictionary then should comprise all aspects of a given source language such as common words, colloquial words, dialectical varieties, archaic words, and etymology of words, words in literature, science, technology, slang, vulgar words and deprecated words, sexist and taboo words. This idea was firstly argued by Zgusta (1984) that a dictionary should contain enough information to allow the dictionary users determine relevant information successfully. In a dictionary, the words that are used in place of another word to explain the thought are referred to as definition. These definitions are what must be covered in a dictionary, and must be accurate, concise and clear. Definitions must be complete enough to satisfy the reader and distinguish several related senses or shades of meaning. According to Kiango (2000) a meaning is referred to as a concept that is built in the mind through the accumulation of experience of various kinds. Kiango further adds that before defining the meaning of a word one should examine whether the word has one or several meanings. Therefore, to determine meanings that a word

Related Papers

Rajakumar Guduru

‘A word is a single unit of language which means something and can be spoken or written’ (Oxford Advanced Learner’s Dictionary, 2010, 8th Edition). It seems the above definition, from the recent 8th edition of the Oxford Advanced Learner’s Dictionary 2010, is inadequate in the global linguistic knowledge point of view because it does not give a comprehensive picture of what a word is. It reads like a layman’s description of what a ‘word’ is. Nevertheless, in lexicology, the definition seems comprehensive as it makes us look at related fields and at times ambiguous and even controversial issues. Thus, when we look at it more closely, defining ‘word’ turns out to be far from simple which we generally think it to be. Therefore, it is necessary to understand what lexicology is and the branches related to it. This paper discusses pertinent problems in defining a word in general. It also highlights some fundamental factors in determining the meaning of word, which a modern linguist should possess in order to understand the complex role of words in a language. In conclusion, it is said that words are governed by social convention and only by going beyond the dictionary definition can lexicologists explore its real meaning.

Michal Kociumbas

Bilingual dictionaries provide direct equivalents of words in the target language and very rarely give any additional information concerning, for example, connotations. The paper discusses components of word meaning and shows why it is difficult to establish the word meaning, contrary to what bilingual dictionaries may suggest.

Dictionary Information and its Importance for a Language Learner

Tomás Mateus

We could gather that dictionaries are very important tools for communication and for personal knowledge. The study which deals with dictionary-making is widely regarded as lexicography. The product of tis painstaking task is a dictionary. This kind of reference book encompasses a wealth of information both linguistic (spelling, pronunciation, meaning, word partners, idioms, phrasal verbs, etc.) and encyclopedic (facts and figures). Thus, before elaborating upon the information or content found in a dictionary, we are going to pave the way for this oldest subfield of linguistics: lexicography.

Selçuk Üniversitesi Edebiyat Fakültesi Dergisi

HERMES - Journal of Language and Communication in Business

Pedro A. Fuertes-Olivera

Ken Sugayama

Arleta Adamska-Sałaciak

ResearchGate

Temidayo J Afolabi

This research focuses on how the dictionary poses as an all-encompassing tool for lexicographic studies and how it achieves this status from the various dimensions it presents the natures, features, meanings, and usages of words or entries from.

Kim Hua Tan

This paper presents the findings of a study aiming to investigate the effectiveness of help-devices in learners' deciphering of meaning of English words among 232 Universiti Kebangsaan Malaysia (UKM) undergraduates in Teaching English as a Second Language (TESL). It focuses on investigating the learners' perspectives on dictionary use including their awareness and realization of the importance of using dictionary to look up the most appropriate meaning of multi-sense words. It also examines the effective role the meaning access device like menu system plays in helping learners easily look up the appropriate sense of a multi-sense-word in English. To collect the required data, the study made use of a questionnaire and two Word Sense Identification Tests (WoSIT). Findings indicated that the participants seemed to be less aware of the importance of meaning access devices in deciphering the meanings of words in English. It was also found that the use of menu as meaning access devices is useful in helping the participants look up the most appropriate meaning of a multi-sense word in English.

International Journal of Linguistics, Literature and Translation

Retna Isti Pratiwi

The dictionary is an important reference source for finding the meaning of a word. The dictionary presents not only denotative meanings but also displays contextual meanings. This study aims to describe the macrostructure of the lemma in the dictionary, the definition of the lemma, and the relation of meaning. The method used in this study is a qualitative method with the subject of the Big Indonesian Dictionary. The object of research is the lemma in the dictionary. The data is selected and recorded in the data card and then identified based on the macrostructure of the dictionary, the definition of the lemma, and the relation of meaning to the definition of the lemma and between the entries. This study uses two approaches, namely the lexicographic approach and the semantic approach. The lexicographic approach was used to analyze the data based on the macrostructure and the lemma definition model. The semantic approach is used to determine the analysis of meaning based on the chara...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

Tamar Manjavidze

Kisyani Laksono , Kisyani-Laksono Sekisman

Respectus Philologicus

Olga Karpova

hoang nguyen

Cornelia Tschichold

Henning Bergenholtz

adam ramejkis

Kim Hua Tan , Nafiseh Zarei

Studii si Cercetari Filologice: Seria Limbi Straine Aplicate

Constantin Manea

Nick Riemer

Terminology

John R Taylor

Papers in Linguistics Presented to László Varga on his 70th Birthday

Péter Lázár, A

Michael Zock

Giovanni Iamartino

Piet Swanepoel

Rufus H. Gouws

Daphné Kerremans

اداب الكـوفة

Nadhum Salami

Patrick Hanks

Joohi Chaturvedi

Novelty Vol 6 No 1

Sead Spuzic

Asian Journal of Research in Social Sciences and Humanities

Dr. Vasanthan R

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

• Search Menu
• Sign in through your institution
• Advance articles
• Editor's Choice
• Special Issues
• Author Guidelines
• Submission Site
• Open Access
• Why Submit?
• About International Journal of Lexicography
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Terms and Conditions
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

1. introduction, 2. literature review, 3. research questions, 4. methodology, 6. discussion, 7. conclusion, acknowledgements, [references marked with an asterisk indicate studies that were included in the meta-analysis.], the effects of dictionary use on second language vocabulary acquisition: a meta-analysis.

• Article contents
• Figures & tables
• Supplementary Data

Songshan Zhang, Hai Xu, Xian Zhang, The effects of dictionary use on second language vocabulary acquisition: A meta-analysis, International Journal of Lexicography , Volume 34, Issue 1, March 2021, Pages 1–38, https://doi.org/10.1093/ijl/ecaa010

• Permissions Icon Permissions

This study reports on the results of a meta-analysis which investigates the effects of dictionary use on second language (L2) vocabulary acquisition, as well as the magnitude of the moderating effect of a number of moderator variables on the effectiveness of dictionary use in improving L2 vocabulary knowledge. A total of 125 effect sizes were gleaned from 44 studies, which represented 87 independent samples and included 3,475 participants. A random-effects model of the meta-analysis shows that the overall effect of dictionary use on L2 vocabulary acquisition is g = 2.10 ( p < .01) for within-group studies and g = 1.03 ( p < .01) for between-group studies respectively, which are both large in the domain of second language research. Subsequent moderator variable analysis reveals how treatment-related variables (timepoint, dictionary form, dictionary type, target lexical unit, vocabulary knowledge type, learning condition, and research setting), methodological variables (type of experimental design, presence of pretest, and assessment type), as well as learner-related variables (age and proficiency) might contribute to the variation across studies. Implications of the findings are discussed.

Dictionary use has become a burgeoning field of research that has garnered considerable scholarly attention over the past three decades. One major concern is the relationship between dictionary use and L2 learning ( Hartmann 2001 ; Lew 2019 ), an area that lies at the interface between lexicography and second language acquisition. Of particular interest to researchers is the effectiveness of dictionary use in L2 vocabulary acquisition. This is probably because, on the one hand, dictionaries are mainly and, above all, about vocabulary; on the other hand, dictionary use is widely acknowledged by L2 vocabulary researchers as an important vocabulary learning strategy ( Fan 2003 ; Gu and Johnson 1996 ; Nation 2001 ; Webb and Nation 2017 ).

Empirical studies have been conducted on the relationship between dictionary form (e.g. Chen 2010 ; Chiu and Liu 2013 ; Dziemianko 2010 , 2011 , 2012b , 2017 ; Kobayashi 2006 ), dictionary type (e.g. Laufer and Hadar 1997 ; Laufer and Levitzky-Aviad 2006 ; Lew 2002 ), use of corpus examples (e.g. Al-Ajmi 2008 ; Frankenberg-Garcia 2012 , 2014, 2015 ), dictionary use under intentional/incidental conditions (Chen 2012 ; Hill and Laufer 2003 ; Ma and Cheon 2018 ; Peters 2007 ; Zou 2016 ; see also Ronald 2003a , b for a review of earlier studies), and L2 vocabulary knowledge improvement, as well as the effectiveness of dictionary use on learning multi-word units (Alzi'Abi 2017 ; Chen 2017 ; Dai et al. 2019 ; Dziemianko 2014 ; Laufer 2011 ; Li and Xu 2015 ; Margarita and Marcos 2019 ; Szczepaniak and Lew 2011 ) and various other aspects of word knowledge ( Chan 2012a , b , c , 2014 , 2017 ; Liu et al. 2014 ).

However, despite the continued growth of empirical work and ongoing interest in this area, no systematic and comprehensive meta-analysis has been undertaken to synthesize the research findings quantitatively. It remains unclear to what extent the effectiveness of dictionary use is in assisting L2 learners to expand their lexical repertoire. Although almost all empirical studies have reported on a positive effect of dictionary use, there are large variations pertaining to the size of the effect across studies. For example, an extremely small effect size of g = .07 was generated by an independent sample from Liu et al. (2014) , while an amazingly huge averaged effect size of g = 11.24 was revealed by Chen (2010) . Currently, there have been two narrative reviews on vocabulary acquisition through dictionary use under intentional/incidental learning conditions ( Ronald 2003a , b ) and a general list of studies on dictionary-induced vocabulary learning ( Welker 2010 ), but none of them have systematically examined the sources of variations among studies. This meta-analysis is, therefore, conducted with a two-fold purpose: to estimate the overall effectiveness of dictionary use on L2 vocabulary acquisition, and to reveal the moderating effects of various moderator variables.

This section examines and evaluates the facilitating or debilitating effects of potential moderator variables in the process of L2 vocabulary learning through dictionary use. These variables, which were all extracted from extant literature, could roughly be categorized into treatment-related variables, methodological variables, and learner-related variables. They were selected either because they serve to resolve contradictions and conflicts across research findings or because their potential moderating effects remain unclear and await to be explored. Alternatively, for some variables, such as methodological ones, their effects are better examined via a meta-analytic approach rather than experimental studies. Also, published meta-analyses on L2 vocabulary acquisition (e.g. Abraham 2008 ; Chen et al. 2018 ; Chiu and Yi-Hui 2013 ; De Vos et al. 2018 ; Huang et al. 2012 ; Lee et al. 2019 ; Mahdi 2017 ; Tsai and Tsai 2018 ; Uchihara et al. 2019 ; Yun 2011 ) were referenced in the choice of moderator variables in order to present a systematic and comprehensive synthesis of research findings on the relationship between dictionary use and L2 vocabulary learning.

2.1 Treatment-related variables

2.1.1 dictionary form.

The relevance of dictionary form to L2 vocabulary acquisition is, to a very large extent, a consequence of technological developments. The advancement of technology has brought about a profound change in learners’ choice of language learning tools, with more of them becoming increasingly dependent on electronic dictionaries instead of paper dictionaries. What comes with this technological convenience is the debate whether electronic dictionaries outperform their paper counterparts in scaffolding L2 vocabulary learning. Dziemianko (2010) investigated the relative usefulness of the paper and electronic versions of a monolingual dictionary in helping Polish learners to acquire L2 lexical knowledge, and she found that the online version was more useful both in the immediate learning and delayed retention of vocabulary knowledge. This finding lent credibility to the results of an earlier study by Kobayashi (2006) , who also reported that electronic pocket dictionaries had an advantage over their paper counterparts. However, like Koyama and Takeuchi (2003) , who reported on a zero effect of dictionary form on immediate learning of L2 vocabulary, Dziemianko (2011 , 2012b , 2017 ) failed to replicate the findings of Dziemianko (2010) that electronic dictionary use led to greater immediate vocabulary learning gains, though she documented a dictionary medium effect on the delayed retention of vocabulary knowledge ( Dziemianko 2017 ). Another study by Chen (2010) even revealed a null effect of dictionary form both in the short-term and long-term retention of vocabulary knowledge. It could be observed that there are controversies over the role of dictionary medium in L2 vocabulary learning, and that even studies from the same researcher have arrived at contradictory assertions.

2.1.2 Dictionary type

Dictionary type in this context is a two-level categorical variable, indicating whether a dictionary is monolingual or bilingual(ized). According to a large-scale study conducted in European countries by Atkins and Knowles (1990) , monolingual dictionaries were often more effective in assisting learners to locate the necessary information as opposed to bilingual ones, although an overwhelming proportion of the participants preferred to use bilingual dictionaries. They also claimed that monolingual dictionaries were superior to bilingual ones for that they usually conveyed information with higher precision and that they could avoid the potential misunderstandings caused by simple translation equivalents common in bilingual dictionaries. Béjoint and Moulin (1987) also argued that monolingual dictionaries had the additional benefit of familiarizing learners with the L2 lexical system. In a controlled study, Laufer and Hadar (1997) investigated the effectiveness of monolingual, bilingual, and bilingualized dictionaries in the receptive and productive learning of L2 new words. They found that overall bilingualized dictionary use yielded greater vocabulary learning gains on the comprehension, production and the whole (comprehension plus production) tasks than monolingual dictionary use, but bilingual dictionaries only outperformed their monolingual counterparts in terms of the production task. When they further grouped the participants into unskilled, average and good dictionary users, results showed that there was only a dictionary type effect on the comprehension task among average users where bilingualized dictionaries outperformed bilingual ones in improving vocabulary knowledge, and a dictionary type effect on the production task among unskilled learners where both bilingualized and bilingual dictionaries led to greater learning gains than their monolingual counterparts. In addition, language proficiency in their study was not found to have a mediational effect as pre-advanced and advanced learners achieved similar learning gains under all the three dictionary use conditions. However, Lew (2002) demonstrated that monolingual dictionaries produced the lowest vocabulary test scores compared with bilingual(ized) ones. He also revealed that proficiency had a positive effect on learning gains and that there was an interaction effect between proficiency and dictionary type. Results of more recent studies suggested that both monolingual and bilingual dictionaries were effective in helping L2 learners acquire various aspects of vocabulary knowledge ( Chan 2012a , b , c , 2014 , 2017 ), although no direct comparison was made between monolingual and bilingual dictionaries in these studies. Apparently, in terms of dictionary type effect in L2 vocabulary acquisition, there were mixed findings.

