Cal State Fullerton News California State University, Fullerton
Public Affairs
800 N. State College Blvd. Fullerton, CA 92632
657-278-2414 Fax 657-278-5226
www.fullerton.edu/news/

 

back

County Residents Support CenterLine Plans

 

Updated September 25, 2003

Fifty-five percent of Orange County residents favor building a light rail line from Santa Ana to John Wayne Airport, according to survey findings reported last week by the Center for Public Policy in cooperation with the Orange County Business Council.

In a telephone survey of 506 county residents, polled between Aug. 26 and Sept. 10, respondents agreed or strongly agreed that light rail should be part of the mix of Orange County transportation and that building light rail, even for a short distance, could serve as a starter for the future. The survey was conducted by the university’s Social Science Research Center.

“Perhaps the most interesting observation is the agreement with the concept of light rail, on what we might call a philosophical level,” said Keith Boyum, center director and associate vice president of academic programs. “Respondents agree that light rail should be part of the mix and six out of 10 say they’d like to be trolley-riders themselves. However, these are not riders of public transportation. They like the idea of light rail.

“What people like in concept is frequently not what they want in their backyards. In Irvine, residents voted no on CenterLine – not as a concept but as a proposed set of rails in neighborhoods.”

“When people are presented with details about the light rail proposals, Orange County residents are supportive of such a transportation option, even when their personal experiences have been conditioned by an ‘automobility’ of freeway dependency,” added Ray Young, professor of geography.

Although the survey was thorough, it cannot predict a hypothetical vote for rail. Variables such as voter turnout, budgets, the line-up of candidates and other issues on the ballot could affect the outcome of the CenterLine project, noted Boyum, a professor of political science. However, respondent attitudes were probed with researchers mimicking, at least in part, some of the information that a campaign about CenterLine might evoke. In fact, many residents weren’t that familiar with the CenterLine proposal. Respondents also were more likely to support light rail if it required no new taxes and if later extensions would serve more areas of Orange County.


Latest Findings, CSUF – OCBC Quarterly Survey:

County Residents Support CenterLine Plans:
• Would vote 55% “yes” on hypothetical ballot measure
• Do not themselves use public transportation
• Six out of ten rate current public transportation positively
• Disapprove of Legislature

 

Orange County Residents Say ‘Yes’ to CenterLine


Some 55% of Orange County residents would – if the question were put before them today – vote in favor of building a light rail line from Santa Ana to John Wayne Airport, according to the latest survey findings from Cal State Fullerton’s Center for Public Policy in partnership with the Orange County Business Council.

Of the 506 Orange County residents reached in a telephone survey conducted between Aug. 26 and Sept. 10, 2003, 45% said they would vote against such a proposal.

The survey employed the question wording proposed by Supervisor Chris Norby, as follows:

Now I will read you a possible ballot question, which is this: “Shall the County of Orange support the building of a $1 billion light rail line from Santa Ana to John Wayne Airport, with possible future exten-sions?” If you were to vote today on that ballot question, would you vote yes to approve building light rail, or no to oppose building light rail?

Norby, a CenterLine skeptic / opponent, proposed his ballot measure language in the context of a debate in which he took the view that Orange County voters, if asked, would defeat a CenterLine plan.

“This survey cannot with certainty predict a hypothetical election, of course,” noted Keith Boyum, director of the CSUF Center for Public Policy. “Variables including turnout, the budgets available to groups favoring and groups opposing CenterLine, the lineup of other candidates and issues on the ballot, and many other standard political factors would surely enter in, and cannot be accounted for here.

“However,” continued Boyum, “in this survey, we tried to be especially thorough. We sought to probe respondent attitudes in a variety of ways, and to provide pro / con information for them to consider – mimicking, thus, at least a part of the information exchange that a campaign about CenterLine might evoke.”

In this survey, Democrats were somewhat more likely to say that they would vote “yes” than Republicans. Democrats broke 61% - 39% in favor of the hypothetical ballot measure. Republicans were evenly divided, at 50% - 50%.

Orange County residents do not ride buses, commuter trains. Beginning the series of questions that preceded the question about a hypothetical ballot measure, the survey inquired first:

How often would you say that you take either a bus or a commuter train for travel in Orange County?

