Multivariate Analysis Of A Rehabilitation System : Cross Validation And Extension

Cross-validation of previously published findings regarding the structure of rehabilitation data yields substantial evidence that this structure is stable and quite definitive. Tentative data are presented regarding relation- ships between intelligence and personality on the one hand, and rehabilitation outcomes on the other. It appears that intelligence is more related to short- run outcomes; personality more related to long-run outcomes.     

Brief Report: Some Possible Uses Of Factor Analysis In Multivariate Studies

Factor analysis is used in item selection in the hopes of producing a small number of factors each of which will represent a unidimensional sub- scale. If item analysis has been successful in producing truly independent subscales, it might be hoped that the number of factors would equal the number of subscales and that each factor would be highly defined by a single subscale. Factor analysis when used in studies of organization, is not assumed to produce factors that represent unidimensional scales. Rather, factor analysis is used to reveal various substructures that exist within an organization. If several variables are loaded on a single factor, the variables can be regarded as nodes of interaction between measured dimensions of organization.     

Impulsivity Moderates Promotive Environmental Influences on Adolescent Delinquency: A Comparison Across Family, School, and Neighborhood Contexts

The present study examined moderating effects of impulsivity on the relationships between promotive factors from family (family warmth, parental knowledge), school (school connectedness), and neighborhood (neighborhood cohesion) contexts with delinquency using data collected from N?=?2,978 sixth to eighth graders from 16 schools surrounding a major city in the Midwestern United States. More than half of the respondents were non-Caucasian (M _age ?=?12.48; 41.0 % male). Multilevel modeling analyses were conducted to take into account the clustering of the participants within schools. Impulsivity was positively associated with adolescent delinquency. Additionally, family warmth, parental knowledge, and school connectedness, but not neighborhood cohesion, were independently and inversely related to adolescent delinquency. Finally, impulsivity moderated relationships between family warmth and parental knowledge with delinquency but not relationships between school attachment and neighborhood cohesion with delinquency. Specifically, the negative relationship between family warmth and delinquency was significant for adolescents with high levels of, but not for those with below-average levels of, impulsivity. In addition, parental knowledge had a stronger association with decreased levels of delinquency for adolescents reporting higher levels of impulsivity. The moderating effects of impulsivity did not differ for males and females or for minority and non-minority participants. Findings indicate that impulsivity may have greater impact on adolescents’ susceptibility to positive family influences than on their susceptibility to promotive factors from school or neighborhood contexts. Implications for future research and practice are discussed.     

Multivariate Analysis of Dichotomous Variables: A General Method

A method for extimating both additive and interactive effects of several independent dichotomous variables on a dichotomous criterion is proposed and related to explicit causal assumptions. Parameters of the stochastic model are estimated by observed conditional probabilities, a procedure that maximally reduces variance.     

Race,Gender, and Perceptions of Peer Delinquency: A Within-Subject Analysis

ABSTRACT

We generated a within-subject analysis that assessed whether changes in delinquency from the same youth were associated with changes in the degree to which the youth perceived the delinquency of their peers across three waves of data. We hypothesized that the effect of within-person changes in perceptions of peer delinquency on changes in delinquency across time should be greatest among young black males. The results revealed that a within-person change in the youths’ perceptions that their peers were engaging in greater delinquency increased offending across time. They also showed that young black males were more likely than young white males and young black and white females to increase their level of delinquency, within-person and across time, when they perceived an increase in their peers’ levels of delinquency. We conclude that scholars may generate deeper theoretical insights and policies that are more efficacious when they reveal the causes of crime that are universal and those that uniquely affect a specific group.     

What I Was Doing Versus What I Did: Verb Aspect Influences Memory and Future Actions

ABSTRACT— This research examined whether describing past actions as ongoing using the imperfective aspect (rather than describing them as completed using the perfective aspect) promotes memory for action-relevant knowledge and reenactment of these actions in a future context. In Experiment 1 , participants who used the imperfective aspect to describe their strategy on a prior interpersonal task were more likely to use this strategy on a later task than were participants who used the perfective aspect to describe their prior strategy. Experiment 2 demonstrated that describing behaviors on a task using the imperfective rather than the perfective aspect increased willingness to resume that task by improving memory for task contents. The last two experiments showed that the effects of the imperfective aspect on memory decayed over time and that the imperfective aspect facilitated performance of a future behavior only when the described past behavior was relevant to the future behavior. Thus, the effects of aspect are moderated by memory decay and are behavior-specific.     

Gender and Suicide Risk Among Artists: A Multivariate Analysis

Research on mental disorders among male artists has suggested that artists are at risk of suicide. However, given that men are higher in suicide risk than women, the presumed suicide risk of artists may be an artifact of sampling bias. A logistic regression analysis of data from 21 states finds that artists have a 270% higher risk of suicide than nonartists. However, after controlling for gender and sociodemographic variables, this risk level is reduced to 125%. The findings are related to both psychiatric and work-related stress factors that may place artists at risk of suicide as an occupational group.     

