Depending on the question you have you can format survey in dif ways. Surveys are helpful for what people think they are doing or what they think in uences their behaviour. A survey might not be the best for capturing what people are really doing or what really in uences their behaviour - instead you could observe them designing surveys and polls: 1st holocaust question - hard to answer - may get wrong data. 2nd holocaust question- adjust it to be easier. Types of survey questions: open-ended- ex. short answer close-ended - ex. multiple choice - can be informative but have to select option questions carefully: rating scales- Likert scales- degree of agreement (strongly disagree- strongly agree: semantic differential scale- rate something with adjectives (strong- weak, decisive-indecisiveetc) Loaded?- form problem in particular way for parti. response. 1st- leading suggesting there is a problem 2nd- leading question implying When you ask two questions at the same time.