[pt] Nesta dissertação, investigo o processo de aprendizagem lexical em segunda língua no contexto de uma sala de aula. Meu objetivo é entender a relação entre a interação em sala de aula e as atividades de vocabulário do material didático e como esses dois mediadores influenciam a construção de conhecimento lexical dos alunos. Para tal investigação, conduzi trabalho de campo em duas turmas de inglês básico constituídas de alunos entre 12 e 13 anos. Gravei e transcrevi em áudio e vídeo aulas de dois módulos do programa para a análise da interação. Analisei também as atividades de vocabulário do material impresso utilizado nas aulas (livro didático e material de apoio). Dentre outros achados, a análise indicou que alunos e professores recorrem com freqüência a práticas conversacionais como a conversa instrucional, o andaimento e a contribuição individual para a construção de significado. Porém, isso normalmente acontece quando há um afastamento da atividade proposta do livro. Também foi observado que o material didático cria o mapa semântico para alunos e professor, provendo o vocabulário e oportunidades para a recuperação deste. De uma forma geral, as atividades de vocabulário do livro didático abrangem todos os aspectos e níveis de conhecimento da palavra. No entanto, as atividades de vocabulário cuja proposta é o uso em contextos múltiplos estão restritas à apresentações orais e encenações, o que restringe o uso espontâneo do vocabulário aprendido. / [en] This dissertation investigates the process of learning vocabulary in a second language in a classroom environment. Its objective is to understand the relationship between the classroom interaction and the vocabulary activities found in the textbook, as well as how these mediators affect the construction of lexical knowledge. To achieve this, two classes of basic English formed by 12 and 13 year-old students have been analysed. The data include the printed exercises of vocabulary used in class (textbook and other supporting material), field observation, video-tapes and transcription of interactions. Among other results, the analyses have indicated that students and teachers often use conversational strategies, like instructional conversation, scaffolding and individual contribution, to construct meaning. However, these strategies usually happen when there is a break in the activity suggested by the book. They have also shown that the textbook creates a semantic map for students and teachers, providing the vocabulary and opportunities to review it. Generally, the vocabulary activities in the textbook cover all the aspects and levels of knowledge of words, but the activities of generative use seem to be restricted to oral presentations and roleplays, which implies non-spontaneous use of the learned vocabulary.

abstract: Globalization is driving a rapid increase in motivation for learning new languages, with online and mobile language learning applications being an extremely popular method of doing so. Many language learning applications focus almost exclusively on aiding students in acquiring vocabulary, one of the most important elements in achieving fluency in a language. A well-balanced language curriculum must include both explicit vocabulary instruction and implicit vocabulary learning through interaction with authentic language materials. However, most language learning applications focus only on explicit instruction, providing little support for implicit learning. Students require support with implicit vocabulary learning because they need enough context to guess and acquire new words. Traditional techniques aim to teach students enough vocabulary to comprehend the text, thus enabling them to acquire new words. Despite the wide variety of support for vocabulary learning offered by learning applications today, few offer guidance on how to select an optimal vocabulary study set. This thesis proposes a novel method of student modeling which uses pre-trained masked language models to model a student's reading comprehension abilities and detect words which are required for comprehension of a text. It explores the efficacy of using pre-trained masked language models to model human reading comprehension and presents a vocabulary study set generation pipeline using this method. This pipeline creates vocabulary study sets for explicit language learning that enable comprehension while still leaving some words to be acquired implicitly. Promising results show that masked language modeling can be used to model human comprehension and that the pipeline produces reasonably sized vocabulary study sets. / Dissertation/Thesis / Masters Thesis Software Engineering 2020

