Using LSA and text segmentation to improve automatic Chinese dialogue text summarization

Automatic Chinese text summarization for dialogue style is a relatively new research area. In this paper, Latent Semantic Analysis （LSA） is first used to extract semantic knowledge from a given document, all question paragraphs are identified, an automatic text segmentation approach analogous to Text＇filing is exploited to improve the precision of correlating question paragraphs and answer paragraphs, and finally some ＂important＂ sentences are extracted from the generic content and the question-answer pairs to generate a complete summary. Experimental results showed that our approach is highly efficient and improves significantly the coherence of the summary while not compromising informativeness.     

What makes a skilled writer? Working memory and audience awareness during text composition

This study investigated the role of working memory capacity as a factor for individual differences in the ability to compose a text with communicative efficiency based on audience awareness. We analyzed its differential effects on the dynamics of the writing processes, as well as on the content of the finished product. Twenty-five graduate students composed a procedural text explaining how to assemble a model turbine. They were free to consult a documentary source, featuring captioned pictures of turbine parts and assembly steps, at any time. Graphomotor and eye movements were recorded using ‘Eye and Pen’ software with an eye-tracker and digitizing tablet. Results showed that high WM capacity writers used a different strategy to explore the visual source, making longer writing pauses and producing more detailed procedures, and achieved the communicative goal more efficiently, by introducing more reader supports. In conclusion, we discuss the feasibility of audience awareness training.     

Explaining the modality effect in multimedia learning: Is it due to a lack of temporal contiguity with written text and pictures?

The study examined whether the modality effect is caused by either high visuo-spatial load or a lack of temporal contiguity when processing written text and pictures. Students (N = 147) viewed pictures on the development of tornados, which were accompanied by either spoken or written explanations presented simultaneously with, before, or after the pictures. For verbal recall no modality effect was observed, whereas for transfer the influence of modality varied as a function of phonological working memory capacity. For pictorial recall the results showed a modality effect that was limited to simultaneous presentation, apparently in line with the temporal contiguity explanation. However, spoken simultaneous presentations were not superior to spoken sequential presentations, which contradicts the temporal contiguity explanation for the modality effect. Rather, it seems as if learners with simultaneous presentation of written text and picture concentrated more on the text and ignored the pictures, resulting in worse pictorial recall.     

How Far Are the Theories about Collaborative Writing Usable in a Chinese Learner Context?

With the development of communication,the role of English is changing from being the language merely for the understanding of other cultures to being a tool of communication. English writing is regarded as equally important as speaking,listening and reading. This study aims to focus on collaborative writing as a effective teaching and learning tool in a Chinese Learner Context.     

Using concept mapping to improve poor readers’ understanding of expository text

The present study examined whether the use of concept mapping is more effective in teaching expository material in comparison to a traditional, lecture only, approach. Its objective was threefold. First, to determine if multimedia concept mapping produces differential learning outcomes compared to digital text-based concept mapping. Secondly, to compare the above experimental approaches to traditional teaching and finally to demonstrate if there are significant differences in the level of satisfaction reported by children under the three instructional approaches used. Thirty fifth grade children participated from which three groups of poor readers were formed, matched on age, gender and reading ability. The two experimental groups were taught with the use of digital text-based and multimedia concept maps respectively and the control group was subjected to the traditional teaching method of expository material. Each group had three sessions of instruction on three different expository texts. The results indicated that concept mapping can be an effective means for teaching expository content. However, there were no differential learning outcomes when adding multimedia elements to concept maps. Moderate satisfaction was reported for all three approaches. The findings coincide with the body of literature emphasizing the importance of concept mapping in the learning and instruction of children with reading difficulties. The study’s results inform potential educators on the use of multimedia with poor readers in elementary school.     

Low test scores in Latin America: poor schools,poor families or something else?

Latin American students consistently score low on international tests of cognitive skills. In the PISA 2012 results, students in seven Latin American countries had an average score of 395 in mathematics, or about 100 points lower than the average score of 497 in four Scandinavian countries. We estimate the effect of socioeconomic characteristics on student scores in Latin America and Scandinavia and find that 50 points of the difference are explained by Latin American parents’ lower average educational and socioeconomic characteristics, 25 points are explained by Latin America’s weak cultural orientation toward reading books and the remaining 25 points are explained by the lower effectiveness of Latin American educational systems in teaching cognitive skills.     

Learning by explaining orally or in written form? Text complexity matters

In this experiment, we examined whether linguistic text complexity affects effects of explaining modality on students’ learning. Students (N = 115) read a high-complex and a low-complex text. Additionally, they generated a written or an oral explanation to a fictious peer. A control group of students retrieved the content. For the low-complex text, we found no significant differences between conditions. For the high-complex text, oral explaining yielded better comprehension than writing explanations. The retrieval condition showed the lowest performance. Mediation analyses revealed that the effect of explaining modality while learning from the high-complex text was mediated by the personal references and the comprehensiveness of the generated explanations. Our findings suggest that the effect of explaining modality emerges when students are required to learn from difficult texts. Furthermore, they show that oral explaining is effective as, likely due to increases of social presence, it triggers distinct generative processes during explaining.     

