The complexity for being at the forefront regarding information retrieval systems are constantly increasing. Recent technology of natural language processing called BERT has reached superhuman performance in high resource languages for reading comprehension tasks. However, several researchers has stated that multilingual model’s are not enough for low-resource languages, since they are lacking a thorough understanding of those languages. Recently, a Swedish pre-trained BERT model has been introduced which is trained on significantly more Swedish data than the multilingual models currently available. This study compares both multilingual and Swedish monolingual inherited BERT model’s for question answering utilizing both a English and a Swedish machine translated SQuADv2 data set during its fine-tuning process. The models are evaluated with SQuADv2 benchmark and within a implemented question answering system built upon the classical retriever-reader methodology. This study introduces a naive and more robust prediction method for the proposed question answering system as well finding a sweet spot for each individual model approach integrated into the system. The question answering system is evaluated and compared against another question answering library at the leading edge within the area, applying a custom crafted Swedish evaluation data set. The results show that the fine-tuned model based on the Swedish pre-trained model and the Swedish SQuADv2 data set were superior in all evaluation metrics except speed. The comparison between the different systems resulted in a higher evaluation score but a slower prediction time for this study’s system.

The purpose of this study was to explore the impact of school library access on student reading comprehension as measured by the Easy CBM universal screener test. The population consisted of 5th and 6th grade students enrolled in one intermediate school in an urban school system in East Tennessee. Data were analyzed for the population of the 2016-2017 academic school year. Specifically, this study considered the impact of school library access on students who qualify for free or reduced meals. A quantitative study was used to find correlations and differences in the reading comprehension data and library checkout data for fifth and sixth grade students who do or do not qualify for free or reduced meals in an urban East Tennessee school district. A quasi-experimental design was selected because preexisting data were collected on 1,003 fifth and sixth grade students enrolled and assessed during the 2016 – 2017 school year. The reading comprehension levels from the Easy CBM universal screener were collected for each of the students in the study. The number of class library visits for each student’s language arts teacher as well as the total checkout numbers for each student were also collected for each student in the study. The results of the study indicated a significant positive correlation in the number of checkouts and class library visits for all students. Results also indicated a significant positive correlation in the number of checkouts for students who qualify for free and reduced meals when these students are taken to the library as part of their language arts instructional block. Additional results of the study indicated a significant positive correlation between the number of books checked out and reading comprehension scores when all students are considered. There was significant difference in the number of books checked out by students who did not qualify for free or reduced meals, with those students checking out approximately one more book than students who do qualify for free or reduced meals. Finally, results of the study indicate a significant difference in reading comprehension scores for student who students who do not qualify for free or reduced meals and students who do qualify. Students who do not qualify for free or reduced meals scored significantly higher than students who do qualify.

The purpose of this study was to refine a rating procedure used to assess intermediate elementary school students' ability to orally retell what they had read from two expository passages. Oral retellings from 28 fourth grade students were tape-recorded and rated on two different occasions by each of 4 raters. A four-facet (passage, day of test administration, rater, and rating occasion) generalizability study was conducted using a partially nested design. The six largest sources of variability identified in the G-study included (a) students, (b) the student-by-day interaction, (c) the interaction of passage with rater (nested within student and day), (d) the student-by-day-by-occasion interaction, (e) the passage-by-raters (nested within students and day)-by-occasion interaction, and (f) the residual. A D-study was conducted to predict the values of the error variances and generalizability indices for both relative and absolute decisions. The results show how the error variance and the generalizability coefficients vary as a function of the number of passages, days of test administration, raters, and rating occasions. The results of the D study indicate that adding an extra reading day would produce a greater increase in reliability than asking the students to read more passages, or using more raters or more rating occasions. To achieve the greatest gain in generalizability, teachers should have students read at least two passages on at least two separate days and have their retelling rated by at least two raters and then compute a mean rating for each student averaged across the various passages, testing days, and raters.

