Choosing among statistics: ◦ ◦ ◦ ◦ ◦ 1. 2. 3. 4. 5. number of independent variables level of measurement (nominal to ratio) number of dependent variables level of measurement (nominal to ratio) other considerations (normality) Marketing Research 7/17/2016 1 Cross-tabulation 1 Independent variable (categorical) ◦ Men versus women 1 Dependent variable (categorical) Bought or didn’t buy Marketing Research 7/17/2016 2 How do you measure the covariation? 1. “inter-ocular impact” test ◦ does it look big? Marketing Research 7/17/2016 3 Male Female Bought [Did not] 70% [30%] 40% [60%] Marketing Research 7/17/2016 4 Covariation one categorical independent variable e.g., male or female one categorical dependent variable e.g., blonde, not blonde ◦ 10% of males are blonde, 40% of females ◦ significant covariation? Marketing Research 7/17/2016 5 2. Statistics (exact probability) calculate the difference between: ◦ observed rates ◦ expected rates (chance) if observed is different than chance, non-independence between variables Marketing Research 7/17/2016 6 OBSERVED Male Female Blonde 10 40 • TOTAL 50 Not | 90 | 60 150 | TOTAL 100 100 200 Marketing Research 7/17/2016 7 Calculate Chi-square: Sum of (Observed-Expected) 2 Expected [do this for each cell] Marketing Research 7/17/2016 8 Male Female TOTAL Blonde 10 25 [9] 40 25 [9] 50 Not 90 75 [3] 60 75 [3] 150 | TOTAL | 100 | | 100 [12] [12] 200 X2= 24 Marketing Research 7/17/2016 9 2. Interpret result: ◦ to X2 w/ (k-1)*(n-1) “degrees of freedom” (Table X2 1 d.f. p < .05 = 3.84) ◦ Because X2 obs = 24 > 3.84, REJECT Ho Marketing Research 7/17/2016 10 Crosstabulation Count SEX Total male female DIANAFUN yes no 83 94 239 95 322 189 maybe 67 83 150 Marketing Research Total 244 417 661 7/17/2016 11 Crosstab % within SEX SEX Total male female DIANAFUN yes no 34.0% 38.5% 57.3% 22.8% 48.7% 28.6% maybe 27.5% 19.9% 22.7% Marketing Research 7/17/2016 Total 100.0% 100.0% 100.0% 12 Chi-Square Tests Pearson Chi-Square N of Valid Cases Value 34.365a 661 df 2 Asymp. Sig. (2-sided) .000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 55.37. Marketing Research 7/17/2016 13 Crosstabulation Count SEX Total male female DIFUN1 none yes 94 83 95 239 189 322 Marketing Research Total 177 334 511 7/17/2016 14 Crosstabulation % within SEX SEX Total male female DIFUN1 none yes 53.1% 46.9% 28.4% 71.6% 37.0% 63.0% Marketing Research Total 100.0% 100.0% 100.0% 7/17/2016 15 Chi-Square Tests Pearson Chi-Square N of Valid Cases Value 30.197b 511 df 1 Asymp. Sig. (2-sided) .000 a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 65.47. Marketing Research 7/17/2016 16 Marketing Research 7/17/2016 17