Estimating efficacy in a proposed randomized trial with initial and later non-compliance

Summary.  A controversial topic in obstetrics is the effect of walking on the probability of Caesarean section among women in labour. A major reason for the controversy is the presence of non-compliance that complicates the estimation of efficacy, the effect of treatment received on outcome. The intent-to-treat method does not estimate efficacy, and estimates of efficacy that are based directly on treatment received may be biased because they are not protected by randomization. However, when non-compliance occurs immediately after randomization, the use of a potential outcomes model with reasonable assumptions has made it possible to estimate efficacy and still to retain the benefits of randomization to avoid selection bias. In this obstetrics application, non-compliance occurs initially and later in one arm. Consequently some parameters cannot be uniquely estimated without making strong assumptions. This difficulty is circumvented by a new study design involving an additional randomization group and a novel potential outcomes model (principal stratification).     

Organizational Change,Politics, and the Official Statistics of Punishment

Most organizational researchers characterize politics as a force organizations experience from outside and administration as the practices managers adopt in response to the uncertain environment. To assess that approach, this paper examines a crisis that changed California criminal justice agencies' administrative practices and their communication of statistical information to outsiders. In 1976, managers and wardens supported conservative reform bills that effectively ended professional criminal rehabilitation and quantitative evaluation studies, and they shifted administration toward bureaucratic controls, using routine population counts. Changes in agency statistical activities clarify how members of government organizations act politically to maintain or change administrative practices.     

Attachment Quality and Post-treatment Functioning Following Sexual Trauma in Young Adolescents: A Case Series Presentation

Children and adolescents who suffer sexual abuse evidence symptoms that may persist into adulthood. Attachment theory may lend some insight into the psychological sequelae following sexual trauma. A clinical case series presents four sexually abused young adolescents from the first author's clinical caseload. Each case is rated regarding the quality of attachment, the severity of trauma, and the child's functioning at the time of follow-up. Ratings are made by the first author and qualified by a second clinician who is familiar with each case. Findings suggest that those youngsters who had the benefit of a strong and secure attachment evidence more positive outcome at follow-up.     

Influences on the effectiveness of peer interaction: children's level of cognitive development and the relative ability of partners

Studies of peer interaction among children have generally shown that this experience aids performance, or at the very least does not result in poorer performance in relation to control conditions. Furthermore, working with a more able partner has been found to be a particularly effective form of peer interaction. In contrast, a model proposed by Karmiloff-Smith suggests that at certain phases of cognitive development children may ignore feedback and information from task activities. These two ideas were tested in a study where pairs of 6–7 year old children had to balance a beam on a fulcrum during a computer task. The performance on a post-test did not support either viewpoint. Children who were working with a more able partner were found to perform significantly worse than other children. The findings of a second study suggested that this effect may have been due to the more able children having an incomplete understanding of the computer task and they may have dominated social interaction thereby restricting the progress of the less able pupil. The findings from these two studies indicate that peer interaction can result in poorer learning outcomes, and that Karmiloff-Smith's model should include the possibility of peer interaction effects.     

To err is social: Network prominence and its effects on self-estimation

In studying professional productivity, researchers find that actors do not always accurately perceive their productivity levels. Yet the literature does not provide an adequate explanation as to why these errors occur. This study examines the effects of network prominence on the self-estimation of productivity. I contend that network prominence is indicated through a variety of dimensions—visibility, connectedness, and professional affiliation—each of which has varying influence on the self-estimation process. This influence is best understood by exploring actors' internal vs. external attributions of prominence. Specifically, when actors attribute their prominence to internal efforts, they will tend to overestimate their productivity. Conversely, when actors attribute their network prominence to external circumstances, they will underestimate their productivity. The distinction between internal and external attributions of prominence relies, in large part, on the process of social comparison.     

For love or money: Costs of child care by relatives

A substantial proportion of employed mothers of young children, especially low-income mothers, use relatives to provide child care. This study uses data on interhousehold exchange of gifts, loans, and household services from the National Survey of Families and Households to examine monetary and nonmonetary costs of child care by relatives. Results show that mothers who use relatives for child care are more likely to give services and to have given gifts or loans to other relatives living outside the household than mothers using other forms of child care or mothers who are not employed. Monetary payments for child care by relatives are made more often for full-time than part-time care and less often to grandparents than to other relatives providing care. Implications for government child care assistance programs are discussed. Preparation of this paper was supported in part by the Illinois Agricultural Experiment Station. The author acknowledges the helpful comments of Marianne Ferber, Andrea Beller, and Sharon Y. Nickols. Her research interests include issues relating to women's employment, including child care and house-hold time allocation. She received her Ph.D. in Family and Consumption Economics from the University of Illinois. Send all correspondence to author.     

Factory women,redundancy and the search for work: toward a reconceptualisation of employment and unemployment

A number of social scientists (see for example, Allen et al., 1986; Brown, 1984; Purcell, 1985; Purcell et al., 1986; Roberts et al., 1985) have drawn attention to the way in which patterns of working have changed radically in recent years suggesting that we need to rethink and reconceptualise with regard to work, employment and unemployment. This paper examines these issues in Sweden where recession in the early eighties and economic restructuring, similar to the experience in other Western countries, have affected individuals’relation to the labour market. It is our view that the effects of these developments (have) affect(ed) women and immigrants differently. It is furthermore our contention that welfare state policies related to the functioning of the labour market (inadvertently) play a role in the maintenance of inequality.     

Adjusted variable plots for Cox's proportional hazards regression model

Adjusted variable plots are useful in linear regression for outlier detection and for qualitative evaluation of the fit of a model. In this paper, we extend adjusted variable plots to Cox's proportional hazards model for possibly censored survival data. We propose three different plots: a risk level adjusted variable (RLAV) plot in which each observation in each risk set appears, a subject level adjusted variable (SLAV) plot in which each subject is represented by one point, and an event level adjusted variable (ELAV) plot in which the entire risk set at each failure event is represented by a single point. The latter two plots are derived from the RLAV by combining multiple points. In each point, the regression coefficient and standard error from a Cox proportional hazards regression is obtained by a simple linear regression through the origin fit to the coordinates of the pictured points. The plots are illustrated with a reanalysis of a dataset of 65 patients with multiple myeloma.     

MASTER FRAMES, POLITICAL OPPORTUNITIES, AND SELF-DETERMINATION:c The Åland Islands in the Post-WWI Period

While the literature on master frames has drawn attention to the crucial role of ideas in cycles of protest, reliance on the creation of frame resonance to account for the success or failure of a social movement within a cycle can be problematic. Applying propositions adapted from McAdam, McCarthy, and Zald (1996), this article traces how political opportunities interacted with framing processes during the emergence and development of the Åland Islands secessionist movement of the post-WWI period. The Ålanders aligned their claims with early representations of the "selfdetermination master frame" that underlay the cycle of protest that emerged after the war in a way that resonated with the Allied leaders adjudicating their case. Shifts in political conditions, however, helped to foster an intense "framing contest" among contenders that in the end undermined the Ålanders' representation of the master frame and their ability to achieve desired ends. Although the case reveals certain shortcomings in the propositions, they nevertheless provide a useful starting point when documenting the complex interplay of political conditions and framing processes in an instance of collective action.