No abstract

No abstract

The literature in the evaluation of built environments has had a tendency to be long on techniques and on examples of applications but short on theory. This is not surprising, as evaluation is a practical activity concerned mainly with the performance of existing environments in use. Its aim is primarily that of providing information which can be applied to improve unsatisfactory environments. Within this limited scope, theory would appear at first glance to be of little relevance, just as it is not important for the driver of an automobile to be conversant with the theory of internal combustion engines or that of aerodynamics in order to drive safely and successfully.So, for the evaluator interested simply in applying certain techniques in order to gather information there is little reason to dwell on theoretical considerations. However, theory, conceptual frameworks, paradigmatic assumptions, and general philosophical perspectives take on fundamental importance for evaluators interested in using information to plan and design new satisfactory environments, and for those concerned with developing appropriate techniques, understanding their uses and limitations, and assessing their utility for the purposes they are claimed to serve. In any field, advances are impossible in a vacuum of theory. The evaluation of the built environment is no exception. As Canter and Kenny (1982) have noted, the value of any empirical work is ultimately a function of the theoretical formulations on which it is based:''Unless there is an understanding of the role that the physical environment plays in people's lives it is extremely difficult to know which aspects of that environment to measure and how to argue for the significance of any relationships which are found between the environment and human actions or experience.

No abstract

For survival probabilities with censored data, Rothman (1978, Journal of Chronic Diseases 31, 557-560) has recommended the use of quadratic confidence limits based on the assumption that the product of the 'effective' sample size at time t and the life-table estimate of the survival probability past time t follows a binomial distribution. This paper shows that the proposed confidence limits are asymptotically correct for continuous survival data. These intervals, as well as those based on the arcsine transformation, the logit transformation and the log(--log) transformation, are compared by simulation to those based on Greenwood's formula--the usual method of interval estimation in life-table analysis. With large amounts of data, the alternatives to the Greenwood method all produce acceptable intervals. On the basis of overall performance, the intervals suggested by Rothman are preferred for smaller samples. Any of these methods may be used to generate confidence sets for the median survival time or for any other quantile.