Download presentation
Presentation is loading. Please wait.
Published byAllison Shields Modified over 7 years ago
1
Presenting the Results of a Contingency Table Analysis
2
Verdict x Defendant Physical Attractiveness Mock jurors were significantly more likely to find the defendant guilty when he was unattractive (75.7%) than when he was attractive (65.0%), 2 (1, N = 145) = 6.229, p =.013, =.207, odds ratio = 2.45, 95% CI [1.20, 4.99]. Please note that with significant results one should emphasize the direction of effect.
3
Nonsignificant Results One should NOT mention direction of effect, unless having tested directional hypotheses.
4
Direction x Device People were significantly more likely to take the stairs when going down (24.3%) than when going up (6.1%), 2 (1, N = 3,217) = 217.22, p <.001, =.26, odds ratio = 4.90, 95% CI [3.91, 6.13]. SPSS gave me an odds ratio of.204 and a CI of [.163,.256]. I inverted these numbers to get ratios greater than one.
5
Figure 1. Device x Weight
6
Choice of device was significantly associated with weight of patron, 2 (2, N = 3,217) = 11.752, p =.003, =.06. As shown in Table 1, obese individuals used the stairs considerably less often than did others.
7
Pairwise Comparisons Use of the stairs did not differ significantly between overweight and normal individuals, 2 (1, N = 2,907) = 1.034, p =.31,, =.02, odds ratio = 1.12, 95% CI [0.90, 1.38]. Individuals of normal weight used the stairs significantly more often than did obese individuals, 2 (1, N = 2,142) = 9.062, p =.003,, =.065, odds ratio = 1.94, 95% CI [1.25, 3.00].
8
SPSS will mess up if you use a dichotomous predictor that is coded with numbers other than consecutive integers, such as Weight = 1 (obese) and 3 (normal). If you declare Weight to be categorical, SPSS works fine. Overweight individuals used the stairs significantly more often than did obese individuals, 2 (1, N = 1,385) = 11.815, p =.001, =.092, odds ratio = 2.16, 95% CI [1.38, 3.38].
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.