Heterogeneous factor analysis models: A bayesian approach

Multilevel factor analysis models are widely used in the social sciences to account for heterogeneity in mean structures. In this paper we extend previous work on multilevel models to account for general forms of heterogeneity in confirmatory factor analysis models. We specify various models of mean and covariance heterogeneity in confirmatory factor analysis and develop Markov Chain Monte Carlo (MCMC) procedures to perform Bayesian inference, model checking, and model comparison.We test our methodology using synthetic data and data from a consumption emotion study. The results from synthetic data show that our Bayesian model perform well in recovering the true parameters and selecting the appropriate model. More importantly, the results clearly illustrate the consequences of ignoring heterogeneity. Specifically, we find that ignoring heterogeneity can lead to sign reversals of the factor covariances, inflation of factor variances and underappreciation of uncertainty in parameter estimates. The results from the emotion study show that subjects vary both in means and covariances. Thus traditional psychometric methods cannot fully capture the heterogeneity in our data.     

A Defining Presidential Moment: 9/11 and the Rally Effect

Public approval ratings of George W. Bush surged after the 11 September 2001 terrorist attacks. This study used a quasi–experimental, within–respondents design to investigate the relative contribution of five factors to this classic rally effect: the stimulus event itself, Bush's speech that evening, media exposure, partisan support, and gender effects. Respondents were pretested on the morning of the attacks; one group was posttested immediately after the speech, another group 41 hours later. Stability of effects was examined through an additional study of Bush's 20 September 2001 speech to a joint session of Congress. The findings indicate that Bush's 11 September speech was the critical factor in this rally effect; none of the other factors contributed significantly.     

Cognitive Skills as Predictor of Attitudes Toward Political Conflict: A Study of Polish Politicians

The relationship between cognitive development and attitudes toward a current political conflict was examined in a sample of 46 Polish politicians. This relationship was examined in a control group in a neutral condition and in an experimental group after participants were presented with a hostile attack on their position on the conflict. Politicians with less advanced cognitive skills tended to use competitive attitudes in both conditions. In contrast, those who possessed more advanced skills tended to use cooperative attitudes in the neutral situation; after the emotional attack, they tended to avoid further involvement in the conflict and sought to exit it. The key difference in participants' cognitive functioning was their ability to differentiate perspectives and to transcend their own point of view in the conflict.     

Authoritarianism and Political Impoverishment: Deficits in Knowledge and Civic Disinterest

Past research shows that authoritarian individuals hold strong opinions about a variety of political and social issues, such as race relations and military conflict. What has not been established, though, is the amount of general political knowledge that authoritarians possess. In this study, three groups of college students were administered Altemeyer's Right-Wing Authoritarianism (RWA) scale; most of them also received items assessing general political knowledge and specific knowledge about the 2000 presidential election, as well as items assessing interest in politics. Relative to students with low RWA scores, those with high scores possessed less political knowledge; moreover, they expressed less interest in learning about politics. In general, authoritarianism was unrelated to how individuals got their political information or how credible they found their sources. The implication that authoritarians hold strong attitudinal beliefs with weak political knowledge is discussed.     

Confirming the Factor Structure of Attention Deficit Hyperactivity Disorder Symptoms in Adult, Nonclinical Samples

Three separate models have been proposed to describe the factor structure of attention deficit hyperactivity disorder (ADHD) in the past 20 years. The Diagnostic and Statistical Manual of Mental Disorders-III (DSM-III, 1980) proposed 3 separate factors of inattention, hyperactivity, and impulsivity. The DSM-III-R (1987) proposed a single factor. The DSM-IV (1994) described the disorder as having 2 factors: hyperactivity/impulsivity and inattention. The following 2 studies employed confirmatory factor analysis to compare each of these models and 3 alternative models. University students completed questionnaires that assessed each of the 18 symptoms listed in the DSM-IV for ADHD. The 3-factor model fit the data significantly better than each of the other models in both studies. These findings suggest that a 3-factor model of inattention, hyperactivity, and impulsivity describes adult ADHD symptoms better than current alternatives.     

Gratitude and Good Government

I attempt to show that it is notphilosophically incompetent to ground politicalobligation in feelings of gratitude. But theargument needs to be stated carefully.Gratitude must be distinguished fromreciprocity. It applies only to good governmentwhich provides benefits to citizens for whichthey ought to feel grateful. It applies only tocitizens who accept that their feelings ofgratitude are properly demonstrated by anacceptance on their part of the duties ofcitizenship. It does not apply to citizenswhose benefits are purchased at the expense ofthe unjust treatment of fellow citizens.     

Is Political Cognition Like Riding a Bicycle? How Cognitive Neuroscience Can Inform Research on Political Thinking

Our understanding of political phenomena, including political attitudes and sophistication, can be enriched by incorporating the theories and tools of cognitive neuroscience—in particular, the cognitive neuroscience of nonconscious habitual cognition (akin to bicycle riding). From this perspective, different types of informational "building blocks" can be construed from which different types of political attitudes may arise. A reflection-reflexion model is presented that describes how these blocks combine to produce a given political attitude as a function of goals, primes, expertise, and inherent conflict in considerations relevant to the attitude. The ways in which neuroimaging methods can be used to test hypotheses of political cognition are reviewed.     

Mars and Venus at Twilight: A Critical Investigation of Moralism, Age Effects, and Sex Differences

Analysts have long sought to understand whether women and men have different ethical orientations. Some researchers have argued that women and men consistently make fundamentally different ethical judgments, especially of corruption; others have found no such disparities. This study considered whether an individual's age may also play a role in determining his or her moral judgment. A statistical investigation of interactive effects between gender and age in a nationally representative data set from Japan shows that this interaction functions better as a predictor of moralism than do education or gender alone. Older individuals of both sexes were found to have similar strict moral perceptions; as women and men age, their ethical judgments converge.     

儿童早期数学认知能力的结构及其特点

从北京市10所幼儿园中选取234名3、4岁的儿童为被试,采用个别测查的方法对儿童早期数学认知能力的结构及其特点进行了考察。经验证性因素分析发现：(1)儿童早期数学认知能力的结构模型是合理的,可接受的,具有较好的构想效度,具体讲,数、计算、测量、空间／几何和模式认知能力五个维度共同解释着儿童早期的数学认知能力;(2)不同年龄儿童早期数学认知能力的结构具有稳定性,但是结构模型并不完全一致,某些项目的解释率有所不同;(3)男、女儿童早期数学认知能力的结构模型具有一致性。     

Validation of the Multidimensional Blood/Injury Phobia Inventory: Evidence for a Unitary Construct

The Multidimensional Blood/Injury Phobia Inventory (MBPI) was developed from a theoretical framework to characterize a broad range of feared stimuli and phobic reactions associated with this pathology. The MBPI consists of 40 items that cross 4 types of stimulus content (injections, hospitals, blood, injury), 5 types of phobic responses (fear, avoidance, worry, fainting, disgust), and a self versus other focus. This study reports on administration of the MBPI to 558 undergraduates, 9 of whom had blood/injury phobia. The instrument had a Cronbach's alpha of .91 and demonstrated good concurrent validity, convergent validity, and discriminant validity. One large factor emerged in an unrotated principal components analysis, suggesting that blood/injury phobia is a unitary psychometric construct. Exploratory factor analyses revealed a 6-factor solution defined by stimulus content domains and fainting, each of which may be important to consider clinically when assessing the unique concerns of treatment-seeking individuals.