Appendix C—Survey Methodology and Data Reliability

Sample Selection

The sampling frame for the FRSS Survey on Advanced Telecommunications in U.S. Public Schools, K-12, was the 1991-92 list of public schools compiled by the National Center for Education Statistics (NCES). This complete file contains about 85,000 school listings and is part of the NCES Common Core of Data (CCD) School Universe. This frame includes 57,935 regular elementary schools, 18,673 secondary schools, and 1,785 combined schools in the 50 states and the District of Columbia. All regular elementary, middle, and secondary schools in the 50 states and the District of Columbia were included in the sampling frame. Special education, vocational, and alternative/other ungraded schools, schools in the outlying territories, and schools with the highest grade level below 1st grade were excluded from the frame prior to sampling. With these exclusions, the final sampling frame consisted of approximately 78,393 eligible schools.

The sample was stratified by instructional level (elementary, secondary, combined) and by geographic region (northeast, southeast, central, and west). Within each of the major strata, schools were sorted by metropolitan status (city, urban fringe, town, rural) and minority status (less than 50 percent white enrollment, 50 to 79.9 percent white enrollment, and 80 percent or more white enrollment). The allocation of the sample to the major strata was made in a manner that was expected to be reasonably efficient for national estimates, as well as for estimates for major subclasses.

Response Rates

In October 1994, survey forms (see appendix F) were mailed to 1,502 public school principals. Principals were asked to forward the questionnaire to the computer or technology coordinator or whomever was most knowledgeable about the availability and use of advanced telecommunications at the school. The accompanying instructions indicated that the data were being obtained by telephone and requested that the respondent complete the form in preparation for the telephone interview. Twelve schools were found to be out of the scope of the study (because of closings), leaving 1,490 eligible schools in the sample. Telephone interviews were conducted from mid-October through late November with 1,380 schools completing the survey by the end of data collection. The survey response rate was 92.6 percent (1,380 schools divided by the 1,490 eligible schools in the sample). The weighted response rate was 93.5 percent.

Sampling and Nonsampling Errors

The responses were weighted to produce national estimates. The sample weights were the inverse probability of selection adjusted for nonresponse. The findings of this report are estimates based on the sample selected and, consequently, are subject to sampling variability.

The survey estimates are also subject to nonsampling errors that can arise because of nonobservation (nonresponse or noncoverage) errors, errors of reporting, and errors made in collection of the data. These errors may result in biased data. Nonsampling errors may include such problems as the differences in the respondents' interpretation of the meaning of the questions; memory effects; misrecording of responses; incorrect editing, coding, and data entry; differences related to the particular time the survey was conducted; or errors in data preparation.

While general sampling theory can be used in part to determine how to estimate the sampling variability of a statistic, nonsampling errors are not easy to measure and, for measurement purposes, usually require that an experiment be conducted as part of the data collection procedures or that data external to the study be used. To minimize the potential for nonsampling errors, the questionnaire was pretested with school principals and computer/technology coordinators like those in the survey population.

During the design of the survey and the survey pretest, an effort was made to check for consistency of interpretation of questions and terms and to eliminate ambiguous items or instructions. The questionnaire and instructions were extensively reviewed by the National Center for Education Statistics. Manual and machine editing of the questionnaire responses were conducted to check the data for accuracy and consistency. Cases with missing or inconsistent items were recontacted by telephone. Final item nonresponse ranged from 0.0 to 3.5 percent (for nearly all items, nonresponse rates were less than 1 percent). No items were imputed. All data were keyed with 100 percent verification.

Variances

The standard error isa measure of the variability of estimates due to sampling. It indicates the variability of a sample estimate that would be obtained from all possible samples of a given design and size. Standard errors are used as a measure of the precision expected from a particular sample. If all possible samples were surveyed under similar conditions, intervals of 1.96 standard errors below to 1.96 standard errors above a particular statistic would include the true confidence interval. For example, the estimated percentage of schools reporting that they have access to the Internet is 35 percent, and the estimated standard error is 1.5 percentage points. The 95 percent confidence interval for the statistic extends from [35 - (1.5 times 1.96)] to [35 + (1.5 times 1.96)], or from 32.1 to 37.9 percent.

Estimates of standard errors were computed using a technique known as jackknife replication. As with any replication method, jackknife replication involves constructing a number of subsamples (replicates) from the full sample and computing the statistic of interest for each replicate. The mean square error of the replicate estimates around the full sample estimate provides an estimate of the variance of the statistic (see Welter 1985, Chapter 4; see Appendix E).

To construct the replication, 40 stratified subsamples of the full sample were created and then dropped one at a time to define 40 jackknife replicates. A proprietary computer program (WESVAR), available from Westat, Inc., was used to calculate the estimates of standard errors.

The software runs under IBM/OS and VAX/VMS systems.

Top