Question

Im trying to see what test in SPSS would I need to run to see if there is a significant relationship between the type of counseling and the outcome. Ive hypothesized that patients who are in counseling for substance abuse will complete their sessions if they have a supportive network whereas if they don't they will discontinue their counseling session

Table A Outcome Completion Premature Termination 12 Type of Counseling Family 22 34 Individual 31 53 45 57 76 N 110

Answer 1

Following steps were performed to carry out Chi-square test of independence in SPSS:

1. Click Analyze > Descriptives Statistics > Crosstabs... on the top menu

2. You will be presented with the Crosstabs dialogue box:

3. Transfer one of the variables into the Row(s): box and the other variable into the Column(s): box. In our example, we will transfer the Types of Counseling variable into the Row(s): box and Outcome into the Column(s): box.

4. Click on the button. You will be presented with Crosstabs: Statistics dialogue box

Statistics...

5. Select checkbox corresponding to the Chi-square and Phi and Cramer's V options

Crosstabs: Statistics Chi-square □ Correlations Nominal Ordinal Contingency coefficient Gamma Somers' d □ Kendall's tau-b Lambda Uncertainty coefficient Kendall's tau-c Kappa Risk McNemar Nominal by Interval Eta Cochran's and Mantel-Haenszel statistics Test common odds ratio equals: 1 Help Continue Cancel

6. Click the button.

Continue

7. Click the button. You will be presented with Crosstabs: Cell Display dialogue box

Cells...

8. Select checkbox corresponding to Observed from the –Counts– area, and Row, Column and Total from the –Percentages– area

Crosstabs: Cell Display Counts z-test Observed Compare column proportions Expected Hide small counts Less than 5 Adjust p-values (Bonferroni method) Percentages Residuals Unstandardized Column Standardized | |Total Adjusted standardized Noninteger Weights O Round cell counts Round case weights Truncate cell counts Truncate case weights No adjustments Continue Cancel Help

9. Click the button.

Continue

10. Click the button to generate your output.

OK

Following is the outcome we obtained using data provided:

Case Processing Summary Cases Missing Tota Valid Percent Percent Percent Counseling* Outcome 1101 100.0% 0.0% 1101 100.0% Counseling Outcome Crosstabulation Outcome Premature ompletionTermination 12 35.3% 21.1 % 10.9% 45 59.2% 78.9% 40.9% 57 51.8% 100.0% 51.8% Tota Count 96 within Counseling 96 within outcome % of Total Counseling Family 34 1 00.0% 30.9% 30.9% 76 100.0% 69.1% 69.1 % 110 1 00.0% 1 00.0% 1 00.0% 64.7% 41.5% 20.0% 31 40.8% 58.5% 28.2% 53 48.2% 100.0% 48.2% | Individual Count 96 within Counseling 96 within outcome % of Total Count 96 within Counseling 96 within outcome % of Total | | Total

Chi-Square Tests Asymp. Sig (2-sided) Exact Sig. (2 Exact Sig. (1 Value df sided) sided) 5.382a 4.466 5.433 020 035 020 Pearson Chi-Square Continuity Correction" Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases 024 017 5.333 021 110 a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 16.38 b. Computed only for a 2x2 table Symmetric Measures Value Approx. Sig 020 020 Nominal by Nominal Phi Cramer's 221 221 110 N of Valid Cases

Interpretation

When reading this table we are interested in the results of the "Pearson Chi-Square" row. We can see here that χ(1) = 5.382, p = 0.020. Since p-value is less than 0.05, it can be inferred that there is statistically significant association between the type of counseling and the outcome.