Correlational analysis of ordinal data: from Pearson’s r to Bayesian polychoric correlation

Correlational analyses are one of the most popular quantitative methods, yet also one of the mostly frequently misused methods in social and behavioral research, especially when analyzing ordinal data from Likert or other rating scales. Although several correlational analysis options have been developed for ordinal data, there seems to be a lack of didactically written literature illustrating the appropriate use and differences among them. The purpose of this paper is to provide a synthesis of correlational analysis options when analyzing ordinal data. These options span from the traditional methods, such as Pearson’s r, to more recent developments, such as Bayesian estimation of polychoric correlations. An illustration of these methods utilizing a contemporary dataset is provided.     

Can EFL Speakers Communicate in English-mediated Classes?: A Case of a Liberal Arts Class for Engineering Students in Korea

The purpose of this paper is twofold: (1) to explore what factors prevent English as a foreign language (EFL) students from participating in English-medium instruction classes; and (2) to examine how dialogic teaching techniques enhance students’ participation in classroom interaction. Utilizing the notions of “communicative competence” and “truncated repertoire,” we first investigated the barriers to communication in classrooms. We next examined how EFL speakers are able to communicate in English-medium instruction (EMI) classes. The findings suggest that when the instruction was designed to reduce cultural barriers and to facilitate authentic discussion, EFL students were able to participate in classroom interaction, and they perceived discussion as a way to learn an unfamiliar subject taught in English.     

Designing a business model for financial products for cultural heritage in the Korean market

There are many contingent valuation (CV) studies to estimate the economic benefits of cultural heritage, but few provided advice on the design of financial products for cultural heritage assets. This paper conducted conjoint and willingness-to-pay (WTP) analyses for providing the design implication of these financial products. We calculated WTP for each attribute. The result of conjoint analysis shows that people prefer a short period of investment, high-expected rate of return and small amount of money invested. Only the amount invested and the rate of return on investment are significant to estimate WTP for these financial products.     

Assessing cultural values: developing an attitudinal scale

Effective measurement of cultural value is often elusive because of its multidimensional nature. It is also influenced by sociodemographic characteristics (manifest variables) and attitudinal characteristics (latent variables) of populations. While the former is easily available to researchers, the latter has not been fully studied. This paper suggests the use of a cultural worldview scale that was developed to measure cultural attitudes of people, using factor and cluster analysis. Four factors comprise the scale: cultural linkages, recognition of cultural values, cultural loss and preservation of traditions and customs. Some advantages of using this scale are demonstrated, and relationships with sociodemographic variables are investigated. Managerial and policy implications are discussed.

Orthogonal nonnegative matrix tri-factorization for co-clustering: Multiplicative updates on Stiefel manifolds

Matrix factorization-based methods become popular in dyadic data analysis, where a fundamental problem, for example, is to perform document clustering or co-clustering words and documents given a term-document matrix. Nonnegative matrix tri-factorization (NMTF) emerges as a promising tool for co-clustering, seeking a 3-factor decomposition X≈USV^?$\mathit{X}\approx {\mathit{USV}}^{?}$ with all factor matrices restricted to be nonnegative, i.e., U?0,S?0,V?0.$\mathit{U}?0\text{,}\mathit{S}?0\text{,}\mathit{V}?0\text{.}$ In this paper we develop multiplicative updates for orthogonal NMTF where X≈USV^?$\mathit{X}\approx {\mathit{USV}}^{?}$ is pursued with orthogonality constraints, U^?U=I,${\mathit{U}}^{?}\mathit{U}=\mathit{I}\text{,}$ and V^?V=I${\mathit{V}}^{?}\mathit{V}=\mathit{I}$, exploiting true gradients on Stiefel manifolds. Experiments on various document data sets demonstrate that our method works well for document clustering and is useful in revealing polysemous words via co-clustering words and documents.     

Naturalistic intervention for Asperger syndrome – a case study

On the basis of their cognitive abilities, children with Asperger syndrome are attractive candidates for inclusive education and, in Australia, most are in integrated settings. However, social interaction between children with Asperger syndrome and their peers remains problematic, with the children with Asperger syndrome often being left alone despite being among or near classmates. For classroom teachers, effective interventions in terms of class management are critical to facilitate positive social interaction between these two groups of children, and case studies can tell us a great deal about what works. In practice, a combination of intervention techniques may be best, in case a single one is ineffective, and in order to take advantage of potential synergies. In this article, Serene Choi, of the School of Education at the University of Newcastle (Australia) and Timo Nieminen, of the Centre for Biophotonics and Laser Science in the Department of Physics at the University of Queensland, report a naturalistic multiple-component intervention used to develop the social skills of a boy with Asperger syndrome in a mainstream school in Australia. This combined intervention, making use of social stories, simple peer modelling and individual lessons, appears to be a practical teaching method for inclusive education.     

Self-efficacy for self-regulated learning, academic self-efficacy, and internet self-efficacy in web-based instruction

Effects of student motivation on performance in Web-based instruction (WBI) were examined. In particular, applicability of the self-efficacy theory to WBI contexts was tested. A total of 152 junior high school students in Seoul, Korea, participated in WBI during regular science classes. Participants completed motivational surveys before the onset of WBI and took the written and search tests at the end of WBI. Path analyses revealed that students' self-efficacy for self-regulated learning positively related to their academic self-efficacy, strategy use, and Internet self-efficacy. Academic self-efficacy predicted students' performance on the written test, which comprised problems on topics covered during the previous WBI sessions. Students' scores on the WBI search test were significantly and positively predicted by their self-efficacy in using the Internet. More interesting, students' academic self-efficacy beliefs were not able to predict their search test performance, whereas students' Internet self-efficacy beliefs were not able to predict their written test performance.This research is based on data collected as part of Ha-Jeen Choi's master's thesis completed at Ewha Womans University.The authors thank Steven Ross and three anonymous reviewers for their helpful comments on earlier versions of this article.     

Age Differences in Online Social Networking: Extending Socioemotional Selectivity Theory to Social Network Sites

This article extends socioemotional selectivity theory to online social networking by examining age differences in the size and composition of Facebook networks across a wide age range of Facebook users (18 to 93 years old) in a nationally representative sample. Findings suggest increasing selectivity of Facebook social partners with age. Compared to younger adults, friend networks of older adults are smaller but contain a greater proportion of individuals who are considered to be actual friends. Moreover, a higher proportion of actual to total Facebook friends is associated with lower levels of social isolation and loneliness across the life span.     

Assessing reflective thinking in solving design problems: The development of a questionnaire

Reflection is a critical factor in solving design problems. Using good methods to observe designers' reflection is essential to inform the design of the learning environments that support the development of design problem‐solving skills. In this study, we have developed and validated a novel self‐reporting questionnaire as an efficient instrument to explore reflection, called Assessing Reflective Thinking in Solving Design Problems (ARTiD). This questionnaire has been developed based on the three‐dimensional model for reflective thinking: the timing, the objects and the levels of reflection. A total of 294 undergraduate and graduate students in the fields of engineering, computer science and instructional design were recruited to participate in one of four iterative, formative tests through which reliability and validity analyses were performed to revise and confirm the questionnaire. Examples of participants' reflection patterns were demonstrated to show the practical value of this questionnaire. In conclusion, the final version of ARTiD was presented as a valid instrument to explore students' reflection for research and educational practice. In addition, we found that ARTiD can be used as an effective learning tool to guide student designers' reflections toward better learning and performance.