An analytic method for randomized trials with informative censoring: Part 1

Consider a randomized trial in which time to the occurrence of a particular disease, say pneumocystis pneumonia in an AIDS trial or breast cancer in a mammographic screening trial, is the failure time of primary interest. Suppose that time to disease is subject to informative censoring by the minimum of time to death, loss to and end of follow-up. In such a trial, the censoring time is observed for all study subjects, including failures. In the presence of informative censoring, it is not possible to consistently estimate the effect of treatment on time to disease without imposing additional non-identifiable assumptions. The goals of this paper are to specify two non-identifiable assumptions that allow one to test for and estimate an effect of treatment on time to disease in the presence of informative censoring. In a companion paper (Robins, 1995), we provide consistent and reasonably efficient semiparametric estimators for the treatment effect under these assumptions. In this paper we largely restrict attention to testing. We propose tests that, like standard weighted-log-rank tests, are asymptotically distribution-free -level tests under the null hypothesis of no causal effect of treatment on time to disease whenever the censoring and failure distributions are conditionally independent given treatment arm. However, our tests remain asymptotically distribution-free -level tests in the presence of informative censoring provided either of our assumptions are true. In contrast, a weighted log-rank test will be an -level test in the presence of informative censoring only if (1) one of our two non-identifiable assumptions hold, and (2) the distribution of time to censoring is the same in the two treatment arms. We also extend our methods to studies of the effect of a treatment on the evolution over time of the mean of a repeated measures outcome, such as CD-4 count.     

"SECRET BALLOTS" AND SELF-REPORTS IN AN EXIT-POLL EXPERIMENT

A controlled exit-poll experiment on Election Day (November3, 1992) shows that refusals to answer questions and other evasiveforms of responding were significantly lower when respondentswere given a self-administered "secret-ballot" questionnairethan when they were interviewed face-to-face. While there weresome suggestive interactions of this mode-of-data collectioneffect with age and sex, and with the partisan atmosphere ofthe precinct in which the interviews were conducted, they wereborderline in significance, inconsistent in pattern, or difficuitto interpret. More important, comparisons with official electionreturns (the truth) indicated that the secret-ballot techniquewas more accurate than were face-to-face interviews in estimatingthe final outcome on the most socially sensitive of three self-reportedvotes: a vote for or against a tax levy for elderly services.Using a genuine secret ballot in the tradition of the olderGallup preelection polls can thus increase the validity of self-reportsin exit polls today.     

An evaluation of population projection errors for census tracts

"In this article we evaluate the accuracy and bias of projections of total population and population by age group for census tracts in three counties in Florida. We use [U.S. census] data from 1970 and 1980 and several simple extrapolation techniques to produce projections for 1990; we then compare these projections with 1990 census counts and evaluate the differences. For the total sample, we find mean absolute errors of 17%-20% for the three most accurate techniques for projecting total population and find no indication of overall bias. For individual age groups, mean absolute errors range from 20%-29%." This is a revised version of a paper presented at the 1993 Annual Meeting of the Population Association of America.     

Implementing partial least squares

Partial least squares (PLS) regression has been proposed as an alternative regression technique to more traditional approaches such as principal components regression and ridge regression. A number of algorithms have appeared in the literature which have been shown to be equivalent. Someone wishing to implement PLS regression in a programming language or within a statistical package must choose which algorithm to use. We investigate the implementation of univariate PLS algorithms within FORTRAN and the Matlab (1993) and Splus (1992) environments, comparing theoretical measures of execution speed based on flop counts with their observed execution times. We also comment on the ease with which the algorithms may be implemented in the different environments. Finally, we investigate the merits of using the orthogonal invariance of PLS regression to improve the algorithms.     

A simple,automatic and adaptive bivariate density estimator based on conditional densities

The standard approach to non-parametric bivariate density estimation is to use a kernel density estimator. Practical performance of this estimator is hindered by the fact that the estimator is not adaptive (in the sense that the level of smoothing is not sensitive to local properties of the density). In this paper a simple, automatic and adaptive bivariate density estimator is proposed based on the estimation of marginal and conditional densities. Asymptotic properties of the estimator are examined, and guidance to practical application of the method is given. Application to two examples illustrates the usefulness of the estimator as an exploratory tool, particularly in situations where the local behaviour of the density varies widely. The proposed estimator is also appropriate for use as a pilot estimate for an adaptive kernel estimate, since it is relatively inexpensive to calculate.     

