I would like to know what the appropriate statistical test is for investigating an association between a nominal variable and an ordinal variable assuming normal distribution? It''s also a 2 x 4 table. Would it be a Pearson''s chi-squared test or a linear-by-linear chi-squared test and why? I would also like to know what the appropriate statistical test is for testing whether something is equal to some hypothesized value. For instance, I have a question relating to 2 independent variables; the first is back pain (nominal): 0 = no or little pain and 1 = Troublesome or severe back pain. The second variable is weight (continuous). I am asked to test whether the mean weight for the troublesome or severe back pain group is equal to 70kg. So, what statistical test is appropriate to use in SPSS?