For the examples below, indicate the scale and suitable statistical test for the given variables. Are the samples dependent or independent? If no scale is given, the highest possible scale is presumed.

(a) Do men and women differ with regard to disposable income? The sample has a size of n = 1000.

(b) Twenty men and women rate how much they like a certain product on a scale from one to five. Are their preferences gender-specific?

(c) Twenty people rate how much they like a certain product on a scale from one to five both before and after hearing a presentation about its features. Did the presentation change respondent’s opinion of the product?

(d) Do three programmes of study differ with regard to the proportion of male and female students enrolled?

(e) Do purchasers and non-purchasers of a product differ with regard to age? The mean age is assumed to approximate a normal distribution.

(f) Do two vehicle models differ with regard to having and not having special features?

(g) People are asked to rate a product on a scale from one to five before and after an ad campaign. Did the ad campaign change their opinion of the product?

(h) You are tasked with finding out whether parents’ attitudes towards wood toys differ depending on whether they have sons or daughters. For the time being, you ignore parents with both sons and daughters. You measure their attitudes by asking them to rate wooden toys on a five-point scale (1 = “strongly dislike” to 5 = “strongly like”).

(i) Which approach should you use if you include parents with sons and daughters in the survey?