Some 83% responded “almost never,” with another 8% responding “Rarely.”

“With about nine out of ten in the survey reporting public transportation ridership varying from rarely to almost never, an analysis of these data as between those who ride buses and commuter trains and those who do not will not make sense,” noted Boyum. “These data should thus be taken as a fair Orange County sample – we surveyed mostly folks who commute via car. But here we asked them a series of questions about public transportation.

Six out of ten give generally favorable ratings to Orange County public transportation. The survey continued with a rating question, as follows:

Still thinking about buses and commuter trains, would you rate public transportation in Orange County as very poor, poor, good, or very good?

As Table One indicates, Orange County public transportation won more positive than negative judgments – as rated by this sample of non-users of public transportation.


Table One

Ratings of Public Transportation in Orange County

 
Very Poor
14 %
Poor
29 %
Good
46 %
Very Good
11 %

County residents unfamiliar with CenterLine proposals. Turning specifically to the CenterLine idea, 39% of Orange County residents reported that they were either “somewhat” familiar [29%] or “very” familiar [10%].

“These data say clearly that most people don’t know much about a CenterLine proposal, all in all,” noted Boyum. “That’s not a surprising finding, since we understand the low level of information about local issues held by ordinary people – and particularly, ordinary people in a county whose media are largely based in a different county, in Los Angeles. Indeed, in anticipation of just such a finding of low familiarity, the survey asked a substantial series of further questions and probes into respondent attitudes.

”Pros and cons of light rail; alternatives to light rail; light rail as part of a desired public transportation mix. Respondents were next asked a series of questions that included two focused on alternatives to light rail [taking the form, “Rather than build light rail, we should …”] and two mildly supportive, together with a question about personal interest in riding light rail. The order of the questions was rotated, meaning that each question had an equal chance of being asked first in order, second in order, etc.

Responses are shown in Table Two.

Table Two

Agree/Disagree With Building Light Rail and Alternatives to Light Rail*

 
Strongly
Agree
Agree
Disagree
Strongly
Disagree
Light rail should be part of the mix of transportation in Orange County.
43%
38%
9%
11%
We should build light rail even for a short distance, as a starter for the future.
33%
40%
11%
16%
Rather than build light rail, we should widen the 22 freeway, or extend the 57 freeway to the 405.
31%
22%
22%
25%
Rather than build light rail, we should improve the bus system.
21%
29%
30%
21%
I would like to use light rail myself.
27%
35%
13%
26%
*Rounding errors may make some totals equal other-than 100%

“Perhaps the most interesting observation about these numbers is the agreement with the concept of light rail, at what we might call a philosophical level,” noted Boyum. “Light rail should be part of the mix, respondents agree. Let’s start for the future, they agree, even if that means accepting a short line. And six out of ten say they’d like to be trolley-riders, themselves. All in all, respondents seem to have something of an ideal mix of local public transportation in mind.”

“Remember, these are not riders of public transportation,” commented Phillip Gianos, CSUF professor of political science. “Yet clearly they like the idea of light rail; and only a tepid majority arises in these data for what might be considered attractive alternatives for a car-driving population: improving or extending freeways.”

Julie Puentes, executive vice president, Orange County Business Council, commented: “Respondents are saying clearly that transportation and congestion remain top issues for Orange County and that a comprehensive mix of solutions, including both rail and highway improvements, is needed. Support for rail is extraordinarily strong, especially considering that respondents say they won't use it.”

“Ms. Puentes is right,” commented Gianos. “It raises an intriguing further question – about which we can speculate. It is whether our respondents are pursuing an idea about the overall good for the general public. Call this a "public-regarding" attitude or point of view. Or to the contrary: are our respondents looking out for themselves-- thinking that if someone else uses public transportation and I don't, congestion will be relieved for me as a by-product?”

“I think we also have to take NIMBY – ‘not in my back yard’ – into account,” said Boyum. “People in Irvine voted no on CenterLine – not as a concept, but as a proposed set of rails in neighborhoods. What people like in concept is frequently not what they want in their back yards.”

