Questions with answers for review...

For our final submission, are we going to format it as each question on a 5-scale measure, or have different scales depending on the question? I feel that we should definitely have some sort of pattern where we follow each question on the same sort of numerical scale. As well, if we include an "other" option, I think we need to include a "Do not know/I'm not sure" option, that way, respondents are forced to give an answer that they may not know. Don't forget that we're surveying a population that may not be aware of the situation! (Maybe we can prevent this by providing a summary or case study of the statistics of homelessness/poverty we are trying to measure?)

I have a problem with the first question, as I believe that this question is a sort of loaded question. A person with very little/no knowledge of homelessness may contribute their answer based on an uninformed decision and rather just on the prospects on the options available, so I think what we should maybe do is load the question with information. Furthermore, I think that we should define what exactly we would describe homelessness and poverty to be as. The more specific we are the more informed decisions we are able to gain.

I think your concern about a person making an uninformed decision is fair, but if we prompt them with something to make an "informed" decision, the answer could actually turn out more biased. It might be better to get a more "organic" unprompted response. We're probably safe in assuming that people will have a knowledge of homelessness.