Publication bias and outcome reporting bias: agomelatine as a case example

Publication bias and outcome reporting bias contribute to distorted perceptions of drug efficacy and the underreporting of adverse events. To demonstrate these biases, this article describes how the clinical profile of the antidepressant agent agomelatine (Valdoxan(?)) has been presented in the literature. Agomelatine has been systematically assessed in 10 short-term placebo-controlled studies and three long-term placebo-controlled relapse prevention studies. Five published trials demonstrated clinically modest but statistically significant benefits over placebo. Five unpublished trials did not find agomelatine more effective than placebo, but in two of these studies the active comparison drug (fluoxetine [Prozac(?)] or paroxetine [Paxil(?)]) was more effective than placebo. Agomelatine was more effective than placebo in one of three relapse prevention studies, but only the positive study was published. Based on what is evident in the entire published and unpublished dataset, agomelatine does not have a tremendously superior sleep and sexual effects profile. The risk of liver toxicity is also not prominently highlighted in the published literature.     

Sentencing outcomes of convicted child sex offenders

This research examines the sentencing outcomes of convicted child sexual offenders from data collected over an eight year period. Multiple regression and nominal log linear regression are used to examine length of prison sentence, length of probation sentence, and whether the convicted offender is actually sent to prison or to probation. While many independent variables appear to be related to sentence outcome, they fall into three categories: characteristics of the offender, characteristics of the victim, and characteristics of the crime. Additionally, while many variables appear related at the bivariate level, when multivariate analysis is applied, fewer variables remain significant, and these are mostly from the characteristics of the offense.     

Responsive Survey Design,Demographic Data Collection,and Models of Demographic Behavior

To address declining response rates and rising data-collection costs, survey methodologists have devised new techniques for using process data (“paradata”) to address nonresponse by altering the survey design dynamically during data collection. We investigate the substantive consequences of responsive survey design—tools that use paradata to improve the representative qualities of surveys and control costs. By improving representation of reluctant respondents, responsive design can change our understanding of the topic being studied. Using the National Survey of Family Growth Cycle 6, we illustrate how responsive survey design can shape both demographic estimates and models of demographic behaviors based on survey data. By juxtaposing measures from regular and responsive data collection phases, we document how special efforts to interview reluctant respondents may affect demographic estimates. Results demonstrate the potential of responsive survey design to change the quality of demographic research based on survey data.     

The NIP graph of a social welfare function

We consider the fraction of pairs of m distinct alternatives on which a social welfare function f may be nondictatorially independent and Pareto when the domain of f satisfies the free k-tuple property. When k = 4 we improve the existing upper bound to . When there are at least 26 alternatives and we obtain an original upper bound, . To obtain these results we define and analyze the graph formed from the nondictatorial independent and Pareto pairs and combine the results of this analysis with known results from extremal graph theory. The authors extend special thanks to the two reviewers and the editor for their comments.     

The Divergent Meanings of Life Satisfaction: Item Response Modeling of the Satisfaction with Life Scale in Greenland and Norway

Cultural differences in response to the Satisfaction With Life Scale (SWLS) items is investigated. Data were fit to a mixed Rasch model in order to identify latent classes of participants in a combined sample of Norwegians (N = 461) and Greenlanders (N = 180). Initial analyses showed no mean difference in life satisfaction between the two subsamples. After transforming the ordinal raw scores into interval scales while simultaneously controlling for response bias, different results appeared. First, approximately 80% of the participants in the Greenlandic subsample fit a latent class with a large degree of random responding to the SWLS. Second, relative to the Norwegians, more Greenlanders were using extreme categories in responding to the SWLS. After statistically controlling for this tendency, Norwegians were in general more satisfied with their lives than Greenlanders. Third, Greenlanders who belonged to one specific latent class were more satisfied than their Norwegian counterparts. A salient feature of this class was the relative unwillingness of respondents to change the circumstances of their lives if they were given such an opportunity. The above results are a reminder of the care that must be used in analyzing survey data across cultures. The analytical strategy applied in the article offers an improved approach to handling such data.     

Remembering the lizard: reconstructing sexuality in the rooms of narcotics anonymous

The crack epidemic was devastating to poor American communities in part because of the destruction wrought by the system of exchanging sex for drugs, which was a key feature of the crack-use culture. Sex-for-drugs exchanges were often conducted under unsafe circumstances and were linked to the spread of AIDS and other STDs, as well as unplanned pregnancies. The existence of this alternative system of sexual relationships threatened the economic viability of established commercial sex work and undermined the status and power of women. Narcotics Anonymous (NA) meetings helped men and women recover from crack addiction through a well-described 12-step process. Described as the rooms, these time- and space-specific encounters helped people become sober in the context of neighborhoods that were centers of the drug trade. Because of the key role of sex in the crack culture, transformation of sexual relationships was essential to establishing and maintaining sobriety. The manner in which the rooms of NA influence the sexuality and lifeworld of addicted people is explored using Barker's theory of ecological psychology.     

Pollsters Under Attack: 2004 Election Incivility and Its Consequences

The 2004 election was remarkable for a number of reasons, includingthe harsh, personal attacks from all parts of the politicalspectrum on a number of media pollsters. The idea of "killingthe messenger" has been around long enough for the phenomenonto have its own name, but it appears to have intensified muchmore than in the past. The article details the experiences oftwo polls and their pollsters, one national and one statewide.These attacks are part of a growing practice of trying to mitigateperceived damage by any message in the political marketplace.The article suggests that while there are positive effects fromthese developments, including heightened awareness among votersof polling methods, negative effects can damage the credibilityof specific polls and their sponsors, as well as the professionin general, including market and other public opinion research.The article ends with a call for researchers to be more openwith their methods and measures, and to strongly defend properlydone research against critics; for journalists to be more discerningin evaluating poll criticisms before publishing them; and forprofessional organizations to help the public better understandpolling, market research, and other public opinion researchand their benefits to society.