Computerized classification testing (CCT) aims to classify persons into one of two or more possible categories to make decisions such as mastery/non-mastery or meet most/meet all/exceed. A defining feature of CCT is its stopping criterion: the test terminates when there is enough confidence to make a decision. There is abundant research on CCT with a single cut-off, and two common stopping criteria are the sequential probability ratio test (SPRT) statistic and the generalized likelihood ratio statistic (GLR). However, there is a relative scarcity of research extending the SPRT to the multi-hypothesis case for when there is more than one cut-off. In this paper, we propose a new multi-category GLR (mGLR) statistic as well as a stochastically curtailed version of the CCT with three or more categories. A simulation study was conducted to show that the mGLR statistic outperformed the existing stopping rules by generating shorter average test length without sacrificing classification accuracy. Results also revealed that the stochastically curtailed mGLR successfully increased test efficiency in certain testing conditions. 相似文献
Theories relating to self-efficacy have developed rapidly since Bandura first proposed the concept in 1977. In the past two decades, psychologists have carried out numerous studies to research the cultural and psychological changes in social development. The research topic of this study is whether self-efficacy changes over time. This study uses a meta–meta analysis and includes 13 meta-analyses, including 536 effect sizes, with a total sample size of 421,880. We find that individual self-efficacy increases over time, which may be related to social development trends. However, the effects of interventions on self-efficacy remain similar (Qmodel = 1.807, df = 1, p > .05), and a possible explanation is that time effects of self-efficacy confuse the effects of intervention, because both in the intervention group and control group, the average of self-efficacy increases over time. And we find that a general decline in the predictive effects of self-efficacy (Qmodel = 5.117, df = 1, p = .024), especially the ability to predict relatively objective variables (e.g. job performance, teaching effectiveness, and transfer of training). A possible explanation is that as social development people tend to overestimate their self-efficacy. Another possible explanation is that the effect sizes in the original studies being overrated, may due to intentional selective reporting or unintentional statistical errors. 相似文献
Psychometrika - This study revisits the parameter estimation issues in multidimensional item response theory more thoroughly and investigates some computation details that have seldom been... 相似文献
Three experiments investigated whether and why sharing experiences of social exclusion or social acceptance with others strengthens social bonds. Participants experienced either social exclusion or social acceptance alongside another co‐participant who either also experienced the same outcome, or experienced a different outcome, as them. Multilevel modeling results showed that participant dyads who shared the experience of social exclusion or social acceptance felt closer to each other than those who experienced different outcomes, and that perceived similarity mediated the effect of shared experiences on social bonds. Interestingly, participants felt closer to one another after having shared social acceptance, more so than when they have shared social exclusion. Implications of the present findings are interpreted in light of theories of social exclusion, shared experiences, and social bonding. 相似文献
Existing knowledge on remote working can be questioned in an extraordinary pandemic context. We conducted a mixed-methods investigation to explore the challenges experienced by remote workers at this time, as well as what virtual work characteristics and individual differences affect these challenges. In Study 1, from semi-structured interviews with Chinese employees working from home in the early days of the pandemic, we identified four key remote work challenges (work-home interference, ineffective communication, procrastination, and loneliness), as well as four virtual work characteristics that affected the experience of these challenges (social support, job autonomy, monitoring, and workload) and one key individual difference factor (workers’ self-discipline). In Study 2, using survey data from 522 employees working at home during the pandemic, we found that virtual work characteristics linked to worker's performance and well-being via the experienced challenges. Specifically, social support was positively correlated with lower levels of all remote working challenges; job autonomy negatively related to loneliness; workload and monitoring both linked to higher work-home interference; and workload additionally linked to lower procrastination. Self-discipline was a significant moderator of several of these relationships. We discuss the implications of our research for the pandemic and beyond. 相似文献
To investigate the influence of the home environment, defined as family socioeconomic status (SES) (parent education level, household income), student resource-mediated SES (access to nutritional resources and cognitively stimulating experiences), reading ability, and difficulty with homework on quality of life in children and adolescents residing in urban and suburban areas in Hangzhou City, Zhejiang Province, China. This study included 3200 Grade 3–6 students from 8 elementary schools in Hangzhou City. Assessments included questionnaires that evaluated student quality of life, family SES, resource-mediated SES (dietary behavior and the home literacy environment), reading ability, and difficulty with homework. The effects of the home environment on student quality of life were analyzed by univariate analysis, multiple linear regression analysis, and structural equation modeling. Overall, 80.6% of students had a medium or better quality of life. Young age (Grade 3 or 4), female sex, household income of 10000–15000 RMB, high breakfast consumption, daily intake of fruit, a balanced diet, and good reading habits were positively correlated with student quality of life (P?<?0.05), while overuse of electronic devices was negatively correlated with quality of life (P?<?0.05). Dietary behaviors, home literacy environment, and student reading ability and difficulty with homework directly affected quality of life. Family SES indirectly affected student quality of life. Children and adolescents in China should have access to good nutrition and cognitively stimulating experiences to enhance their well-being and provide them with social and academic advantages.
Journal of Child and Family Studies - We examined hypothalamic–pituitary–adrenal stress reactivity as an indicator of differential susceptibility in moderating the links between family... 相似文献