Are parents reliable in reporting child victimization? Comparison of parental and adolescent reports in a matched Chinese household sample

There is ongoing debate about the reliability of parental reports on child victimization. Some studies have shown that they are useful, whereas some others have provided contrary evidence suggesting that parents are not accurate in reporting child victimization, especially when they are the one who inflicted the violence. This study aimed to (a) examine the reliability of parental reports of adolescents’ experiences of victimization, including that inflicted by parents as well as others, by comparing them with self-reports using a parent–child matched sample from China; and (b) explore the possible reasons underlying any disagreement between the parental and adolescent reports. A total of 2,624 parent–adolescent pairs were recruited during 2009 and 2010 in 6 cities in China. Parents were asked to report the victimization experiences of their child using of the Juvenile Victimization Questionnaire, and these reports were matched with the adolescents’ self-reports of victimization. Low levels of parent–adolescent agreement in reporting were found (Cohen's kappa = .04–.29). Except for sexual violence, parents were significantly less likely to report all types of victimization. Overall, lower levels of agreement were found in the reporting of (a) less severe types of victimization, (b) victimization outside the family, and (c) victimization involving parents as perpetrators. Intimate partner violence between parents was significantly associated with discrepancies between reports. The findings suggest that parents might not be reliable as a single source of information on certain types of adolescent victimization.     

Are university students' component reading skills related to their text comprehension and academic achievement?

Among developing and poor readers, text comprehension is strongly related to word decoding accuracy and reading fluency. However, among relatively skilled adult readers, these aspects of reading skill are largely independent of one another. The reading of 193 second- and third-year students enrolled in an educational psychology course at a public research university was described well by three orthogonal principal components: Word Decoding Accuracy, Reading Speed, and Text Comprehension Accuracy. Furthermore, component reading skills were not associated with individual differences in the extent to which students in this sample had succeeded in postsecondary education. Decoding Accuracy was most related to spelling accuracy and phonemic awareness. Only Text Comprehension Accuracy was associated even minimally with grade-point average. A subgroup of 15 students who reported various reading problems did not have a low mean grade-point average.     

Are you happy in New Year?

宝宝在过去的一年里学了不少本领。新年到了,秀一秀!爸爸妈妈也一起来参加,为客人们献上精彩的亲子歌舞秀!     

Sourcing in professional education: Do text factors make any difference?

The present study investigated the extent to which the text factors of source salience and emphasis on risk might influence readers’ attention to and use of source information when reading single documents to make behavioral decisions on controversial health-related issues. Participants (n = 259), who were attending different bachelor-level professional programs at a university college, generally disregarded source information irrespective of textual manipulations, especially sources cited or embedded within other documents, and mainly relied on their own personal experiences and opinions when making behavioral decisions on the issues. Theoretical as well as educational implications of the findings are discussed.     

‘Are you listening please?’ The advantages of electronic audio feedback compared to written feedback

Feedback on students' work is, probably, one of the most important aspects of learning, yet students' report, according to the National Union of Students (NUS) Survey of 2008, unhappiness with the feedback process. Students were unhappy with the quality, detail and timing of feedback. This paper examines the benefits of using audio, as opposed to written, feedback in an attempt to overcome student criticisms. Using the Audacity audio software MP3 feedback files were created and sent to 60 students either via a VLE or email. The students were asked to complete an online survey on audio feedback. Twenty‐six students responded. The results were, generally, very positive. The use of audio feedback seemed to have overcome the problems reported by the NUS survey. Students are at least 10 times more likely to open audio files compared to collecting written feedback. The paper concludes with reflections, and advice, on introducing audio feedback.     

Are boys discriminated in Swedish high schools?

Girls typically have higher grades than boys in school and recent research suggests that part of this gender difference may be due to discrimination of boys in grading. We rigorously test this in a field experiment where a random sample of the same tests in the Swedish language is subject to blind and non-blind grading. The non-blind test score is on average 15% lower for boys than for girls. Blind grading lowers the average grades with 13%, indicating that personal ties and/or grade inflation are important in non-blind grading. But we find no evidence of discrimination against boys in grading. The point estimate of the discrimination effect is close to zero with a 95% confidence interval of ±4.5% of the average non-blind grade.     

Are teacher effects larger in small classes?

Teachers spend most of their time in school in classrooms, and their instruction and teaching practices may be affected by classroom context such as class size. We examine whether teacher effects interact with classroom context such as class size. Specifically, we seek to determine whether teacher effects are more pronounced in small classes than in regular size classes in early grades. The results point to some evidence that teacher effects are more variable in small classes than in regular size classes. However, the pattern does not seem to be systematic.