2.1.3 Target lexical unit

Although Schmitt (2010 : 8-12) pointed out that formulaic sequences were as important as single words in vocabulary instruction and advocated that vocabulary researchers should incorporate both multi-word units and individual words in their work, Lew (2011) noted that the role of collocations has long been neglected both in lexicography and language learning. Lew’s observation could be extended to other multi-word units such as phrasal verbs and idioms. Only a handful of studies (Alzi'Abi 2017 ; Chen 2017 ; Dai et al. 2019 ; Dziemianko 2014 ; Laufer 2011 ; Li and Xu 2015 ; Szczepaniak and Lew 2011 ) have been conducted to explore the relationship between dictionary use and L2 acquisition of multi-word units. As no direct comparison was made in these afore-mentioned studies, whether using dictionaries to learn single words and multi-word units was equally effective remains to be investigated. Two studies, though conducted in the domain of second language vocabulary acquisition, are relevant here. Laufer and Girsai (2008) compared the effects of three types of form-focused instruction on the incidental learning of individual words and collocations. Their results indicated that participants picked up more knowledge about collocations than single words across all the three instructional conditions, a result tending to suggest that collocations were easier to be learned than individual words. However, as Laufer and Girsai (2008) acknowledged, the collocation tests in their study were too easy as almost all the collocations were semantically transparent. There is empirical evidence that intra-lexical factors such as semantic transparency would influence L2 learning difficulty of collocations ( Gyllstad and Wolter 2016 ; Peters 2016 ). Therefore, the findings of Laufer and Girsai (2008) should be interpreted with caution. In a study that investigated how repetition and time of post-test administration would influence the form recall of single words and collocations, Peters (2014) concluded that collocations were more difficult to be learned than single words. It could be seen that even these two studies, both conducted under incidental learning conditions, yielded inconsistent findings about the learning difficulty of single words and multi-word units. Therefore, type of target lexical units was included as a moderator variable to measure its potential moderating effect on dictionary-induced vocabulary learning.

2.1.4 Type of vocabulary knowledge

Since vocabulary knowledge is a multi-dimensional complex construct ( Schmitt 2010 ), it is often operationalized differently and assessed with distinct types of instruments across studies. Among these various frameworks of vocabulary knowledge ( Anderson and Freebody 1981 ; Daller et al. 2007 ; Henriksen 1999 ; Meara and Wolter 2004 ; Nation, 2001 ; Read 2000 ), the most relevant one to this study is Anderson and Freebody's (1981) distinction of vocabulary breadth and vocabulary depth. Vocabulary breadth knowledge, defined by Anderson and Freebody (1981) as the number of words that a learner has at least some certain aspects of knowledge, is usually operationalized as form-meaning mapping knowledge ( Schmitt 2014 ). Laufer and Goldstein (2004) categorized form-meaning mapping knowledge into passive recognition (to recognize the meaning of a word), active recognition (to recognize the form of a word), passive recall (to produce the word meaning when given its form) and active recall (to produce the word form when given its meaning). These four terms used by Laufer and Goldstein (2004) were relabeled by Schmitt (2010) as meaning recognition (for passive recognition), form recognition (for active recognition), meaning recall (for passive recall) and form recall (for active recall), respectively. We will use Schmitt’s (2010) terms to conceptualize different types of form-meaning mapping knowledge (viz. vocabulary breadth knowledge) throughout this paper as these terms make the form-meaning relationships more explicit.

Vocabulary depth knowledge denotes how well a word is known ( Qian 2002 ) and it is concerned with the quality or depth of understanding ( Anderson and Freebody 1981 ). In practice, vocabulary depth knowledge is usually operationalized as morphological knowledge and word associational knowledge (e.g. collocations, synonyms, antonyms; see Qian 1999 , 2002 ; Read 1993 , 2000 ; Schmitt 2014 for details).

As there is evidence to show that different dimensions of vocabulary knowledge do not develop at the same pace under instructional settings (e.g. Van Zeeland and Schmitt 2013 ), whether the effects of dictionary use on L2 vocabulary acquisition might differ depending on different types of vocabulary knowledge being measured awaits to be testified.

2.1.5 Learning condition

Incidental and intentional vocabulary acquisition are two widely researched topics in instructed L2 vocabulary learning ( Schmitt 2008 ). The former refers to the learning of lexical knowledge that occurs as a by-product of activities not designed with the explicit purpose of vocabulary learning ( Ellis 1999 ; Hulstijn 2001 , 2003 ; Laufer and Hulstijn 2001 ), while the latter refers to activities that are purposefully geared to commit word knowledge into memory ( Hulstijn 2001 , 2003 ). Accordingly, it could be argued that studies on dictionary-assisted vocabulary leaning were almost all conducted under incidental and intentional learning conditions. Currently there is empirical evidence suggesting that intentional vocabulary learning would lead to better short-term learning outcomes than incidental vocabulary learning ( Laufer 2003 ; Webb 2002 ), but the effects of learning conditions on long-term gains remain unclear ( Webb and Nation 2017 ). Whether this conclusion could be generalized to dictionary-induced vocabulary learning awaits to be confirmed. In addition, some researchers (e.g. Gass 1999 ; Lew and Doroszewska 2009 : 240) doubted the validity of the distinction between incidental and intentional vocabulary learning and argued that there was no clear-cut boundary in between. Incorporating learning conditions as a moderator variable would help to demonstrate the validity of the controversial incidental/intentional vocabulary learning distinction. The effect of leaning conditions on short-term and long-term vocabulary learning gains, therefore, needs to be further explored.

2.1.6 Research setting

The variable of research settings concerns whether a study is conducted in a laboratory or in a classroom. It is usually believed that laboratory-based studies yield larger effect sizes than classroom-based studies, as in a laboratory, researchers could exercise greater experimental control to isolate the true effects from contaminating variables (see Plonsky and Oswald 2014 for a detailed discussion). We included research settings as a moderator variable in this meta-analysis to examine whether this hypothesis is true in the case of dictionary-induced vocabulary acquisition.

2.2 Methodological variables

2.2.1 type of research design and presence of pretest.

Different types of experimental design were adopted by the studies in our sample, and the major differences among them mainly resided in the presence/absence of a pretest or a control group. Consequently, effect sizes should be calculated from four types of studies—studies that used posttest scores only with a within-group design, studies that used posttest scores only with a between-group design, studies that used pretest-posttest scores with a within-group design, or studies that used pretest-posttest scores with a between-group design where a comparison was made between the experimental group and the control group. Although it is statistically feasible to include studies with different research designs in one meta-analysis ( Borenstein et al. 2009 ), Vacha-Haase and Thompson (2004) pointed out that effect sizes depend upon the type of design from which they are derived. Morris and Deshon (2002) also warned that it would be inappropriate to combine effect sizes from within-group studies and between-group studies if there are systematic differences between the two designs. What matters here is whether these methodological features would moderate the magnitude of the effectiveness of dictionary use on L2 vocabulary acquisition, and whether effect sizes for the two research designs should be presented separately. To investigate this issue, two moderator variables, the presence of a pretest (Yes/No) and the inclusion of a control group (Yes/No), were included in the data analysis.

2.2.2 Assessment type

In literature pertaining to the effects of dictionary use on vocabulary acquisition, researchers are also interested in whether the effects would differ when different types of assessment (receptive vs. productive tests) are used (e.g. Chen 2010 ; Dziemianko 2010 , 2011 , 2012b , 2017 ). In practice, receptive vocabulary knowledge (reception) is usually measured via a multiple-choice test format, requiring test-takers to select the correct word form or meaning; while productive vocabulary knowledge (production) is often evaluated through open-ended questions such as blank-filling or sentence construction, where participants need to supply the appropriate word form or meaning or compose a sentence using the target word. It is a common conviction among vocabulary researchers (e.g. Laufer et al. 2004 ; Nation 2001 ; Schmitt 2010 ; Webb 2005 ) that productive vocabulary tests usually yield lower scores than receptive vocabulary tests. Therefore, there are good reasons to believe that assessment type would be a source of variance.

2.3 Learner-related variables

Age is a widely discussed topic in second language acquisition, and it has been shown to be relevant to various aspects of language learning such as vocabulary and grammar (e.g. Birdsong 2005 ; Ellis 2008 ; Ortega 2009 ). Age matters here in that it concerns the usefulness and usability of a dictionary, that is, whether a dictionary is compiled catering to the needs of users from different age brackets. De Schryver and Prinsloo (2011) presented a detailed review of comparative studies on dictionaries and found that such studies usually made comparisons among dictionaries of the same type or different types, but the dictionaries under comparison were almost all aimed at the same target user group. Therefore, they investigated the effectiveness of three dictionaries intended for different age groups. Their findings seemed disappointing, as all the dictionaries proved to be challenging for their users. To our knowledge, De Schryver and Prinsloo (2011) were the only one of its kind that examined the usefulness of dictionaries among various age groups. However, their study was conducted on L1 learners, with a focus on their understanding of dictionary definitions. How age would moderate dictionary-induced L2 vocabulary learning outcomes remains to be an open issue.

2.3.2 Proficiency

Proficiency is another factor that is relevant to the effectiveness of dictionary use. It is a potential moderator, worth investigating for two reasons. On the one hand, two studies that examined dictionary type effect yielded conflicting results pertaining to the role of proficiency on vocabulary learning through dictionary use. Laufer and Hadar (1997) revealed that language proficiency was not a predictor of vocabulary learning outcomes, as the mean scores of the pre-advanced learners and advanced learners in their study did not differ significantly. However, Lew (2002) showed that high proficiency learners tended to achieve better learning gains. Moreover, he also found that there was an interaction between proficiency and dictionary type, indicating that proficiency might mediate the magnitude of the effectiveness of different types of dictionaries in assisting vocabulary learning. On the other hand, learners’ foreign language proficiency is highly correlated with their vocabulary knowledge (e.g. Alderson 2005 ), and learners’ prior vocabulary knowledge in turn has been proved to predict vocabulary learning (see Uchihara et al. 2019 for a detailed discussion). This is corroborated by Knight (1994) , who explored how dictionary use while reading affected vocabulary acquisition and reading comprehension, and reported that advanced learners picked up significantly more words than low-level learners. Therefore, it is reasonable to deduce that language proficiency would mediate the effectiveness of dictionary-induced vocabulary acquisition.

As can be seen from the literature review above, various moderator variables may contribute to the variations across the empirical findings. To settle the controversies in existing studies and explain the variations quantitatively is clearly beyond the scope of a single experimental study or a narrative review. Meta-analysis offers us a viable solution to summarize the overall effect size and quantify the heterogeneity across empirical findings (see Norris and Ortega 2000, 2006 for detailed discussions on the advantages of meta-analysis over narrative and vote-counting reviews). Therefore, this study aims to address two research questions:

What is the overall effect size of dictionary use on L2 vocabulary acquisition?

What variables moderate the strength of the relationship between dictionary use and L2 vocabulary acquisition?

4.1 Literature retrieval

As meta-analysis has become an increasingly popular quantitative approach in the field of applied linguistics, researchers have given due attention to formulating the procedures for conducting meta-analyses (e.g. Oswald and Plonsky 2010 ; Plonsky and Oswald 2015 ). Accordingly, the first step in the process of data collection for the current meta-analysis was to gather a comprehensive, if not exhaustive, set of meta-analyzable empirical studies which addressed the relationship between dictionary use and L2 vocabulary acquisition. Li et al. (2012) pointed out five most frequently used techniques of literature retrieval and provided a detailed description of each, while In'nami and Koizumi (2012) outlined the guidelines to follow for database selection while doing meta-analysis in the domain of applied linguistics. Primary studies included in this meta-analysis were, therefore, retrieved following the recommendations given by Li et al. (2012) and In'nami and Koizumi (2012) .

Due to the broad coverage and time-saving convenience, six frequently used electronic databases were initially consulted for literature search. They were the Web of Science, ERIC (Education Resources Information Center), LLBA (Linguistics and Language Behavior Abstracts), PsycINFO, PQDT (ProQuest Dissertations and Theses Full Texts), and DDS (Dissertation Discovery System). The search fields used to launch the database search were a number of keywords and their possible combinations, including dictionary, monolingual/bilingual/bilingualized dictionary, dictionary use, dictionary form, dictionary type, paper/printed/mobile/online/electronic/digital dictionary, dictionary user, foreign/second language learner, vocabulary acquisition/learning, incidental/intentional vocabulary acquisition/learning, second/foreign language vocabulary acquisition/learning, vocabulary knowledge, learner’s dictionary, pedagogical dictionary, language production/comprehension (reception), language decoding/encoding, receptive/productive use, and dictionary/corpus example. In addition, prestigious academic journals that cover the topics of lexicography and vocabulary acquisition, such as International Journal of Lexicography and Lexikos , were manually searched issue by issue to locate studies that were not identified through automatic electronic database search. The references of relevant research articles, reviews, books or book sections were also browsed to trace target studies. To cast the net wider, results returned from Google Scholar were also examined to ensure the sampling reached adequate representativeness and inclusiveness.

4.2 Inclusion criteria

In this meta-analysis, a study was considered for inclusion if it satisfied the following eligibility criteria:

It must be reported in English, and should have been published before the cutoff point of data collection, viz. May 2019.

It must investigate the effects of dictionary use on learners’ vocabulary knowledge.

It selected the target vocabulary to be learned from a second/foreign language to the participants.

It must contain at least one lexical test that taps into a certain aspect of learners’ vocabulary knowledge.

It should report by-participant means and standard deviations of participants’ scores in the vocabulary test so that standardized measures of effect sizes, such as Cohen’s d or Hedges’ g , could be calculated. In case it did not report such values, its original datasets should be available. Otherwise, its author(s) should respond to our request for data elicitation, and supply the by-participant means and standard deviations.

It should control for prior knowledge about the target vocabulary by adopting a pretest-posttest within-group design or a between-group design with a control group. Alternatively, it should carefully choose the target vocabulary to ensure that the participants had no prior knowledge before the experiment.

It should explicitly report information about the moderator variables.

It should involve normal L2 learners without cognitive dysfunctions or language disorders as its participants.

A total of more than 3,100 studies that might be relevant to this meta-analysis were identified through the literature search, and the abstracts, sometimes the full texts, of these studies were screened by the researchers to make sure each individual study met the above-mentioned criteria and contained the information needed. In the end, 44 studies were chosen to be included in this meta-analysis.

4.3 Coding rubrics and moderator variables

The coding rubrics for this meta-analysis were developed by following the procedures formulated by Lipsey and Wilson (2001 : 73-90), as well as by referencing the practices of other relevant meta-analytic studies (e.g. Boulton and Cobb 2017 ; Kang and Han 2015 ; Lee et al. 2019 ). Four types of variables, including publication-related factors ( n = 4), participant-related factors ( n = 5), treatment-related factors ( n = 13), and methodological factors ( n = 3), were identified, listed, and elaborated (see Table 1 ).

Coding rubrics for this meta-analysis

Moderator variables, which might moderate the magnitude of effectiveness of dictionary use on L2 vocabulary acquisition, include 7 treatment-related factors (dictionary form, dictionary type, learning condition, target lexical unit, vocabulary knowledge type, research setting, and timepoint), 3 methodological factors (presence of pretest, type of experimental design, and assessment type), and 2 participant-related factors (age and proficiency). Timepoint of the vocabulary test was also included as a moderator variable with the aim of comparing the difference in effect sizes between the immediate posttest and the delayed posttest, although only 15 studies (representing 30 independent samples, corresponding to 40 effect sizes) adopted a pretest-posttest-delayed posttest design. In CMA, this comparison can be made by treating each timepoint as independent of the other timepoint. It should be noted that age was coded as a categorical rather than continuous variable, as only a handful of the sampled studies reported on the exact mean age of the participants. This is also the practice of other meta-analyses such as De Vos et al. (2018) and Uchihara et al. (2019) .

