Wilson’s Margin in O.C. Falling Short of Victory : Election: Times Poll projects his vote total well below that captured twice in the county by Gov. Deukmejian.

In California’s most Republican county, GOP Sen. Pete Wilson is well ahead of Democrat Dianne Feinstein in the race for governor, but a Times Orange County Poll found his lead still below the level that experts consider necessary to win statewide.

With about eight weeks remaining until the Nov. 6 election, more than 1 in 4 of the county’s voters were still undecided about their choice for governor. Even among Orange County Republicans--some of the most loyal voters in the state--nearly a quarter said they had not made a selection.

Among all county voters, Wilson led Feinstein 48% to 22% and he ran ahead in every age, income, sex and geographic category.

“Even though Wilson holds a better than 2-to-1 edge over Feinstein, the high number of undecideds means that the size of the margin of victory is an uncertainty in this race,” said Mark Baldassare, who conducted the poll for The Times. “Wilson’s current vote margin is well below Deukmejian’s in 1982 and 1986.”

In the race for lieutenant governor, the poll also found Orange County’s native daughter--Marian Bergeson--running ahead of incumbent Democrat Leo T. McCarthy but far short of the winning margin she should have in California’s Republican mother lode.

Bergeson, a state senator from Newport Beach, was ahead 35% to 26% and she was leading in all of the age, sex and geographic categories. McCarthy was slightly ahead, however, among voters earning annual incomes under $35,000.

Bergeson’s position was particularly weak considering that 4 of 10 voters from the home territory she has served since 1984 were still undecided.

The telephone poll of 486 Orange County registered voters was conducted Sept. 4 through 6 by Mark Baldassare & Associates. The margin of error for the entire sample is plus or minus 4.5%.

Orange County is such a stronghold of the Republican Party that GOP strategists consider it a key benchmark of their performance in statewide campaigns.

Generally, they figure statewide candidates need a victory margin in Orange County of at least 200,000 votes in order to offset Democratic strongholds in San Francisco and Los Angeles.

Baldassare said Wilson’s 26-point lead in The Times Orange County Poll represents about 174,000 votes. In comparison, Wilson carried Orange County by 323,000 votes when he won reelection to the U.S. Senate in 1988 over Democrat McCarthy.

Bergeson is running in her first statewide general election, and Baldassare said her nine-point lead over McCarthy represents less than a 10,000-vote edge.

A statewide poll conducted by The Times two weeks ago showed McCarthy leading Bergeson 35% to 17% with nearly half of the voters undecided.

Statewide in the governor’s race, The Times Poll found that Wilson and Feinstein were running dead even, both with 39% of the vote. That poll also found that the strong support from women that Feinstein enjoyed in the primary had faded.

Interestingly, The Times Orange County Poll found sex to be a slight advantage for Feinstein but no help for Bergeson.

Feinstein was supported by 28% of the county’s women voters, six points ahead of her support among all voters. But Wilson had a strong lead among men. Even a surprising 31% of Democratic men favored Wilson. He was also backed by 73% of the Republican men.

“The gender advantage (for Feinstein) fades among Republican women, where only 13% intend to vote for Feinstein and 56% support Wilson,” Baldassare said.

Bergeson, meanwhile, had almost identical support from men and women.

One possible reason for the flat performance by Bergeson among women voters could be the issue of abortion. Bergeson is opposed to legalized abortion while McCarthy supports abortion rights.

Both Feinstein and Wilson also support abortion rights.

Feinstein’s support improved slightly among middle-income and South County voters. Wilson’s lead, on the other hand, expanded significantly (57%-18%) among voters earning annual incomes over $75,000.

Wilson was also supported by 65% of the Republican voters but the poll found fewer than half of the Democratic voters (44%) backing Feinstein. Nearly 1 in 3 Democrats was undecided in the governor’s race.

In the lieutenant governor’s race, the poll also found that Bergeson may have been right when she claimed in her primary that Wilson would benefit from her placement on the GOP ticket.

“In what could be a factor in this expected close election, lieutenant governor candidate Bergeson could end up assisting Wilson win votes more than McCarthy helps Feinstein,” Baldassare said. “Three in four Bergeson supporters would vote for Wilson for governor, while only 45% of McCarthy voters say they will back Feinstein.”

TIMES ORANGE COUNTY POLL

“If the election for governor were held today, whom would you vote for?” Pete Wilson: 48% Dianne Feinstein: 22% Other/ Don’t know: 30% All Voters CALIFORNIA GOVERNOR’S RACE

POLITICAL AFFILIATION

Dianne Pete Other/ Feinstein Wilson Don’t know Democrats 44% 22% 34% Republicans 10 65 25 Independents/others 25 31 44 SEX Men 17% 57% 26% Women 28 39 33

All Voters Marian Bergeson: 35% Leo McCarthy: 26% Other/ Don’t know: 39% CALIFORNIA LT. GOVERNOR’S RACE

POLITICAL AFFILIATION

Leo Marian Other/ McCarthy Bergeson Don’t know Democrats 50% 13% 37% Republicans 14 50 36 Independents/others 19 21 60 SEX Men 24% 36% 40% Women 27 35 38

Source : Times Orange County Poll How the Poll Was Conducted

The Times Orange County poll was conducted Sept. 4 through 6 by Mark Baldassare & Associates. The telephone survey of 486 registered voters was conducted on weeknights using a random sample of listed and unlisted Orange County telephone numbers.

The survey included a question screening for registered voters. An equal number of men and women were interviewed. The sample was statistically weighted to reflect the actual population distribution of Orange County registered voters.

For a sample of this size, the margin of error is plus or minus 4.5% for the total sample. For subgroups such as Democrats and Republicans, it would be larger.

Sampling error is just one type of error that can affect opinion polls. Results also can be affected by question wording, survey timing and other variables.