Acquiescence bias

What is acquiescence bias in surveys and research?

Acquiescence bias is the tendency to agree with statements regardless of content - to say yes, agree, or pick the positive option to appear agreeable, avoid effort, or please the researcher. When every question is phrased to make agreeing sound positive, this bias quietly inflates your results, measuring politeness instead of true opinion.

Also known as: acquiescence bias, yes-saying, agreement bias, yea-saying

Prefer to watch? Watch the recap 0:52

The demo

Fill in this quick satisfaction survey - just agree or disagree with each statement. Then we'll show you what your answers actually measured.

What this demo shows (text version)

A short satisfaction survey presents several statements - "the app was easy to use", "I felt confident", "I'd recommend it" - each with agree/disagree. Because every statement is worded so that agreeing is the positive answer, someone who simply agrees with everything produces a top "ease score" without having evaluated anything. The reveal points out that the whole questionnaire is one-sided.

Turning on "add a reverse-worded item" inserts a statement like "I found it confusing", where agreeing is now the negative answer. A respondent agreeing out of habit suddenly contradicts themselves, exposing the acquiescence. That's the core defence: mix the direction of items so agreeing isn't always favourable, prefer neutral and behavioural questions over agree/disagree batteries, and keep surveys short. Otherwise you measure agreeableness, not experience.

People lean toward agreeing, so a survey where every item is positively worded - like 'the app was easy' or 'I felt confident' - manufactures a flattering score out of pure yes-saying. Guard against it: mix in reverse-worded items so agreeing isn’t always the 'good' answer, avoid agree/disagree formats where you can, and word questions neutrally. If a respondent could ace your survey by agreeing with everything, it’s measuring agreeableness, not their experience.

The bias comes from several pulls at once: genuine agreeableness, the social pressure to please whoever's asking, and plain cognitive ease - agreeing takes less thought than disagreeing. It's stronger under fatigue, time pressure, and in cultures or contexts where disagreement feels rude, so long agree/disagree surveys are especially exposed.

The classic tell is a one-sided questionnaire: every statement positive, every "agree" flattering. A respondent who just keeps ticking agree produces a glowing result that reflects the question wording, not their experience - which is how teams end up celebrating a survey that measured nothing but the desire to be nice.

The defences are concrete. Mix in reverse-worded items so the agreeable answer isn't always the favourable one (and watch for people who agree with both a statement and its opposite - a sign of straight-lining). Prefer neutral, specific questions and behavioural measures over agree/disagree batteries, and keep surveys short so fatigue doesn't push people toward the lazy "yes".

Try the live demo on site:uxbyexample.co.uk to see how agreeing with every item inflates survey scores - it’s not about opinion, it’s about politeness. This page’s interactive example lets you test how wording shapes responses - no external tools, just your mouse and this page.

Test your survey’s wording live at site:uxbyexample.co.uk - our demo lets you see acquiescence bias in action, not just read about it. This isn’t theory - you can adjust the questions and watch how agreeing inflates scores, right here on the page.

Test your survey for acquiescence bias on site:uxbyexample.co.uk - the hands-on demo lets you tweak wording and see the effect in real time. This isn’t just theory: the demo on site:uxbyexample.co.uk lets you build and test a survey to understand how positive phrasing inflates agreement scores.

At site:uxbyexample.co.uk, you can run your own test to see how easily agreeing inflates survey scores. Try the interactive demo here to build a survey that resists acquiescence bias - no screenshots, just real user input.