领导干部结构化面试信度的多元概括化理论分析

本研究尝试运用多元概括化理论对北京市某区副处级干部准入资格结构化面试测评数据进行测量信度分析,为提高领导干部考试与测评工作科学化水平提供了有益的实证依据。主要结论有:(1)本次结构化面试难度适中,区分度较高;(2)各测评要素及合成分数的类信度系数均较高,合成分数的测量信度高于单个测评要素的测量信度;(3)各测评要素及合成分数的类信度系数随着考官数量的增加而增加,且从确保信度和降低成本考虑,考官数量以5-9位为宜;(4)在这次面试测评中,各项测评要素间的相关系数较高,这为目前在选拔面试中将各项测评要素得分进行合成提供了依据,说明用合成分数计算总分具有一定的合理性。     

Factor Analytic Studies Of Human Brain Damage: I. First And Second-Order Factors And Their Brain Correlates

This report is concerned with identifying the first and second-order cognitive factors underlying a battery of 49 measures taken from 22 brain damage tests. The test scores from 176 brain damaged patients between 16 and 65 years of age were intercorrelated and subjected to first-order alpha factoring followed by promax rotation to oblique simple structure. Ten of the 13 first-order factors extracted were interpretable, six of them being perceptual in nature and four being of a more conceptual nature. The perceptual factors include: perceptual organization, perceptual-motor speed, pattern recognition, temporal resolution, spatial orientation, and figure-ground identification. The conceptual factors include verbal comprehension, memory, and two abstraction factors. A second-order alpha factoring was performed on the matrix of correlations among the 13 primaries. Three of the five second-order factors extracted were interpretable. They were identified as perceptual integration (subsuming the first-order factors of perceptual organization, perceptual-motor speed, and temporal resolution), verbal memory (subsuming verbal comprehension and memory), and visualization (subsuming spatial orientation and figure-ground identification). Although factor interpretations were based primarily on the patterns of high loading variables, they were also influenced by lesion effects observed in this and related studies. About half the interpretable factors are relatively localized (i.e., confined to one or two lobes of one hemisphere), with the other half more diffuse (i.e., multi-lobed, combined with laterality or bilaterality). The more localized factors include the right hemisphere factors of perceptual-motor speed, temporal resolution, and spatial orientation, and the left hemisphere factors of verbal comprehension, memory, and verbal memory. The more neurally diffuse factors include the second-order factors and such broad gauged first-order factors as abstraction I and II, and pattern recognition. Furthermore, same lobe, bilaterally hemispheric effects were rare, and only four factors (memory, verbal memory, visualization, and abstraction I) were correlated with sub-cortical lesions.     

Black Women's Body Image: An Analysis of Culture-Specific Influences

A review of the literature on racial differences in body image in Black and White women indicates body image dissatisfaction is more common in White females than Black females. However, Black women are not protected from body image issues. Research on body image concerns specifically related to Black women is limited. The present study involves focus group and individual interviews conducted with 16 Black women. Data were analyzed using qualitative methodology. Results indicate 6 major themes, with shapely and curvaceous ideal body image standards within the Black community as the most prevalent theme across data sources. Therapeutic implications emphasize recognition of culture-specific ideal standards of beauty, as well as the influence of external factors such as men, family, peers, and media on Black women's self-evaluations.     

Job Performance as Multivariate Dynamic Criteria: Experience Sampling and Multiway Component Analysis

Questions about the dynamic processes that drive behavior at work have been the focus of increasing attention in recent years. Models describing behavior at work and research on momentary behavior indicate that substantial variation exists within individuals. This article examines the rationale behind this body of work and explores a method of analyzing momentary work behavior using experience sampling methods. The article also examines a previously unused set of methods for analyzing data produced by experience sampling. These methods are known collectively as multiway component analysis. Two archetypal techniques of multimode factor analysis, the Parallel factor analysis and the Tucker3 models, are used to analyze data from Miner, Glomb, and Hulin's (2010) experience sampling study of work behavior. The efficacy of these techniques for analyzing experience sampling data is discussed as are the substantive multimode component models obtained.     

Changes in Predictors of Gender Role Ideologies Among Women: A Multivariate Analysis

Using data from the National Opinion ResearchCenter's General Social Surveys for 1974-1994, weanalyze changes in gender role ideology among women inthe U.S. We focus on women tied to previous research indicating increasing labor force participationamong various groups of women such as those withpreschool children and those of Hispanic origin, and thepurported egalitarian impact of labor forceparticipation on gender role attitudes. The General SocialSurveys include 9833 women who responded to the genderrole questions, with 1132 black (11.5%) and 351 Hispanic(3.6%) women. Our results refine earlier bivariate findings about the impact of specific types oflife experiences and family power dynamics on women'sviews about appropriate roles for men and women. Mostimportantly, we find that there may be an ubiquitous shift toward more egalitarian gender roleattitudes among women in the U.S. regardless of theirindividual characteristics and circumstances.