METHODOLOGY |
The table below provides an overview of the methodology utilized for the project.
Table 2 - Overview of Project Methodology
| Method: |
Telephone Survey |
| Universe: |
77,856 Residents 18 years and older within the City of Carlsbad |
| Number of Respondents: |
1,001 Residents Completed a Survey |
| Average Length: |
20 minutes |
Field Dates:
|
October 1 – November 11, 2007
(Interviewing was suspended for a week and a half during the fires) |
Margin of Error:
|
The maximum margin of error for questions answered by all 1,001 respondents was +/-3.08% at the 95% level of confidence. |
Research Objectives
Prior to beginning the project, BW Research Partnership met with the City of Carlsbad’s Performance Measurement Resource Team to determine the research objectives for the study. The main research objectives of the study were to:
- assess residents’ perceptions regarding satisfaction with city services, quality of life, sense of community, city government, safety, and city-resident communication;
- evaluate residents’ satisfaction with a number of specific city services;
- identify residents’ perceptions regarding land use in the City as well as open space preferences for the land near the strawberry fields, and
- assess residents’ experience visiting the Carlsbad Village as well as preferences for the types of stores and businesses that could be located throughout the City
Questionnaire Design
Through an iterative process, BW Research worked closely with the City to develop a survey instrument that met all the research objectives of the study. In developing the instrument, BW Research utilized techniques to overcome known biases in survey research and minimize potential sources of measurement error within the survey.
Sampling Method
A random digit dial (RDD) methodology was utilized to interview a representative sample of residents 18 years and older within the City of Carlsbad. The RDD methodology is based on the concept that all residents with a telephone in their home have an equal probability of being called and invited to participate in the survey.
The RDD method includes both the listed and unlisted phone numbers that fall into the active telephone exchanges within a City (the exchange includes the area code and first three digits of the phone number). Since telephone exchanges often overlap with neighboring cities, screener questions were utilized at the beginning of the survey to ensure that the residents who participated in the survey lived within the City boundaries.
Data Collection
Prior to beginning data collection, BW Research conducted interviewer training and also pre-tested the survey instrument to ensure that all the words and questions were easily understood by respondents.
Interviews were generally conducted from 5:00 pm to 9:00 pm Monday through Friday and 10:00 am to 2:00 pm on Saturday and Sunday to ensure that residents who commuted or were not at home during the week had an opportunity to participate.
Throughout data collection, BW Research checked the data for accurateness and completeness and monitored the percentage of residents with language barriers to determine whether or not the survey should be translated into a language other than English. Since only 0.2 percent of calls were identified as having a language barrier, translating the survey into languages other than English were not necessary to ensure representativeness of results.
Data Processing
Prior to analysis, BW Research examined the demographic characteristics of the 1,001 respondents who completed a survey to the known universe of residents 18 years and older using the San Diego Association of Government’s (SANDAG) 2007 demographic estimates for the City of Carlsbad. It is estimated that among Carlsbad’s 101,337 residents, 77,856 are 18 years and older. After examining the dimensions of zip code, gender, ethnicity, and age, the data were weighted to appropriately represent the universe of adult residents and ensure generalizability of the results.
A Note About Margin of Error And Analysis of Sub-Groups
Although the overall margin of error for the study, at the 95% level of confidence, is between +/-1.85 percent and +/- 3.08 percent (depending on the distribution of each question) for questions answered by all 1,001 respondents, it is important to note that questions asked of smaller groups of respondents (such as questions that were only asked of residents who reported low satisfaction) or analysis of sub-groups (such as examining differences by length of residence or gender) will have a margin of error greater than +/-3.08 percent, with the exact margin of error dependant on the number of respondents in each sub-group. BW Research has utilized statistical testing to account for the margin of error within sub-groups and highlight statistically significant sub-group differences throughout this report.
|
