Yes. Most telephone poll interviews in the US are now completed on cell phones, because most people don’t have anything else. In the second half of 2024, 78.7% of American adults lived in homes with no landline, according to the CDC’s National Health Interview Survey. A poll that called only landlines would miss most young and middle-aged adults entirely.
Calling cells is slower and more expensive than calling landlines used to be, and very few people pick up. That’s why most polls you see today mix phone calls with texts and online surveys.
Who can pollsters reach on a landline anymore? #
Very few people, and they’re not typical. The CDC found that wireless-only living peaks among adults in their late 20s through early 40s, at about 88%, and among renters. Landline households skew older and more likely to own their homes. Pollsters have called cell phones as part of their samples since the late 2000s, and cells now make up most of the phone numbers dialed in a typical live-interviewer poll.
Can pollsters robocall your cell phone? #
Not without your consent. The Telephone Consumer Protection Act of 1991 (TCPA) bans calls to cell phones made with an autodialer or a prerecorded or artificial voice unless the person has agreed to receive them. Poll calls are unsolicited, so pollsters can’t blast recorded surveys to cell numbers.
The Supreme Court narrowed the definition of an autodialer in Facebook v. Duguid (2021) to equipment that stores or produces numbers using a random or sequential number generator. Since random digit dialing is literally a list of randomly generated numbers, many pollsters still have a live interviewer start each call to a cell phone rather than risk a lawsuit. That keeps costs high: an interviewer might dial dozens of numbers, most of which go unanswered, for each completed interview.
The ban on prerecorded calls without consent is why automated “press 1” polls, known as IVR polls, can’t simply switch to cell phones. IVR pollsters typically pair their landline calls with an online or text sample to reach people without landlines.
How do pollsters get your cell phone number? #
They don’t buy telemarketing lists. The main methods are:
| Method | How it works | Strength | Weakness |
|---|---|---|---|
| Random digit dialing (RDD) | Generate random numbers within working area codes and exchanges | Reaches anyone with a phone, registered or not | Many numbers are dead, business or out of area |
| Registration-based sampling (RBS) | Start from the state voter file, then match phone numbers from commercial data | Know who you’re calling, their district and vote history | Misses voters with no matched phone or a new number |
| Address-based sampling (ABS) | Randomly sample postal addresses, invite by mail to answer online or by phone | Covers almost every household | Slow and costly |
Registration-based sampling has become popular for election polling because it solves an old headache: people keep their phone numbers when they move. Someone with a 212 area code might live in Georgia. A pollster using the Georgia voter file already knows where that person lives, while an RDD pollster has to ask and screen out people outside the target area.
Why do so few people answer poll calls? #
Phone response rates have collapsed. The Pew Research Center reported that its typical telephone survey response rate fell from 36% in 1997 to 6% in 2018, and rates across the industry have kept falling since. Several things drive it:
- Call labeling. Carriers and phones flag unfamiliar numbers as “Spam Likely,” so many poll calls never ring through or get ignored. Pollsters now register their numbers with carriers to reduce this.
- Habits. Most people no longer answer calls from numbers they don’t know.
- Time. A 15-minute survey is a big ask.
Low response doesn’t automatically make a poll wrong. What matters is whether the people who answer differ from those who don’t in ways that affect their views. That’s the core of why falling response rates matter, and it’s why weighting has become so important.
How do pollsters fix a skewed phone sample? #
They weight it. If the people who answered are older, more educated or more likely to be women than the population, each respondent’s answers are scaled up or down until the sample matches known totals for age, sex, race, education, region and often past vote. Heavy weighting has a cost: when a few people stand in for a large group, the poll’s real margin of error grows. We explain the mechanics in how pollsters weight survey data.
What does a modern “phone poll” actually look like? #
Very few polls are phone-only now. A typical design combines several modes:
- Live calls to cell phones and landlines, often from a voter-file sample.
- Text-to-web. You get a text from a named pollster with a link to a survey. This reaches younger people who won’t answer a call.
- Online panels, either probability-based panels recruited by mail or opt-in panels. See are online opt-in polls reliable? for how those differ.
Mixing modes lowers costs and widens reach, but people sometimes answer differently depending on whether a human is asking. Our explainer on the polling mode effect covers that.
When you look at a new poll, it helps to know how it was done. Election Tracker shows every published poll for a race with its pollster, sample size, population (likely voters, registered voters or adults) and field dates, and opens the pollster’s own release in an in-app browser, which is where the method is disclosed. The app’s 30-day average for each race weights newer polls more and counts each pollster once per window. It’s free on iPhone.
Frequently asked questions #
Can pollsters legally call my cell phone? #
Yes, as long as they don’t use an autodialer or a prerecorded voice without your consent. Most legitimate poll calls to cell phones are started by a live interviewer. You can decline or ask not to be called again.
How do I know a poll call or text is real? #
A real pollster will name its organization, won’t ask for money, and won’t ask for your Social Security number, bank details or passwords. If a “poll” asks for a donation or keeps feeding you negative claims about one candidate, it may be a fundraising pitch or a push poll, which is persuasion dressed up as research.
Why have I never been called by a pollster? #
The odds are low. There are more than 260 million adults in the US, and a national poll interviews about 1,000 of them. Even a pollster that ran a national survey every day would reach only a tiny fraction of adults in a year.
Do phone polls still work? #
They can. Live-interviewer polls reach people who don’t join online panels, and several of the most accurate pollsters still use them. Their challenge is cost and low response, which is why most combine phone calls with text and online interviews.