4.4 Coding reliability

Two coders were involved in coding the studies included in this meta-analysis. Five studies were first coded collaboratively by both coders. The aim was to ensure that a consensus could be reached as to how each variable should be coded. Then the remaining 39 studies were coded multiple times by one coder until maximal intra-coder reliability was reached. Subsequently, one third of these 39 studies were randomly assigned to the second coder for coding. Following the reporting practices of previous meta-analyses (e.g. Boulton and Cobb 2017 ; Plonsky 2011 ), we reported the percentage of agreements on all values as the inter-coder reliability measure, and the reliability turned out to be 93%, which was high. The agreement ratios for specific variables were as follows: age, 92%; proficiency, 87%; dictionary form, 94%; dictionary type, 94%; learning condition, 96%; target lexical unit, 97%; type of vocabulary knowledge, 88%; mean, 91%; standard deviation, 91%; sample size, 91%; presence of pretest, 93%; type of research design, 98%; and assessment type, 99%.The minor disagreements between the two coders were settled through negotiation. In the end, the first coder re-examined the coding results to ensure overall consistency.

4.5 Data analysis procedures

This study was conducted using Comprehensive Meta-Analysis (CMA) ( Borenstein et al. 2005 ), a widely used professional software package for doing meta-analysis.

Usually two statistical models are involved in performing a meta-analysis—the fixed-effect model (or the common-effect model, a more descriptive term) versus the random-effects model. The underlying assumption of the former is that the true effect size is the same across all studies, and that sampling error is the only source of variance in observed effects ( Borenstein et al. 2009 ). In contrast, the latter assumes that the true effect size may differ across studies, and differences in observed effect sizes could be a corollary to two sources of variance—both between-study variations and within-study sampling errors ( Borenstein et al. 2009 ). The problem here, when choosing the appropriate model, is whether there is a common effect size across the sampled studies, and whether sampling error is the only source of heterogeneity in observed effect sizes. As both participant- and intervention-related factors would moderate the magnitude of the effectiveness of dictionary use on L2 vocabulary acquisition, it was, therefore, reasonable to adopt the random-effects model as the appropriate statistical model when performing this meta-analysis.

Effect sizes in this study were calculated from means, sample standard deviations, and sample sizes in the immediate outcome measure reported in the sampled studies. As the scale of outcome measures differs from study to study in our sample, it would be meaningless to estimate the summary effect based on raw (unstandardized) mean difference. In this case, the correct way to gauge the effect size is to use standardized mean difference, which is called Cohen’s d ( Cohen, 1962 ). However, Cohen’s d proves to have a slight bias that would overestimate the effect sizes when samples are small ( Borenstein et al. 2009 ; Lipsey and Wilson 2001 ). A solution to avoid such a bias is to use Hedges’ g ( Hedges, 1981 ), which is an unbiased estimate of effect sizes and a simple correction of Cohen’s d ( Borenstein 2009 ; Borenstein et al. 2009 ). Therefore, the effect size statistic used in this study is Hedges’ g . It should be noted that in our analysis with CMA, the results are automatically calibrated by assigning more weight to studies with higher precision, namely, studies with smaller sampling variances.

Another point that needs to be clarified concerns how to deal with studies that yielded more than one effect size. This problem occurred when we were working with studies that included more than one treatment group or studies that adopted multiple outcome measures. In such cases, the common practice to avoid sample size bias is to retrieve only one effect size from one study, unless the study has more than one independent sample. For studies with multiple independent samples, aggregated effect sizes generated from each independent sample should be included ( Lipsey and Wilson 2001 ). This is probably not difficult to understand, as each independent sample carries distinctive information and should be treated individually as one study ( Borenstein et al. 2009 ). A case in point is Szczepaniak and Lew (2011) , where four treatment groups were included and each group of participants were administered a form recall task as well as a meaning recognition task. Accordingly, this study could be seen as including four independent samples. Although each group of participants was measured both for their recognition and recall knowledge of idioms, effect sizes generated from these two tests were averaged when aggregating the overall effect size of dictionary use on L2 vocabulary acquisition, as these two tests both measured the same construct—vocabulary knowledge. Therefore, only four effect sizes from this study were finally taken into consideration when estimating the summary effect.

It is also worth mentioning that the purpose of this meta-analysis is not simply to estimate a summary effect of dictionary use on L2 vocabulary acquisition, but rather to identify and quantify heterogeneity in true effect sizes. Therefore, moderator variable analysis was also performed to examine how the 12 potential moderator variables, as discussed earlier, would moderate the magnitude of effectiveness of dictionary use on L2 vocabulary acquisition. Q- between tests were used to analyze the variables as all of them were categorical. Whenever a decision had to be made on which effect size should be picked in the moderator analysis, we chose the one that appeared first in the coding scheme. For example, we always entered the effect sizes for receptive tests into the analysis when making the reception/production comparison.

Unlike meta-analyses in second language research that only focused on studies with a single type of research design, we included both within-group design studies and between-group design studies in the meta-analysis, following the practices of researchers such as Boulton and Cobb (2017) , Grgurović et al. (2013) , and Lee et al. (2015) . Results of the moderator analyses for the two types of research design are summarized and displayed separately as systematic differences between the two research designs are found (see Table 3 for details).

In total, there were 44 studies in our sample, published between 1992 and 2019. The studies consisted of 87 independent samples, generating 125 effect sizes from a total of 3,475 participants. Participants’ first language (L1) background was mainly Arabic, Chinese, Dutch, English, German, Hebrew, Japanese, Korean, Malay, Polish, and Portuguese, while their L2 only included English, German, and Spanish. The mean length (number of target words) of the vocabulary tests was 12.8 (SD = 6.3, MIN = 5, MAX = 40). Out of the 125 times the vocabulary knowledge measures were administered, the reliability index was only reported 24 times (19%). Among the reported figures, 11 were internal reliability measures and 13 were external reliability (interrater reliability) measures. A detailed description can be found in Table 2 .

Reliability index of the vocabulary knowledge measures

5.1 Outlier diagnosis and overall effect size

Random-effects model of meta-analysis was adopted to gauge the weighted means of the observed effect sizes. The aggregated effect sizes for within-group design studies and between-group design studies are tabulated in Table 3 . It should be noted that before running the analysis, the effect sizes for studies with the two types of research design were first transformed into Z scores, and then entered into SPSS 22.0 for outlier diagnosis. By default, the heuristic used to detect outliers by SPSS is a boxplot-based procedure where data points falling at least 1.5*IQR (Interquartile Range) outside of the IQR box are identified as outliers. For within-group design studies, four extreme cases, all generated by the two independent samples from Chen (2010) , were detected as outliers. For between-group design studies, two outliers from two of the three independent samples (generating 6 effect sizes in total) of Frankenberg-Garcia (2012) were identified. Removal of these extremely large values led to a slight decrease of the aggregated effect size ( g from 2.29 to 2.10 for within-group design studies; g from 1.08 to 1.03 for between-group design studies), but accompanied by higher precision of the summary effect (indicated by the narrower confidence intervals), smaller sampling errors ( Q from 931.60 to 724.13 for within-group design studies; Q from 134.84 to 120.25 for between-group studies), and fewer inconsistencies among the observed effect sizes ( I 2 from 93.99 to 92.54 for within-group studies; I 2 from 77.01 to 74.22 for between-group studies). It could be argued that the model fit was much improved. These outliers were, therefore, filtered out in all statistical analyses that followed.

Overall effect size of dictionary use on L2 vocabulary acquisition

= p < .01; SE = Standard Error; CI = Confidence Intervals; g = Hedges’ g .

5.2 Publication bias

The most direct measure to test for the presence of publication bias is through the use of funnel plots. A funnel plot visually displays the observed effect sizes on the horizontal axis and the standard error or precision of each effect size on the vertical axis. Ideally, if there is no publication bias, the observed values would be nested symmetrically around the averaged effect size, which is indicated by the center line in the funnel plot. The funnel plot of standard error by Hedges’ g for within-group design studies and between-group design studies are shown in Figure 1 and Figure 2 , respectively. A visual inspection of the funnel plots reveals that there is no pronounced asymmetry of data distribution for between-group design studies, while for within-group design studies, there seems to be a slight bias. Following the common practice, Duval and Tweedie’s Trim and Fill ( Duval and Tweedie 2000 ) was also used to assess the potential effect of publication bias on the summary effects. It was found that the point estimates for the two groups of studies remained unchanged when we looked for missing studies on the left of the center line in the funnel plot. However, when imputed studies to the right of the mean were taken into consideration, the point estimate increased slightly from 2.10 (95% CI [1.85, 2.34]) to 2.36 (95% CI [2.07, 2.65]) for within-group design studies; while for between-group design studies, there is only a minute difference between the observed overall effect size and the re-imputed (or adjusted) overall effect size, as the point estimate changes from 1.03 (95% CI [0.84, 1.22]) to 1.06 (95% CI [0.86, 1.25]). Critically important in this context is that for within-group design studies, an effect size of g = 2.10 has the same substantive implications as the adjusted value g = 2.36, as according to Plonsky and Oswald (2014) , both of them are large. In addition, the Classic fail-safe N and Orwin’s fail-safe N were computed for the two groups of studies respectively. For within-group design studies, Classic fail-safe N ( N = 43,918) suggests that 798.5 missing studies are needed for every observed study for the effect to be nullified, while the latter (Orwin’s fail-safe N = 600) indicates that 466 studies with mean Hedges' g of 0 are needed to bring the combined Hedges' g under 0.2 (the benchmark for a trivial Hedges' g ). For between-group design studies, the number of missing studies needed to nullify the effect and bring the effect size to be non-trivial is 3,404 (indicated by Classic fail-safe N ) and 130 (indicated by Orwin’s fail-safe N ), respectively. It could be seen that all these values are larger than the threshold value 5 k + 10 ( k represents the number of sampled primary studies in a meta-analysis; in our case, k = 30 for within-group design studies and k = 14 for between-group design studies) set by Rosenthal (1991) , indicating that publication bias is not a big issue both for within-group design studies and between-group design studies in this meta-analysis.

The Funnel Plot of Standard Error by Hedges’s g for Within-group Design Studies Note : The filled diamond in blue at the bottom of the funnel plot indicates the observed overall effect size, while the filled diamond in red at the bottom of the funnel plot indicates the re-imputed (or adjusted) overall effect size.

The Funnel Plot of Standard Error by Hedges’s g for Between-group Design Studies Note : The filled diamond in blue at the bottom of the funnel plot indicates the observed overall effect size, while the filled diamond in red at the bottom of the funnel plot indicates the re-imputed (or adjusted) overall effect size.

5.3 Moderator variable analysis

An important aim of this study was to isolate and quantify the variations in true effect sizes. To this end, a homogeneity of variance test was conducted to test the assumption of homogeneity in effect sizes. As Table 3 shows, the Q statistic and its p -value (for within-group design studies, Q = 724.13, p < .01; for between-group design studies, Q = 120.25, p < .01), which function as a test of significance, demonstrate that there are large variations among the observed effect sizes. Therefore, the null hypothesis is rejected. Another measure of heterogeneity, the I 2 statistic, which is an index of the proportion of true heterogeneity to the total variances in sampled effect sizes, reveals that 92.54% of the observed heterogeneity for within-group design studies and 74.22% of the observed heterogeneity for between-group design studies are attributable to between-study variations.

In view of all the above, it is necessary to conduct a moderator variable analysis to further investigate how potential moderator variables would predict the effects of dictionary use on L2 vocabulary acquisition. Therefore, subgroup analyses that used Q -between heterogeneity tests were performed on categorical variables, which included 7 treatment-related factors, 2 participant-related factors, and 3 methodological factors, as mentioned earlier. In addition, several interaction effects among the moderator variables were also investigated, including the interactions between two treatment-related factors (dictionary form and dictionary type) and one methodological factor (assessment type). These interactions were singled out mainly because in studies that examined the effects of dictionary form and dictionary type on L2 vocabulary acquisition, an additional factor of assessment type (receptive or productive tests) was usually introduced to investigate whether the effects would differ under different task conditions (see Sections 6.2.1.1 and 6.2.1.2 for detailed discussions).

5.3.1 Treatment-related moderator variables

Results of the analyses for treatment-related moderator variables for within-group design studies and between-group design studies are presented in Tables 4 and 5 , respectively. As can be seen, the effect sizes for immediate post-tests and delayed post-tests are both large across the two research designs, indicating that dictionary use is a very effective vocabulary learning strategy in helping learners achieve immediate learning gains and retain the acquired knowledge. For within-group studies, the effect size for immediate post-tests is significantly larger than that for delayed post-tests ( Q = 9.22, p < .01), indicating that a certain amount of the acquired knowledge is subject to attrition. For between-group studies, no significant difference in effect sizes is detected between immediate post-tests and delayed post-tests. It is interesting to find that the effect size for delayed post-tests is even larger than that for immediate post-tests (such a result is understandable given that effect sizes are calculated from the means and standard deviations of test scores in the immediate post-tests and delayed post-tests respectively).

Analyses of treatment-related moderators (within-group studies)

= p < .05;

= p < .01; (a), (b), and (c) indicate that the analyses were based only on studies that included a delayed posttest.

Analyses of treatment-related moderators (between-group design studies)

Dictionary form is not found to be a significant moderator in either of the two groups of studies. This is the case when we make comparisons based either on the immediate test (for within groups, Q = 0.81, p > .05; for between groups, Q = 3.77, p > .05) or on the delayed test (for within groups, Q = 2.99, p > .05; for between groups, Q = 2.03, p > .05). Regarding the interaction effects between dictionary form and assessment type, the moderating effect of dictionary form is revealed to be significant when productive vocabulary tests are used in within-group studies, but such an effect is not found under any other three conditions (Dictionary form*Reception in within-group studies, Dictionary form*Reception and Dictionary form*Production in between-group studies). However, it should be noted that the p value for the Dictionary form*Production interaction effect in within-group studies is approaching threshold ( p = .04, close to .05). Given that a ubiquitous null effect of dictionary form is observed across all other comparisons and the effect sizes for paper and electronic dictionary use under productive test conditions in within-group studies are not categorically different (both are large according to Plonsky and Oswald 2014 ), we tend to believe that dictionary form is not a significant moderator in dictionary-assisted L2 vocabulary acquisition.