In order to learn a language, it is important to develop a vocabulary because it facilitates the language skills: reading, listening, writing and speaking. According to the Swedish curriculum, students must be able to develop these skills in English. However, the national tests show that students have poor results in reading and reading comprehension in English. Therefore, as future teachers of languages in Sweden, we chose to investigate how students can develop and strengthen their vocabulary through reading. Our research questions are: What cognitive processes and strategies are used when learning vocabulary? What does research say about incidental vocabulary learning through reading? What relationship does vocabulary have with reading comprehension? We have answered our questions by synthesizing and analyzing empirical studies which have been divided into two categories: cognitive processes and strategies, and vocabulary acquisition through reading. The findings show that, when encountering new vocabulary, learners use memory, determination, social and metacognitive strategies. Learning vocabulary happens both incidentally and intentionally and during this procedure words go through cognitive processes that determine where in our knowledge system they belong but this procedure can be affected by both internal and external factors. Results show that learners find reading and reading while listening to be a good method of learning vocabulary and these approaches also show good results in incidental vocabulary acquisition. Learners acquire new vocabulary incidentally through reading but the number of words they learn varies greatly. Learner’s prior vocabulary knowledge and the level of the target text is important for the outcome of new vocabulary acquisition. If the learner does not have an adequate prior vocabulary, associations and connections cannot be made and coherence not completed. In summary, the relationship between incidental vocabulary acquisition and reading comprehension is mutually beneficial. Reading provides context during a learner’s integration process and this leads to comprehension and vocabulary growth. There is a reciprocal relationship between comprehension and vocabulary growth, where both build on one another. However, since all the empirical studies did not take the same factors into consideration, the results have varied. In other words, external and internal factors can have a major impact on incidental vocabulary acquisition through reading. We suggest further research to investigate these factors in order to get a clearer picture of how we as teachers can improve strategies and instructions for vocabulary acquisition through reading.

Vocabulary plays a fundamental role in language learning acquisition. With a sufficientvocabulary, learners are able to express themselves in various ways, communicate and share their thoughts and opinions with others. In contrast, an inadequate vocabulary gives rise to misunderstanding and poor communication. For many years, one of the most common tools used to enhance and develop learners’ vocabulary in language classrooms have been textbooks. However, to what extent textbooks actually help facilitate and promote vocabulary acquisition has been questioned and debated. Therefore, the present study investigates how English textbooks promote vocabulary acquisition for learners at upper secondary schools in Sweden, and this was done by analysing two English textbooks, namely Blueprint and Solid Gold, used for English 5 in Swedish schools. Firstly, the study examined whether vocabulary in the texts is repeated in the tasks, and if it is, to what extent the vocabulary is repeated. Secondly, the study examined to what extent vocabulary in the textbooks corresponds to CEFR levels targeted in English 5. Lastly, the study examined what types of tasks there are in the textbooks, and whether the textbooks have a focus on explicit or incidental learning. The results showed that there is a lack of repetition in the textbooks. On the other hand, both textbooks corresponded well to CEFR levels targeted in English 5. With respect to different types of tasks, the closed exercise type is the most frequent exercise type in both textbooks. Moreover, there was also a large proportion of communicative tasks in the textbooks. The results also showed that both textbooks seem to have a focus on incidental learning.

The increased need for digital education in upper secondary schools of Sweden during the COVID-19 outbreak showcased that there is a lack of professional knowledge in regards to what tools to use in order to vary the experience for the students in a virtual teaching environment. In this paper, we investigate whether Mobile-Assisted Language Learning (MALL) could assist English vocabulary acquisition of Swedish upper secondary students and what their attitudes are towards it. By finding and synthesizing studies that have been published on MALL in regards to vocabulary acquisition, we hoped to gauge the possible outcomes of applying it in Swedish classrooms as well. The majority of the analyzed studies showed positive results on both vocabulary acquisition and the students' attitudes. However, the methods of the experiments and the attitude questionnaires, along with the innate cultural differences, made us question whether we would see the same results in Swedish schools, but when considering the curriculum, the empirical evidence and learning theories, we still found enough relevancy and validity in the method to claim that MALL could work to assist vocabulary acquisition in the English classroom of Swedish upper secondary school.

