This first Crimson survey is the result of 397 telephone interviews with randomly selected undergraduates.

The Crimson selected students for the sample from a computer-generated list of random dormitory telephone numbers, and conducted the interviews in the late afternoons and evenings last week.

The Crimson staff members who did the nearly 400 four-minute interviews had only the telephone numbers of the people called, so all respondents remain anonymous.

Almost without exception, undergraduates were willing to take the time to answer the 15 Crimson questions.

Statistically, the results of the survey have a standard error of between 3 and 6 per cent at a 95 per cent confidence interval. This means that for 19 of every 20 samples, the results received will be within a few per cent of the actual figure for the complete undergraduate body.

The Crimson is reporting differences in the responses for crosstabulated variables only if they are significant at a level of less than .05. This means, for example, that a difference in preference for equal access admissions broken down by sex was reported only if there was a less than one out of twenty chance that the difference could have occurred by chance.

The key demographic characteristics of The Crimson sample for sex, House, class and concentration are all within 5 per cent of the actual figures for the undergraduate population.