If you’re uncertain about local support for Oregon State University-Cascades campus’ developing west-side location, an online poll by Truth in Site won’t clarify the situation. By most measures of good polling, it comes up short.

Good polls, those that accurately reflect a population’s opinion about something, require at least three things, according to Russell Renka, a political science professor at Southern Missouri University. Miss any one of the three, he says, and results will be invalid.

No. 1: The questions asked must be clear, written in neutral language and provide a range of answers to chose from. The Truth in Site survey language, at times, seems to push a response. One question is: “Do you feel that Central Oregon might be better served by a campus that is located closer to Highway 97 and more accessible to those who live outside of Bend?” Well, of course, it might.

Meanwhile, Now for Bend, a group of OSU-Cascades backers, say they paid for a statistically valid survey and have so far been unwilling to make their survey’s questions public. That’s too bad, for it denies the public a means of judging the survey’s impartiality.

No. 2: The people questioned must be randomly selected — every person in the community must have an equal chance of being chosen to take the poll. As Renka puts it, “Any self-selected sample is basically worthless as a source of information beyond the population (answering) itself.”

Online surveys like Truth in Site’s do not involve random selection. Now for Bend says its survey was of randomly selected adults.

No. 3: The sample involved must be large enough to keep the margin of error fairly small, about 5 percent. That’s generally 400 or so people, Renka says. The number of respondents doesn’t matter much here, however, because the Truth in Site survey already violates the first two requirements. Now for Bend’s survey involved 300 adults with a margin of error of 5.6 percent.

We don’t know when Truth in Site plans to announce the results of its online survey, but don’t put much stock in it.