Applied sociology? Even musicians give concerts

Applied sociology is basically what sociologists do for nonsociologists and sometimes for themselves. Applied sociology includes the teaching of sociology as one of the liberating arts and sciences. It also involves the practice of sociology outside academia in the public and private sectors. Either way, applied sociology needs support groups, and state sociological associations need useful things to do beyond their traditional interests in academic teaching and research. Professional sociological associations, and especially those that serve at the local, state level, can become important support groups for applied sociology. This article suggests five types of applied sociology projects appropriate for state associations. These are volunteering applied sociology; doing applied sociology through consulting; making the value of sociological applications more visible; identifying applied sociology jobs for our baccalaureate, master's, and doctoral graduates; and helping to improve the socioeconomic outlook for our academic colleagues and, in turn, ourselves. State associations provide an organizational base, proximate members, and local opportunities for applying sociology. Ron Wimberley, teaches sociological research methods and does research on the southern Black Belt and other topics. He also attempts applied sociology through volunteer work, consulting, and occasional leaves from his university position. Catherine Harris of Wake Forest University is appreciated for suggesting the topic. The author is responsible for the views expressed in the article.     

The contemporary crisis and the social relations department at Harvard: A case study in hegemony and disintegration

The history of sociology is marked by periods of theoretical pluralism and hegemony. Their interplay has resulted in the slow and uneven development of the discipline. Today, however, bodies of theory and practice have become so diverse that many scholars worry that sociology is in a state of disintegration. The theoretical career of Talcott Parsons provides a microcosm in which to explore similar processes. Building on the success of The Structure of Social Action, Parsons led a movement that transformed Harvard's Department of Sociolgy into the Department of Social Relations. There he attempted a grander, but failed synthesis in Toward A General Theory of Action. His case provides a situation in which to explore features of disciplinary integration and fragmentation. The comparison stimulates one to fashion an informed sense of whether sociology is now facing its latest crisis or may be in the process of decomposition as a corpus of study. This research was supported by grants from the American Sociological Association's Fund for the Advancement of the Discipline and by Indiana University Northwest through a sabbatical leave.     

Home,work and community: skilled international migration and expatriate women in Singapore

While skilled labour migration across international borders is a phenomenon of increasing significance in the age of globalization and an important component in the production of global cities, it has not been given sufficient attention in traditional migration analyses.
Recent research has focused on institutional mechanisms regulating the patterns of skill transfer rather than the individual experience of being part of the international labour circuit. Women, in particular, have usually been relegated to the role of "trailing spouses" and are generally invisible in the migration process.
Using a questionnaire survey and in-depth interviews, this article attempts to reinstate the importance of women's roles by portraying them as active agents who adopt a range of strategies in negotiating the move and coming to terms with the transformations wrought by the move in the domains of home, work and community. It argues that skilled labour migration is a strongly gendered process, producing different sets of experiences for the men and women involved in it.
While international circulation often represents "career moves" for expatriate men, their spouses often experience a devalorization of their productive functions and a relegation to the domestic sphere. As an adaptive strategy, expatriate women often turn to the social and community sphere to reach for grounding in their lives.
The article also points to the diversity of "expatriate experiences": while "western" expatriates tend to recreate a more exclusive world by drawing on strong institutional support, "Asian" expatriates find that they have to navigate much finer social and cultural divides between themselves and the host society.     

Community,Family, and Race/Ethnic Differences in Health Status in Rural Areas1

Abstract Many race-specific differences in health outcomes that have been observed in previous research have been attributed to class and racebased group differences which either facilitate or constrain health opportunities and behaviors. These include such variables as different rates of poverty, health insurance coverage, and access to medical care. However, these relationships have been inadequately examined in rural communities where minority status may be even more detrimental to health than in urban areas, due to various constraints on access to health care. We present an analysis that assesses the effects of community, family structure, sociodemographic, and medical care variables on self-reported health status among Hispanics, Mrican Americans, and non-Hispanic whites in six rural communities in Florida. Community structural characteristics had a significant effect on self-reported health, as did some of the measures of how respondents “experience” community. These relationships held even when other sets of variables were added to the models. Family/household characteristics and sociodemographic and medical care variables were less important in explaining self-reported health status. These findings suggest that community continues to be important in explaining differences in health status in rural areas.