This study investigates the relationship between vocabulary knowledge and reading comprehension of authentic Arabic texts; in particular, it attempts to investigate the percentage of vocabulary coverage (known words) readers need to ensure reading comprehension of two reading passages from online Arabic news source. Data was collected from twenty-three Arabic as-foreign language (AFL) learners at Brigham Young University, who ranged from Intermediate Low to Intermediate Mid in both productive and receptive skills. Two reading comprehension tests, circling the unknown words in texts and a lexical coverage test for each passage texts were given to the subjects. A linear regression analysis of the data shows that there is a correlation coefficient of 0.7 and 0.6 between the percentage of known words and students'comprehension of the two reading texts. The results indicate that the subjects needed to know approximately 90% of running words to adequately comprehend the first passage and around 86% to comprehend the second passage. Based on the findings, this study suggests that there is a lexical threshold for AFL learners,below which adequate comprehension of authentic texts might not be possible.

[pt] Ler com vista a alcançar a compreensão ainda é uma dificuldade a ser superada por muitos alunos. A compreensão leitora e, mais especificamente, as inferências precisam se tornar objeto de ensino, cujo propósito seja formar leitores capazes de produzir sentido para o que leem. Assim, se as inferências demandam um ensino sistematizado, nosso objetivo com esta pesquisa foi analisar e descrever o conhecimento dos professores de quarto e quinto ano do ensino fundamental sobre o ensino de inferências. Participaram do presente estudo 67 professores que atuam ou já atuaram no quarto e/ou quinto ano em escolas do Rio de Janeiro. Esses professores responderam um formulário eletrônico contendo os instrumentos: 1) questionário sobre tipos de perguntas inferenciais e não inferenciais; 2) Escala de Estratégias para Ensino de Inferências (EEEI); 3) questionário de perfil dos docentes. Os achados receberam um tratamento quantitativo, e se estruturou a análise dos dados em duas etapas: análise fatorial exploratória e análise das frequências de respostas. Os resultados foram discutidos à luz da teoria cognitiva da leitura. De uma maneira geral, os resultados referentes às perguntas inferenciais e não inferenciais, bem como às estratégias de compreensão leitora apontaram que os professores conhecem tanto perguntas adequáveis ao ensino de inferências quanto estratégias de compreensão que possuem relevância em tal ensino. À vista desse conhecimento dos professores, uma perspectiva de ensino de compreensão que entende a leitura como uma atividade de produção de sentido se sobressaiu. Sobre a realização de perguntas, os professores declararam conhecer perguntas de processo e de produto, não se restringindo às perguntas literais. A importância dos conhecimentos prévios do leitor para o ensino de inferências parece integrar o conhecimento dos professores. Por meio desta pesquisa, foi possível ressaltar e discutir a importância da produção de inferências para a formação de leitores proficientes e autônomos, bem como dispor estratégias de compreensão leitora e refletir sobre elas como caminhos favoráveis para ensinar os alunos a estabelecerem inferências autorizadas pelo texto. Por fim, através do que os professores declararam conhecer a respeito de ensino de inferências, conseguimos analisar como questões relacionadas à compreensão leitora e, mais especificamente, à inferência são concebidas pelos docentes. / [en] Reading with a view to achieving comprehension is still a difficulty to be overcome by many students. Reading comprehension and, more specifically, inferences need to become an object of teaching, whose purpose is to form readers capable of producing meaning for what they read. Thus, if inferences demand a systematized teaching, our objective with this research was to analyze and describe 4th and 5th grade teachers knowledge about the teaching of inferences. In this study, 67 teachers who work or have worked in the 4th and/or 5th grade in schools of Rio de Janeiro participated. These teachers answered an electronic form containing the following instruments: 1) a questionnaire about the types of inferential and non-inference questions; 2) the Inferential Teaching Strategies Scale (EEEI); 3) a questionnaire about the teachers profile. The findings received a quantitative treatment, and the data analysis was structured in two stages: exploratory factor analysis and analysis of the frequency of responses. The results were discussed in the light of the cognitive theory of reading. In general, the results concerning inferential and non-inference questions, as well as reading comprehension strategies, indicated that teachers know both questions suitable for teaching inferences and comprehension strategies that have relevance in such teaching. In view of this teachers knowledge, a comprehension teaching perspective that understands reading as a meaning-making activity stood out. Regarding questioning, the teachers stated that they were familiar with process and product questions, not restricted to literal questions. The importance of the reader s prior knowledge to teach inferences seems to be part of the teachers knowledge. Through this research, it was possible to highlight and discuss the importance of the production of inferences for the formation of proficient and autonomous readers, as well as to have reading comprehension strategies and reflect on them as favorable ways to teach students to establish inferences authorized by the text. Finally, through what the teachers declared they knew about the teaching of inferences, we were able to analyze how issues related to reading comprehension and, more specifically, to inference are conceived by the teachers.

