An article in iMedia Ethics looks at how questions can affect the results of polls. It examines two Gallop polls attempting to gauge American opinion on gun control. In the span of one month, the two polls showed different results. The article cites the different types of questions as being behind the different outcomes.
Other ways that polls can produce different results is in the sampling in the first place, with many polls tilting the sampling towards groups that would traditionally favor the outcome the poll might be designed to produce. Polling is an effective tool to use what I call Social Influence as a way to influence the action of others.
If most people think a certain way, it causes many to adapt that belief. This is what polls are often designed to do. This is not to say that polls are all corrupt or that no poll should be trusted. This is to say, though, that take polls with a grain of salt, and even if they are accurate (unless you’re still engaged in the voting game, or if the poll is measuring market response to a product or service you’re selling), don’t base your position, or defend your position, on how well it scores in the polls.
In November, Gallup reported that just a bare majority of Americans would rather the government approve new gun laws instead of only enforcing existing laws.
The previous month, however, referring to the very same poll (but different questions), Gallup came to a starkly different conclusion: “The great majority of Americans are in favor of more stringent regulation of the sale and ownership of guns in three ways that go beyond current law in most states.” (emphasis added)
How did Gallup find barely a majority of Americans supporting new gun laws and, in the same poll, find virtually everyone supporting at least one new gun law, with large majorities supporting each of the three new laws included in the poll?
Different Questions, Different Results
The two Gallup reports in October and November were based on the same poll that was conducted Oct. 5-11. Numerous questions about guns (as well as other items) were included in the poll.
Q19 in the survey asked:
“In terms of gun laws in the United States, which of the following would you prefer to see happen — enforce the current gun laws more strictly and NOT pass new gun laws [47%] or pass new gun laws in addition to enforcing the current laws more strictly [51%]?”
Several questions later, Q24 asked:
“Please say whether you favor or oppose each of the following: [RANDOM ORDER]
- “Requiring background checks for all gun purchases (96% – 4%)
- “Enacting a 30 day waiting period for all gun sales (75% – 24%)
- “Requiring all privately owned guns to be registered with the police (70% – 29%)”
All of the items in Q24 represent potential “new” gun laws. As Gallup acknowledges, while federal law requires background checks for most gun purchases, there are significant exceptions – and most states have not prohibited such exceptions. Also, there is no federal 30-day waiting period, nor a federal gun registry.
Read More at iMedia Ethics