How the National Survey Was Taken

This is the most extensive opinion survey of the nation’s Catholic priests since a Los Angeles Times poll of priests and nuns conducted over several months in 1993 and 1994.

Opinion surveys of the general population usually are conducted by telephone, using random-digit dialing. For a small population such as priests, however, that method is impractical.

Instead, The Times Poll chose a sample of 5,000 active and retired priests across the country in 80 dioceses and mailed surveys to them. The packet included a cover letter promising that all results would remain anonymous.

The sample was designed to be geographically distributed in the same way as the 45,382 priests who make up the total U.S. priest population. Diocesan priests and those in religious orders were included in proportion to their overall share of the total priest population in each geographic area.

Addresses and population counts were taken from directories of dioceses and religious orders and from the Official Catholic Directory published by P.J. Kennedy & Sons.

To achieve the largest possible return rate, the poll contacted respondents four times, mailing questionnaires June 27 and July 25, a reminder postcard Aug. 5, and a final set of questionnaire packets Sept. 4. The end date of the survey was Oct. 11, giving it a 16-week field period.

The poll was criticized by some Catholic organizations that said it asked inappropriate questions. Nonetheless, it achieved a response rate of 37%, for a total of 1,854 respondents nationwide, which is considered statistically representative.

Response rates were acceptable in all the dioceses surveyed, and ranged from 30% in some dioceses in the South to 44% in some parts of the Midwest. The margin of error for the sample is plus or minus 3 percentage points.

The sample of priests who responded to the survey slightly over-represents active diocesan priests as compared with those who belong to religious orders. It also slightly over-represents priests in the South as compared with those in the East and Midwest. The final results were adjusted slightly to account for the differences.

Besides the survey questions, the poll invited respondents to submit additional written comments. Some priests also agreed to speak with reporters for follow-up interviews.

In the written comments, some priests praised the survey. Others criticized the wording of some of the 67 questions.

The most common complaint was that some questions required a more complicated answer than simply yes or no.

Others suggested that certain items were designed to produce responses that would embarrass the church.

The survey was supervised by Los Angeles Times Poll Director Susan Pinkus along with Associate Director Jill Darling Richardson and Field Director Roger Richardson. Claudia Vaughn was the data management supervisor, and Ray Enslow was the publications coordinator.

Further information on this study is available by writing to Los Angeles Times Poll, 202 W. 1st St., Los Angeles, CA 90012-4105. In addition, Times Poll data and question wording may be found online at www.latimes.com/home/news/polls.