How the Study was Conducted
The survey was sponsored by The Dyson Foundation in partnership with The Marist College Institute for Public Opinion. 4,520 adults, eighteen years of age and older residing in New York State, were interviewed by telephone.
A random digit dial (RDD), stratified probability design was used to draw the telephone numbers for the survey. Through the combining of counties, nine analytic regions were created. Telephone numbers were selected based upon a list of telephone exchanges from throughout New York State. The exchanges were selected to ensure that each county was represented in proportion to its population. To increase coverage, this land-line sample was supplemented by respondents reached through random dialing of cell phone numbers. The two samples were then combined. The landline numbers were generated through ASDE, Inc; cellular telephone numbers were obtained from Survey Sampling, Inc.
4,520 total respondents across the following nine regions were interviewed:
The Capital Region - 500
The Adirondacks - 501
Western New York - 504
The Finger Lakes - 502
Central New York - 503
The Mid-Hudson Valley - 501
The Lower Hudson Valley - 501
New York City - 502
Long Island - 506
The map below details the counties included in each region.
To validate correct inclusion in the survey, respondents were screened by first being asked whether they are at least 18 years of age or older, then in which county in New York State they live. Adult respondents who do not currently live in the State were not included in the study.
The questionnaire and telephone sample were programmed for computer assisted telephone interviewing (CATI). Interviewing was conducted from a centralized telephone facility at Marist College in Poughkeepsie, New York using trained interviewers who were specifically briefed on this study. Interviews were conducted between February 7 and March 3, 2011 and at the preference of the respondent, in either English or Spanish. An 8 call attempt design for the landline frame and 5 call attempt design for the cell phone frame was implemented. Interviewers attempted to contact households during both daytime and evening hours; both weekdays and weekends.
Demographic information on the gender, race, income, and age of adult residents was available for each county. Data were combined to yield population estimates for weighting for each of the nine regions. Thus, the dataset was weighted by county of residence, gender, race, income, and age to reflect the demographic distribution of households across the population of New York State and within each of the nine regions.
The margin of error for these survey results for all 4,520 residents interviewed is ±1.5 percentage points for percentages near 50% at a confidence level of 95%. For the approximately 500 residents in each of the regions, the margin of error is ±4.5 percentage points. The error margin increases as the number of interviews for a particular group or sub-group within the sample declines.