Explaining default intuitions using maximum entropy

While research into default reasoning is extensive and many default intuitions are commonly held, no one system has yet captured all these intuitions nor given a formal account to motivate them. This paper argues that the extended maximum entropy approach which incorporates variable strength defaults provides a benchmark for default reasoning that is not only objectively motivated but also satisfies all the accepted default intuitions. It is shown that the behaviour of the approach coincides with a wide range of default intuitions taken from examples in the literature, and can be used to explain why some examples have led to confusion. Moreover, analysing the solutions produced by the maximum entropy approach enables clearer differentiation between the default knowledge they contain and the default inferences required of the reasoning system. This suggests that the maximum entropy approach can be used as a benchmark both for eliciting default knowledge when building a knowledge base and, by comparison, for clarifying the underlying biases of other default reasoning systems.     

Super-ordinary bias in Japanese self-predictions of future life events

Japanese are said to value being ordinary and emphasize similarity with others. We theorized that Japanese tend to perceive themselves as being ordinary, so much so that their self-predictions about future life events are biased (super-ordinary bias). Specifically, it was expected that Japanese overestimate the likelihood of experiencing common events (such as getting married or catching the flu) and underestimate the likelihood of experiencing rare events (such as winning a lottery or being murdered). We examined the effects of commonness and desirability of future life events on the relative-likelihood estimates. Our expectation was supported by three studies, involving a questionnaire, a laboratory experiment, and a mail survey. Findings are consistent with the assumed tendency to view oneself as being super-ordinary. The super-ordinary bias was also found to be independent of unrealistic optimism or pessimism.     

Tests of homogeneity of means and covariance matrices for multivariate incomplete data

Existing test statistics for assessing whether incomplete data represent a missing completely at random sample from a single population are based on a normal likelihood rationale and effectively test for homogeneity of means and covariances across missing data patterns. The likelihood approach cannot be implemented adequately if a pattern of missing data contains very few subjects. A generalized least squares rationale is used to develop parallel tests that are expected to be more stable in small samples. Three factors were varied for a simulation: number of variables, percent missing completely at random, and sample size. One thousand data sets were simulated for each condition. The generalized least squares test of homogeneity of means performed close to an ideal Type I error rate for most of the conditions. The generalized least squares test of homogeneity of covariance matrices and a combined test performed quite well also.Preliminary results on this research were presented at the 1999 Western Psychological Association convention, Irvine, CA, and in the UCLA Statistics Preprint No. 265 (http://www.stat.ucla.edu). The assistance of Ke-Hai Yuan and several anonymous reviewers is gratefully acknowledged.     

Influence analysis of ranking data

This paper develops diagnostic measures to identify those observations in Thurstonian models for ranking data which unduly influence parameter estimates that are obtained by the partition maximum likelihood approach of Chan and Bentler (1998). Diagnostic measures are constructed by employing the local influence approach that uses geometric techniques to assess the effect of small perturbations on a postulated statistical model. Very little additional effort is required to compute the proposed diagnostic measures, because all of the necessary building blocks are readily available after a usual fit of the model. The work described in this paper was partially supported by the grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (RGC Ref. No. CUHK4186/98P and RGC Direct Grant ID2060178). The authors are grateful to the Editor and four anonymous referees for their helpful comments.     

Book Reviews

Books reviewed:
George E. Marcus, W. Russell Newman, and Michael MacKuen, Affective Intelligence and Political Judgment
Micha Popper, Hypnotic Leadership:Leaders, Followers, and the Loss of Self
James W. Davis, Jr., Threats and Promises: The Pursuit of International Influence
Mark R. Warren, Dry Bones Rattling: Community Building to Revitalize American Democracy
Tali Mendelberg, The Race Card: Campaign Strategy, Implicit Messages, and the Norm of Equality     

Asymptotic bias of normal-distribution-based maximum likelihood estimates of moderation effects with data missing at random

