In each of several projects in this Superfund Hazardous Substances Basic Research Program, epidemiological data will be collected from a variety of sources. There is a need that the data, once collected, be computerized and organized into a database with quality control procedures to monitor the accuracy and completeness of the data. Once the database has been established, exploratory analyses of the data need to be done to examine each variable singly before considering its relationship with other variables. As the exploratory process continues, unusual values are identified and need to be checked for accuracy. After data-cleaning operations have been carried out and a thoroughly explored data set is established, statistical modeling of relationships among multiple variable is required. Finally, after statistical models have been developed, they must be assessed for their adequacy of fit to the observed data. Only after developed models have been shown to adequately fit the data is the investigator in a position to make statistical inferences and to draw conclusions from the data. This core has the resources and technical expertise to provide these data management and statistical services to investigators in the program project. Core staff will use statistical software packages, together with custom written programs for data entry, to establish and maintain databases. The biostatisticians will work closely with the investigators in analyzing and interpreting the data and in preparing tabular and graphical displays of results for publication. Cost effectiveness, economy of scale, and heightened coordination and support will be achieved through the activities of this Core.
Showing the most recent 10 out of 50 publications
