Yes, political pollsters absolutely call cell phones, and in fact, the vast majority of telephone interviews conducted today are completed on mobile devices. Because over 70% of American adults now live in wireless-only households, reaching cell phone users is essential for pollsters to produce accurate, representative data.
To understand how modern polling works, you have to understand how the industry adapted to the death of the landline. If pollsters only called landlines today, their samples would skew dramatically older, wealthier, and more conservative than the actual voting public. Calling cell phones is no longer optional—it is the backbone of high-quality telephone polling.
However, calling mobile phones is vastly different, more expensive, and more legally complicated than calling traditional landlines. Here is a look behind the scenes of how pollsters dial cell phones, how they find your number, and how they make sense of the data they collect.
The Legal and Financial Reality of Cell Phone Polling #
For decades, telephone polling was relatively simple and cheap. Pollsters used automated dialers to call random landline numbers, and when someone picked up, a live interviewer or an automated voice system would conduct the survey.
But when cell phones became ubiquitous, the legal landscape shifted.
The TCPA and Manual Dialing #
Under the Telephone Consumer Protection Act (TCPA) of 1991, it is illegal to use an “automatic telephone dialing system” (commonly called an autodialer or robocaller) to call a cell phone without the recipient’s prior express consent.
Because political polling calls are unsolicited, pollsters cannot use automated systems to dial cell phones. Instead, every single mobile number must be dialed manually by a human being. A live interviewer must sit at a workstation, look at the phone number on a screen, and physically press the buttons or click to initiate the call.
The Soaring Cost of Polling #
This manual dialing requirement makes cell phone polling incredibly expensive.
- Landline calls can be automated and cost very little per dial.
- Cell phone calls require paying live, trained interviewers for every minute they spend manually dialing numbers, waiting for rings, and navigating voicemail greeting messages.
Because the vast majority of Americans do not answer calls from unknown numbers, interviewers have to dial hundreds of cell phone numbers just to get a single completed interview. This has driven the cost of a high-quality, live-caller telephone poll of 1,000 registered voters up to $50,000 or even $100,000.
To manage these costs, most reputable polling organizations now use a “mixed-mode” approach, combining a smaller sample of live cell phone calls with online surveys and text-to-web invitations. When you track real-time presidential polling averages, you are looking at the combined results of these various methodologies, each trying to solve the puzzle of reaching a representative sample of Americans.
How Do Pollsters Get Your Cell Phone Number? #
Pollsters do not just guess phone numbers, nor do they buy random lists of “leads” from telemarketers. Instead, they rely on two primary scientific sampling methods to build their calling lists.
1. Random Digit Dialing (RDD) #
Random Digit Dialing is the traditional gold standard of scientific polling. A computer program generates random phone numbers using active area codes and telephone exchanges (the first three digits after the area code).
- The Benefit: RDD captures everyone with a phone number, including people who have unlisted numbers, people who recently moved, and people who are not registered to vote.
- The Downside: RDD is highly inefficient. Many generated numbers are disconnected, belong to businesses, or belong to people who are ineligible to vote (such as non-citizens or minors).
2. Registration-Based Sampling (RBS) #
In recent years, many pollsters have shifted to Registration-Based Sampling. Instead of dialing random numbers, they pull calling lists directly from official state voter registration files. These files contain public records of registered voters, including their names, addresses, party registration, and voting history.
Commercial data vendors match these voter files with consumer databases to append landline and cell phone numbers to each voter’s profile.
- The Benefit: Pollsters know exactly who they are calling. They can target “likely voters” based on past turnout history and ensure they are calling people who live in specific congressional districts.
- The Downside: If a voter has not updated their phone number on their registration, or if they are a newly registered voter without a consumer footprint, they might be left off the list.
The Crisis of Low Response Rates #
Even though pollsters are calling millions of cell phones every week, actually getting someone to talk is harder than ever.
In the 1990s, telephone polls regularly achieved response rates of 30% to 40%. Today, the typical response rate for a live-caller telephone poll is between 1% and 2%.