Statements pro and con about CenterLine. Pressing on, the survey posed to respondents a series of statements about light rail in Orange County. This served two purposes. First, it allowed further probes into the apparent interest in building light rail in a county and region known for freeways. Second, taken as a package, the statements could be taken as foreshadowing – if imperfectly – some of the information pro and con that would emerge in a campaign if a measure about CenterLine were to be placed on the ballot.

We asked respondents the following question.

Now I want to read you some statements about the proposed CenterLine light rail. Please tell me whether each statement makes you much less, a little less, a little more, or much more likely to support the proposed construction.

The questions were rotated during the administration of the survey, meaning that the order in which they were asked changed from respondent to respondent, with each question having equal turns to be asked first, to be asked second, etc.

Results are shown in Table Three.

Table Three

More / Less Likely to Support Light Rail, Given Each Statement*

Each statement (below)
makes respondent . . . . . >
Much more likely to support
More likely to support
Less likely to support
Much less likely to support
Light rail will require no new local taxes.
50%
30%
8%
13%
Later extensions of light rail, if built, would serve more areas of the county.
44%
37%
7%
11%
About 84% of the cost for light rail would be paid by the federal and the state governments.
40%
32%
13%
15%
The local share of the light rail cost would come from taxes already being collected.
37%
32%
1%
17%
People in Irvine voted against CenterLine for their city.
30%
22%
26%
22%
Light rail may get very few people to stop driving their cars.
23%
24%
27%
26%
Initially, light rail would run only from Santa Ana to John Wayne Airport.
19%
29%
23%
30%
*Rounding errors may make some totals equal other-than 100%

“People obviously react well to the idea that no new local taxes would be required to pay for CenterLine,” observed Gianos.

“Seven or eight out of ten among our respondents reacted very positively to the generally supportive statements about building a light rail line,” continued Gianos. “Meanwhile, however, the fact that the line would be short, and not get people out of cars, seemed to make only small majorities of our respondents think a less-positive thought about trolleys. The fact that Irvine voters spurned light rail got only about half of our respondents to say that, yes, that statement made them think less well of the rail proposal.”

Ray Young, professor of geography at CSUF, is the author of Commuting in Orange County, a report published by the Cal State Fullerton Center for Demographic Research (December 2002). Contact: 657-278-3528

Young comments about this survey:

When presented with details about the light rail proposals, Orange countians are supportive of such a transportation option, even when their personal experiences have been conditioned by an ‘automobility’ of freeway dependency.

It is noteworthy that the public is more receptive to a light rail future than to bus system expansion. That may be linked to the prospect of greater speed as well as the novelty of a rail starter line, linking major activity nodes and job centers in the county.

Remember that the light rail concept was approved as part of the Measure M funding initiative in 1990, a process that led to substantial freeway widenings and bus line improvements. Considerable dollars have already been expended on the planning phases for light rail options as part of a diversified transportation system to serve the county’s growing population and projected job increases.

_________________________

Previous results of CSUF/OCBC quarterly surveys are conveniently accessible on the OCBC web site. See: http://www.ocbc.org/resourcesf.htm
_________________________

These data result from a Random Digit Dialed survey of 506 households in Orange County. The survey was administered by telephone between Aug. 26 and Sept. 10, 2003, by the California State University, Fullerton Social Science Research Center (SSRC) for the university’s Center for Public Policy. The SSRC director is Gregory Robinson.

The population of inference is heads of household or their spouses or domestic partners, 18 years of age or older, residing in households with telephones in Orange County. Interviews typically ran about 11 minutes.

Computer-assisted telephone interviewing software was utilized, supporting highly accurate call management. For example, up to 21 call-back attempts were made in some cases to obtain completed interviews. Telephone interviews were generally conducted Monday through Thursday from 4 p.m. until 9 p.m., and on Saturdays and Sundays from 2 - 8 p.m.. All interviews were conducted in English.

The response rate for this telephone survey is 67.66%, calculated as completed interviews as a proportion of eligible respondents. Calculated conservatively, the margin of error for a random sample of this size is plus or minus 4.45 percent. The margin of error will be larger when sub-groups of the sample are analyzed.