Component selection norms for principal components regression

Multicollinearity or near exact linear dependence among the vectors of regressor variables in a multiple linear regression analysis can have important effects on the quality of least squares parameter estimates. One frequently suggested approach for these problems is principal components regression. This paper investigates alternative variable selection procedures and their implications for such an analysis.     

The application of generalized ridge and stein-like general minimax rules to multicollinear data

We consider the use of minimax shrinkage estimators for the linear regression mcjel under several loss functions when severe multicollinearity is present. The examples considered illustrate that little or no departure from the least squares estimates is permitted in many cases when the data is highly multicollinear and/or shrinkage is toward a point in the parameter space that does not closely agree with the sample data     

Gender-Specificity of Women's and Men's Self-Reported Attention to Sexual Stimuli

Men's sexual arousal is largely dependent on the actor's gender in a sexual stimulus (gender-specific), whereas for women, particularly androphilic women, arousal is less dependent on gender (gender-nonspecific). According to information-processing models of sexual response, sexual arousal requires that attention be directed toward sexual cues. We evaluated whether men's and women's self-reported attention to sexual stimuli of men or women were consistent with genital responses and self-reported arousal. We presented gynephilic men (n = 21) and women (n = 22) and androphilic men (n = 16) and women (n = 33) with audiovisual stimuli depicting men or women engaged in sexual activities. Genital responses were continuously recorded and, following each stimulus, participants reported the amount of attention paid to the video and feelings of sexual arousal. Self-reported attention was gender-specific for men and gender-nonspecific for women, and generally mirrored genital responses and self-reported arousal. Gender-specificity of genital responses significantly predicted gender-specificity of self-reported arousal; however, for men only, this effect was significantly mediated by gender-specificity of self-reported attention. Gender differences in gender-specificity of sexual arousal may be partially accounted for by differences in gender-specificity of self-reported attention, although attention may play a greater role in men's sexual arousal than women's.     

<Emphasis Type="Italic">Si,Se Puede</Emphasis>: Organizing Latino Immigrant Workers in South Omaha’s Meatpacking Industry

Faced with declining union membership and a growing immigrant workforce, the US labor movement has started to realize the importance of organizing immigrant workers. Yet the conventional wisdom among many within the movement is that immigrant workers are “unorganizable.” Based on a case study of a collaborative effort between the United Food and Commercial Workers Union and Omaha Together, One Community to organize an estimated 4,000 Latino immigrant meatpacking workers, I demonstrate not only the “organizability” of immigrant workers, but also the fact that they have been organizing themselves, with the help of a community-based organization, in the absence of union efforts. This case study suggests that in order to facilitate successful organizing campaigns among immigrant workers, unions need to reach out to community-based organizations and institutions that have established relationships with immigrant workers.

The Importance of Sex in the Lives of Women Living with HIV: A Critical Quantitative Analysis

The authors explored the importance of sex for 1,289 women living with HIV in Canada. Approximately half of women viewed sex as “very” (19.6%) or “somewhat” important (32.3%) and the remaining reported “neither important or unimportant” (22.0%), “somewhat unimportant” (5.4%), or “not at all important” (20.1%). Women who had a regular sex partner, identified as African, Caribbean, or Black, were more educated, believed HIV treatment prevents transmission, or had better physical health-related quality-of-life reported greater importance of sex, whereas those who were older, used illicit drugs, or experienced violence in adulthood reported lesser importance. Findings underscore the diversity of women's perspectives within the context of their lives.     

Work Activation Regimes and Well‐being of Unemployed People: Rhetoric,Risk and Reality of Quasi‐Marketization in the UK Work Programme

Well‐being and employment activation have become central and intertwined policy priorities across advanced economies, with the mandation of unemployed claimants towards employability interventions (e.g. curriculum vitae preparation and interview skills). Compelled job search and job transitions are in part justified by the well‐being gains that resulting employment is said to deliver. However, this dominant focus within the activation field on outcome well‐being – the well‐being improvement triggered by a transition to paid work – neglects how participation in activation schemes can itself affect well‐being levels for unemployed people – what we term ‘process well‐being’ effects. Combining theoretical literature with empirical work on the UK's large‐scale quasi‐marketized Work Programme activation scheme, we develop the limited existing academic discussion of process well‐being effects, considering whether and how activation participation mediates the negative well‐being effects of unemployment, irrespective of any employment outcomes. We further relate variation in such process well‐being effects to the literature on activation typologies, in which ‘thinner’ work‐first activation interventions are linked to weaker process well‐being effects for participants compared to ‘thicker’ human capital development interventions. Confirming these expectations, our empirical work shows that Work Programme participants have, to date, experienced a largely ‘thin’ activation regime in which participants are both expected to, and empirically demonstrate, similar if not lower levels of process well‐being than those who are openly unemployed. These concerning findings speak to all nations seeking to promote the well‐being of unemployed people and particularly those perusing ‘black box’ activation schemes based around quasi‐marketization, devolution and New Public Management.