Using plausible values in secondary analysis in large-scale assessments

Plausible values are typically used in large-scale assessment studies, in particular, in the Trends in International Mathematics and Science Study and the Programme for International Student Assessment. Despite its large spread, there are still some questions regarding the use of plausible values and how such use affects statistical analyses. The aim of this paper is to demonstrate the role of plausible values in large-scale assessment surveys when multilevel modeling is used. Different user strategies concerning plausible values for multilevel models as well as means and variances are examined. The results show that some commonly used user strategies give incorrect results while others give reasonable estimates but incorrect standard errors. These findings are important for anyone wishing to make secondary analyses of large-scale assessment data, especially those interested in using multilevel models to analyze the data.     

Why does zeno the stoic hold his breath? "Zenoism" as a new variable for studying suicide

A single misfortune may be seen as just that, or may be interpreted as Zeno the Stoic did, as a "sign" of cosmic proportion (zenoizing), which paradoxically provides a missing meaning structure. This article presents two studies testing out this new variable (zenoism): in Study One (n = 233), we explore the moderating effects of religiosity and gender on zenoism, depression, demoralization, and suicidality. In Study Two (n = 137), we explore these same moderating effects on zenoism, fear of dependency, value of life, and favorability toward physician-assisted suicide (PAS). Results indicate: 1) men and non-religious respondents (especially men) zenoize more than women and religious respondents; 2) non-religious respondents, (especially men) are more generally favorable to PAS and more suicidal; 3) Zenoism is negatively related to depression, demoralization, and value of life for non-religious respondents, and for men; 4) Zenoism is positively related to favorability toward PAS toward one's self and overall suicidality; and finally 5) general and self-specific favorability toward PAS are distinct, but both relate to overall suicidality.     

Continuing complexity: The university careers of a scientific elite in relation to their class origins and schooling

We report on continuing research on the UK scientific elite, intended to illustrate a proposed new approach to elite studies and based on a prosopography of Fellows of the Royal Society born from 1900. We extend analyses previously reported of Fellows' social origins and secondary schooling to take in their university careers as under- and postgraduates. The composite term ‘Oxbridge’, as often applied in elite studies, is called into question, as members of the scientific elite prove to have been recruited more from Cambridge than from Oxford. Particular interest then attaches to the relation between Fellows' social origins and schooling and their attendance at Cambridge. Among Fellows whose university careers were made at Cambridge, those of more advantaged class origins and those with private schooling are over-represented, although in this, as in various other respects, including Fellows' field of study, family influences persist independently of schooling. One suggestive interaction effect exists in that being privately educated increases the probability of having been at Cambridge more for Fellows from managerial than from professional families. Private schooling leading on to both undergraduate and postgraduate study at Cambridge can be identified as the educational ‘royal road’ into the scientific elite; and Fellows coming from higher professional and managerial families alike have the highest probability of having entered the elite in this way. But the most common route turns out in fact to be via state schooling and attendance at universities outside of ‘the golden triangle’ of Cambridge, Oxford and London; and this route is far more likely to have been followed by Fellows of all other class origins than higher professional. The relation between the degree of social skew in the recruitment of an elite and the degree of social homogeneity among its members can be more complex than has often been supposed.