This six-week quasi-experimental study compared the effectiveness of two English storybook reading styles (preview-review reading versus concurrent translation) on the vocabulary development of kindergarteners ( n =50) in China, whose primary language is Chinese, who learn English as a second language. The children (aged 5 to 6) were pretested to evaluate their knowledge of general and targeted words using a researcher-designed instrument tailored to the content of three selected English stories. Validity evidence based on contrasted groups and test-retest reliability had been previously gathered for the 24-word instrument (based on 8 words per story) in a pilot study. The children in the concurrent translation group listened to the English storybook with the reader using both English and Chinese interchangeably to assist children in understanding the story. With the children in the preview-review group, the reader built background knowledge and discussed difficult vocabulary in Chinese before reading the storybook only in English then reinforcing key points in Chinese after reading the storybook. Each of three English stories was read three times by the researcher/reader separately to both groups. Two days after the third reading of each story, children were post-tested on the 8 words associated with that specific story. Total posttest scores, based on combining the three 8-word posttest scores, were compared to the 24-word pretest scores to examine vocabulary gains. Results suggest that both methods of reading English storybooks to Chinese preschoolers were effective. Results also indicate that children in the concurrent translation group performed significantly better than children in the preview-review group, even after initial levels of vocabulary, based on the pretest, were controlled (Cohen’s ƒ 2 = .42, which is considered to be a large effect). Additionally, an Aptitude x Treatment Interaction model was tested, but insufficient evidence was found to suggest that the differential effectiveness of the two methods depended on the child’s initial vocabulary level. Implications for teachers, parents, and early childhood education policymakers are discussed and suggestions for further research are offered.

This thesis documents the process of developing and evaluating a curriculum written for adult literacy students transitioning from their current literacy goals to academic and occupational pursuits. The curriculum, titled the Basic College Skills Transition Curriculum (BCS Transition Curriculum), was written for students at Project Read, a non-profit adult literacy program located at the Provo Library in Provo, Utah. The model used to design this curriculum is a variation on the ADDIE model (analysis, design, development, implementation, and evaluation), including the reiterative steps of research and evaluation. The lesson plans and materials in the BCS Transition Curriculum include the topics of U.S. History Vocabulary, Textbook Reading Skills, Lecture Note-Taking Skills, Study Organization Skills, Test-Taking Skills, Computer Skills, and Time-Management Skills. The curriculum is content-based, with each skill being taught through the content subject of U.S. History. This allows vocabulary to be recycled through each lesson plan and creates more opportunities for vocabulary acquisition. This also ties the curriculum together in an integrated experience that simulates an academic course in U.S. History. After the lesson plans were developed, tutors and students used the lesson plans and participated in an evaluation of the curriculum. Their feedback gave direction for revisions and provided ideas for the future development of this curriculum. Some lessons learned during this project about curriculum design include the importance of 1) including the curriculum designer's needs in the needs assessment; 2) clearly communicating needs and constraints to collaborators and stakeholders; 3) planning for problems and accepting limitations in the product; and 4) collaborating with others.

Students enrolled in a college preparatory reading class at one particular community college were categorized based on language origin. Native English speaking students comprised one group and foreign students formed two additional groups--students whose language origin was Latin-based (i.e. Romance languages) and students whose language origin was not Latin-based (i.e. Japanese). A pretest assessment measure was used to quantify the extent that pre-existing knowledge of Latinate word parts and morphologically complex vocabulary differed among groups based on language origin. The identical instrument served as a posttest to measure the extent that direct instruction in morphological analysis resulted in change among the same groups after one semester of instruction. Two sections on both the pretest and posttest yielded a total of four distinct mean scores that formed the primary basis for comparison. Categorizing students within the college preparatory reading class based on language origin revealed distinctive strengths and weaknesses relative to group identity when learning Latin-based word parts and vocabulary. Results of a one-way fixed-factor analysis of variance, in conjunction with multiple comparison procedures, indicated that the Latin-based group performed the strongest. This group had the greatest mean score on all four measurements; however, only for the word part section of the pretest was the difference statistically significant. The non Latin-based group performed the poorest as evidenced by scoring the lowest on three of the four measures, with a statistically significant difference for the vocabulary pretest. Additionally, a disproportionately large number of students within the native English-speaking group had difficulty mastering word parts. Though the lower group mean was statistically significant for the word part section of the posttest, practical significance was not observable from the descriptive data. A follow-up frequency tabulation revealed a dichotomization within the native English speaking group between those who proceeded to master word parts and those who did not. Furthermore, results from a pretest/posttest comparison for each respective group indicated that all three groups made significant gains on both sections of the test instrument as a result of direct instruction in Latinate word parts and vocabulary. However, there was an incongruity between word part and vocabulary mastery as all three group means were markedly better on the word part section of the instrument. The results of this study suggest that college preparatory students, regardless of their language origin, enter higher education with limited knowledge of Latinate word parts and vocabulary. The results further suggest that students comprising the heterogeneously populated college preparatory reading class can profit from direct instruction in morphological analysis--regardless of language origin. Prior research has demonstrated that college-level content words tend to be morphologically complex, singular in meaning, and likely to be Latinate in origin. Reading is the salient skill utilized across the curriculum and often the primary means of content dissemination. Reading, in turn, is principally linked to the extent of one's vocabulary. Consequently, teaching morphologically complex vocabulary at the college preparatory level along with providing a working knowledge of morphemes can assist students toward college readiness.

