1. Planning the survey research As the initial step in this research project, seven focus groups were conducted: two in San Francisco, California on March 16, 1998; two in Dallas, Texas on March 17, 1998; two in New York City on March 18, 1998; and, one in Louisville, Kentucky on March 30, 1998. The groups were composed of men and women, all of whom were voters representing a spectrum of ages, occupations, incomes, and including both parents and non-parents for a total of 69 parents and non-parents. The groups were further separated by race: two African-American and Latino groups, one each, in Dallas and New York; two Caucasian groups, one each, in San Francisco and Louisville; and a group of Asian- Americans in San Francisco. The guide for the discussions was prepared in consultation with Ann Baker, Katrina Kelley, and the working group for the National School Board Foundation Initiative. The groups were led by BR&S partner Nancy Belden and Donita Buffalo, a consulting African-American moderator. The sessions were videotaped and transcripts of the discussions were prepared. The findings from the groups were key to the development of the questionnaire for the surveys. 2. National surveys of urban public and school board members Two related surveys were conducted simultaneously: one among 100 school board members in urban areas and one with 1,000 residents of urban areas nationwide. Questionnaire development The questionnaires used for the school board members and for the urban public were each designed by BR&S in consultation with the Steering Committee of the Council of Urban Boards of Education (CUBE) and the Working Group for the National School Boards Foundation Urban School Boards Initiative, Ann Baker of National School Boards Foundation, and Katrina Kelley of National School Boards Association. A draft of each questionnaire was subjected to a pretest which resulted in modifications. Sample and interviewing for the survey of the general public The universe for the urban public survey is all adult residents in the largest urban centers in the United States, as determined by CUBE eligibility, which is defined as: school boards which serve a core city population of at least 100,000 or are urban in character. The boards in the largest 100 cities qualify. However, districts with smaller core school district populations are allowed to join on a case by case basis. The sample was selected in two stages, using probability-proportional- to-size sampling. In the first stage, the sampling frame was a list of randomly created phone numbers (a technique known as random digit dial or RDD) created by SDR, Inc., in the counties that contain the school districts that comprise the CUBE member list and the CUBE eligible list. Telephone numbers were selected at random from this frame. The second stage of sampling was selection at the household level. In residences where working telephones were reached, the survey respondents were selected using a random probability method, i.e., the interviewer requested to speak with the adult 18 years or older in the household who had the most recent birthday. The survey includes of a total of 1,000 completed interviews. The margin of sampling error for the entire survey is plus or minus 3.1 percentage points at the 95% level of confidence. The margin of sampling error for smaller subgroups within the sample will be larger. For example, the margin of sampling error for men (n=450) is plus or minus 4.6 percentage points, and for women (n=550) is plus or minus 4.2 percentage points. Other non- sampling error may also contribute to total error. The fieldwork was conducted by telephone using a computer-assisted telephone interviewing (CATI) system, from May 20 to May 31, 1998 by a team of professional, fully trained and supervised telephone interviewers. A briefing session familiarized the interviewers with the sample specifications and the instrument for this study. The interviews averaged approximately 14 minutes in length. BR&S monitored the interviewing and data collection at all stages to ensure quality and that a proper mix of interviews was conducted among various demographic groups. The demographic characteristics of the sample, obtained via the selection methods described above, were matched to population estimates for the urban population produced by SDR and based on the most recent Census data available. The data collected have been weighted statistically to bring gender, race and age into their proper proportions for this population. The following table shows the demographic composition of the survey respondents.
Sample and interviewing for the sample of school board members The universe for the school board survey is a list of members of school boards which are members of CUBE or are eligible for CUBE membership supplied by National School Boards Association. The sample of school board members was selected using a systematic probability technique. An interval for selecting a school district for an interview was found by dividing the total number of school districts on the CUBE member and eligible lists (172), by the number of interviews desired (100). The school districts were arranged in alphabetical order and were selected by moving through the list and rounding to the next interval. Once a school district had been identified, a particular board member was selected for interviewing from among the total members of that board using a random number chart. When the randomly selected member was reached by phone, an interviewer screened to make sure that he or she was still a member of the school board for the district indicated. In the event that a particular individual could not be reached within the dates that the survey was in the field or was no longer a board member, the next school board member on the list became the target individual. A total of 100 interviews with urban school board members were conducted by a select team of professional interviewers by telephone from May 26 to May 30, 1998. The interviews were closely monitored to ensure that board members were treated professionally and to allow the participants to request further contact or copies of the report. The interviews lasted approximately twelve to thirteen minutes. Ninety five of the 100 respondents were elected members of boards, and five were appointed. Because the number of appointed members was small, the answers from appointed and elected respondents were not compared in the analysis. The following table shows the regional distribution of respondents.
Reading the crosstabs All the questions in this study have been cross tabulated by demographic and other characteristics, such as gender, age, education, and income. When looking at the computer generated cross tabulations, most of the categories used are self explanatory. Those that benefit from extra explanation follow. Categories used in both surveys Race: The race variable includes the respondents' self-identification of their race or ethnicity. The white category is white non- Hispanics, black is non-Hispanic blacks or African-Americans, Hispanic includes participants from all races who self-identify as Hispanic, and other includes Asian- Americans, Native Americans, and those who answered "something else." Region: Regional categories are based on the Census definition of geographic region in the United States. The regional breaks are as follows:
Marital: Married includes respondents who report that they are married, single includes those who have never been married, while dsw includes those who are divorced, separated, or widowed. Categories used in the urban public survey only Voters: There are two categories of voters in this banner: school board voters and other voters (i.e. Congressional voters but excluding school board voters). Both include only those respondents who report that they are registered to vote. Nonvoters are those who are not registered to vote. PTA activist: This banner point includes only those people who report that they are currently active in the PTA. School board lobbyist: This variable includes people who say that they have attended a school board meeting in the last five years to accomplish something specific. School employee: Those respondents who are currently employed in, or have been employees of, a school or school district are divided into educators—those who have worked in a teaching or administrative capacity—and others who hold or have held clerical, service, or school jobs of other non-academic types. Private sector: This variable includes those who report that they own or currently work (or in the case of retired respondents that they used to work) for a business or corporation. Category used in the school board members survey only
Occupation: This banner point was recoded from
the question which asked board members to name
their occupation. From these open-ended
responses, we created the following categories:
educators, full time school board members,
homemakers, and those who work in the private
sector. Private sector employees are comprised
of white collar professionals, small business
owners, and those in managerial positions.
|
