Local model uncertainty and incomplete-data bias (with discussion)

Summary.  Problems of the analysis of data with incomplete observations are all too familiar in statistics. They are doubly difficult if we are also uncertain about the choice of model. We propose a general formulation for the discussion of such problems and develop approximations to the resulting bias of maximum likelihood estimates on the assumption that model departures are small. Loss of efficiency in parameter estimation due to incompleteness in the data has a dual interpretation: the increase in variance when an assumed model is correct; the bias in estimation when the model is incorrect. Examples include non-ignorable missing data, hidden confounders in observational studies and publication bias in meta-analysis. Doubling variances before calculating confidence intervals or test statistics is suggested as a crude way of addressing the possibility of undetectably small departures from the model. The problem of assessing the risk of lung cancer from passive smoking is used as a motivating example.     

Getting the offer: Sex discrimination in hiring

The hiring process is currently the least understood aspect of the employment relationship. It may be the most important for understanding the broad processes of stratification with allocation of demographic groups to jobs and firms. The lack of knowledge is due to difficulty of assembling data on the processes that occur at the point of hire. Against this background we analyze data on all applicants to positions in one of the largest Scandinavian banks in 1997–1998, providing what we believe to be the first study using applicant pool data and information about extended offers in a private-sector European firm, adding to the record of about half a dozen such U.S. studies. The hiring agents in the organization are fully conscious and concerned about the nonconscious biases and gender schemas they carry when making hiring decisions. Their effects on hiring are considered to be beyond dispute: women are at a clear disadvantage. For actual hiring practices we found that the opposite is true: women are not at a disadvantage and may even be at an advantage in getting offers. Two organizational practices may lead to female advantage. The hiring agents had been educated about the role of nonconscious biases, which perhaps mitigated their effects. But they had also been instructed to search actively for qualified females in the applicant pool. With no qualified females in the first pass, they go through the pool a second and third time hoping to find one. We discuss reasons why the interpretations and meanings the hiring agents attribute to the hiring process are at odds with what actually occurs.     

行为金融学理论述评

20世纪80年代以来,大量的金融研究与市场异象对传统资本市场理论提出了挑战。行为金融理论在整合多学科研究成果,特别是心理学研究的基础上,从投资者的实际决策行为出发,分析与解释金融市场。通过文献研究,文章回顾了有关行为金融学的基本理论与研究,包括前景理论及其他相关理论的发展。此外,对国内有关行为金融学研究特点进行总结分析,并对其今后研究发展提出了设想。     

通用性指标偏好及其实验研究——平衡计分卡在多部门企业中的应用

文章分析了通用性指标偏好产生的原因、影响和解决途径,并通过实验研究,考察了中国实验者在平衡计分卡应用中是否也存在通用性指标偏好,以及集体决策方式对通用性偏好程度的影响。决策者组成的多元化,是集体决策发挥作用的重要因素,这一结论对于部门平衡计分卡在现实中的应用和企业绩效评价实践具有启示作用。     

Small-sample results for maximum-likelihood estimation in branching processes

In this article, small sample properties of the maximum-likelihood estimator (m.l.e.) for the offspring distribution (pk) and its mean m are considered in the context of the simple branching process. A representation theorem is given for the m.l.e. of (Pk) from which the m.l.e. of m is obtained. The case where p0 + p1 + p2 = 1 is studied in detail: numerical results are given for the exact bias of these estimators as a function of the age of the process; a curve fitting analysis expresses the bias of m? as a function of the mean and the variance of the offspring distribution and finally an “approximate m.l.e.” for (pk) is given.     

Assessing the Reliability and Validity of the Jump$tart Survey of Financial Literacy

Financial education represents an area of popular interest, owing largely to the Jump$tart surveys of financial literacy. However, while the surveys represent indicators of financial knowledge among high school seniors, these measures have not been statistically validated. This article describes an assessment of the surveys reliability (internal consistency), and validity. It reports a moderately high degree of consistency overall, however, discloses low to moderate internal consistencies among subscales. It also finds significant response differences to one quarter of comparable items between surveys. The researcher observes challenges to affirming the surveys validity and offers statistics suggesting social bias among survey items. He calls for further research into measures of financial literacy.The author appreciates the guidance and support of Dr. Duane Giannangelo, Dr. Jean Steiz, Dr. Lou Franceschini, and Dr. Steven Ross.     

LOGISTIC REGRESSION COMPARED TO NORMAL DISCRIMINATION FOR NON-NORMAL POPULATIONS'

An investigation is undertaken of the logistic regression procedure for estimating the posterior probability of an object belonging to one of two populations. The asymptotic bias and mean square error associated with the procedure are derived for univariate populations whose distributions satisfy the general Day-Kerridge model for which the logistic form is valid for the posterior probability. These properties are compared with those of the normal discrimination method based on the classical assumption of normal populations with common variances. The asymptotic relative efficiency of logistic regression is considered on the basis of asymptotic mean square error.     

On making causal claims: A review and recommendations

Social scientists often estimate models from correlational data, where the independent variable has not been exogenously manipulated; they also make implicit or explicit causal claims based on these models. When can these claims be made? We answer this question by first discussing design and estimation conditions under which model estimates can be interpreted, using the randomized experiment as the gold standard. We show how endogeneity – which includes omitted variables, omitted selection, simultaneity, common-method variance, and measurement error – renders estimates causally uninterpretable. Second, we present methods that allow researchers to test causal claims in situations where randomization is not possible or when causal interpretation could be confounded; these methods include fixed-effects panel, sample selection, instrumental variable, regression discontinuity, and difference-in-differences models. Third, we take stock of the methodological rigor with which causal claims are being made in a social sciences discipline by reviewing a representative sample of 110 articles on leadership published in the previous 10 years in top-tier journals. Our key finding is that researchers fail to address at least 66% and up to 90% of design and estimation conditions that make causal claims invalid. We conclude by offering 10 suggestions on how to improve non-experimental research.     

社会支配取向量表在中国的信度和效度研究

本研究尝试修订西方的社会支配取向量表.359名大学本科生参与了研究.结果表明,社会支配取向量表修订后包含14个项目,量表具有较好的内部一致性、重测信度和一定的预测效度;中国被试社会支配取向得分普遍比西方的高;中文修订量表不同于西方的单因素或者双因素结构,它包括4个因素,即"反对群体平等"、"赞同优势群体的支配性"、"赞同劣势群体的较低地位"、"赞同维持等级差异".     

民企竞争贷款过程中行为扭曲的经济学分析

现有的融资制度下,国有企业相对于民营企业具有融资上的制度禀赋优势,这导致了民营企业在与国有企业竞争贷款资源过程中的扭曲行为。本文运用两期模型证明了这一命题,并且认为,如果不提高贷款机构对借款人使用贷款行为的监督能力,单纯强调政企分开并不能解决民营企业竞争贷款过程中的行为扭曲问题。