Several factors drive this decline:
- Spam Filtering: Modern smartphones and wireless carriers automatically flag unfamiliar numbers as “Scam Likely” or “Potential Spam,” preventing the call from ever ringing through.
- Behavioral Shifts: People simply do not answer calls from numbers they do not recognize.
- Caller ID: Unlike the early days of telephony, callers no longer have the element of surprise. If a screen does not show a recognized name, the call goes straight to voicemail.
How Pollsters Fix the Data (Weighting) #
With a response rate of 1%, how can a poll be accurate? Pollsters rely on a mathematical process called demographic weighting (or raking).
If a pollster needs to interview 1,000 people to represent a state’s population, but their raw respondents end up being 70% women and only 30% men, they will adjust the mathematical “weight” of the male responses. Each male response will count slightly more, and each female response will count slightly less, until the final data reflects the actual 50/50 split of the population.
Pollsters weight their samples for age, race, education, region, and sometimes past voting behavior. This ensures that even if certain groups (like young men) are harder to reach on cell phones, their voices are still accurately represented in the final numbers. Because weighting is highly complex and varies by pollster, looking at a comprehensive election aggregator is the best way to see the broader consensus of the electorate rather than relying on a single poll’s methodology.
Beyond Voice Calls: The Rise of Mixed-Mode Polling #
Because calling cell phones is so expensive and response rates are so low, the polling industry has rapidly evolved. Today, very few pollsters rely only on live phone calls. Instead, they use a mix of methods to reach you on your mobile device.
- Text-to-Web (SMS): You may have received a text message saying something like, “Hi, this is [Name] from [Polling Firm]. We are conducting a brief survey of voters in your area. Click here to participate.” When you click the link, it opens a secure browser-based survey on your smartphone. This is highly cost-effective and reaches younger demographics who refuse to answer voice calls.
- Online Panels: Pollsters recruit large groups of people to agree to take regular surveys via mobile apps or websites in exchange for small rewards or gift cards.
- Interactive Voice Response (IVR): Often called “robopolling,” these are automated phone surveys where you press buttons on your keypad. Because of TCPA laws, IVR is still largely restricted to landlines, meaning IVR-only polls miss a massive portion of the cell-phone-only population unless they pair their results with online panels.
Frequently Asked Questions #
Can pollsters legally robocall my cell phone? #
No. Federal law prohibits political pollsters (and political campaigns) from using automated dialers to call cell phones without your explicit consent. If you receive a voice call on your cell phone from a legitimate polling organization, there is a live human being on the other end of the line who dialed your number manually. However, automated text messages (text-to-web surveys) are permitted under certain conditions, provided they are sent using platforms that require human intervention to initiate each text.
How do pollsters know my location if I have an out-of-state cell phone number? #
This is a major challenge for modern pollsters. Because people keep their mobile phone numbers when they move, someone living in Georgia might have a New York (212) area code.
If a pollster is using Random Digit Dialing, they might call your New York number thinking you live in NY. Once you answer, the very first question they ask will usually be, “In what state and county do you currently live?” or “What is your current zip code?” If you do not live in the target area, they will politely end the interview. If the pollster is using Registration-Based Sampling, they already know you live in Georgia because they got your number from the Georgia voter file, regardless of your out-of-state area code.
Do online polls work better than cell phone polls? #
Neither is inherently “better”; both have strengths and weaknesses. Cell phone polls with live interviewers are excellent at reaching a truly random sample of the population, but they are expensive and suffer from low response rates. Online polls are much cheaper and can gather large samples quickly, but they can suffer from “self-selection bias” (only people who want to take surveys sign up for them) and require sophisticated weighting to ensure accuracy. The most reliable modern polls often use a hybrid of both.
Why do I never get called by pollsters? #
It is a simple matter of probability. There are more than 260 million adults in the United States. A typical national poll interviews only 1,000 people. Even if a pollster conducted one national poll every single day of the year, they would only interview 365,000 people out of 260 million. The statistical odds of your specific cell phone number being selected in any given year are incredibly low.