The aim of this degree project is to examine how three 6th grade teachers implement reading instruction in their lessons. Furthermore, the aim is to investigate how they describe their work with reading comprehension. Our investigation is based on qualitative semi- structured interviews and observations. The results indicate that teachers do not explicitly teach reading strategies. Furthermore, teachers showed an awareness of students’ interests and took them into consideration during the lessons. For instance, the teachers argued that in order to motivate students to read, they must be provided with interesting texts. The findings also showed that all teachers emphasizes vocabulary and they argue that the bigger the vocabulary is, the more effortless reading becomes.

Mot bakgrund av dagens läroplaners betoning av att elever inom olika utbildningsnivåer och från olika kulturella kontexter ska kunna tillgodogöra sig information i text och bild inom olika ämnen, är syftet för denna uppsats att undersöka bilders betydelse för lärsförståelsen hos vuxna andraspråksinlärare i Sfi - svenskundervisning för invandrare. Metoden jag använt mig av är en kvalitativ metod med uppföljande individuella intervjuer. Informanterna delades upp i två grupper av vilka den ena gruppen fick läsa en text med tillhörande bilder samtidigt som den andra gruppen fick läsa motsvarande text men utan bilder. Efter läsningen fick samtliga informanter svara på ett antal övningsuppgifter till texten. Resultatet visade sig bli att de informanter som läst texten med bilder fick ett bättre utfall på övningsuppgifterna än de som läst texten utan bilder. Resultatet visade sig även ligga i linje med tidigare forskning som gjorts i området; forskning som bl.a. visar på att om det finns redundans mellan text och bild så bidrar detta till en bättre förståelse av innehållet i texten men även i bilderna - framförallt i pedagogiska läromedel. / The purpose of this paper is to investigate the meaning of images for reading comprehension within adult second-language education "Sfi- svenskundervisning för invandrare" i.e. Swedish language teaching for immigrants. Today's teaching programmes outlines that students from different cultural contexts and educational levels should be able to comprehend information in both written texts, from images and also within different topics. In order to achive my results I have used a qualitative method, followed by individual interviews. Initially the informants were divided into two groups of which one group read a text accompanied with pictures while the other group read the same text but without the pictures. After reading, the informants were asked questions relating to the text they had just read.The results showed that the informants who had been reading the text along side with the pictures reached better results than the informants who had read the same text without the pictures. The result also showed to be corresponding to previous research findings within the area, wich shows that if there is clarity between picture and text, the picture will contribute to a better coprehension of the text as well as a better comprehension of the content of the picture; particularly regarding pedagogical teaching materials.