Dictionary type is not found to be significantly associated with the magnitude of the effectiveness of dictionary use in L2 vocabulary acquisition for within-group studies ( Q = 0.84, p > .05) under either receptive ( Q = 2.34, p > .05) or productive ( Q = 0.46, p > .05) vocabulary test use conditions. As shown in Table 4 , both bilingual(ized) and monolingual dictionary use yield large effect sizes, and the difference between the two does not reach statistical significance, although monolingual dictionary use consistently generates larger effect sizes than bilingual(ized) dictionary use in terms of the comparison based on dictionary type, as well as that based on the interaction between dictionary type and assessment type. However, for between-group studies, dictionary type proves to be a significant moderator ( Q = 14.19, p < .01), and monolingual dictionary use leads to a large effect size, while bilingual(ized) dictionary use produces a medium effect size, both when receptive ( Q = 12.64, p < .01) and productive ( Q = 5.24, p < .05) vocabulary tests are used (see Table 5 ).

The effect sizes for using dictionaries to learn single words and multi-word units are both large and not statistically different in within-group studies ( Q = 0.58, p > .05). However, for between-group studies, single words yield a large effect size while multi-word units (though there are only three samples) generate a medium effect size, but the difference does not reach statistical significance ( Q = 2.07, p > .05).

As only one study is found to adopt vocabulary depth measures, which is not meta-analyzable, the comparisons on vocabulary knowledge type are made based on the four levels of form-meaning mapping knowledge. Vocabulary knowledge type proves to be a significant moderator in within-group studies ( Q = 58.62, p < .01) and all the four types of form-meaning mapping knowledge produce large effect sizes. For between-group studies, the moderating effect of vocabulary knowledge type turns out to be non-significant ( Q = 2.06, p > .05) and the effect sizes for the three types of vocabulary breadth knowledge range from small to large. However, it should be pointed out that the number of samples for form recognition in within-group contrasts is only 2. For between-group contrasts, form recall is also under-represented, with only 2 samples, and studies that investigate form recognition knowledge are totally missing.

Both intentional learning and incidental learning yield large effect sizes across the two research designs (note that according to Plonsky and Oswald 2014 , an effect size of g = 0.97 in incidental learning conditions from between-group contrasts could be interpreted as large as it is close to 1.0), and the former consistently generates larger effect sizes than the latter. Learning conditions turns out to be a significant moderator in within-group contrasts ( Q = 5.19, p < .05), while for between-group contrasts, the difference between the two subgroups is not significant ( Q = 3.02, p > .05).

The effect sizes for studies conducted in classroom settings and laboratory settings are both large and not significantly different ( Q = 0.04, p > .05) in within-group contrasts. However, for between-group studies, the effect size for studies conducted in laboratory conditions is medium, while that for classroom settings is large, and the difference is statistically significant ( Q = 6.87, p < .05). But again, the number of samples in laboratory conditions in between-group contrasts is only 5.

5.3.2 Methodological variables

Tables 6 and 7 present the results of moderator analyses on methodological variables. It can be seen that the effect sizes for studies with or without a pretest are large and not significantly different both in within-group contrasts ( Q = 2.49, p > .05) and between-group contrasts ( Q = 0.05, p > .05), indicating that presence of pretest is not a significant moderator variable. Also, both receptive and productive vocabulary tests yield large effect sizes across the two research designs, and receptive tests tend to produce larger effect sizes than productive tests. Although the difference between the two subgroups does not reach statistical significance in either within-group contrasts ( Q = 3.60, p > .05) or between-group contrasts ( Q = 3.39, p > .05), the p values ( p = .06 for both of the two designs when rounded) approach the threshold.

Analyses of methodological variables (within-group studies)

= p < .01.

Analyses of methodological variables (between-group studies)

5.3.3 Participant-related variables

Listed in Tables 8 and 9 are the results of the analyses on participant-related moderator variables. For within-group contrasts ( Table 8 ), it is revealed that both proficiency ( Q = 13.36, p < .01) and age group ( Q = 5.72, p < .05) are significant moderators. The effect sizes for the three proficiency levels are all large, with intermediate learners generating the largest effect size. Also, both the college group and the middle school group produce large effect sizes and the difference between them is significant, indicating that college learners might benefit more from dictionary use in the process of vocabulary acquisition. For between-group contrasts, however, proficiency seems not to be a significant moderator, as the difference among the three proficiency levels does not reach statistical significance ( Q = 3.00, p > .05). The effect sizes for advanced and intermediate learners are large, but that for low level learners is medium. Age is also not found to significantly moderate the relationship between dictionary use and vocabulary acquisition ( Q = 0.61, p > .05). The effect sizes for college learners and middle school learners are both large, and college learners tend to yield a larger (though not significantly) effect size than middle school learners.

Analyses of participant-related variables (within-group studies)

Analyses of participant-related variables (between-group studies)

6.1 Overall effect size and the reliability indices of outcome measures

Echoing the recent trend of research synthesis in second language research (e.g. Norris and Ortega 2006 ; Oswald and Plonsky 2010 ; Plonsky and Oswald 2012 ), this study examines the overall effectiveness of dictionary use on L2 vocabulary acquisition. It is found that the aggregated effect size for within-group studies is Hedges’ g = 2.10, and that for between-group studies is Hedges’ g = 1.03, which are both large in the domain of second language studies. The results indicate that dictionary use leads to an increase of L2 learners’ vocabulary knowledge by 2.10 and 1.03 standard deviations on average for within-group studies and between-group studies, respectively.

According to the guidelines for interpreting effect sizes established by Plonsky and Oswald (2014) , the benchmark value of Cohen’s d for a small, medium, or large effect size for within-group studies is 0.6, 1.1, and 1.4, respectively; while for between-group studies, the value is 0.4, 0.7, and 1.0, respectively. All values approaching the above-mentioned benchmarks could be interpreted as small, medium, and large, respectively. As g is a simple correction of Cohen’s d and is close to it, the estimated summary effect in our study could be interpreted as large in both of the two groups of studies, indicating that dictionary use is very effective in assisting L2 learners to improve their vocabulary knowledge. Table 3 also shows that within-group studies yield a significantly larger effect size than between-group studies ( Q = 43.49, p < .01), a trend that is consistently observed in almost all meta-analyses that included both of the two designs (see Plonsky and Oswald 2014 for a detailed discussion).

It is also interesting to note that the between-group studies in our meta-analysis generate a larger effect size compared with other relevant meta-analytic studies on vocabulary acquisition (e.g. Chen et al. 2018 ; Chiu and Yi-Hui 2013 ; Huang et al. 2012 ; Lee et al. 2018; Mahdi 2017 ; Tsai and Tsai 2018 ; Yun 2011 ) except for Abraham (2008) . On the one hand, this might indicate that dictionary use is a more effective measure in improving L2 vocabulary knowledge than other measures such as digital games and mobile devices. On the other hand, however, this could be attributable to the different learning conditions under which the sampled studies in each meta-analysis were conducted, as both dictionary use under intentional and incidental learning conditions were included in our study and intentional learning is found to yield larger effect sizes than incidental learning (see Section 6.2.1.5). As almost all meta-analyses on vocabulary acquisition have only included between-group design studies in the analysis so far, it is not feasible to compare the results of the within-group studies in our study with those of other studies.

In addition, the overall effect sizes in the immediate and delayed posttests are both large across the two designs (see Tables 4 and 5 ), indicating that dictionary use is effective in both helping learners achieve short-term learning gains and retaining the acquired knowledge in the long run, although a considerable amount of knowledge is subject to attrition in the case of within-group studies.

As shown in Table 2 , the reliability index of vocabulary knowledge measures is only reported 19% (24 out of 125) of the times. It is also found that the medians of internal reliability and interrater reliability are 0.7 and 0.93, respectively. The median internal reliability (0.7) is relatively lower when compared with Plonsky and Derrick (2016) , who meta-analyzed the reliability coefficients in second language research and reported the median internal reliability for vocabulary measures to be 0.83. However, the median interrater reliability (0.93) of vocabulary measures in our study closely approximates the figure (0.95) reported by Plonsky and Derrick (2016) . This might implicate that researchers should place more emphasis on the internal reliability of vocabulary measures when investigating the relationship between dictionary use and L2 vocabulary acquisition in future studies.

6.2 The moderator variables

As indicated by the results of the homogeneity of variance test, large variations among effect sizes are found to exist across studies ( Q = 724.13, p < .01 for within-group studies; Q = 120.25, p < .01 for between-group studies). Subsequent moderator variable analysis generates significant findings that are useful in clarifying the complex issues in dictionary-induced L2 vocabulary learning, as well as contributing to our understanding of the sources of heterogeneity among studies.

As mentioned earlier, we included both within-group studies and between-group studies in the meta-analysis, and presented the results for each research design separately. Doing so, however, may make it difficult to identify a clear pattern of the data, as the two groups of studies sometimes yield inequivalent or contradictory findings. The common practice to tackle this tricky issue is to rely on the more robust findings. As can be seen from the results section, all Confidence Intervals do not cross zero, therefore the results are statistically reliable (see Boulton and Cobb 2017 and Lee et al. 2015 ). There is no widely acknowledged consensus about the minimum number of effect sizes that are meta-analyzable, as Borenstein et al. (2009) argued that even two effect sizes are meta-analyzable theoretically. But according to the seminal review by Hillocks (1986) , analyses should be conducted for each treatment that includes four or more effect sizes. We regard effect sizes generated from 4 or more samples (though see Boulton and Cobb 2017 for a more rigorous criterion of 10 samples) in either of the two research designs as robust. This is done, as we believe that results from either of the two research designs would provide unique information if a certain criterion is met (in our case, results should be generated from 4 samples or more), as Lee et al. (2015: 352) mentioned.

6.2.1 Treatment-related variables

6.2.1.1 dictionary form.

No clear pattern is observed regarding the relative effectiveness of paper/electronic dictionaries in L2 vocabulary acquisition, as across the two designs, paper dictionaries sometimes outperform their electronic counterparts, while at other times the reverse is true. Given that both paper and electronic dictionaries yield medium to large effect sizes across all the conditions under comparison, it could be argued that both types are effective vocabulary learning tools. Q - between tests also show that dictionary form does not significantly predict either short-term or long-term vocabulary learning gains, either in terms of reception knowledge or in terms of production knowledge. This finding is consistent with Chen (2010) , Dziemianko (2011 , 2012b ), and Koyama and Takeuchi (2003 , 2004 ), but in contradiction with Dziemianko (2010 , 2017 ).

First, it is widely acknowledged that electronic dictionaries allow users to retrieve information more conveniently and look up more words within a limited range of time ( Aust et al. 1993 ; De Schryver 2003 ; Dziemianko 2012a ; Guillot and Kenning 1994 ; Nesi 2000b ; Roby 1991 ). Some researchers speculated that more dictionary lookups yield greater vocabulary learning gains ( Komuro et al. 2006 ; Roby 1991 ), which might explain why an advantage was found of electronic dictionaries over their paper counterparts in assisting L2 vocabulary learning. However, this speculation has never been tested empirically ( Nesi 2000b ; Tono 2000 ). The study by Chen (2010) did reveal that learners who used electronic dictionaries looked up words more frequently than those who used paper dictionaries, but still no effect of dictionary form was found on their vocabulary learning gains. Laufer and Hill (2000) even reported that there was no relationship between word learning and the number of times the target word was looked up in a dictionary. Therefore, it is possible that the assumed advantage of electronic dictionaries over paper dictionaries in assisting L2 vocabulary learning actually does not exist at all.

Second, electronic dictionary use is not without disadvantages. As Sharpe (1995) observed, there was a fear among language instructors that fast information retrieval might result in lower retention rates during the language learning process. Nesi (2000a ) also expressed a similar concern in claiming that more easily retrieved information was more likely to be forgotten. In light of the Involvement Load Hypothesis ( Hulstijn and Laufer 2001 ; Laufer and Hulstijn 2001 ), extracting information with more ease in the case of electronic dictionaries might result in smaller task involvement load and shallower processing of target language structures, which in turn might lead to worse learning outcomes. The null effect of dictionary form demonstrated in our meta-analysis points to the possibility that there is a tradeoff between the lookup speed advantage and the shallower processing or smaller involvement load accompanied by electronic dictionary use. Even if more lookups resulted from more convenient information retrieval really led to better learning outcomes, the effect would be counterbalanced, as shallower processing of the looked-up words makes newly acquired words more easily forgotten.

In addition, in summarizing the findings of her series of studies ( Dziemianko 2010 , 2011 , 2012b , 2017 ) on the effect of dictionary form, Dziemianko (2017 : 347) acknowledged that the beneficial effect of electronic dictionary medium observed in Dziemianko (2010 , 2017 ) was ‘indeed dictionary-specific’, indicating that only the use of MEDO in Dziemianko (2017) and COBUILD6 in Dziemianko (2010) led to greater vocabulary learning gains. Even so, the findings from the two studies were not entirely consistent, as Dziemianko (2010) revealed that there was an effect of dictionary form on task type (reception or production), immediate retention, and delayed retention, while Dziemianko (2017) failed to observe such an effect on task type. This might invite us to question the generalizability of the advantage effect of electronic dictionaries. If it is indeed the case that the positive effect of electronic dictionary form is dictionary specific, then our results suggest that this effect could be nullified when a large number of other dictionaries are sampled in the 44 studies (representing 87 independent samples) included in our meta-analysis.

Therefore, it could be argued that overall electronic dictionaries are as effective as their paper counterparts in assisting L2 learners to acquire receptive or productive vocabulary knowledge, and achieving short-term or long-term learning gains.

6.2.1.2 Dictionary type

Overall, monolingual dictionaries consistently produce larger effect sizes than bilingual(ized) dictionaries across the two research designs, both in terms of receptive and productive tasks, indicating that monolingual dictionaries are more effective than bilingual(ized) ones in facilitating L2 vocabulary acquisition. The results lend support to the large-scale study conducted in European countries by Atkins and Knowles (1990) , who revealed a monolingual dictionary advantage. However, it should be noted that the difference in effect sizes between monolingual and bilingual dictionaries does not reach statistical significance in within-group studies, although monolingual dictionaries always outperform bilingual(ized) ones. This might implicate that the monolingual dictionary advantage effect is more revealing if greater experimental controls are exercised, as significant differences across all comparisons between monolingual and bilingual(ized) dictionaries are detected in between-group studies.

This finding is not consistent with Laufer and Hadar (1997) and Lew (2002) . Laufer and Hadar (1997 : 195) exercised caution in interpreting the bilingualized dictionary advantage by stating that the bilingualized dictionary use ‘tends to produce the best scores’, as they found that not all the differences among monolingual, bilingual, and bilingualized dictionaries were statistically significant when the participants were grouped according to their dictionary use skills. More importantly, they pointed out that the relative effectiveness of monolingual and bilingual dictionaries was conditional upon dictionary users and task types. If this is the case, results of our meta-analysis suggest that monolingual dictionary use leads to larger L2 vocabulary learning gains when samples are large, both in terms of the number of dictionaries that are used and participants that are involved.

Another conceivable reason why Laufer and Hadar (1997) discovered an advantage in favour of bilingual(ized) dictionaries under some (but not all) circumstances is that the participants were more familiar with the use of bilingual(ized) dictionaries, as an overwhelming proportion of L2 learners showed preference for this type of dictionaries (e.g. Atkins and Knowles 1990 ; Piotrowski 1989 ). It could well be the case that dictionary use preferences contributed to a source of difference in learners’ test scores. When studies with participants who preferred using monolingual dictionaries, such as Chan (2012a , 2012b , 2012c , 2014 ), were included in the meta-analysis, the potential effects of dictionary use preferences were counter-balanced and the advantage effect of monolingual dictionaries becomes pronounced. This speculation, however, remains to be confirmed empirically.

