SECTION II - METHODOLOGY
Questionnaire Development
Initial Committee Input
Dr. Wacker met with the Compass Planning Committee to determine the overall nature and type of information to be collected. A subcommittee was formed to specifically assist in the process of developing the surveys. The Compass Planning Committee was also instrumental in notifying the Weld County community of the Needs Survey via direct phone contacts with key informants and ads and letters to the editor placed in the Greeley Tribune.
Questionnaire Refinement
The Compass process and written questionnaire were developed by the United Way of America and is used in many communities across the country. The questionnaire was modified slightly so it would reflect the needs of Weld County. There were five different questionnaires used in the Compass process to survey opinions from associations, service providers, business, key informants, and Weld County residents. Dr. Wacker refined all of the questionnaires with assistance from the Data Subcommittee. The final questionnaires were reviewed and approved by both the Data Subcommittee and the Compass Planning Committee (copies of the Household and Key Informant questionnaires are located in the Appendix). The household survey was translated into Spanish by staff at Weld School District 6.
Sample – Household Survey
Sampling Frame
To obtain a representative sample of eligible respondents, a list of households in Weld County was obtained from a database company. The database company randomly selected 50% of the total number of households (n= 52,141) which was 26,070 households. The list indicated each householder’s name, mailing address, and race of each potential respondent. A list of all Hispanic households in Weld County was also obtained and totaled 10,863 households.
In addition, focus groups of Hispanic and/or low income individuals were conducted by Dr. Faye Hummel and April Hummel in various locations around Weld County. Focus groups were asked the same types of questions used in the household questionnaire and discussions were held in Spanish where appropriate.
Sample Stratification
It was important that the total sample demographics closely reflect the demographic characteristics of the county with regard to geographic location and race. Thus, the sample was stratified into rural and urban families and Hispanic households were specifically selected to receive the questionnaire. Hispanic households were randomly “oversampled” in order to gain adequate representation in the sample. In addition, focus groups were primarily comprised of Hispanic respondents in order to add to the representation of Hispanic residents.
To ensure equal chance for inclusion in the study, every 30th name on the database list of urban (Greeley/Evans) and rural households for each age group was selected to be interviewed. Surveys were received from respondents residing in every town in Weld County.
Sample – Key Informant Surveys
The purpose of this aspect of the Compass process was to assess what community key informants believed to be the strengths and needs of Weld County communities. Thus, this portion of the survey process employed a “purposive” sampling process and was not concerned with the generalizability of the results to all key informants living in Weld County.
Sampling Frame
Key informants were randomly selected from city council members, the local hospital, the university and community college, local foundations, Weld County Government employees, school district leaders and board members, and leaders in the faith community. A total of 63 key informants participated in the study.
Methods and Procedures
Survey Procedures
Time Period
Key informant questionnaires were sent during the months of March 2004 and April 2004. Household surveys were mailed during the months of May 2004 through December 2004. Focus groups were conducted from November 2004 to January 2005.
Staff
Dr. Wacker was in charge of the survey process and was assisted by Dr. Susan Collins, Sonja Rizzolo, and Richard Dougherty. Volunteers from RSVP assisted in the mailing of the questionnaires. Dr. Faye Hummel and April Hummel conducted the focus groups.
Random Sample With Replacement
To achieve the designated number of household respondents, a “random sample with replacement” method was used. When respondents refused or could not be reached because of a bad address a replacement was selected from the sampling frame. In order to achieve the target number of households in different geographical locations and by race, the records were divided into urban and rural subgroup and by race.
Interviewing Spanish-Speaking Respondents
The sample frame of Hispanic respondents received a questionnaire in both English and Spanish. The focus groups were also conducted in Spanish when needed.
Focus Groups
Even though the list of households provided by the data base company was considered representative of Weld County, this list may have excluded portions of our community due to resource disparities, and language or literacy issues. As a result, five focus groups were conducted, aimed at ameliorating the identified limitations and designed to supplement the written surveys. Focus groups were used to supplement the data collection process of survey and were conducted to collect qualitative data about the perspectives and contexts of the Weld County residents who participated.
Purposive sampling was used to recruit focus group participants who were monolingual Spanish speaking in rural and urban areas of Weld County as well as English speaking Hispanic and Anglo residents in rural Weld County. The focus groups were purposely selected to capture experiences in and unique knowledge about Weld County. Groups were recruited through English as a Second Language classes, education and community resources and health care services. Five focus groups were conducted in September 2004, in elementary schools, 2 community centers, a church and health care clinic. Three focus groups were conducted in Greeley, one in Southeastern Weld County and one in South Central Weld County. Four focus groups were conducted in Spanish and one was conducted in English. A total of thirty-two persons volunteered to participate in the focus groups. The focus group size ranged from 3 to 15 persons. The focus groups were composed of individuals with various educational and economic backgrounds as well as individuals from urban and rural areas. The duration of each focus group session ranged from sixty to ninety minutes.
The focus group discussions were guided and directed by an interview guide that was created from the items on the household survey tool. The interview guide served two important purposes: (1) as a tool to ensure that broader issues of the community assessment were discussed and (2) to provide a guide around which specific issues could be addressed by the participants in the focus groups. At each focus group, two skilled moderators provided structure and direction to the group discussions. The primary moderator of the Spanish speaking focus groups was Ms. April Hummel who is bilingual. The English speaking focus group was moderated by Dr. Faye Hummel. All focus groups were tape recorded after verbal permission was obtained from the participants.
The Spanish-speaking focus group tapes were translated from Spanish into English and transcribed by Ms. April Hummel, the bilingual moderator of the focus groups. Ms. Hummel also transcribed the English language focus group tape as well.
Coding and Data Entry
Dr. Wacker and Sonja Rizzolo created the coding key for data entry so that all data would be coded exactly the same. Ms. Rizzolo entered the coded all the surveys into SPSS statistical program.
Analysis and Report Writing
Survey Analysis and Reporting
The quantitative data derived from surveys were analyzed by Dr. Wacker and Ms. Rizzolo using the SPSS statistical analysis program. In addition, support staff typed all comments made by respondents during the interview. The responses to the openended questions are reported in the final report. Dr. Wacker, with assistance from Ms. Rizzolo wrote the report describing the data output and created the graphs and tables. The cut-and-sort technique was used in the analysis of the data generated by the focus groups. This method yielded a set of sorted information that is the basis for the development of this summary report. The various pieces of interview transcription were used as supporting materials and incorporated within an interpretative analysis. The analysis relied on two analysts, Dr. Hummel and Ms. April Hummel, to increase reliability of major themes and issues.