Limited health literacy is a barrier to understanding health information. Simplifying text can reduce this barrier and possibly other known disparities in health. Unfortunately, few tools exist to simplify text with demonstrated impact on comprehension. By leveraging modern data sources integrated with natural language processing algorithms, we are developing the first semi-automated text simplification tool. We present two main contributions. First, we introduce our evidence-based development strategy for designing effective text simplification software and summarize initial, promising results. Second, we present a new study examining existing readability formulas, which are the most commonly used tools for text simplification in healthcare. We compare syllable count, the proxy for word difficulty used by most readability formulas, with our new metric ‘term familiarity’ and find that syllable count measures how difficult words ‘appear’ to be, but not their actual difficulty. In contrast, term familiarity can be used to measure actual difficulty.

This thesis examines the scientific language in different subjects by analysing all grade 8 science items from TIMSS 2011, using four characteristic meaning dimensions of scientific language – Packing, Precision and Presentation of information, and the level of Personification in a text. The results, as well as results from established readability measures, are correlated with test performances of different student groups. The TIMSS vocabulary is compared with three Swedish corpora where low frequency words are identified and further analysed. The thesis challenges the notion that there is a single scientific language, as results show that the language use varies between subjects. Physics uses more words, biology shows higher Packing and lower Precision, while physics shows the opposite pattern. Items are generally low in Personification but physics has higher levels, earth science lower. Chemistry often presents information in more complex ways. The use of meaning dimensions manages to connect the language use in science items to student performance, while established measures do not. For each subject, one or more of the meaning dimensions shows significant correlations with small to medium effect sizes. Higher Packing is positively correlated with students’ results in earth science, negatively correlated in physics, and has no significant correlations in biology or chemistry. Students’ performances decrease when placing items in everyday contexts, and skilled readers are aided by higher precision, while less-skilled seem unaffected. Many meaning dimensions that influence low performers’ results do not influence those of high performers, and vice versa. The vocabulary of TIMSS and school textbooks are closely matched, but compared with more general written Swedish and a more limited vocabulary, the coverage drops significantly. Of the low frequency words 78% are nouns, where also most compound–, extra long– and made-up words are found. These categories and nominalisations are more common in biology and, except for made-up words, rare in chemistry. Abstract and generalizing nouns are frequent in biology and earth science, concrete nouns in chemistry and physics.

In today’s multicultural society it is increasingly important that information is made available in a way that allows it to reach as many people as possible. The present study investigates the readability of the information provided in English on a Swedish municipality’s website. While Umeå Municipality sets a good example when it comes to providing information in foreign languages, the question is how easy the information is to read. The methods used to measure the readability of the texts were three automated readability formulas as well as additional analyses focusing on sentence structure and the number of clauses per word. The results show that despite obvious efforts to follow the guidelines for providing public information, more attention needs to be given to the form of the texts themselves. The complexity of the texts as gauged by the reading formulas was in all cases greater than what is recommended for information written for the general public. Some of the texts would seem to require the reader to have a college degree to fully comprehend the information. The supplementary analyses, especially when it comes to the number of clauses per sentence, confirmed the complexity of the texts. The importance of ‘writing for your audience’ thus seems to have been neglected for parts of the analysed material, which implies that some readers may not fully understand their rights and responsibilities regarding the areas addressed on the municipality’s website.

In our information age, it is of the highest importance that information is easy to understand by as many members of the potential target audience as possible. The present study analyses and compares the readability of 20 newspaper articles, half from the tabloid Daily Mail and the other half from the broadsheet The Daily Telegraph. The methods used to analyse the articles are mathematic readability formulas based on sentence and word length, as well as analyses of the use of the active and the passive voice, type-token ratio, number of clauses per sentence, and linking words. The results do not completely align with each other. Three of the five methods – the readability formulas, the use of the passive voice, and clauses per sentence – suggest that the Daily Mail articles are easier to understand, whereas the type-token ratios imply the opposite, and the linking words results did not show a difference in readability.

No description available.

In light of increasing mandates to incorporate close reading of primary source historical documents at the elementary level, this study explored the reading difficulty level of the US Constitution with preservice elementary teachers using a traditional cloze assessment procedure. While best practice pedagogy of social studies has long included thoughtful reading of primary sources, new language arts guidelines situate the analysis of primary documents within formulaic quantifiable frameworks, often problematic to the pre-service teacher. With implications for reading and social studies, this paper explores several relevant issues to both pre-service teachers and the elementary classrooms they will teach in.