Is My Vote Forward just a poll with extra steps?

In important ways, yes. My Vote Forward’s priority rankings and opinions are answers to questions from people who chose to take part, which makes them closer to an opt-in online poll than to a scientific survey or an election, and they carry the same limits. What the framework adds is structure rather than representativeness: answers persist and can be revised over time, each is tied to an exact question version, opinions are kept separate from any vote, and results are published only through scheduled, reviewed releases. Whether that structure makes the results more useful than a well-run poll is an open question.

The objection in its strongest form

The objection is that the framework dresses up a familiar activity in new language. Asking people to rank issues and state opinions is what pollsters have done for decades. Scientific polls already do it better, using random samples that can describe a population, which the framework cannot. Opt-in online polls already do it more cheaply. Calling the result “collective priorities” rather than “poll results” does not change what it is, and may make it sound more authoritative than it deserves.

The strongest version adds that the extra features may make things worse. Letting people revise answers makes results depend on timing. Tying people to accounts creates privacy risks a one-off anonymous poll does not have. And a framework that also discusses voting invites readers to confuse an opinion tally with an election result. On this view, the honest move is to call it what it is, a structured opt-in survey, and stop there.

Why the objection matters

The objection matters because naming shapes trust. If My Vote Forward’s results are presented as something more than an opt-in poll, readers will give them more weight than the method supports, and critics will be right to say the framework overclaims. If they are presented as exactly a poll, the framework must explain why anyone should use it rather than the polls that already exist. Either way, the comparison has to be faced directly.

Defined terms

• Probability poll: a survey in which respondents are selected at random from a defined population, so results can be generalized with known uncertainty.
• Opt-in poll: a survey of people who chose to respond. Results describe respondents, not a population.
• Deliberative poll: a random sample surveyed before and after structured deliberation. A trademark of James S. Fishkin.
• Participatory budgeting: a process in which community members decide how to spend part of a public budget.
• Official ballot: a vote in an election run by a legally responsible authority, with binding effect.
• Published release: in this framework, a scheduled, reviewed, immutable publication of aggregate results. See published releases.

Evidence

What separates a population estimate from an opt-in result. A 2010 task force of the American Association for Public Opinion Research concluded that “Researchers should avoid nonprobability online panels when one of the research objectives is to accurately estimate population values,” because “There currently is no generally accepted theoretical basis” for claiming such results project to the population.⁠¹ That conclusion was not unanimous; one task force member asked to be removed as an author because he disagreed.⁠¹ AAPOR’s disclosure standards now require researchers to state “whether the sample comes from a frame selected using a probability-based methodology” and specify that reports of nonprobability surveys will only provide measures of precision if they are defined.⁠²

Opt-in samples carry quality problems. Pew Research Center’s 2026 report, “No Easy Fix for Bogus Respondents in Online Opt-In Polls,” screened an opt-in survey of 11,114 U.S. adults with trap questions, automated prescreening, and voter file matching; 9,151 passed the trap questions.⁠³

Question wording changes answers. Pew reports that “Even small wording differences can substantially affect the answers people provide” and that offering a list of options changes which answers people give.⁠⁴ Any instrument that asks questions, poll or framework, inherits this.

A deliberative alternative. Stanford’s Deliberative Democracy Lab describes Deliberative Polling as beginning with “A random, representative sample,” which is polled, deliberates, and is “again asked the original questions”; the lab reports “dramatic, statistically significant changes in views” across its studies.⁠⁵ Deliberative Polling changes both who answers (a random sample) and what they know when answering.

A decision-making alternative. The Participatory Budgeting Project defines participatory budgeting as “a democratic process in which community members decide how to spend part of a public budget,” and dates its origin to Porto Alegre, Brazil, in 1989.⁠⁶ Participatory budgeting is self-selected like an opt-in poll, but its votes have consequences for a real budget.

How comparable systems have handled it

The comparable activities differ on three axes: who takes part, what the answer does, and how the result is published.

On who takes part and what the answer does, My Vote Forward is an opt-in poll. It differs on publication rules and on structure over time.

Competing interpretations

One interpretation holds that the framework’s structure is a real contribution. Persistent, revisable answers let people express what they think now rather than at one survey moment; versioned questions prevent answers from being applied to wording people never saw; separating opinions from votes protects ballot secrecy; and scheduled releases with denominators set a higher disclosure bar than most opt-in polls meet. On this view, a better-governed opt-in instrument is worth having even if it is not representative.

A second interpretation holds that none of this addresses the core limit. Without random selection, results cannot describe a population, and structure cannot fix that⁠¹ On this view, the framework should either add a probability sample or be described simply as a structured opt-in survey.

A third interpretation holds that the comparison to polling misses the purpose. Polls measure opinion for others to read; the framework is meant as a participation habit for the person taking part, with private priorities that help them follow issues over time. On this view, the value is to the participant, and the aggregate is a secondary product.

What the framework answers

My Vote Forward’s answer is to accept the poll comparison where it applies and to state what differs:

• Results describe participants and state their denominators, dates, and methods; they are never called public opinion or a mandate (see measurement and representation).
• Each opinion is tied to exact question wording and version, and is marked stale when wording changes (see opinions).
• Priorities are recorded only when confirmed; untouched defaults are never counted (see priorities).
• Opinions are never treated as votes, and no attempt is made to link a named person’s opinion to their ballot (see opinion and choice).
• Results are released on a schedule, the same version to everyone, with no live scoreboard (see published releases).
• The framework does not pay for participation.

Where the objection still wins

The objection wins on representativeness and on novelty. My Vote Forward has no sampling design, so its results cannot describe a population, which is the main thing scientific polls offer. Much of what it promises already exists elsewhere: professional standards already require disclosure of sponsor, exact wording, sample design, and weighting,⁠² and none of the framework’s other features is a new idea. Its combination may be distinctive, but the framework has no evidence that the combination produces more useful or more trusted results, because no real participant data has been collected.

Unresolved questions

• Should a future deployment add a random-sample panel so results can be compared with a probability benchmark?
• Does the ability to revise answers make aggregates more accurate or merely more volatile?
• Is the main value of the framework to individual participants or to readers of aggregates, and should the design choose?
• What evidence would show that this structure adds anything beyond good polling practice?

Related

• Measurement
• Opinion and choice
• Question design
• What if most people never participate?
• Is this an official voting service?
• Hard questions