SURVEY FOOTNOTES

1. The questions were ranked in descending order of favorable answers while the numbers to the left show the order in which they appeared in the questionnaire. In this case the order of asking the questions did not appear to influence the responses.

2. Excludes duplicates and a complete lack of answers to any of the ten [10] key questions. Some respondents did not provide the data to classify them or their association into one or more of the 14 trait categories.  Consequently the total number of answers to a specific question or category usually is less than the 450 usable questionnaires received.

3. Equals the high number divided by the low number and rounded to the nearest integer according to the 5/4^ths rule.

4. Priorities range from 1 to 10 with 1 being most important and 10 being least important.  Priorities only apply to positive answers to questions 1 -10.

5. A standard error is a statistical measure of the variability within a population, a sample or the answers to a specific question. It can be used to find the odds of a particular answer being wrong by a given amount. For instance, how likely is it that the 84.7% YESes to question 3 in the survey actually came from a population where half the people would say YES while the rest would say NO? As all the t-values were greater than 6 there’s less than 1 chance in a million that the population actually opposed any of the proposed legislative changes. The formula to find a standard error is the square root of (P*Q/N) where P is the percent of positive answers, Q is the percent of negative answers or 100-P and N is the number of responses. 

6. A t-test measures the difference between an actual value and a hypothetical one.  The hypothetical value in this analysis was 50% which is what one would expect if respondents actually didn’t care one way or another, that is — they were indifferent.  A t-value of 1 equates to roughly 1 chance in 3 that the respondents don’t care.  Similarly, a t-value of 2 equates to roughly 1 chance in 20 that the respondents are indifferent, a value of 3 equates to roughly 1 in a thousand and a value of 6 equates to about 1 chance in a million that the respondents really were indifferent.  By convention, t-values of 2 or more are deemed significant or notably or meaningfully different.

7. Each of the five regions is a group of Florida counties that together include all 67 counties in the state.  The counties in each region are:

   North —            Alachua, Bay, Bradford, Calhoun, Columbia, Dixie, Escambia, Franklin, Gadsden, 

                               Gilchrest, Gulf, Hamilton, Holmes, Jackson, Jefferson, Lafayette, Leon, Levy,

                               Liberty, Madison, Okaloosa, Putnam, Santa Rosa, Suwannee, Taylor, Union and

                               Wakulla

   Central   —       Citrus, Hardee, Highlands, Lake, Marion, Orange, Osceola, Polk, Seminole and

                               Sumter

   West Coast — Charlotte, Collier, Desoto, Glades, Hendry, Hernando, Hillsborough, Lee, Manatee,

                               Monroe, Pasco, Pinellas and Sarasota

   East Coast —  Baker, Clay, Duval, Flagler, Indian River, Martin, Nassau, Okeechobee, Saint 

                               Johns, Saint Lucie and Volusia

   South —            Broward, Miami-Dade and Palm Beach

8. Membership in CCFJ was determined by comparing respondents’ names with a current official membership list.

9. Gender was deduced from respondents’ names.

10. Current and/or past service on a community’s board of directors was counted as a “YES” even if that service was not on the board of the respondent’s current community.