Using SPSS to Compare Means UDP 520 Lab 4 Lin November 6 th, 2007
Outline Dataset Independent samples Paired samples
Dataset—Weight Lost Training Program (WLTP) 1000 adults aged 18+ (including both males and females) were recruited to study the effectiveness of Weight Lost Training Program. Variables –Sex –BMI_1(before WLTP) –BMI_2(after WLTP) Download dataset from
Questions Question one: –Is BMI significantly different between males and females before WLTP? Question two: –Overall (for both males and females), is WLTP effective?
Question One Step one: Making assumptions and meeting test requirements –Random sampling –Level of measurement is interval-ratio –Sampling distribution is normal Step two: stating the null hypothesis Step three: selecting the sampling distribution and establishing the critical region –Sampling distribution = Z distribution –Alpha=0.05, two-tailed –Z(critical)= 1.96
Question One (cont.) Step four: computing the test statistic in SPSS
Question One (cont.) Step five: making a decision and interpreting the results of the test result or Z (obtained)
Question Two Step one: Making assumptions and meeting test requirements –Random sampling –Level of measurement is interval-ratio –Sampling distribution is normal Step two: stating the null hypothesis Step three: selecting the sampling distribution and establishing the critical region –Sampling distribution = Z distribution –Alpha=0.05, two-tailed –Z(critical)= 1.96
Question Two (cont.) Step four: computing the test statistic in SPSS
Question Two (cont.) Step five: making a decision and interpreting the results of the test result or Z (obtained)
Exercises Is BMI significantly different between males and females after WLTP? Is WLTP effective for males? –Hint (do selection in “Data”— “Select cases”— “if condition is satisfied”— “if”— “sex=0”) Is WLTP effective for females?
Selecting cases in SPSS