Summary of the 1974 Televote trials in San Jose, CA funded by the National Science Foundation


A televote system to aid rapid two-way communication between public officials and large numbers of constituents was developed and demonstrated in the San Jose Unified School District (SJUSD). The objectives of the system are:

  1. to provide citizens and public officials the most relevant information bearing on important community problems;
  2. to provide civic planners accurate knowledge of the current views of citizens so that their decisions will be more responsive to these views;
  3. to offer citizens effective roles in civic and school planning suited to different levels of interest in a given issue.

In a typical televote, statements of opposing views are sent to registered televoters, who then have a week to feed back their own preferences by dialing certain numbers on the telephone. Results showing the number of people preferring each view can be in the hands of planners within two days after the televote ends. Use of the system can help a public agency serve its functions better by bringing the many people affected by agency decisions into the planning process at very low cost in time and money.

After initial development and a pilot study to test equipment and procedures, a district-wide public demonstration of the televote system was conducted seven months during the 1973-74 school year. A committee representing students, school staff, parents a nd other citizens met weekly to decide on which issues communication was most needed. When the committee was satisfied that all sides of the issue were stated well and fairly, the issues were mailed to all televoters and published in the SUN newspapers.

In order to become televoters, citizens were required to register by phone or mail and were sent a unique televoter number as a way of insuring that only one vote from each person was counted. Registrations were solicited mainly through school newsletters and occasional public service announcements on radio and TV.

Nine televotes were conducted during the demonstration, including 30 specific questions. A three-digit number was printed beside each alternative answer, plan or policy. A televoter studied the alternatives, then indicated his preference by calling the te levote line, dialing his own televoter number, and dialing the numbers of the answers he preferred. Televotes were processed by computer and all information from any individual was kept confidential. Televote counts broken down by demographic variables we re published and distributed to all interested individuals and groups, and to the media.

Evaluation Results

Televote results had a significant impact on four educational decisions. The one of greatest consequence was the choice of new courses for a $3 million program addition to the Regional Vocational Center. The courses chosen corresponded closely to the pref erences of televoters. The televote issues which had tangible impact were in most cases those issues initiated and defined by the same school officials who used the results in their planning.

Over 5,500 persons (about 4% of the eligible population) voluntarily registered as televoters, and most of these participated in one or more televotes. An average of about 700 persons voted on a given issue. This rate of participation represents substantially greater input than school authorities usually receive on specific issues, exce pt in public elections. Participation was higher in suburban areas than in the central city areas where the less affluent residents live. The largest minority group in the area is Mexican-American and they participated at a lower rate than Anglo Whites. I nformation was provided in both English and Spanish to this group.

Measures of basic attitudes and habits of communication with schools were administered before and after the demonstration in a design which permitted comparison of changes in SJUSD to changes in a comparable control district nearby. In the suburban areas of SJUSD there was a marked increase in perceived interest of the school in citizen opinions, an increase not found in comparable suburbs of the control district. Also, over 85% of both students and adults felt the school district should ask for their opi nions before making policy decisions. Televoters and other citizens in SJUSD showed a greater increase in awareness of school issues during the demonstration year that did citizens in the control district. From these results it appears that participation in the televote system led to greater awareness of school issues and better relations between citizens and the school district.

All groups questioned about the value of televoting, including selected staff members, student and adult televoters, and a random sample of non-participating adult citizens, evaluated the system favorably on the whole. Each group also offered many specifi c criticisms. Asked how much they would be willing to pay per year to have a televote system continue in San Jose, a random sample of San Jose citizens (none had participated in televotes) said they would be willing to pay on average 62 cents per year, wh ich was more than the estimated tax or subsidy needed to operate a televote system in 1974 dollars. Televoters responded with a mean value of $1.07. Apparently most residents who are told even briefly how televoting works think it has potential value for the community and are willing to pay the small cost of operation.

Postscript (1995)

The project summarized here was conducted in 1973-74 with National Science Foundation support (Grant No. GI-37183) while I was at the American Institutes for Research in Palo Alto, CA. More detailed information on the study is available on microfiche at many libraries as ERIC documents ED095896 (report of the San Jose research), ED095897 (appendices to report) and ED107300 (implementation guidelines).

In the 20 years since then, the technology has evolved so that a highly efficient and reliable system which accommodates a large community can be introduced for less than $20,000 and operated at an annual cost of $5,000 to $10,000, including costs of main tenance, analysis and reporting results to the community. For large cities this cost assumes information on each issue will be disseminated largely by newspapers, with mailing only to special groups, such as a scientifically drawn representative sample of the community.

Vincent Campbell, President
Decision Systems Inc.
8073 Kincross Way
Boulder, CO 80301

Phone (303) 530-9691
Fax (303) 530-9692