项目反应理论观察分数核等值的影响因素

探究带宽选择方法、样本量、题目数量、等值设计、数据模拟方式对项目反应理论观察分数核等值的影响。通过两种数据模拟方式,获得研究数据,并计算局部与全域评价指标。研究发现,在随机组设计中,带宽选择方法表现相似;考生样本量和题目数量影响甚微。在非等组设计中,惩罚法与Silverman经验准则表现优异;增加题目量可降低百分相对误差和随机误差;增加样本量导致百分相对误差变大,随机误差减小。数据模拟方式可影响等值评价。未来应重点关注等值系统评估。     

社会创造性的微观变化：基于虚拟仿真技术的研究

本研究将微观发生法设计和虚拟仿真技术相结合,考察了不同特质社会创造性大学生跨情境、跨时间的状态社会创造性的变化特点。结果表明：（1）在四种具体社会问题情境中,个体策略新颖性的变化表现出了跨情境的多样性特点,而策略实用性的变化则主要表现出了跨情境的一致性特点;（2）个体特质社会创造性对其状态社会创造性的表现的影响主要体现在策略的实用性上。     

Single-case effect size calculation: Comparing regression and non-parametric approaches across previously published reading intervention data sets

Growing from demands for accountability and research-based practice in the field of education, there is recent focus on developing standards for the implementation and analysis of single-case designs. Effect size methods for single-case designs provide a useful way to discuss treatment magnitude in the context of individual intervention. Although a standard effect size methodology does not yet exist within single-case research, panel experts recently recommended pairing regression and non-parametric approaches when analyzing effect size data. This study compared two single-case effect size methods: the regression-based, Allison-MT method and the newer, non-parametric, Tau-U method. Using previously published research that measured the Words read Correct per Minute (WCPM) variable, these two methods were examined by comparing differences in overall effect size scores and rankings of intervention effect. Results indicated that the regression method produced significantly larger effect sizes than the non-parametric method, but the rankings of the effect size scores had a strong, positive relation. Implications of these findings for research and practice are discussed.     

Using generalized additive (mixed) models to analyze single case designs

This article shows how to apply generalized additive models and generalized additive mixed models to single-case design data. These models excel at detecting the functional form between two variables (often called trend), that is, whether trend exists, and if it does, what its shape is (e.g., linear and nonlinear). In many respects, however, these models are also an ideal vehicle for analyzing single-case designs because they can consider level, trend, variability, overlap, immediacy of effect, and phase consistency that single-case design researchers examine when interpreting a functional relation. We show how these models can be implemented in a wide variety of ways to test whether treatment is effective, whether cases differ from each other, whether treatment effects vary over cases, and whether trend varies over cases. We illustrate diagnostic statistics and graphs, and we discuss overdispersion of data in detail, with examples of quasibinomial models for overdispersed data, including how to compute dispersion and quasi-AIC fit indices in generalized additive models. We show how generalized additive mixed models can be used to estimate autoregressive models and random effects and discuss the limitations of the mixed models compared to generalized additive models. We provide extensive annotated syntax for doing all these analyses in the free computer program R.     

Rereading the Kripkean Intuition on Reference

Saul Kripke's thought experiments on the reference of proper names target the theory that the properties which identify a term's referent are the subject of an implicit agreement. Recently, survey versions of the experiments have been thought to show that intuitions about reference are culturally contingent. Proposing a revisionary interpretation, this article argues, first, that Kripke's Cicero/Feynman experiment reveals that every name user knows enough to be capable of identifying the same individual as the name's most informed users. Second, the article shows that Kripke's presentation of the Gödel/Jonah experiment is ambiguous with respect to the properties attributed to the referent. Disambiguated, the experiment fails to reveal that name users may be mistaken in every unique property they attribute. Since the experiment's ambiguity is replicated in survey presentations, cross‐cultural variation in survey response fails to show that intuitions about reference are culturally contingent.     

Within‐subject analysis of a prevention strategy for problem behavior

Although decades of research on functional analysis methodology have identified common contingencies that maintain problem behavior and effective interventions, relatively little research has been conducted on strategies to prevent the initial development of problem behavior. We conducted a 2‐part case study, the purposes of which were to illustrate the use of sensitivity tests as the bases for intervention (Study 1) and subsequently to assess the efficacy of a prevention strategy using a single‐subject design (Study 2). Results showed that the sensitivity tests identified establishing operations that may set the occasion for the development of problem behavior and that interventions based on differential reinforcement prevented increases in the severity of problem behavior relative to untreated and control baselines. Benefits and limitations to this individualized approach to prevention are discussed.     

“Where You Lead,I Will Follow”: Partisan Cueing on High-Salience Issues in a Turbulent Multiparty System

The ability of parties to not only reflect, but actually shape, citizens' preferences on policy issues has been long debated, as it corresponds to a fundamental prediction of classic party identification theory. While most research draws on data from the United States or studies of low-salience issues, we exploit the unique opportunity presented by the 2013 Italian election, with the four major parties of a clear multiparty setting holding distinct positions on crucial issues of the campaign. Based on an experimental design, we test the impact of party cues on citizens' preferences on high-salience issues. The results are surprising: Despite a party system in flux (with relevant new parties) and a weakening of traditional party identities, we find large, significant partisan-cueing effects in all the three experimental issues, and for voters of all the major Italian parties—both old and new, governmental and opposition, ideologically clear or ambiguous.     

Effects of perception on public bike-and-ride: A survey under complex,multifactor mode-choice scenarios

Bicycles are an environmentally-friendly, energy-saving mode of transportation. Public bike-and-ride is a term used to describe the use of public bicycles combined with transit as a means of transportation. A new stated preference (SP) design scheme based on the factors respondents felt were most important was proposed to augment the traditional SP design under complex multi-factor and multi-level situations. iPads were used to automatically administer the survey under each design scheme and collect the choice behavior data for public bike-and-ride. The concept of perceived psychological distance was further proposed and used to estimate Logit models. Comparative analysis of these models shows for the traditional SP design scheme, some factors, including perceived psychological distance, have reduced statistically significant effects on public bike-and-ride choice relative to the new SP design scheme. This indicates that in scenarios with complex decisions, the decision-makers would reduce their consideration of some factors and simplify their decision process making the new SP survey design scheme more suitable to analyze their choice behavior in those cases. Model analysis showed that perceived psychological distance for each traveler is the most critical factor during the decision process for travel behavior analysis. Finally, sensitivity analysis was conducted to determine which factors in the new survey design were important to increase the share of public bike-and-ride users. Practical findings for promoting public bike-and-ride use are presented in addition to theoretical references for improving the traditional survey methods.