The voters; Who Votes? by Raymond E. Wolfinger and Steven J. Rosenstone. Yale University Press. $3.95 (paperback).

The authors, both professor of political science, have written a small book with an explosive wallop. Candidates, campaign managers, convention delegates, and journalists will ignore this volume at their peril.

"Who Votes?" is a most powerful attack on the commonly accepted ideas of American voting behavior. The authors have mined a rich statistical data base, the current Population Survey conducted by the US bureau of the Census. Their findings are iconoclastic and utterly opposed to mainstream political science:

* People with more education are more likely to vote than those with less. While the college educated account for 26 percent of the total adult population, they make up 32.3 percent of the voters.

* "People who are white, well educated, well-to-do, middle-aged, married, Notherners, government employees, and residentially stable account for a proportion of voters larger than their share of the population," they write.

* It is doubtful, had voter turnout been any higher, that the outcome of recent presidential elections would have been different.

* Advocates of both sides of controversial policy issues are represented among voters in proportion to their numbers in the general population.

* Abstention from voting has little to do with evalutation of candidates. People who find both presidential candidates equally attractive or unappealing are as likely to vote as those who have strong preferences.

* Americans who say they are suspicious of politicians and government vote at the same rate as those who say they trust the political system and politicians.

* The idea that more lenient and relaxed registration requirements would change electoral outcomes is unprovable. In fact, the authors argue that the partisan consequences of registration law reform woud be trivial; i.e., the Democratic Party would not benefit, as seems to be believed. The present proportions in the voting public between liberal and conservative would hardly be affected if the voting population were to expand.

Statisticians will long devate the findings of Professors Wolfinger (Berkeley) and Rosenstone (Yale). The authors, however, have provided a paradigm for studying the coming November election. If their conclusions hold, their book may revolutionize long-held theories.