EAllusion wrote:Off the top of my head: I'd use weighted sampling techniques and make sure I was hitting demographically relevant portions of the LDS faith in proportion. The sample size would be much larger. I'd have test-retest, and internal reliability constructs built into the question lists through similarly worded questions attempting to measure the same dimension and rephrasing of specific questions. There would be a clear operational definition of the concepts I'm attempting to measure with a rationale for why each question validly relates to each group. I'd use an ANOVA to make sure there was statistically significant relationships between answer patterns. I'd use Likert scale measuring instead of Yes No I Don't Know, etc.If you had typed up the survey, what would it look like?
Please be specific; please type out the improved survey so that I can copy, paste, and print it.
As it stands, your survey did not allow you to draw any meaningful conclusions from it. The more seriously you take it, the more embarrassing it is.
That's exactly why I don't take it seriously. Recall that I called it--and still call it--an "informal survey." It's only my Internet Mormon friends over at MA&D who take it as gospel truth when they wish to use it against me.