Methodology

Sampling plan

A total of 2,500 residents of Kelowna were selected at random to receive a survey by mail. Each household selected to receive an invitation to participate in the research was mailed a survey package that included: a letter from the Mayor on City letterhead, a copy of the survey and a postage paid reply envelope.

In addition to the mail survey, households that were chosen to receive the survey were able to complete the survey on-line if they preferred. The on-line survey was password protected to ensure that only those households that received a survey could access the survey.

The survey was in-field from June 1, 2008 until July 4, 2008.

Survey response and margin of error

A total of 696 surveys were completed and returned. While the response rate was lower than in prior years, it was still high enough to be considered representative of the population’s opinions.

A sample of this size has a margin of error of +/- 3.7% at the 95% confidence interval. The margin of error expresses the amount of the random variation underlying a survey's results. If this survey were to be repeated 20 times the survey parameter would be within +/- 3.7% of the true population parameter 19 times.

When making comparisons between subgroups of the survey and comparing this year’s survey and past surveys the margin of error increases, sometimes substantially.

Questionnaire design

Readers should note that due to significant changes in the questionnaire and the rating scales used that comparisons to past surveys is not possible. The rating scales used throughout the survey were changed from four point scales to five point scales. The five point scale provides a more discriminating view of satisfaction and other types opinions than do four point scales. Without a midpoint option (i.e. three out of five on a five point scale), respondents often choose a positive response rather than a negative response, creating positively skewed data.

Population and data weighting

 The age distribution of respondents was significantly different than the actual age distribution of the population of Kelowna. As such, the survey data is over representative of older age groups and under representative of younger age groups.

In order to mitigate the impact of age bias in the survey the sample has been weighted to match the age distribution of the entire population of the City of Kelowna. According to 2006 Census data, Kelowna had a total population aged 18 years and older of 87,144. The table shows the distribution of Kelowna residents by age group, as well as the age distribution of the survey sample and weighting of the data.