6.2.1.3 Target lexical unit

Across the two research designs, there is no clear pattern for the relative effectiveness of dictionary use in L2 acquisition of single words and multi-word units and the difference in effect sizes between the two types of lexical units is not significant, suggesting that dictionary use leads to roughly equal learning gains for single words and multi-word units. This finding is somewhat counterintuitive, as it is widely acknowledged that more complex lexical structures like multi-word units often pose challenges to L2 learners (e.g. Boers and Lindstromberg 2012 ; Schmitt 2004 , 2010 ; Wray 2002 , 2008 ).

However, it would be premature to rush to the conclusion that target lexical unit is not a significant moderator, and that single words and multi-word units are equally difficult for learners. A closer examination of the studies included in this meta-analysis shows that out of the 15 studies on the acquisition of multi-word units, 14 of them were conducted under intentional learning conditions. In the case of single words, the number of sampled studies for intentional and incidental learning conditions is 27 and 45 respectively, which is relatively balanced and reaches adequate representativeness for each learning condition. As shown in Tables 4 and 5 , studies conducted under intentional learning conditions yield larger effect sizes than those conducted under incidental learning conditions. Therefore, it could be reasonably argued that the effect size for multi-word units might be inflated as learning conditions are not balanced. To clarify this issue, we propose that future studies should be conducted to investigate the effectiveness of dictionary use on L2 learning of single words and multi-word units both under incidental and intentional learning conditions, so that a direct comparison could be made on their learning difficulties under different learning conditions.

6.2.1.4 Vocabulary knowledge type

Vocabulary knowledge type proves to significantly moderate the relationship between dictionary use and L2 vocabulary acquisition in within-group studies. Among the three levels of form-meaning mapping knowledge, form recall knowledge benefits from dictionary use most, followed by meaning recognition and meaning recall. This suggests that vocabulary form knowledge (note that form recognition knowledge yields the largest effect size of 4.55, 95% CI [3.83, 5.26] if it is not filtered out of the analysis due to the limited number of effect sizes) may be easier to be learned than meaning knowledge with the assistance of dictionaries, and that learners who use a dictionary to learn vocabulary may have to master knowledge of lexical forms before they could establish form-meaning mappings. While for between-group studies, the difference between the three types of form-meaning knowledge (note that there are only 2 samples for form recall, and form recognition is totally missing ) is not significant, and meaning recognition knowledge seems to be more difficult to master than meaning recall knowledge, a pattern that is opposite to that observed in within-group studies. To sum up, our results regarding the effect of vocabulary knowledge type are still inconclusive. Further studies should be conducted to investigate how dictionary use may assist learners to acquire the four different types of form-meaning knowledge.

As mentioned earlier in Section 5.3, only one study ( Dai et al. 2019 ) investigated whether dictionary use would help learners acquire vocabulary depth knowledge (collocational knowledge in their case). It is suggested that future studies may adopt a dimensions approach to vocabulary knowledge to examine the effects of dictionary use in improving other aspects of vocabulary knowledge (e.g. grammatical knowledge, collocational knowledge, and semantic relations) beyond form-meaning mappings.

Although we are unable to pin down how dictionary use may help learners pick up various types of vocabulary knowledge, we believe that more fine-grained classifications of vocabulary knowledge beyond the receptive/productive distinction are necessary and should be introduced into dictionary-induced L2 vocabulary acquisition research. On the one hand, Laufer and Goldstein (2004) have shown that the four types of form-meaning knowledge are of different hierarchies of difficulty. Researchers therefore should be aware which specific type of vocabulary knowledge may benefit from dictionary use most, so that they could design follow-up activities to improve knowledge aspects that are not easily learned through dictionary use. On the other hand, a fine-grained classification of vocabulary knowledge is crucially important to research design. For example, if we were to choose target words that are unfamiliar to learners, a test that measures form recognition knowledge would be the ideal choice because form recognition is easiest among all types of form-meaning knowledge. If a learner has no form recognition knowledge of a given lexical unit, it is quite possible that he/she also has no mastery of other aspects of knowledge about the given lexical unit, then prior knowledge about the target lexical units could be well controlled for. It is also suggested that researchers should try to adopt vocabulary tests that gauge the same specific aspect of vocabulary knowledge in the pretest and (immediate and delayed) posttest, so that test score differences could truly reflect learning gains on a certain knowledge construct, though we are aware that it is sometimes difficult to do so in practice.

6.2.1.5 Learning condition

The fact that the effect sizes for dictionary use under intentional and incidental learning conditions are significantly different in the immediate test in within-group studies but not in between-group studies indicates that the boundaries between the two constructs, intentional learning and incidental learning, might not be that clear-cut, and that the validity of the distinction is not completely confirmed (at least in the case of dictionary-assisted L2 vocabulary acquisition). This might be attributable to the fact that the distinction between the two learning conditions is ad-hoc and methodological, rather than theoretical ( Hulstijn 2003 ; Peters et al. 2009 ). The larger effect sizes for intentional learning conditions than incidental learning conditions across the two designs (though the difference is not significant in the case of between-group studies) also lend credibility to Laufer (2003) and Webb (2002) , who reported that deliberate learning always yielded larger immediate vocabulary learning gains.

Webb and Nation (2017 : 60) also pointed out that it remained unclear whether intentional learning would produce better long-term vocabulary learning gains than incidental learning. Our moderator analysis on delayed retention of vocabulary knowledge suggests that dictionary use under incidental conditions would lead to better long-term retention of the acquired vocabulary knowledge than intentional learning conditions, as the effect sizes for incidental learning change only slightly from g = 1.81 (within-group studies) and g = 0.90 (between-group studies) in the immediate test to g = 1.66 (within-group studies) and g = 0.97 (between-group studies) in the delayed retention test, while in the case of intentional learning conditions, the effect sizes drop sharply from g = 2.34 (within-group studies) and g = 1.27 (between-group studies) to g = 1.49 (within-group studies) and g = 0.75 (between-group studies).

In addition, De Vos et al. (2018 : 930) claimed that it was impossible to compare the effect size for incidental vocabulary learning from spoken input (i.e. the results of their study) with that from written input, as no meta-analysis has been undertaken on the latter yet. Results of our study demonstrate that dictionary use under incidental vocabulary learning from written input conditions produced an effect size of g = 1.81 (within-group studies) and g = 0.9 (between-group studies).As the effect size g = 1.05 revealed by De Vos et al. (2018) was calculated by combining the data from within-group studies and between-group studies, it is not feasible to compare the data in our study with that in their study directly. If we follow De Vos et al. (2018) in calculating the effect size data, the combined effect size for incidental vocabulary learning in our study will be g = 1.41, with 0.36 standard deviations larger than that reported by De Vos et al. (2018) .This might suggest that incidental vocabulary learning from written input produces better learning outcomes than that from spoken input, although in our study learning was facilitated by dictionary use and in the study by De Vos et al. (2018) learners engaged in four different types of activities.

6.2.1.6 Research setting

For within-group studies, the effect size for laboratory settings does not differ significantly from that for classroom settings. The effect sizes for the two research settings are both very large, and there is only a minute difference (0.05) between the two, although classroom settings yield a slightly larger effect size than laboratory settings. For between-group studies, the effect size for classroom settings is large while that for laboratory settings is medium, and the difference between the two is significant. Our results provide counterevidence to Plonsky and Oswald (2014) who claimed that laboratory-based studies always yielded larger effect sizes than classroom-based studies. The tricky effect of research settings is also found in Boulton and Cobb (2017) and Lee et al. (2015) . Boulton and Cobb (2017) revealed that the effect size for laboratory-based studies was larger than that for classroom-based studies in within-group contrasts, and the reverse was true for between-group contrasts. Lee et al. (2015) CR, however, found the opposite pattern with Boulton and Cobb (2017) across the two designs.

6.2.2 Methodological variables

6.2.2.1 type of research design and presence of pre-test.

Methodologically, within-group design studies produce a larger effect size than between-group design studies (see Table 3 ), indicating that the absence of a control group tends to inflate the effect size. This finding is congruent with De Vos et al. (2018) , who also reported on a similar effect of experimental design while meta-analyzing L2 vocabulary learning from spoken input. This is probably not difficult to understand, as in studies with a within-group design, participants usually serve as their own control, which results in less error variance ( Plonsky and Oswald 2014 : 898). Nation and Webb (2011) also pointed out that the inclusion of a control group in the research design might help to adjust for the potential side effects of pretest use. Anyway, the findings point to the importance of including a control group in estimating the effectiveness of dictionary use on L2 vocabulary acquisition. This is especially so in cases similar to ours, where about sixty percent of the studies in our sample adopted a within-group design. Future studies should make improvements in this aspect.

Presence of pretest is not found to be a significant moderator, which goes against our expectation. One possible explanation is that prior knowledge about the target words to be learned was well controlled for in studies without using a pretest. Typically in these studies, a set of words would be tested in a pilot study on a group of participants parallel to those who took part in the main study. Test administrators carefully singled out items that were unknown to participants in the pilot study, with the assumption that they were also unknown to the participants in the main study (e.g. Liu and Lin 2011 ). The target words could also be selected based on the researchers’ intuition that they were problematic to the participants (e.g. Frankenberg-Garcia 2014 ; Laufer 2011 ). In some cases, pseudo words were used to ensure that the participants had no knowledge about them at all (e.g. Peters 2007 ). Thus, the influence of prior vocabulary knowledge could be minimized through the administration of a pilot study and careful selection of target test items. Given that in this meta-analysis the presence of pretest variables was coded depending upon whether there was a pretest in the main study rather than the pilot study, it is probably not difficult to understand why presence of the pretest is not a significant moderator.

6.2.2.2 Assessment type

It is not surprising to find that the effect size generated by receptive tests is consistently larger than that generated by productive tests across the two designs. This finding is consistent with the common assumption that receptive knowledge is easier to acquire. Although the difference in effect sizes between receptive and productive tests is not significant across the two groups of studies, the p values are close to the threshold ( p = .06 for within-group studies and p = .07 for between-group studies). It should also be pointed out that the effect sizes for the two assessment types are both very large (for within-group studies, g = 2.50 for reception and g = 1.95 for production, respectively; for between-group studies, g = 1.33 for reception and g = 0.90 for production, respectively), indicating that dictionary use is a very effective vocabulary learning strategy for improving both receptive and productive word knowledge.

6.2.3 Learner-related variables

6.2.3.1 age.

Both the college group and the middle school group generate large effect sizes across the two designs, indicating that dictionary use is an effective vocabulary learning strategy across the two age groups. Older learners (college group) also consistently yield a larger effect size than younger learners (middle school group), although the difference between the two is significant only in within-group contrasts. It should be pointed out that no studies in our sample have ever tested primary school students to evaluate the effectiveness of dictionary use in improving their L2 vocabulary knowledge. A possible explanation for the age effect could be that older learners might have more L2 learning experience, such as the amount of L2 input and length of L2 exposure, which proves to be closely related to L2 ultimate attainment in vocabulary and collocations (e.g. Dąbrowska 2019 ). It could also be that age is entangled with proficiency. As all the participants in the sampled studies were from formal educational context, it is reasonable to claim that older learners usually attain higher education levels. An implication of this finding is that dictionary compilers should cater to the needs of users at different age brackets to maximize the effectiveness of dictionary use.

6.2.3.2 Proficiency

We mainly rely on the data from within-group contrasts when discussing the effect of proficiency, as there are only 3 samples (less than 5) for advanced learners from between-group studies. Overall proficiency has a positive effect on dictionary-assisted vocabulary acquisition, but the relationship between the two is not linear, as the magnitude of the effectiveness of dictionary use hits its peak when learners reach intermediate level but plateaus at advanced proficiency. One conceivable reason is that advanced learners are more likely to infer the meaning of target words from context correctly (e.g. Montero Perez et al. 2014 ; Peters et al. 2016 ; Peters and Webb 2018 ), therefore they tend to use dictionaries less. If we look at the data in between-group studies despite the limited number of effect sizes for advanced level learners, the results would reveal that more advanced learners tend to pick up more vocabulary knowledge from dictionary use. It should be noted that the inconsistency between the two groups of studies might also be attributable to the inherent difficulty in operationalizing proficiency in meta-analyses ( Lee et al. 2015)

To our knowledge, this is the first meta-analysis conducted in the field of lexicography. The overall effect size for dictionary-induced vocabulary learning is found to be very large. Moderator variable analysis was performed on treatment-related variables, methodological variables, and learner-related variables to quantify the variations across studies. The findings of our study have several implications for dictionary use research as well as L2 vocabulary learning. First, dictionary use should be encouraged as it is shown to be a very effective vocabulary learning strategy. Second, as dictionary form is not a significant predictor of vocabulary learning gains, there is no need for language instructors to be torn between electronic and paper dictionaries when advising their students in the choice of the appropriate dictionary for use. Third, as monolingual dictionary use tends to produce larger effect sizes than bilingual(ized) dictionary use, learners may give top priority to monolingual dictionaries in the learning process. Fourth, although dictionary use under intentional learning condition produces larger immediate learning gains, the acquired knowledge more easily gets forgotten. Post-task activities (see Nation 2001 and Webb and Nation 2017 for detailed discussions of vocabulary learning activities) should be in position to overcome this disadvantage of deliberate learning. In addition, in terms of experimental design, researchers should try to include a control group in their research design.

There are also several limitations in the current study. The first one concerns the population of this study. On the one hand, as mentioned earlier, no studies have been conducted to examine the effect of dictionary use on L2 primary school students’ vocabulary learning. On the other hand, the population in the sampled studies almost all come from formal education settings such as universities and middle schools, which Kosem et al. (2019) , Müller-Spitzer (2014 : 4) and Lew (2015) also noticed. It could be argued that pupils and the more general dictionary-use public are not represented in this study. Future studies should direct attention to these groups of dictionary users. Secondly, only four studies have experimented on learners whose L2 is not English, but all of the target L2 in these four studies are Indo-European languages. Whether the findings from this study could be generalized to learners with a non-Indo-European L2 such as Chinese remains to be further investigated. Third, almost no studies have been conducted on the effect of dictionary use on L2 incidental acquisition of multi-word units. It remains unclear how effective dictionary use is in this context. In addition, our results pertaining to how dictionary use would help learners acquire various aspects of vocabulary knowledge are still inconclusive. More studies are needed to investigate this issue by adopting a dimensions approach to vocabulary knowledge. Finally, a closer examination of the sampled studies shows that only a handful of them (19%) reported the reliability of the vocabulary measures. It remains unclear how the reliability of the outcome measures would influence the learning outcomes. Future studies should make improvements in this regard.