As digitalization continues to advance, digital tools have become omnipresent in the classroom, and Student Response Systems (SRSs) have emerged as a valuable tool for teachers. However, the added value and efficacy of integrating gamified SRSs, compared to non-gamified SRSs, in the classroom has yet to be fully explored. Furthermore, little attention has been paid to non-gamified SRSs. The present study examines how a gamified learning environment in SRSs impacts vocabulary acquisition compared to a non-gamified learning environment. To analyze this effect, a quantitative study was designed using statistical analyses such as the Shapiro-Wilk test, the Friedman test and the Wilcoxon signed-rank test. SRSs, namely Gimkit and Socrative, were selected to measure vocabulary acquisition. A total of 61 high school students studying English were enrolled in this study. Eighteen target words have been picked from two different vocabulary books designed to help students reach the C1 level in English. The selected words have all been identified as adjectives. The students were split into two class-based groups, and each was assigned a SRS to use for two 20-minute study sessions, happening at an interval of three days. To measure the effect of the two SRSs on vocabulary acquisition, three tests were conducted over three weeks. The students took a pre-test, an immediate post-test after the second study session, and a final delayed post-test two weeks later. The statistical analysis and the results of the three tests measuring vocabulary acquisition revealed a statistically significant improvement in vocabulary test scores for both groups in the short- and long-term. However, no statistical significance was shown for the test scores between the two groups: gamified and non-gamified learning environments. Hence, a gamified learning environment using SRSs was shown to have no statistical significance on vocabulary learning and retention compared to a non-gamified one. While digital tools are widely embraced, this study suggests that gamified SRSs, compared to non-gamified SRSs, may not offer a significant advantage in terms of vocabulary acquisition and retention. It is, therefore, important for the teachers to get to know their students and find the type of SRSs that suit their learning style best. Further studies could investigate the impact of gamified and non-gamified SRS on the students’ vocabulary acquisition and motivation over a school year to understand the effects better.

