**How the poll was conducted:** Unless otherwise noted, all tables are based on the March 10, 1992 exit poll of Hispanic voters in the Texas Democratic primary, conducted by the Southwest Voter Research Institute (Now William C. Velasquez Institute). All numbers, except n's are percents.

The poll was conducted among 974 Hispanic voters in 25 polling places and has a margin of error of plus or minus 3 to4 percentage points.

The sampling design, prepared by Karol Krolki, sampling statistician at the Institute for Survey Research, Temple University, Philadelphia, is a two-stage equal probability sample of Hispanic voters.

In the first stage, precincts were selected with probabilities proportional to the number of expected Hispanic voters. Precincts with at least 20 Hispanic voters were included in the sample. This results in an overall coverage rate of 91% for Hispanic voters.

In the second stage of sampling, voters were selected systematically throughout the day until 6 p.m. within each of the 25 precincts drawn in stage one at a rate that gives all voters in a precinct the same chance of being interviewd.

All survey estimates generated from this poll have been weighted to reflect acutal distribution of Hispanic registered voters.

The sampling error for most questions is plus or minus about 3 to 4 percentage points with 95% confidence. That means that in 19 out of 20 cases, the results from the poll should differ by no more than 4 percentage points in either direction from what would have been obtained by interviewing all Hispanic voters in the state.

The sampling error depends, for the most part, on the number of respondents in each group. Therefore, results based on subgroups within the sample will have larger margins of error.

The source for all Texas March 1992 tables are SVRI's 1992 Exit Poll, except where indicated.