The National Organization for Marriage (NOM), an anti-gay group that has fought against gay rights in almost every state, has released a poll stating that 54% of Maryland voters oppose gay marriage, and that 78% of them believe the issue of same-sex marriage should be decided by the people (via referendum) and not the legislature. This contrasts from an earlier poll, which showed Maryland voters favoring gay marriage 51-44%.

Given that the poll was released by NOM and conducted by an anti-gay polling organization, Lawrence Research, LGBT rights supporters are trying to figure out how this poll is biased, if it is. After quickly scanning the Methodology section of the report, I was able to quickly ascertain exactly where the bias comes from.

The key is in this sentence: “Lawrence Research based in Orange County, California conducted this poll using random digit dialing to reach and complete interviews with 600 self-identified registered voters.” Random digit dialing is a way for polling organizations to quickly and randomly select people to poll over the telephone; it is a common tactic used by many polling organizations to conduct polls by phone.

The problem, though, is that it is illegal to include cell phone numbers in random digit dialing lists under the Telephone Consumer Protection Act. And given that over 1/3 of people under the age of 35 (including 1/2 of people between the ages of 25 and 29) use only a mobile phone and not a landline, polls utilizing random digit dialing are, by default, automatically going to be biased towards older respondents.

It is no secret that younger people favor marriage equality more than older people. That means that NOM's poll specifically left out a large segment of the pro-gay population in Maryland: young people. The other poll, conducted by Gonzales Research and Marketing Strategies, also polled by phone, but did not use random digit dialing and thus did not have this problem.

Furthermore, Gonzales' poll had a larger sample size and a typical margin of error of only 3.5%, whereas NOM's poll had a margin of error of 4%, much higher than normal and indicative of a sample that was not truly representative of the population.

NOM's poll should not be regarded as a legitimate or accurate depiction of Maryland residents' feelings towards same-sex marriage. Gonzales' poll is far more indicative of Marylanders' true support for marriage equality.