Thesis (DPhil)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: Though Biblical Hebrew (=BH) is no longer a spoken language, students continue to learn it for the purpose of reading, or at least interacting at a deeper level, with the text of the Hebrew Bible. This suggests that BH shares with any modern language learning course the goal of learning to read. One important part of learning to read is the acquisition of an adequate number of vocabulary items. The purpose of this study is to determine which insights from Second Language Vocabulary Acquisition (=SLVA) research and related fields hold the most promise for a new — and possibly more effective — approach to learning BH vocabulary, to evaluate currently existing BH instructional materials in light of these insights, to develop a new approach based on these insights, and to test aspects of the new approach empirically. Researchers in SLVA have uncovered a number of helpful insights concerning how vocabulary and vocabulary learning should be defined as well as concerning how vocabulary is best learned. On the other hand, BH instructional materials reflect little to no influence from these insights. These materials have continued to define vocabulary narrowly as individual words and continued to conceive of vocabulary learning primarily as pairing form and meaning in contrast to the much more sophisticated definitions found in the SLVA literature. For example, SLVA researchers consider items beyond the word level, such as idioms, to be vocabulary (Moon 1997; Lewis 1993, 1997). BH instructional materials have also failed to include a significant number of beneficial Vocabulary Learning Strategies (=VLSs), while including some VLSs that are either intrinsically problematic or problematic in the ways they are employed. For example, the strategy of learning semantically related items together is common in BH instructional materials, though it has been shown to be problematic in a considerable number of experimental studies (e.g. Nation 2000; Finkbeiner & Nicol 2003; Papathanasiou 2009). Since SLVA research has yet to influence BH instructional materials, a new approach to BH vocabulary learning is warranted. This new approach is based on sound theory concerning what vocabulary is and what it means to learn it, while offering learners as many helpful strategies for learning lexical items as possible. To justify this new approach, a set of experimental studies was run including one longitudinal case study and three larger-scale experiments. This testing was partial in nature since it was only possible to test one variable at a time. The testing revealed a number of important areas for future research into BH vocabulary learning. / AFRIKAANSE OPSOMMING: Alhoewel Bybelse Hebreeus (=BH) nie meer gepraat word nie, hou studente aan om dit aan te leer vir die doel om te lees, of ten minste om op ‘n dieper vlak met die oorspronklike teks van die Hebreeuse Bybel om te gaan. Dit impliseer dat die aanleer van BH net soos met die aanleer van moderne tale, ten doel het om die taal te kan lees. Een belangrike aspek van om te leer lees, is om die woordeskat aan te leer wat vir hierdie doel nodig is. Die doel van hierdie studie is om vas te stel watter insigte, verkry uit Tweede Taal Woordeskat-Aanleer (=TTWA) navorsing en ander verwante studievelde, die meeste belofte inhou vir ‘n nuwe – en moontlik meer effektiewe – benadering tot die leer van BH woordeskat; om huidige BH leermateriaal te evalueer, in lig van hierdie insigte; om ‘n nuwe benadering te ontwikkel gebaseer op hierdie insigte; en om aspekte van die nuwe benadering empiries toets. Navorsers in TTWA het al heelwat insig verkry in hoe woordeskat en die aanleer van woordeskat gedefineer behoort te word, sowel as aangaande die mees effektiewe maniere waarop woordeskat aangeleer kan word. Dit lyk egter of beskikbare BH leermanier niks of baie min by hierdie insigte baatgevind het. Hierdie materiaal handhaaf tipies ‘n baie nou definisie van wat woordeskat is, te wete, individuele woorde. Dit beskou dus die aanleer van woordeskat as die aanleer van hoofsaaklik woordpare met verskillende vorme en betekenisse, in plaas daarvan om die meer gesofistikeerde definisies te gebruik wat in TTWA literatuur gevind word. TTWA navorsers beskou, byvoorbeelde idiome, wat uit meer as een woord kan bestaan, ook as woordskat-eenhede (Moon 1997; Lewis 1993, 1997). BH leermateriaal gebruik selde van die groot getal beskikbare nuttige Woordeskat Leer- Strategieë (=WLS). Daar word eerder dikwels strategieë gebruik wat volgens TTWA navorsing nie baie effektief is nie.’n Goeie voorbeeld in hierdie verband is die strategie om semanties-verwante items saam te leer. Empiriese navorsing het aangetoon dat hierdie strategie sonder twyfel problematies kan wees (bv. Nation 2000; Finkbeiner & Nicol 2003; Papathanasiou 2009). Aangesien TTWA feitelik nog geen beduidende invloed op BH leermateriaal gehad het, is ‘n nuwe benadering tot die aanleer van BH woordeskat waarin TTWA insigte verreken word, geregverdig. Hierdie nuwe benadering is gebaseer op ‘n deeglik begrondige teoriese model aangaande wat woordeskat is, en wat dit beteken om dit te leer. Verder kan geput word uit die skat van nuttige strategieë wat al in TTWS geïdentifiseer is om sodoende aan BH leerder strategieë te bied wat hulleself al bewys het. Om die nuwe benadering empiries te begrond, is ‘n reeks eksperimentele studies geloods. Nie alle aspekte van die nuwe model kon sistematies getoets word nie. Die grondslag vir verdere navorsing is egter gelê.