Tuesday, May 29, 2007

A suspect poll

Siena’s poll results on gay marriage—which found that 47% of New Yorkers opposed and 43% in support—leave out some important pieces of the puzzle.

We know that 620 registered voters were polled, but we don’t know the breakdown of their demographics—a crucial element in determining the validity of the findings because it demonstrates whether or not the poll adheres to the reality of voter demographics in New York State. For example:
  • The poll points out that a majority of Democrats support marriage equality. But it does not tell us how many/what percentage of the respondents were actually Democrats. If the poll is going to accurately represent New York, there needs to be 2.5 registered Democrats polled for every 1 registered Republican.

  • The poll also tells us that a majority of voters under the age of 55 support marriage equality, even though the overall number of those polled oppose rather than support the law. Because we do not know how many/what percentage of the respondents are above/below 55-years old or what the breakdown is for each age group, it's hard to explain this result. So again, more questions here than answers.

The Pride Agenda’s most recent polling shows almost a flip of what Siena reports—that 53% of registered New York voters support marriage equality, while 38% are opposed. Given that we have two polls and only one provides methodology and a demographic breakdown of all respondents (see page 3), one could easily assume that the findings in the Siena poll just might not mirror the reality of NYS voter demographics.

Siena needs to provide a bit more transparency in order for these findings to be respected.


KipEsquire said...

there needs to be 2.5 registered Democrats polled for every 1 registered Republican"

More like 1.6 to 1, but your point is (mostly) valid.

Anonymous said...

Was Siena's poll requested by a particular organization that is not favorable to equality for LGBT folks? Or has Siena made any mention that the transparency concerns would be addressed?
We should wait and see what the answers to those questions are before we jump to conclusions, people...