Opponents of democratic judicial elections seem worried about a new American Justice Partnership Foundation poll that revealed an overwhelming 75% of Americans believe state judges should be elected. They know they’ve got a losing hand and can’t attack the results – so they attack the poll and the pollster.

I’ve already responded to the personal attacks on Ayres, McHenry, one of the most prestigious polling firms in the country. Now let’s tackle the other main criticism – the poll’s sample size of 800, which opponents claim isn’t big enough.

Sample sizes of 800-1200 are common for national surveys. A national USA Today/Gallup poll taken May 31-June 3 had a sample size of 803. A Time magazine national survey in late June had a sample of 805. NPR did a national poll in May with 800 likely voters as its sample size. CBS/NY Times and NBC News/Wall Street Journal also have done recent polls with sample sizes of 601 and 700 respectively. Nobody argues that these polls are somehow illegitimate.

The fact is, there is no one “correct” sample size. The main reason to do a larger sample is to be able to break down the sample into smaller subgroups. The AJPF poll’s margin of error of +/- 3.46, compared to a MoE of +/- 3.10 for a sample size of 1000 demonstrates both the robustness of the results and the shallowness of their criticism.

Another attack focuses on the language used to describe the pro-merit selection arguments, which critics claim wasn’t favorable enough to their side. Opponents of democratic judicial elections go to extreme lengths to conceal their true intentions, cloaking their campaign in good government language. The language used in the poll may not have dutifully repeated their propaganda, but any objective observer will have to conclude that the questions were fair and balanced.