Moderation analysis is useful for addressing interesting research questions in social sciences and behavioural research. In practice, moderated multiple regression (MMR) models have been most widely used. However, missing data pose a challenge, mainly because the interaction term is a product of two or more variables and thus is a non-linear function of the involved variables. Normal-distribution-based maximum likelihood (NML) has been proposed and applied for estimating MMR models with incomplete data. When data are missing completely at random, moderation effect estimates are consistent. However, simulation results have found that when data in the predictor are missing at random (MAR), NML can yield inaccurate estimates of moderation effects when the moderation effects are non-null. Simulation studies are subject to the limitation of confounding systematic bias with sampling errors. Thus, the purpose of this paper is to analytically derive asymptotic bias of NML estimates of moderation effects with MAR data. Results show that when the moderation effect is zero, there is no asymptotic bias in moderation effect estimates with either normal or non-normal data. When the moderation effect is non-zero, however, asymptotic bias may exist and is determined by factors such as the moderation effect size, missing-data proportion, and type of missingness dependence. Our analytical results suggest that researchers should apply NML to MMR models with caution when missing data exist. Suggestions are given regarding moderation analysis with missing data.     

两种新的计算机化自适应测验在线标定方法

在线标定技术由于具有诸多优点而被广泛应用于计算机化自适应测验(CAT)的新题标定。Method A是想法最直接、算法最简单的CAT在线标定方法, 但它具有明显的理论缺陷--在标定过程中将能力估计值视为能力真值。将全功能极大似然估计方法(FFMLE)与“利用充分性结果”估计方法(ECSE)的误差校正思路融入Method A (新方法分别记为FFMLE-Method A和ECSE-Method A), 从理论上对能力估计误差进行校正, 进而克服Method A的标定缺陷。模拟研究的结果表明：(1)在大多数实验条件下, 两种新方法较Method A总体上可以改进标定精度, 且在测验长度为10的短测验上的改进幅度最大; (2)当CAT测验长度较短或中等(10或20题)时, 两种新方法的表现与性能最优的MEM已非常接近。当测验长度较长(30题)时, ECSE-Method A的总体表现最好、优于MEM; (3)样本量越大, 各种方法的标定精度越高。     

Profile-likelihood Confidence Intervals in Item Response Theory Models

Confidence intervals (CIs) are fundamental inferential devices which quantify the sampling variability of parameter estimates. In item response theory, CIs have been primarily obtained from large-sample Wald-type approaches based on standard error estimates, derived from the observed or expected information matrix, after parameters have been estimated via maximum likelihood. An alternative approach to constructing CIs is to quantify sampling variability directly from the likelihood function with a technique known as profile-likelihood confidence intervals (PL CIs). In this article, we introduce PL CIs for item response theory models, compare PL CIs to classical large-sample Wald-type CIs, and demonstrate important distinctions among these CIs. CIs are then constructed for parameters directly estimated in the specified model and for transformed parameters which are often obtained post-estimation. Monte Carlo simulation results suggest that PL CIs perform consistently better than Wald-type CIs for both non-transformed and transformed parameters.     

LGM模型中缺失数据处理方法的比较:ML方法与Diggle-Kenward选择模型

追踪研究中缺失数据十分常见。本文通过Monte Carlo模拟研究,考察基于不同前提假设的Diggle-Kenward选择模型和ML方法对增长参数估计精度的差异,并考虑样本量、缺失比例、目标变量分布形态以及不同缺失机制的影响。结果表明:(1)缺失机制对基于MAR的ML方法有较大的影响,在MNAR缺失机制下,基于MAR的ML方法对LGM模型中截距均值和斜率均值的估计不具有稳健性。(2)DiggleKenward选择模型更容易受到目标变量分布偏态程度的影响,样本量与偏态程度存在交互作用,样本量较大时,偏态程度的影响会减弱。而ML方法仅在MNAR机制下轻微受到偏态程度的影响。     

网络广告的心理传播效果及其理论探讨

网络广告现已得到众多商家的重视与青睐。研究发现, 网络广告的心理传播效果明显受到广告自身特点(如网络广告的形式、互动性、情感元素、产品类型)、受众状态(如受众的期待、卷入度、先前经验、性别)、以及网络环境等因素的影响。学者们利用修正的精细加工可能性模型、互动广告模型、网络广告心理效果模型等对该类广告效应进行理论解释。但现有研究也依然存在一些问题, 如影响因素有待拓展、研究方法有待创新、因变量指标有待规范、理论总结不力等。