This study was supported by the National Social Science Fund of China (Grant No. 15BYY062), and by the MOE Project of Key Research Institute of Humanities and Social Sciences at Universities in P.R. China (Project No. 17JJD740004). We would like to thank Professor Robert Lew and two anonymous reviewers for their valuable comments and suggestions. Professor Robert Lew, Dr. Alice Yin Wa Chan and Dr. Yuanjun Dai kindly provided us with the raw data of their studies. Correspondence should be addressed to Professor Hai Xu.

Supporting Information

Additional Supporting Information may be found in the online version of this article at the publisher’s website. The EXCEL file contains the following materials:

Appendix 1 : Effect sizes for between group studies without a pretest Appendix 2 : Effect sizes for between group studies with a pretest Appendix 3 : Effect sizes for within group studies without a pretest Appendix 4 : Effect sizes for within group studies with a pretest Appendix 5 : Study Features

Abraham L. B.   2008 . ‘ Computer-Mediated Glosses in Second Language Reading Comprehension and Vocabulary Learning: A Meta-Analysis .’ Computer Assisted Language Learning   21 . 3 : 199 – 226 .

Google Scholar

Al-Ajmi H.   2008 . ‘ The Effectiveness of Dictionary Examples in Decoding: The Case of Kuwaiti Learners of English . ’ Lexikos   18 : 15 – 26 .

*Al-Shehri S. , Gitsaki C. . 2010 . ‘ Online Reading: A Preliminary Study of the Impact of Integrated and Split-Attention Formats on L2 Students’ Cognitive Load .’ ReCALL   22 . 3 : 356 – 375 .

Alderson J. C.   2005 . Diagnosing Foreign Language Proficiency: The Interface between Learning and Assessment . London : Continuum .

Google Preview

*Alzi'Abi S. E.   2017 . ‘ Guessing Verb–Adverb Collocations: Arab EFL Learners' Use of Electronic Dictionaries .’ Lexikos   27 : 50 – 77 .

Anderson R. C. , Freebody P. . 1981 . ‘Vocabulary Knowledge’ In Guthrie J. T. (ed.), Comprehension and Teaching: Research Reviews   Newark, DE : International Reading Association , 77 – 117 .

Atkins B. T. S. , Knowles F. E. . 1990 . ‘Interim Report on the EURALEX/AILA Research Project into Dictionary Use’ In Magay T. , Zigány J. (eds), Budalex ’88 Proceedings. Papers from the 3rd International Euralex Congress, Budapest, 4-9 September 1988 . Budapest : Akadémiai Kiadó , 381 – 392 .

Aust R. , Kelley M. J. , Roby W. B. . 1993 . ‘ The Use of Hyper-Reference and Conventional Dictionaries .’ Educational Technology Research & Development   41 . 4 : 63 – 73 .

Béjoint H. B. , Moulin A. . 1987 . ‘The Place of the Dictionary in an EFL Programm’ In Cowie A. P. (ed.), The Dictionary and the Language Learner . Tübingen : Niemeyer , 381 – 392 .

Birdsong D.   2005 . ‘Interpreting Age Effects in Second Language Acquisition’ In Kroll J. F. , de Groot A. M. B. (eds), Handbook of Bilingualism: Psycholinguistic Approaches . Oxford : Oxford University Press , 109 – 127 .

Boers F. , Lindstromberg S. . 2012 . ‘ Experimental and Intervention Studies on Formulaic Sequences in a Second Language .’ Annual Review of Applied Linguistics   32 : 83 – 110 .

Borenstein M.   2009 . ‘Effect Sizes for Continuous Data’ In Cooper H. , Hedges L. V. and C. Valentine J. (eds), The Handbook of Research Synthesis and Meta-Analysis  (2nd edition.). New York : Russell Sage Foundation , 221 – 235 .

Borenstein M. , Hedges L. V. , Higgins J. P. T. , Rothstein H. R. . 2005 . Comprehensive Meta-Analysis (Version 2) . Engelwood, NJ : Biostat .

Borenstein M. , Hedges L. V. , Higgins J. P. T. , Rothstein H. R. . 2011 . Introduction to Meta-Analysis . Chichester, UK : John Wiley & Sons .

Boulton A. , Cobb T. . 2017 . ‘ Corpus Use in Language Learning: A Meta‐Analysis .’ Language Learning   67 . 2 : 348 – 393 .

*Chan A. Y. W.   2012 a. ‘ Advanced Cantonese ESL Learners' Use of a Monolingual Dictionary for Language Production .’ Lexikos   22 : 319 – 326 .

*Chan A. Y. W.   2012 b. ‘Cantonese ESL Learners’ Use of Grammatical Information in a Monolingual Dictionary for Determining the Correct Use of a Target Word .’ International Journal of Lexicography   25 . 1 : 68 – 94 .

*Chan A. Y. W.   2012 c. ‘ The Use of a Monolingual Dictionary for Meaning Determination by Advanced Cantonese ESL Learners in Hong Kong .’ Applied Linguistics   33 . 2 : 115 – 140 .

*Chan A. Y. W.   2014 . ‘ Using LDOCE5 and COBUILD6 for Meaning Determination and Sentence Construction: What Do Learners Prefer? ’ International Journal of Lexicography   27 . 1 : 25 – 53 .

*Chan A. Y. W.   2017 . ‘ The Effectiveness of Using a Bilingualized Dictionary for Determining Noun Countability and Article Selection .’ Lexikos   27 : 183 – 213 .

*Chang Y. H. , Liu T. C. , Paas F. . 2018 . ‘ Cognitive Resources Allocation in Computer-Mediated Dictionary Assisted Learning: From Word Meaning to Inferential Comprehension .’ Computers & Education   127 : 113 – 129 .

Chen M. H. , Tseng W.-T. , Hsiao T.-Y. . 2018 . ‘ The Effectiveness of Digital Game-Based Vocabulary Learning: A Framework-Based View of Meta-Analysis .’ British Journal of Educational Technology   49 . 1 : 69 – 77 .

*Chen Q.   2015 . Role of Dictionary Use in the Production and Retention of Formulaic Sequences in SLA . M.A. Thesis, Soochow University.

*Chen Y.   2010 . ‘ Dictionary Use and EFL Learning. A Contrastive Study of Pocket Electronic Dictionaries and Paper Dictionaries .’ International Journal of Lexicography   23 . 3 : 275 – 306 .

*Chen Y.   2012 . ‘ Dictionary Use and Vocabulary Learning in the Context of Reading .’ International Journal of Lexicography   25 . 2 : 216 – 247 .

*Chen Y.   2017 . ‘ Dictionary Use for Collocation Production and Retention: A CALL-Based Study . ’ International Journal of Lexicography   30 . 2 : 225 – 251 .

Chiu Y. H.   2013 . ‘ Computer-Assisted Second Language Vocabulary Instruction: A Meta-Analysis .’ British Journal of Educational Technology   44 . 2 : E52 – E56 .

*Chiu L.-L. , Liu G.-Z. . 2013 . ‘ Effects of Printed, Pocket Electronic, and Online Dictionaries on High School Students’ English Vocabulary Retention .’ The Asia-Pacific Education Researcher   22 . 4 : 619 – 634 .

Cohen J.   1962 . Statistical Power Analysis for the Behavioral Sciences . New York : Academic Press .

Dąbrowska E.   2019 . ‘Experience, Aptitude, and Individual Differences in Linguistic Attainment: A Comparison of Native and Nonnative Speakers .’ Language Learning   69 . S1 : 72 – 100 .

*Dai Y. , Wu Z. , Xu H. . 2019 . ‘ The Effect of Types of Dictionary Presentation on the Retention of Metaphorical Collocations: Involvement Load Hypothesis vs. Cognitive Load Theory .’ International Journal of Lexicography   32 . 4 : 411 – 431 .

Daller H. , Milton J. , Treffers-Daller J. . 2007 . ‘Editor's Introduction’ In Daller H. , Milton J. and Treffers-Daller J. (eds), Modelling and Accessing Vocabulary . Cambridge : Cambridge University Press , 1 – 32 .

De Schryver G. M.   2003 . ‘ Lexicographers' Dreams in the Electronic‐Dictionary Age .’ International Journal of Lexicography   16 . 2 : 143 – 199 .

De Schryver G. M. , Prinsloo D. J. . 2011 . ‘ Do Dictionaries Define on the Level of Their Target Users? A Case Study for Three Dutch Dictionaries .’ International Journal of Lexicography   24 . 1 : 5 – 28 .

De Vos J. F. , Schriefers H. , Nivard M. G. , Lemhöfer K. . 2018 . ‘ A Meta-Analysis and Meta-Regression of Incidental Second Language Word Learning from Spoken Input .’ Language Learning   68 . 4 : 906 – 941 .

Duval S. , Tweedie R. . 2000 . ‘ Trim and Fill: A Simple Funnel-Plot-Based Method .’ Biometrics   56 . 2 : 455 – 463 .

Dziemianko A.   2010 . ‘Paper or Electronic? The Role of Dictionary Form in Language Reception, Production and the Retention of Meaning and Collocations .’ International Journal of Lexicography   23 . 3 : 257 – 273 .

Dziemianko A.   2011 . ‘Does Dictionary Form Really Matter?’ In Akasu K. , Uchida S. (eds), ASIALEX2011 Proceedings. Lexicography: Theoretical and Practical Perspectives . Kyoto : The Asian Association for Lexicography , 92 – 101 .

Dziemianko A.   2012 a. ‘On the Use(Fulness) of Paper and Electronic Dictionaries’ In Granger S. , Paquot M. (eds), Electronic Lexicography . Oxford : Oxford University Press , 319 – 341 .

Dziemianko A.   2012 b. ‘ Why One and Two Do Not Make Three: Dictionary Form Revisited .’ Lexikos   22 : 195 – 216 .

Dziemianko A.   2014 . ‘ On the Presentation and Placement of Collocations in Monolingual English Learners’ Dictionaries: Insights into Encoding and Retention .’ International Journal of Lexicography   27 . 3 : 259 – 279 .

Dziemianko A.   2017 . ‘ Dictionary Form in Decoding, Encoding and Retention: Further Insights .’ ReCALL   29 . 3 : 1 – 22 .

Ellis R.   1999 . Learning a Second Language through Interaction . Amsterdam : John Benjamins .

Ellis R.   2008 . The Study of Second Language Acquisition (2nd edition.). Oxford : Oxford University Press .

Fan M. Y.   2003 . ‘ Frequency of Use, Perceived Usefulness, and Actual Usefulness of Second Language Vocabulary Strategies: A Study of Hong Kong Learners.’ The Modern   Language Journal   87 . 2 : 222 – 241 .

*Fischer U.   1994 . ‘ Learning Words from Context and Dictionaries: An Experimental Approach .’ Applied Psycholinguistics   15 . 4 : 551 – 574 .

*Frankenberg-Garcia A.   2012 . ‘ Learners’ Use of Corpus Examples .’ International Journal of Lexicography   25 . 3 : 273 – 296 .

*Frankenberg-Garcia A.   2014 . ‘ The Use of Corpus Examples for Language Comprehension and Production .’ ReCALL   26 . 2 : 128 – 146 .

*Frankenberg-Garcia A.   2015 . ‘ Dictionaries and Encoding Examples to Support Language Production .’ International Journal of Lexicography   28 . 4 : 490 – 512 .

Gass S.   1999 . ‘ Discussion: Incidental Vocabulary Learning .’ Studies in Second Language Acquisition   21 . 2 : 319 – 333 .

Grgurović M. , Chapelle C. A. and M. C. Shelley. 2013 . ‘ A Meta-Analysis of Effectiveness Studies onComputer Technology-Supported Language Learning .’  ReCALL   25.2 : 165–198 .

Gu Y. , Johnson R. K. . 1996 . ‘ Vocabulary Learning Strategies and Language Learning Outcomes . ’ Language Learning   46 . 4 : 643 – 679 .

Guillot M. N. , Kenning M.-M. . 1994 . ‘ Electronic Monolingual Dictionaries as Language Learning Aids: A Case Study .’ Computers & Education   23 . 1-2 : 63 – 74 .

Gyllstad H. , Wolter B. . 2016 . ‘ Collocational Processing in Light of the Phraseological Continuum Model: Does Semantic Transparency Matter? ’ Language Learning   66 . 2 : 296 – 323 .

Hartmann R. R. K.   2001 . Teaching and Researching Lexicography . Harlow : Person Education Ltd .

Hedges L. V.   1981 . ‘ Distribution Theory for Glass’s Estimator of Effect Size and Related Estimators .’ Journal of Educational Statistics   6 . 2 : 107 – 128 .

Henriksen B.   1999 . ‘ Three Dimensions of Vocabulary Knowledge .’ Studies in Second Language Acquisition   21 . 2 : 303 – 317 .

*Hill M. , Laufer B. . 2003 . ‘ Type of Task, Time-on-Task and Electronic Dictionaries in Incidental Vocabulary Acquisition .’ IRAL - International Review of Applied Linguistics in Language Teaching   41 . 2 : 87 – 106 .

Hillocks G.   1986 . Research on Written Composition: New Directions for Teaching . Urbana, IL : National Council of Teachers of English .

Huang S. , Willson V. , Eslami Z. . 2012 . ‘ The Effects of Task Involvement Load on L2 Incidental Vocabulary Learning: A Meta‐Analytic Study .’ The Modern Language Journal   96 . 4 : 544 – 557 .

Hulstijn J. H.   2001 . ‘Intentional and Incidental Second-Language Vocabulary Learning: A Reappraisal of Elaboration, Rehearsal and Automaticity’ In Robinson P. (ed.), Cognition and Second Language Instruction . Cambridge : Cambridge University Press , 258 – 286 .

Hulstijn J. H.   2003 . ‘Incidental and Intentional Learning’ In Doughty C. , Long M. H. (eds), The Handbook of Second Language Acquisition . Malden, MA : Blackwell Publishing , 349 – 381 .

Hulstijn J. H. , Laufer B. . 2001 . ‘ Some Empirical Evidence for the Involvement Load Hypothesis in Vocabulary Acquisition .’ Language Learning   51 . 3 : 539 – 558 .

In'nami Y. , Koizumi R. . 2012 . ‘ Database Selection Guidelines for Meta‐Analysis in Applied Linguistics .’ TESOL Quarterly   44 . 1 : 169 – 184 .

*Iu Kam-man V.   2003 . The Impact of Dictionary Use in Four Different Conditions on Incidental Vocabulary Learning . M.A. Thesis, The University of Hong Kong.

Kang E. , Han Z. . 2015 . ‘ The Efficacy of Written Corrective Feedback in Improving L2 Written Accuracy: A Meta‐Analysis .’ The Modern Language Journal   99 . 1 : 1 – 18 .

*Kim S.   2011 . ‘"Minimum Input, Maximum Output, Indeed!" Teaching Collocations through Collocation Dictionary Skills Development .’ Lexikos   27 : 265 – 286 .

*Knight S.   1994 . ‘Dictionary Use While Reading: The Effects on Comprehension and Vocabulary Acquisition for Students of Different Verbal Abilities.’   The Modern Language Journal   78 . 3 : 285 – 299 .

Kobayashi C.   2006 . The Use of Pocket Electronic Dictionaries as Compared with Printed Dictionaries by Japanese Learners of English . Ph.D. Thesis, The Ohio State University.

