They might be. I think they have lists of pre-screened poll participants in their databases and are able to pull a scientific sample online from known participants. In these days, they might be more able to obtain a sufficient number of responses from a balanced demographic than firms still relying on telephone contact.
The polling industry is in a state of flux. Pollsters who conduct surveys for partisan clients and for media organizations are struggling to figure out how to understand the American electorate, especially at a moment when more of us are eschewing landline phones in favor of cell phones.
At Morning Consult, we’re embracing a new way of contacting voters, one that we think gives us the ability to survey voters and model the electorate in a timely, accurate manner. And we want to be transparent about that process. To that end, we offer answers to these frequently asked questions about how we conduct our polls:
Morning Consult regularly runs surveys containing both Morning Consult and client content. When we have room for additional questions (which happens in almost every poll we run), our editorial and polling teams independently write questions. Those questions are never vetted or approved by any clients. The data produced from questions that the Morning Consult editorial and polling teams write is proprietary to Morning Consult.
In many polls, especially statewide surveys, we use both live telephone interviews conducted using random digit dialing (RDD) and online interviews using several well-known, well-respected national vendors to provide a broad cross-section of registered voters across the country. For our weekly 2,000 person national polls, we use several national online vendors to recruit a broad cross-section of Americans.
In every story based on Morning Consult polling, we will disclose how the poll was conducted, whether by phone, online or a mix of both.
Nope, our telephone interviews use live callers and are conducted using random digit dialing (RDD).
No, but we make up for those we miss on the phone with our online surveys. Here’s how it works: The landline telephone interviews are used to target landline only, landline mostly and mixed landline/cell individuals, while the online interviews are used to target cell only, cell mostly and mixed landline/cell individuals. We use the distribution of landline and online interviews to approximate the true proportion of cell-phone only, mixed cell and landline, and landline-only individuals based on estimates from the National Health Interview Survey (NHIS).
The internet component of our polls are based on opt-in panels compiled by several large national firms. That gives us access to millions of potential respondents about whom we know a lot — where they live, their gender, their educational attainment and ethnicity, things like that. That means we can refine our panels to make them representative of the country at large, which helps us build more accurate samples.
For our large, weekly national polls, we include a few additional variables for weighting such as educational attainment and region. These population figures also come from the 2012 Current Population Survey. For our mixed methodology landline and online poll, we weight survey results based on age, gender, and race/ethnicity to population parameters from the 2012 Current Population Survey.
Yes. Online polling is increasingly common among political professionals. In just the last few years, the Democratic Congressional Campaign Committee, the National Republican Congressional Committee, the National Republican Senatorial Committee and other prominent groups have used online polling to supplement their data from elsewhere. And other respected media outlets like The Economist, The New York Times and CBS have partnered with YouGov to conduct their own online polls.
We’re hardly the first, or the only, group conducting these types of polls.
“I think they have lists of pre-screened poll participants in their databases and are able to pull a scientific sample online from known participants. In these days, they might be more able to obtain a sufficient number of responses from a balanced demographic than firms still relying on telephone contact.”
For two presidential campaigns, Bush II’s, I was on two pre-screened polling companies’s lists.
I knew that I would be called and what the questions were before the polling calls. They called ahead re the date and time and the questions.
We have a couple of younger relatives without landlines, and they are contacted via their cell phones by legit pollsters.