Building on the past couple of weeks and the talk about surveys, i wanted to share some thoughts about using Likert scales.
First, a Likert Scale is a rating scale that’s often used in surveys. It’s popular because it lets you offer a range of answers from one extreme to another (strongly disagree to strongly agree), while still being much easier to analyze than an open-ended question.
- group questions on a single topic together
- offer enough choices to be accurate, but not too many that people feel overwhelmed (often people use 5-point scales)
- odd-number of choices allow for a neutral choice (neither agree nor disagree)
- even-number of choices force the respondent to choose an answer. this may make them think deeper, but they might also be annoyed that they have to choose when they really are neutral, and making them choose could skew the data
- be consistent in how you order choices. consider going from low to high, or bad to good (as a way to cancel out two biases - read more here)
- when it’s possible that respondents don’t know the answer or that a questions isn’t applicable, do offer that choice. again if you don’t this could skew your results
hope that’s helpful! please share any other tips you may have!