Komuro Y. , Shitara-Matsuo Y. , Ishii Y. , Uchida S. , Kawamura A. , Kanazashi T. . 2006 . ‘An Analysis of the Oxford Advanced Learner’s Dictionary of Current English, Seventh Edition, with Special Reference to the CD-ROM.’   Lexicon   36 : 55 – 146 .

Kosem I. , Lew R. , Müller-Spitzer C. , Ribeiro Silveira M. , Wolfer S. , Dorn A. , Gurrutxaga A. , Ceberio K. , Etxeberria E. , Lefer M.-A. , Geeraerts D. , Štrkalj Despot K. , Stojanov T. , Ljubešić N. , Škrabal M. , Štěpánková B. , Vodrážková V. , Lorentzen H. , Trap-Jensen L. , Kallas J. , Tuulik M. , Koppel K. , Langemets M. , Heinonen T. , Thomas I. , Margilitadze T. , Markantonatou S. , Giouli V. , Mulhall C. , Kernerman I. , Ben-Moshe Y. , Sadan T. , Abel A. , Nied Curcio M. , Tanturovska L. , Nikovska B. , Tiberius C. , Grønvik O. , Hovdenak M. , Berg-Olsen S. , Karlsen K. E. , Smith Ore C.-E. , Biesaga M. , Zingano Kuhn T. , Silvestre J. , Isabelle Tamba E. , Haja G. , Clim M.-R. , Patrascu M.-I. , Tasovac T. , Petrović S. , Arhar Holdt Š. , Valcarcel Riveiro C. , Domínguez Vázquez M. J. , Volodina E. , Pilán I. , Sköldberg E. , Holmer L. , Nesi H. . 2019 . ‘ The Image of the Monolingual Dictionary across Europe. Results of the European Survey of Dictionary Use and Culture .’ International Journal of Lexicography   32 . 1 : 92 – 114 .

*Koyama T. , Takeuchi O. . 2003 . ‘ Printed Dictionaries vs. Electronic Dictionaries: A Pilot Study on How Japanese EFL Learners Differ in Using Dictionaries . ’ Language Education & Technology   40 : 61 – 79 .

Koyama T. , Takeuchi O. . 2004 . ‘Comparing Electronic and Printed Dictionaries: How the Difference Affected EFL Learning . ’ JACET Bulletin   38 : 33 – 46 .

Laufer B.   2003 . ‘ Vocabulary Acquisition in a Second Language: Do Learners Really Acquire Most Vocabulary by Reading? Some Empirical Evidence .’ Canadian Modern Language Review   59 . 4 : 567 – 587 .

*Laufer B.   2011 . ‘ The Contribution of Dictionary Use to the Production and Retention of Collocations in a Second Language .’ International Journal of Lexicography   24 . 1 : 29 – 49 .

Laufer B. , Elder C. , Hill K. , Congdon P. . 2004 . ‘ Size and Strength: Do We Need Both to Measure Vocabulary Knowledge? ’ Language Testing   21 . 2 : 202 – 226 .

Laufer B. , Girsai N. . 2008 . ‘ Form-Focused Instruction in Second Language Vocabulary Learning: A Case for Contrastive Analysis and Translation .’ Applied Linguistics   29 . 4 : 694 – 716 .

Laufer B. , Goldstein Z. . 2004 . ‘ Testing Vocabulary Knowledge: Size, Strength, and Computer Adaptiveness .’ Language Learning   54 . 3 : 399 – 436 .

*Laufer B. , Hadar L. . 1997 . ‘ Assessing the Effectiveness of Monolingual, Bilingual, and “Bilingualised” Dictionaries in the Comprehension and Production of New Words.’   The Modern Language Journal   81 . 2 : 189 – 196 .

*Laufer B. , Hill M. . 2000 . ‘ What Lexical Information Do L2 Learners Select in a Call Dictionary and How Does It Affect Word Retention? ’ Language Learning & Technology   3 . 2 : 58 – 76 .

Laufer B. , Hulstijn J. . 2001 . ‘ Incidental Vocabulary Acquisition in a Second Language: The Construct of Task-Induced Involvement .’ Applied Linguistics   22 . 1 : 1 – 26 .

*Laufer B. , Levitzky-Aviad T. . 2006 . ‘ Examining the Effectiveness of “Bilingual Dictionary Plus” – A Dictionary for Production in a Foreign Language .’ International Journal of Lexicography   19 . 2 : 135 – 155 .

Lee H. , Warschauer M. , Lee J. H. . 2019 . ‘ The Effects of Corpus Use on Second Language Vocabulary Learning: A Multilevel Meta-Analysis .’ Applied Linguistics   40 . 5 : 721 – 753 .

Lee J. , Jang J. and L. Plonsky.  2015 . ‘ The Effectiveness of Second Language Pronunciation Instruction: A Meta-Analysis .’  Applied Linguistics   36.3 :  345–366 .

*Lee S.   2001 . The Effect of Topic Interest on Dictionary Use and Its Effect on Vocabulary Learning: The Case of Korean EFL Learners . Ph.D. Thesis, Southern Illinois University at Carbondale.

*Lee S. J.   2008 . A Comparison of the Effects of Bilingualised Dictionary and No Dictionary Use on Reading Comprehension and Vocabulary Learning of Hong Kong Lower Secondary Students in a Chinese Medium of Instruction School . M.A. Thesis, The University of Hong Kong.

Lew R.   2002 . ‘A Study in the Use of Bilingual and Monolingual Dictionaries by Polish Learners of English: A Preliminary Report’ In Braasch A. , Povlseh C. (eds), Proceedings of the Tenth EURALEX International Congress, EURALEX 2002, Copenhagen, Denmark, August 13-17, 2002 . Copenhagen : Centre for Sprogteknologi , 759 – 763 .

Lew R.   2011 . ‘ Studies in Dictionary Use: Recent Developments .’ International Journal of Lexicography   24 . 1 : 1 – 4 .

Lew R.   2015 . ‘ Research into the Use of Online Dictionaries .’ International Journal of Lexicography   28 . 2 : 232 – 253 .

Lew R.   2019 . ‘Introduction to the Virtual Issue Dictionaries and Language Teaching.’   International Journal of Lexicography . Accessed on 20 April 2019. https://academic.oup.com/ijl/pages/dictionaries_and_language_teaching

*Lew R. , Doroszewska J. . 2009 . ‘ Electronic Dictionary Entries with Animated Pictures: Lookup Preferences and Word Retention .’ International Journal of Lexicography   22 . 3 : 239 – 257 .

*Li L. , Xu H. . 2015 . ‘ Using an Online Dictionary for Identifying the Meanings of Verb Phrases by Chinese EFL Learners .’ Lexikos   25 : 191 – 209 .

Li S. , Shintanani N. , Ellis R. . 2012 . ‘ Doing Meta-Analysis in SLA: Practices,   Choices, and Standards.’ Contemporary Foreign Language Studies   384 . 12 : 1 – 17 .

Lipsey M. W. , Wilson D. B. . 2001 . Practical Meta-Analysis . Thousand Oaks, CA : SAGE .

*Liu T.-C. , Fan M. H.-M. , Paas F. . 2014 . ‘ Effects of Digital Dictionary Format on Incidental Acquisition of Spelling Knowledge and Cognitive Load During Second Language Learning: Click-on vs. Key-in Dictionaries .’ Computers & Education   70 : 9 – 20 .

*Liu T.-C. , Lin P.-H. . 2011 . ‘ What Comes with Technological Convenience? Exploring the Behaviors and Performances of Learning with Computer-Mediated Dictionaries .’ Computers in Human Behavior   27 . 1 : 373 – 383 .

*Liu O.   2007 . Research on the Relationship between Lexical Information L2 Learners Select in CALL Dictionaries and Word Retention . M.A. Thesis, Chongqing University.

*Luppescu S. , Day R. R. . 1993 . ‘ Reading, Dictionaries, and Vocabulary Learning .’ Language Learning   43 . 2 : 263 – 279 .

*Ma J. H. , Cheon H. J. . 2018 . ‘An Experimental Study of Dictionary Use on Vocabulary Learning and Reading Comprehension in Different Task Conditions . ’ International Journal of Lexicography   31 . 1 : 29 – 52 .

Mahdi H. S.   2017 . ‘ Effectiveness of Mobile Devices on Vocabulary Learning: A Meta-Analysis .’ Journal of Educational Computing Research   56 . 1 : 134 – 154 .

Margarita A.-R. , Marcos G. S. . 2019 . ‘ Testing the Use of a Collocation Retrieval Tool without Prior Training by Learners of Spanish . ’ International Journal of Lexicography   32 . 4 : 480 – 497 .

Meara P. , Wolter B. . 2004 . ‘V_Links: Beyond Vocabulary Depth’ In Albrechtsen D. , Haastrup K. and Henriksen B. (eds), Angles on the English-Speaking World . Copenhagen : Museum Tusculanum Press , 85 – 96 .

Montero Perez M. , Peters E. , Clarebout G. , Desmet P. . 2014 . ‘ Effects of Captioning on Video Comprehension and Incidental Vocabulary Learning .’ Language Learning & Technology   18 . 1 : 118 – 141 .

Morris S. B. , Deshon R. P. . 2002 . ‘ Combining Effect Size Estimates in Meta-Analysis with Repeated Measures and Independent-Groups Designs .’ Psychological Methods   7 . 1 : 105 – 125 .

Müller-Spitzer C.   2014 . ‘Introduction’ In Müller-Spitzer C. (ed.), Using Online Dictionaries. (Lexicographica Series Maior 145 .). Berlin : Walter de Gruyter , 1 – 10 .

Nation I. S. P.   2001 . Learning Vocabulary in Another Language . Cambridge : Cambridge University Press .

Nation I. S. P. , Webb S. . 2011 . Researching and Analyzing Vocabulary . Boston : Heinle .

*Nesi H.   1996 . ‘ The Role of Illustrative Examples in Productive Dictionary Use .’ Dictionaries: Journal of the Dictionary Society of North America   17 . 1 : 198 – 206 .

Nesi H.   2000 a. ‘Electronic Dictionaries in Second Language Vocabulary Comprehension and Acquisition: The State of the Art’ In Heid U. , Evert S. , Lehmann E. , Rohrer C. (eds), Proceedings of the Ninth EURALEX International Congress, EURALEX 2000, Stuttgart, Germany, August 8th-12th, 2000 . Stuttgart : Institut für Maschinelle Sprachverarbeitung, Universität Stuttgart , 839 – 847 .

Nesi H.   2000 b. ‘On Screen or in Print? Students’ Use of a Learner’s Dictionary on CD-ROM and in Book Form’ In Howarth P. , Herington R. (eds), EAP Learning Technologies . Leeds : Leeds University Press , 106 – 114 .

*Nesi H. , Tan K. H. . 2011 . ‘ The Effect of Menus and Signposting on the Speed and Accuracy of Sense Selection .’ International Journal of Lexicography   24 . 1 : 79 – 96 .

Norris J. M. , Ortega L. . 2006 . ‘The Value and Practice of Research Synthesis for Language Learning and Teaching’ In Norris J. M. , Ortega L. (eds), Synthesizing Research on Language Learning and Teaching . Philadelphia, PA : John Benjamins , 3 – 50 .

*Nurmukhamedov U.   2017 . ‘ The Contribution of Collocation Tools to Collocation Correction in Second Language Writing . ’ International Journal of Lexicography   30 . 4 : 454 – 482 .

Ortega L.   2009 . Understanding Second Language Acquisition . Oxford : Oxford University Press .

Oswald F. L. , Plonsky L. . 2010 . ‘Meta-Analysis in Second Language Research: Choices and Challenges .’ Annual Review of Applied Linguistics   30 : 85 – 110 .

*Peters E.   2007 . ‘ Manipulating L2 Learners' Online Dictionary Use and Its Effect on L2 Word Retention .’ Language Learning & Technology   11 . 2 : 36 – 58 .

Peters E.   2014 . ‘ The Effects of Repetition and Time of Post-Test Administration on EFL Learners’ Form Recall of Single Words and Collocations.’ Language   Teaching Research   18 . 1 : 75 – 94 .

Peters E.   2016 . ‘ The Learning Burden of Collocations: The Role of Interlexical and Intralexical Factors.’   Language Teaching Research   20 . 1 : 113 – 138 .

Peters E. , Heynen E. , Puimège E. . 2016 . ‘ Learning Vocabulary through Audiovisual Input: The Differential Effect of L1 Subtitles and Captions .’ System   63 : 134 – 148 .

Peters E. , Hulstijn J. H. , Sercu L. , Lutjeharms M. . 2009 . ‘ Learning L2 German Vocabulary through Reading: The Effect of Three Enhancement Techniques Compared .’ Language Learning   59 . 1 : 113 – 151 .

Peters E. , Webb S. . 2018 . ‘ Incidental Vocabulary Acquisition through Viewing L2 Television and Factors That Affect Learning .’ Studies in Second Language Acquisition   40 . 3 : 1 – 27 .

Piotrowski T.   1989 . ‘Monolingual and Bilingual Dictionaries: Fundamental Differences’ In Tickoo M. (ed.), Learners' Dictionaries: State of the Art . Singapore : SEAMEO Regional Language Center , 72 – 83 .

Plonsky L.   2011 . ‘ The Effectiveness of Second Language Strategy Instruction: A Meta‐Analysis .’ Language Learning   61 . 4 : 993 – 1038 .

Plonsky L. , Derrick D. J. . 2016 . ‘ A Meta-Analysis of Reliability Coefficients in Second Language Research.’ The Modern   Language Journal   100 . 2 : 538 – 553 .

Plonsky L. , Oswald F. L. . 2012 . ‘How to Do a Meta-Analysis ’ In Mackey A. , Gass S. M. (eds), Research Methods in Second Language Acquisition: A Practical Guide   West Sussex, UK : Blackwell , 275 – 295 .

Plonsky L. , Oswald F. L. . 2014 . ‘ How Big Is “Big”? Interpreting Effect Sizes in L2 Research .’ Language Learning   64 . 4 : 878 – 912 .

Plonsky L. , Oswald F. L. . 2015 . ‘Meta-Analyzing Second Language Research’ In Plonsky L. (ed.), Advancing Quantitative Methods in Second Language Research . New York and London : Routledge , 106 – 128 .

Qian D. D.   1999 . ‘ Assessing the Roles of Depth and Breadth of Vocabulary Knowledge in Reading Comprehension .’ Canadian Modern Language Review   56 . 2 : 282 – 308 .

Qian D. D.   2002 . ‘ Investigating the Relationship between Vocabulary Knowledge and Academic Reading Performance: An Assessment Perspective .’ Language Learning   52 . 3 : 513 – 536 .

Read J.   1993 . ‘ The Development of a New Measure of L2 Vocabulary Knowledge .’ Language Testing   10 . 3 : 355 – 371 .

Read J.   2000 . Assessing Vocabulary . Cambridge : Cambridge University Press .

Roby W. B.   1991 . Glosses and Dictionaries in Paper and Computer Formats as Adjunct Aids to the Reading of Spanish Texts by University Students . Ph.D. Thesis, University of Kansas.

