Politics
By ROBERT KIECKHEFER

Who counts the votes?

THOSE vote totals that you will hear on radio and see on television and in the newspapers on the night of November 4 will not be exactly accurate. They will also be incomplete and unofficial.

Vote counting is one of the last mysteries of election nights. Polls and other predictive devices often tell viewers, listeners and readers well ahead of time who has won which race. But you can still win an election-night bet in most taverns by waggering you know better than your companion about who counts the votes.

At least you can after you finish reading this column.

In Illinois, the only organizations which gather statewide vote totals on election night and make them available to the public are News Election Service and the Illinois State Election Service. Neither is a branch of government. Both are run by the news media.

The News Election Service (NES) is a national organization — a cooperative formed by the three television networks and the two major wire services, United Press International and Associated Press. NES operates nationwide, counting votes for president in each state, for governor, for U.S. senators and for U.S. representatives. It also tabulates the vote on other questions and offices if its board of directors approves.

Vote totaling works like this in Illinois: in each of the state's 102 counties, NES hires a "stringer" to staff the central vote-gathering point — usually the county courthouse. That stringer is paid a fee to keep track of vote totals as they come in and to relay those results periodically to NES. The procedure is slightly different in Cook County and involves more stringers.

At NES headquarters, the stringer reports are fed into a computer, which keeps up-to-date counts of the vote in each race. Those returns then are made available to ABC, CBS, NBC, AP and UPI to be used as each organization sees fit.

In the case of UPI, the NES figures are fed automatically into the UPI central computer in Dallas, where they are —  also automatically — turned into constantly updated wire stories. As a result, wire editors in each state always have available a fresh "tab" on each top-of-the-ticket race.

The figures also are available in county-by-county tables, which are used to determine voting trends and patterns.

Of course, each news organization supplements the NES figures with other sources of its own.

The ISES works in a similar fashion, although its operations usually are manual, rather than computerized. The ISES — operated by AP and UPI —  counts votes for lesser state offices and the state legislature and makes them available to the two wires.

In no way, however, are those totals official. Mistakes can and do occur, sometimes at the county level and sometimes in the NES or ISES counting. On more than one occasion, the computers have worked their reverse magic to confuse things.

A prime example was the 1978 comptroller's race. Roland Burris, checking figures on his own, was sure he had won. But the figures ISES was providing showed the election almost in a deadlock. Burris, understandably, became rather upset as the misunderstanding persisted well into Wednesday afternoon.

It was finally resolved when AP and UPI reporters discovered a huge error in the Madison County totals which had given Burris' opponent more than 20,000 extra votes.

The incident made it crystal clear, especially to those who deal with the election-night counting process, how little outsiders know about it. Aside from the emotional and psychological strain involved, it really made no difference to Burris whether the news media ever caught their error and declared him the winner. All he had to worry about were the figures certified to the state Board of Elections by the counties. The vote total is official only when the board members sign the tally sheets weeks after election day.

All this is of special interest this year because the whole system may change by the 1982 election.

The State Board of Elections is very interested in setting up an election night vote-counting system of its own, which might make totals available much more rapidly than is now possible.

The board is discussing modeling the system after California's. Political reporters in California say the system works well and they rely heavily on it. There are a couple of problems with implementing a similar procedure in Illinois, however.

One is that California, despite its size, has fewer than half the 102 counties that make up Illinois. That means there would be more than double the number of reporting points — and more than double the number of potential trouble spots. Also, the California counties are mandated by law to report results quickly to the state on election night. A similar law would be needed to make the system work in Illinois — and would not be popular with some Illinois county officials.

Finally, all counties in California use some type of automated vote counting. The few Illinois counties still on paper ballots would have to be upgraded to make a speedy statewide counting system work.

34/November 1980/Illinois Issues