method was used. The total number of usable survey was 301.1 In terms of statistical analysis, both
parametric (regression) and non-parametric (chi-square) techniques were employed.
III. Results and Discussions
3.1 Respondent Characteristics
Not all 301 respondents answered all questions in the survey, therefore sample size in some variables
is less than 301. For instance, four respondents did not respond to the gender question. Among the
remaining 297 respondents, about 37 percent of the respondents were male. The average age among
the respondents (N=294) was about 40 years. A breakdown of the age of the respondents show that
almost 80% of the respondents were in the 21-50 age group and 54.7% were in the 31-50 age group,
and only 18% of the respondents were above 50 years old. That is, most of the respondents were
relatively young.
A majority of the respondents were employed fulltime (73%) and only 13% were unemployed at the
time of the survey. In terms of their education status, 45% had at least a college degree and almost
28% had high school diploma, i.e., respondents were generally educated. Most of the respondents
(58%) reported their annual household income at $50,000 or above, i.e., the population was relatively
wealthy. Only 7% reported income below $25,000 which brings them close to the national poverty line.
In terms of total number of persons in a household, 25% of the respondents lived alone. About half of
the households (50.2%) reported having two people in the household. Slightly over 35% of the
1 We use the confidence interval method to determine our sample (Burns and Bush, 2006, p.
372). In this study, the sample had an error of ±5.65 at 95% level of confidence. Note that
a higher level of sampling error, say ±10%, would have resulted in a sample size to 96 only.
So, the sample size used in this study is representative of the population.
4