Ronald J.   2003 a. ‘ A Review of Research into Vocabulary Acquisition through Dictionary Use. Part 1: Intentional Vocabulary Learning through Dictionary Use . ’ Studies in the Humanities and Sciences   44 . 1 : 285 – 307 .

Ronald J.   2003 b. ‘ A Review of Research into Vocabulary Acquisition through Dictionary Use. Part 2: Incidental Vocabulary Learning through Dictionary Use . ’ Studies in the Humanities and Sciences   44 . 2 : 67 – 97 .

Rosenthal R.   1991 . Meta-Analytic Procedures for Social Research . Newbury Park, CA : Sage Publications .

Schmitt N. (ed.). 2004 . Formulaic Sequences: Acquisition, Processing and Use . Amsterdam/Philadelphia : John Benjamins .

Schmitt N.   2008 . ‘ Instructed Second Language Vocabulary Learning .’ Language Teaching Research   12 . 3 : 329 – 363 .

Schmitt N.   2010 . Researching Vocabulary: A Vocabulary Research Manual . Basingstoke, UK : Palgrave Macmillan .

Schmitt N.   2014 . ‘ Size and Depth of Vocabulary Knowledge: What the Research Shows .’ Language Learning   64 . 4 : 913 – 951 .

Sharpe P.   1995 . ‘ Electronic Dictionaries with Particular Reference to the Design of an Electronic Bilingual Dictionary for English-Speaking Learners of Japanese .’ International Journal of Lexicography   8 . 1 : 39 – 54 .

*Shu Z.   2015 . The Effects of Dictionary Use During Reading on L2 Vocabulary Learning: A Small-Scale Empirical Study . M.A. Thesis, Nanjing University.

*Szczepaniak R. , Lew R. . 2011 . ‘ The Role of Imagery in Dictionaries of Idioms .’ Applied Linguistics   32 . 3 : 323 – 347 .

*Tono Y.   1992 . ‘ Effect of Menus on EFL Learners' Look-up Processes .’ Lexikos   2 : 230 – 252 .

Tono Y.   2000 . ‘On the Effects of Different Types of Electronic Dictionary Interfaces on L2 Learners’ Reference Behaviour in Productive/Receptive Tasks’. In Heid U. , Evert S. , Lehmann E. and Rohrer C. (eds), Proceedings of the Ninth EURALEX International Congress, EURALEX 2000, Stuttgart, Germany, August 8th-12th, 2000 . Stuttgart : Institut für Maschinelle Sprachverarbeitung, Universität Stuttgart , 855 – 861 .

Tsai Y. L. , Tsai C. C. . 2018 . ‘ Digital Game-Based Second-Language Vocabulary Learning and Conditions of Research Designs: A Meta-Analysis Study .’ Computers & Education   125 : 345 – 357 .

Uchihara T. , Webb S. , Yanagisawa A. . 2019 . ‘ The Effects of Repetition on Incidental Vocabulary Learning: A Meta-Analysis of Correlational Studies .’ Language Learning   69 . 3 : 559 – 599 .

Vacha-Haase T. , Thompson B. . 2004 . ‘ How to Estimate and Interpret Various Effect Sizes .’ Journal of Counseling Psychology   51 . 4 : 473 – 481 .

Van Zeeland H. , Schmitt N. . 2013 . ‘ Incidental Vocabulary Acquisition through L2 Listening: A Dimensions Approach .’ System   41 . 3 : 609 – 624 .

Webb S.   2002 . Investigating the Effects of Learning Tasks on Vocabulary Knowledge . Ph.D. Thesis, Victoria University of Wellington.

Webb S.   2005 . ‘ Receptive and Productive Vocabulary Learning: The Effects of Reading and Writing on Word Knowledge .’ Studies in Second Language Acquisition   27 . 1 : 33 – 52 .

Webb S. , Nation I. S. P. . 2017 . How Vocabulary Is Learned . Oxford : Oxford University Press .

Welker H. A.   2010 . Dictionary Use: A General Survey of Empirical Findings . Brasília : Author's Edition. Accessed on 20 November 2010. http://www.let.unb.br/hawelker/ dictionary_use_research.pdf .

Wray A.   2002 . Formulaic Language and the Lexicon . Cambridge : Cambridge University Press .

Wray A.   2008 . Formulaic Language: Pushing the Boundaries . Oxford : Oxford University Press .

Yun J. W.   2011 . ‘ The Effects of Hypertext Glosses on L2 Vocabulary Acquisition: A Meta-Analysis .’ Computer Assisted Language Learning   24 . 1 : 39 – 58 .

*Zou D.   2016 . ‘ Comparing Dictionary-Induced Vocabulary Learning and Inferencing in the Context of Reading .’ Lexikos   26 : 372 – 390 .

Supplementary data

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1477-4577
• Print ISSN 0950-3846
• Copyright © 2024 Oxford University Press
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• The Oxford English Dictionary
• OED Researchers Advisory Group
• The OED Community
• Research sub
• Language Datasets
• Indian Language Datasets
• Oxford Dictionaries API
• Lexical Data for NLP
• Dictionary Solutions
• Pronunciations Data
• Educational Assessment Platforms
• Assisted Writing Solutions
• Gamified Learning
• Semantic English Language Database

Dictionaries

• Dictionaries for your products

Dictionary Apps

• Oxford Dictionaries Premium
• For English Learners
• For Children
• Creating Content
• Word of the Year
• Branding resources
• Case Studies
• Our dictionary data

A number of Oxford’s dictionaries are available as mobile apps, covering multiple languages and areas of language learning.

Among these is the Oxford Dictionary of English, the leading English dictionary mobile app on the market, and Oxford Dictionaries of French, Chinese, German, Urdu, and Russian.

A number of Oxford thesauri are available for download, offering unparalleled access to synonyms and example sentences.

Specialist dictionaries such as the Oxford dictionaries of Etymology, Finance, Geology, and more, offer insight into specific fields. The apps offer an immersive learning experience with features such as audio pronunciations, offline mode for easy access at all times, and a Word of the day feature to expand your vocabulary in everyday life.

Enjoy the wealth of linguistic knowledge in app form with just one click!

View all Oxford dictionary apps.

• Dictionaries home
• American English
• Collocations
• German-English
• Grammar home
• Practical English Usage
• Learn & Practise Grammar (Beta)
• Word Lists home
• My Word Lists
• Recent additions
• Resources home
• Text Checker

Definition of study noun from the Oxford Advanced Learner's Dictionary

activity of learning

• a room set aside for private study
• academic/literary/scientific study
• It is important to develop good study skills .
• Physiology is the study of how living things work.
• distance learning
• further education
• higher education
• qualification
• Now that her children are all at school, she's going to take up full-time study again.
• Students do a foundation year before specializing in their chosen field of study.
• The course integrates academic study and practical training.
• This grammar book is suitable both for classroom use and for independent study.
• The scientific study of American dialects began in 1889.
• The company allows its staff to take paid study leave.
• A study group meets every Sunday at the church.
• This work is the result of decades of study and research.
• Students have great opportunities through study abroad and internship programs.
• programme/​program
• an area of study
• a field of study

Join our community to access the latest language learning and assessment tips from Oxford University Press!

• to continue your studies
• Many undertake further studies after college.
• She returned to her studies when her children reached school age.
• When he has completed his studies, he'll travel around the world.
• He needed more leisure to pursue his studies.

academic subject

• business/media/American studies
• He lectures in management studies.
• She's doing women's studies at Liverpool University.
• the university's cultural studies course
• to conduct/undertake a study
• Research studies carried out in Italy confirmed the theory.
• study on something He has recently done a study on the effects of traffic and aircraft noise.
• study of something a detailed study of how animals adapt to their environment
• scientific studies of fishing grounds and methods of fishing
• This study shows/finds/suggests that…
• the purpose/aim of a study
• the results/findings of a study
• in a study the methods used in the present study
• in previous/earlier studies
• according to a study According to a recent study published in the Journal of Sleep Research…
• formulate/​advance a theory/​hypothesis
• build/​construct/​create/​develop a simple/​theoretical/​mathematical model
• develop/​establish/​provide/​use a theoretical/​conceptual framework
• advance/​argue/​develop the thesis that…
• explore an idea/​a concept/​a hypothesis
• make a prediction/​an inference
• base a prediction/​your calculations on something
• investigate/​evaluate/​accept/​challenge/​reject a theory/​hypothesis/​model
• design an experiment/​a questionnaire/​a study/​a test
• do research/​an experiment/​an analysis
• make observations/​measurements/​calculations
• carry out/​conduct/​perform an experiment/​a test/​a longitudinal study/​observations/​clinical trials
• run an experiment/​a simulation/​clinical trials
• repeat an experiment/​a test/​an analysis
• replicate a study/​the results/​the findings
• observe/​study/​examine/​investigate/​assess a pattern/​a process/​a behaviour
• fund/​support the research/​project/​study
• seek/​provide/​get/​secure funding for research
• collect/​gather/​extract data/​information
• yield data/​evidence/​similar findings/​the same results
• analyse/​examine the data/​soil samples/​a specimen
• consider/​compare/​interpret the results/​findings
• fit the data/​model
• confirm/​support/​verify a prediction/​a hypothesis/​the results/​the findings
• prove a conjecture/​hypothesis/​theorem
• draw/​make/​reach the same conclusions
• read/​review the records/​literature
• describe/​report an experiment/​a study
• present/​publish/​summarize the results/​findings
• present/​publish/​read/​review/​cite a paper in a scientific journal
• A new study shows that fewer students are studying science.
• A preliminary study suggested that the product would be popular.
• An independent study was commissioned by the department.
• He has made a special study of the way that birds communicate with one another.
• In a recent study, 40% of schools were found to be understaffed.
• The company undertook an extensive feasibility study before adopting the new system.
• The present study reveals an unacceptable level of air pollution in the city centre.
• The study aims to examine bias in television news coverage.
• The study compares the incidence of bone cancer in men and women.
• The study group was selected from a broad cross section of the population.
• The study highlighted three problem areas.
• The study provided valuable insight into the development of the disease.
• The ten-year study covered 13 000 people aged 15-25.
• This phenomenon has been observed in both laboratory and field studies.
• A comparative study was carried out into the environmental costs of different energy sources.
• They undertook a study of a sample of 992 first year students in 13 secondary schools.
• Most previous studies have focused on the elderly.
• Clinical studies show that adding calcium to your diet can help people lose weight.
• Animal studies have linked the chemical to certain cancers.
• According to the study, respondents preferred the strawberry-flavoured drink.
• Studies conducted in Iowa have produced similar results.
• In a study of 400 patients, half of those given the medication saw the number of their headaches halved.
• Study participants developed no major side effects.
• aim at something
• aim to do something
• according to a/​the study
• in a/​the study
• under study
• the author of a/​the study
• a study of Jane Austen's novels
• Shakespeare is the subject of a new study by Anthony Bryan.
• These proposals deserve careful study.
• under study the biochemical process under study
• She crossed the hallway and opened the door to her private study.
• a study of Chartres Cathedral
• a nude study
• The novel is a character study of a city and its people.
• (also étude ) [countable] a piece of music designed to give a player practice in technical skills Topics Music c2

perfect example

• His face was a study in concentration.
• (old-fashioned, British English) thinking deeply so that you do not notice what is happening around you

Other results

• the Institute for Advanced Study
• Institute for Advanced Study
• in a brown study

Nearby words

Advertisement

Chaos and Confusion: Tech Outage Causes Disruptions Worldwide

Airlines, hospitals and people’s computers were affected after CrowdStrike, a cybersecurity company, sent out a flawed software update.

• Share full article

By Adam Satariano Paul Mozur Kate Conger and Sheera Frenkel

• July 19, 2024

Airlines grounded flights. Operators of 911 lines could not respond to emergencies. Hospitals canceled surgeries. Retailers closed for the day. And the actions all traced back to a batch of bad computer code.

A flawed software update sent out by a little-known cybersecurity company caused chaos and disruption around the world on Friday. The company, CrowdStrike , based in Austin, Texas, makes software used by multinational corporations, government agencies and scores of other organizations to protect against hackers and online intruders.

But when CrowdStrike sent its update on Thursday to its customers that run Microsoft Windows software, computers began to crash.

The fallout, which was immediate and inescapable, highlighted the brittleness of global technology infrastructure. The world has become reliant on Microsoft and a handful of cybersecurity firms like CrowdStrike. So when a single flawed piece of software is released over the internet, it can almost instantly damage countless companies and organizations that depend on the technology as part of everyday business.

“This is a very, very uncomfortable illustration of the fragility of the world’s core internet infrastructure,” said Ciaran Martin, the former chief executive of Britain’s National Cyber Security Center and a professor at the Blavatnik School of Government at Oxford University.

A cyberattack did not cause the widespread outage, but the effects on Friday showed how devastating the damage can be when a main artery of the global technology system is disrupted. It raised broader questions about CrowdStrike’s testing processes and what repercussions such software firms should face when flaws in their code cause major disruptions.

How a Software Update Crashed Computers Around the World

Here’s a visual explanation for how a faulty software update crippled machines.

How the airline cancellations rippled around the world (and across time zones)

Share of canceled flights at 25 airports on Friday

50% of flights

Ai r po r t

Bengalu r u K empeg o wda

Dhaka Shahjalal

Minneapolis-Saint P aul

Stuttga r t

Melbou r ne

Be r lin B r anden b urg

London City

Amsterdam Schiphol

Chicago O'Hare

Raleigh−Durham

B r adl e y

Cha r lotte

Reagan National

Philadelphia

1:20 a.m. ET

CrowdStrike’s stock price so far this year

We are having trouble retrieving the article content.

Please enable JavaScript in your browser settings.

Thank you for your patience while we verify access. If you are in Reader mode please exit and  log into  your Times account, or  subscribe  for all of The Times.

Thank you for your patience while we verify access.

Already a subscriber?  Log in .

Want all of The Times?  Subscribe .

Words and phrases

Personal account.

• Access or purchase personal subscriptions
• Get our newsletter
• Save searches
• Set display preferences

Institutional access

Sign in with library card

Sign in with username / password

Recommend to your librarian

Institutional account management

Sign in as administrator on Oxford Academic

Information

Many of the OED blog posts have been moved to our Discoverability hubs. In these hubs, you will find overviews on the history of English, commentaries on the etymology and semantic development of words, World Englishes resources, and a series of word lists on a wide range of topics. Explore all of our Discoverability hubs below.

History of English

Embark on a journey through the English language in our History of English hub , charting the history of the language from Old English to the present day.

Word Stories

Explore the etymology and semantic development of words such as taffety tarts, dungarees, and codswallop in our Word Stories hub .

World Englishes

Our World Englishes hub is home to our content and resources on World Englishes in the  OED  including webinars, teaching resources, pronunciation, and more!

In our Word Lists hub , you will find a collection of articles and correlated word lists on an array of fascinating topics, from film-based coinages and Christmassy words to place names and Tex-Mex terms.

Commentaries

In our Commentaries hub , you will find commentaries on the OED quarterly updates, including new words notes, articles on etymology, and overviews on World English additions.

Browse information