Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

Ranting Randy

(101 posts)
Fri Oct 25, 2024, 03:52 PM Oct 25

Why We Have Polling Errors (Don't wring your hands, ring your neighbors)

I worked in market research for a few years, and have a master’s degree focused on research, sampling, statistics, etc. I want to explain why polling results are so incredibly erratic.

The idea when developing a poll that represents your community is that polling companies have to contact enough people who represent the demographics in your community. You want this percentage of men, that percentage of women; a specific percentage of specific minorities; a specific percent of people who are certain ages, religions, political parties, zip codes, incomes; etc. Now the question is, how many calls do you need to make to develop a survey group who will represent all these different sub groups?

Remember, these sub groups and all the demographics vary from city to city, across the country. You will find a different community in South Florida than you will find in Northern Minnesota. Another and possibly worse problem is that a huge number of people don’t answer their phones any more. Anyone with a cell phone can see if the caller is someone you know and if the number is hidden. So we skip the call. Plus younger people don’t answer phones as much and especially don’t like to take time out to talk with strangers. They like to text. Their voice mail is often full. So it’s hard to get the majority of Americans to participate in polling, and not quite impossible to get enough younger people’s opinions into a poll to make them representative of reality.

Soooo, which groups do answer their phone? Generally it’s those with land lines, those who who have time to talk, those who aren’t working and are retired. And this group tends to skew more conservative than those who are younger.

Now you know why polling is often skewed. So what is the answer​?

For polling companies the answer is to dramatically increase the survey’s sample size. This will increase your reliability. This is expensive and time consuming; they can’t do it. For those of us who read the results of the polls, please reflect back on how horribly wrong the polls have been in just the past election. We were told a red wave, a red tsunami was coming; and quite frankly, they were wayyyyy off base. We actually had a blue wave.

So, what do we do with this information? We vote, we get our neighbors to the polls, we make calls, we talk with our neighbors, and no matter what the election results are … no matter what happens between the election, inauguration day, and the next election, our actions are the same. If we keep involved, we will make a difference.

The last time we had this type of turmoil was during the late 60s and early 70s. Good people stayed involved in the election process, and as a result we got the Voting Rights Act, The Clean Water Act, the Civil Rights Act, The Clean Air Act, and a half dozen pieces of major legislation that made our country a better place.

I guarantee that Trump will not get the support of the majority of America. He’s come in second in both of his previous election campaigns, and not even for a minute while he was in office did he have the support of a majority of Americans. Not once during his term in office was his popularity even close to 50%.

So for those of you who are concerned, my answer is to channel your anxiety and get busy. Don’t wring your hands and worry, ring your neighbors and make sure they vote.

Latest Discussions»Issue Forums»Campaigns»Why We Have Polling Error...