Change search
ReferencesLink to record
Permanent link

Direct link
Reading Success and Vocabulary Knowledge among advanced professionals with English as their second language (L2): A comparative study of Russian and Swedish medical professionals in Sweden
Stockholm University, Faculty of Humanities, Department of English.
2009 (English)Independent thesis Basic level (degree of Bachelor), 80 credits / 120 HE creditsStudent thesis
Abstract [en]


It is vital you have advanced knowledge in English as a second language (L2) if you work and/or research as a medical specialist at the Swedish academic university hospitals in Sweden. Otherwise it will be impossible to communicate within your area of interest, either orally or in writing, or even by means of reading, not only internationally but also between co-workers. All communication between academic professionals from different countries as well as textbooks, articles, instructions, lectures and exchange of information are in English.

Interviews and tests for this essay were made with advanced medical researchers and specialists at Karolinska Institutet and Karolinska University Hospital in Stockholm, Sweden.


The aim of this research is to find out how closely estimated vocabulary size is related to successful reading.

There has not been much research on advanced academic intellectuals when it comes to advanced reading comprehension. Some studies on children and younger adults have shown that reading proficiency is based on the size of their vocabulary, the bigger vocabulary they have the higher their reading proficiency, while other studies have shown that the more they read the more they understand and automatically their vocabulary increases. Are these the only reasons for their reading proficiency, or are there other aspects involved? Do their vocabulary sizes affect their understanding when they read? Or what other reasons help reading and comprehending the text?

In comparing readers’ understanding of different domain-specific texts, it can be hypothesized that there are differences in comprehension between general and more specialized texts. It can also be thought that it is easier for Swedes to read and comprehend English as the same alphabet is used in English and Swedish, than for Russians, who are used to a different language structure with other typographical factors.


Factors that Can Affect Reading Success

- The Linguistic Threshold Hypothesis

- Background knowledge

- Prior knowledge

- Interest and attitude

General Questions

1. How important is a L2 reader’s vocabulary size to his/her understanding of a text? Could other factors be more important?

2. Is it possible for professionals with poor linguistic proficiency in English to read and understand domain specific texts due to their expert knowledge?

3. Is professionals’ receptive proficiency similar within their domain specific areas and in general areas?

4. Is there an effect of having lived in an English-speaking country on reading comprehension or vocabulary size?


1. The L1 typology would make a difference.

2. The Swedes would have a better vocabulary and a better general comprehension of general English, and thereby a better understanding of the domain-specific texts as well.


Professionals often have English as their L2. They are assumed to read English texts as efficiently as they read texts in their L1. In this essay the focus was set on their reading ability and their vocabulary knowledge. Vocabulary was assessed as academic or infrequent, and reading was assessed by means of tests of word recognition, single sentence comprehension, evaluation of logical arguments based on two sentence recognitions, and reading domain-specific, medical articles as well as general articles.

Data was collected from medical researchers and specialists during this empirical research. The specialists were presumed to read English journal articles on a daily basis in their area of expertise, medicine, and use English as their working language. Seven Russians and seven Swedes were interviewed between February and April 2007. They had chosen the date, time and place for their interviews.

The testing took about 90 minutes and was divided into


Each subject was asked to fill out a questionnaire about his/her educational and social background. This data could not be used in this essay, but would be interesting to use for research about the impact of social background on the proficiency of English as a L2.

Then they were asked to continue with paper diagnostic tests, the Proficiency Test, testing their syntax skills and vocabulary size. The Proficiency Test is a part of DIALANG, a self-assessment test developed by the Project of the European Commission for use on the Internet, and it was used along with a test of academic vocabulary, and a test of infrequent vocabulary, divided into synonyms and antonyms.

The computerized test, the SuperLab test, was divided into two sections. For the first section, Reading Proficiency, articles within the medical domain as well as various general domains were read and then retold orally. The retold stories were taped and the language used when retelling the story, L1 or L2, was decided by the subject, so they would feel comfortable when speaking. These interviews were transcribed and analyzed. The second part of the computerized sub skills, Reading Comprehension, consisted of word recognition and reading comprehension of one and more sentences.


The Proficiency Test

The Proficiency test tested vocabulary knowledge. Overall most of the subjects thought Part 2 Synonyms and Part 3 Antonyms were the most difficult parts of the whole test. Although they knew other languages, it was not possible to use that knowledge in this part of the test. One Swede thought the Part 3 Antonyms was the most difficult. Some words were recognised from earlier, from looking them up in dictionaries several times and then forgetting them again. Sometimes no alternative seemed to be correct. Another Swede commented that it was not obvious which alternative was the correct answer. Sometimes the synonym or antonym could be explained in his own words, but not with any of the alternatives. A third Swede commented that in reading a regular text you do not have to know every single word to get coherence.

The SuperLab Test

The computerized reading test, the SuperLab test, was to show reading comprehension of texts, sentences and word recognition. Since sentence comprehension did not correlate well with vocabulary size here, these results suggest that the size of the vocabulary above a certain level does not have an immediate impact on the logical judgement or comprehension of sentences At the same time it has to be remembered that the subjects are advanced academic professionals, and might not only use their vocabulary knowledge and size in a L2 when it comes to making conclusions and decisions.