Scientific literacy has been at the forefront of science education reform for the past 20 years, particularly for students from culturally and linguistically diverse (CLD) backgrounds (Lee et. al., 2005; Pearson, Moje & Greenleaf, 2010). The ability to extract meaning from text is an important skill. Yet many students struggle with effectively comprehending what they read, particularly in content areas of science, math and history. According to the National Assessment Educational Progress (NAEP, 2013) report, adolescents are not acquiring advanced literacy skills needed to succeed in the workplace and academic setting. Literacy experts have called for the use of disciplinary literacy approaches to engage learners with the content in ways that mirror what scientists, historians and mathematicians do to gain understanding in their disciplines (Moje, 2006; Shanahan & Shanahan, 2008). Although disciplinary literacy instruction is promising, there is limited empirical research on the effectiveness of discipline-specific literacy approaches. The present study examined the effects of a metalinguistic approach on the comprehension of science text among African American 5 and 6th grade students. The focus of the instructional protocol was to explicitly teach adverbial clauses and assist students to unpack adverbial clauses through the use of a graphic organizer. The process of unpacking complex sentences aimed to facilitate comprehension of science text by engaging the participants in analysis and discussion of the meaning obtained from the adverbial clauses. This study employed an experimental single-case multiple-probe across participants design. Visual Analysis (VA) and the Improvement Rate Difference (IRD) were used to analyze the data. The results of VA and IRD indicated that all participants demonstrated progress between baseline and treatment phases. Overall, the results of the investigation suggest that it is possible for 5th and 6th grade African American students to benefit from instruction that closely analyzes language. Clinical implications and future research directions are discussed.

The aim of this paper is to investigate students’ understanding of the learning requirements for grade E in English in year 9 at a secondary school in the south of Sweden. Additionally, the student’s teacher is also interviewed to provide a context and setting for the study. The student data was conducted using open-ended questionnaires while the teacher was interviewed using a semi-structured qualitative approach. In their responses to the questionnaire, the students showed different degrees of understanding of parts of the syllabus. Some students were able to answer all questions and were capable to express themselves, however some students did not answer all the questions. Furthermore, a majority of the students felt that the language in the syllabus was difficult to understand. The teacher believed that the language in Lgr11 makes it difficult for students to understand the text. Additionally, she believed that the language is difficult even for teachers.

Läsningen är en viktig färdighet för att kunna fungera och ta del av samhället. Läsningen lärs ut i en tidig ålder men det är dock inte alla som har positiva upplevelser av läsningen. Många elever möter stora svårigheter och utmaningar i sin läsning, vilket kan bidra till en förminskad tilltro till läsandet. En förminskad tilltro kan bidra till en negativ spiral av elevens läsupplevelse och den riskerar att eleven inte vill läsa eller ta del av textvärlden. Goda läskunskaper är en viktig grund för att eleven ska kunna fungera i samhället. Ett arbetssätt som möter och arbetar med elevers läsning är läshund och pedagogisk tjänstehund. Läshund är en hund som är specialutbildad och används som pedagogisk resurs i arbetet med elevers läsning. Elever läser för läshunden medan hunden sitter bredvid och lyssnar. Även pedagogisk tjänstehund kan användas i arbetet med elevers läsning, dock med större fokus på elevers kognitiva förmågor (Skolan för Pedagogiska Tjänstehundar).Syftet med studien är att få kunskap om hur elevers läsning med läshund kan påverka deras upplevelse av läsning. Tidigare forskning visar att läshund har positiv påverkan på elevers läsning gällande självförtroende, självkänsla och motivation. Forskningen lyfter också fram betydelsen av relationen mellan hund och läsare. Det teoretiska perspektivet som forskningen utgår ifrån är det sociokulturella perspektivet, med fokus på att vi lär oss när vi kommunicerar och samspelar med varandra. Metoden som valdes var intervjuer med elever där ljudinspelning användes i syfte att säkerhetsställa empirin. En observation genomfördes också som ett komplement till intervjuerna. Informationen från intervjuer och tidigare forskning visar att arbete med hundar, som en pedagogisk resurs för elever, bidrar till ett positivt förhållningssätt till läsning. Resultatet visar även att motivationen för att läsa fler texter ökar i samverkan med läshund.