MDPH Health Study Design

Home Winthrop Health Study Mass State DPH Health Study About Winthrop About Us Web Research Contact Us
Logan Health Study CAC Mass DPH Health Study Plans Mass DPH Health Study Design Mass DPH Health Study Timetable Mass DPH Health Study Outlook Mass DPH Health Study Status Wide Area Map Five Mile Radius Map
Mass DPH
Health Study
Exposure Identification
Mass DPH
Health Study
Survey
Instrument
Design
Mass DPH Health Study
Conducting
The
Interviews
About The Public Health
Resource
Group

The purpose of this section is to provide an introduction and collection point for the design considerations and decisions for the MDPH Logan Airport Health Study.

Design Overview
In order to comply with the Logan Health Study legislative mandate, and to remain within budgetary constraints, there are several design areas to be considered. These are:

  • Determination of the number of people to be surveyed.
    This has been set at 2700 people. This number will allow an appropriate statistical sample size, and the setting up of control groups to perform the analysis of the Health Study Data. More ...
  • Defining the Exposure Opportunities i.e. those geographical areas (neighborhoods) that will enable a comparison of the health status of people with different levels of exposure. This is a very complex task. Click here to get more info.
  • Defining the Survey Instrument. This includes the details of the survey questions, the methods of contacting responders, and the procedures for conducting the interviews with the responders. This is a highly complex effort. Click here to get more info.
  • Selecting and Contacting the Responders. This includes the methods used to choose the Responders within the neighborhoods of given exposures, random sampling techniques, and methods of ensuring anonymity. Click here to get more info.

Sample Size of ~2700.
Results of statistical power calculations suggest a sample-size of approximately 2700 individuals.

  • Sample size of ~2700 is necessary to have sufficient (80%) power to detect a statistically significant effect.
  • Level of statistical significance will most likely be p < .05 or type-I error rate of 5%.

 

Return to Top of Page

 

E-Mail