Reading Comprehension - Analysis

The Reading Comprehension part of the test consisted of six different texts: three general texts and three medical texts. For each text there were a number of propositions that had to be correctly recalled to get points. The first text was on paper and the last five read and timed on the computer. It has to be kept in mind that the text read on paper, Text 1, presented the subjects with an opportunity to read as they usually do. Most of them seemed to read the title, skimmed the text and then read and reread the interesting or missed parts several times, until the message was clear. When the other five texts were read on the computer screen on the other hand, only one sentence was shown at a time, and disappeared when the next sentence was keyed for. This computerized way of reading the text forced the reader to remember what was read after each sentence, without knowing what came next. So, the ordinary way of reading a text was changed.

Reading Comprehension - Results

A conclusion of all propositions indicated that the Russians as a group had higher results than the Swedes. When looking at the individual answers a pattern can be shown. The Swedes on an average were a more homogenous group with a smaller span between the top result and the lowest. The Russians were a more heterogeneous group.

In this essay the focus has been on the reading quality. The results showed that there are other circumstances than simple language and reading proficiency to be aware of. There is a difference in their background knowledge. How much of their ‘recall’ do they actually remember from the text and how much do they know from beforehand?

For example, the Reading Comprehension Test with the retelling of the five texts read in an unnatural way did not only test how much the subjects understood and recalled, but also how much they remembered. While reading the texts on the computer, one sentence at a time, the subjects had to focus on the meaning of each sentence in the text, memorize it and then read the next sentence. For the three first texts there was no demand for any previous knowledge to be able to read and understand the words and sentences separately. The subjects did not use any dictionaries, which they seldom do in their ordinary, everyday work either.

The results also show that interest and background knowledge were important factors to be able to understand the whole contents of each text. The meaning of the text was more easily understood the more background knowledge the subjects had.


Hypothesis 1 - The L1 typology would make a difference.

The tests did not show that the language background made a difference. The Russians lived in Sweden and spoke Russian among themselves but English with everybody else. They were well-educated academics, not only proficient in English within their own area, medical texts, but also very good at understanding general texts. When it came to reading comprehension the background knowledge seemed to be more important than the vocabulary skills.

Hypothesis 2 - The Swedes would have a better vocabulary and a better general comprehension of general English, and thereby a better understanding of the domain-specific texts as well.

It was not indicated at any time during the research for this essay that the Swedes had a better vocabulary and thereby would understand both general and domain-specific texts better. The understanding and comprehension seemed to be more based on attitude and background knowledge of the different texts.

Tsui and Fullilove (1998) found in their studies that it was extremely difficult to process information which contradicted what they already knew. This particularly showed during this essay, with some of the Russians questioning and discussing the contents of the medical texts, exactly as Steffensen and Joag-Dev (1979) found in their study. The authors stated that readers seemed to dismiss information they found unimportant, add information they thought should be there, and focus on what they found important, all based on their world view and their opinion.

Most of the Russians seemed to be more involved and questioning than the Swedes in this investigation. This study showed that the more background knowledge the subjects had, the more they understood the texts.

During the testing he Swedes made a point of the difficulties of reading the texts on the computers, which did not agree with their reading strategies. These Swedes seemed to attempt to use more normal strategies when reading the texts on the computer screen than the Russians. The Russians appeared to read more slowly overall. Slow readers usually had difficulties putting things together towards the end of sentences and paragraphs. Probably the Russians used more background information than English reading proficiency and understanding when doing the test.

Vocabulary Size and Reading Success

During the work with this essay it was not proven that vocabulary size was directly linked to reading proficiency. The readers tested were advanced academic professionals, and might not only use their vocabulary but also their background knowledge when understanding the content of the texts used. The Russians were not a random selection, because those who participated were more confident with their English knowledge than those that did not participate.


There were of course some limitations during this study. One was fatigue among the subjects due to their hectic working environment and the long duration of the tests. Another issue was that most of the texts were on the screen, to be read sentence by sentence, and hence unnaturally read.

Out of the results of these tests the conclusion could be drawn that even if you have a larger vocabulary, it would not necessarily mean you understand more.

After having worked with the material for this essay and interviewing the scientists, brought up in an academic environment, I came to the conclusion that vocabulary knowledge and reading skills shown during a test did not show everything. Speed and accuracy of reading often correlate, but that was not the case in this study. Low proficiency did not seem to make an impact on the understanding of the text. The background knowledge was very important, but at the same time the attitude and interest of the subjects during the testing also had an enormous impact on their results.


Steffensen, M.S., Joag-Dev, C. and Anderson, R.C. 1979. A cross-cultural perspective on reading comprehension. Reading Research Quarterly, Vol. 15, No. 1 (pp. 10-29).

Tsui, A.B.M. and Fullilove, J. 1998 Bottom-up or top-down processing as a discriminator of L2 listening performance. Ap

Place, publisher, year, edition, pages
2009. , 25 p.
Keyword [en]
successful reading, vocabulary size, advanced reading comprehension, second language, English, domain-specific text, academic professionals, background knowledge, prior knowledge, interest, attitude, the linguistic threshold hypothesis
National Category
General Language Studies and Linguistics
URN: urn:nbn:se:su:diva-8711OAI: diva2:175228
2008-06-14, 00:00 (English)
Humanities, Theology
Available from: 2009-08-11 Created: 2004-01-01 Last updated: 2009-08-11Bibliographically approved

Open Access in DiVA

No full text

By organisation
Department of English
General Language Studies and Linguistics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 50 hits
ReferencesLink to record
Permanent link

Direct link