Rosner 13.11 and 13.12 In Example 13.23, Rosner examines the effect of passive smoking on cancer risk
Rosner 13.11 and 13.12
In Example 13.23, Rosner examines the effect of passive smoking on cancer risk (see pages 612-620). Passive smoking is exposure to the smoking by a spouse who smokes at least a cigarette a day over the last six months. A possible confounding variable is smoking by the participants. This SPSS analysis controls for a participant’s smoking in the study while examining the relationship between passive smoking and cancer risk. Please note that I have rotated the 2 X 2 Tables 13.11 and 13.12 so that cancer cases and controls are in the columns needed for SPSS.
| Participants’ | Status: | Nonsmoker | |
| Cancer Case | Control | Total | |
| Passive Smoke Exposure | 120 [1,1,2] | 80 [2,1,2] | 200 |
| No Passive Smoke | 111 [1,2,2] | 155 [2,2,2] | 266 |
| Total | 231 | 235 | 466 |
| Participants’ | Status: | Smoker | |
| Cancer Case | Control | Total | |
| Passive Smoke Exposure | 161 [1,1,1] | 130[2,1,1] | 291 |
| No Passive Smoke | 117[1,2,1] | 124 [2,2,1] | 241 |
| Total | 278 | 254 | 532 |
Compute the Mantel-Haenszel common odds ratio with stratification by the participants’ smoking status. Use the Mantel-Haenszel Chi Square test statistic to determine if we can conclude that there is an association between the risk factor of passive smoking and cancer risk. Let alpha = .05.
Ho : There is no association between the exposure of passive smoking and cancer risk in families with participants who themselves are smokers.
H A : There is an association between the exposure of passive smoking and cancer risk in families with participants who themselves are smokers.
Step One: Create the Variables
- Click the Variable View tab
- Enter the variables Cancer , Passivesmoke , and Selfsmoker , and Frequency and create value labels for the first three. Create value labels [1 = cancer case, 2 = control], [1 = passive smoke exposure, 2 = no passive smoke exposure], and [1 = smoker, 2 = nonsmoker].
Step two: Enter the Data
- Click the Data View tab.
- Select Data>Weight Cases to pull open the Weight Cases box. Select Weight Cases by the variable Frequency and move it to the Frequency variable box.
- Click OK .
- In the Analyze>Descriptives>Crosstabs dialog box, select the variable Passivesmoke and move it to the Rows box. Select the variable Cancer and move it to the Columns box. Layer by Selfsmoker .
- Click Continue .
- Click Statistics , and select Chi square , Risk , Cochrane’s and Mantel=Haenszel .
- Click Cells . Under Counts , select expected and under percentages , select Row .
- Click Continue .
- Click OK .
Deliverable: Word Document
