ANNEX 3 RESULTS OF MULTIVARIATE ANALYSES
The logistic regression model commented on in this report is presented below. Logistic regression uses binary dependent variables where the value of interest is coded 1 and the rest of the cases are coded 0. The binary dependent variable coding is shown at start of the table. This analysis method compares the parameter estimates for each category within a variable to a "reference" category. The reference category for each variable is shown in brackets. Two asterisks (**) denote significance at the 1% level, one (*) denotes significance at the 5% level.
Table 1 Negative attitudes towards young people
Dependent variable coding:
1= Most negative attitudes towards young people, 0=or not | Odds Ratio (Exp (B)) | 95% Confidence Intervals for Exp (B) |
|---|
| | Lower | Upper |
|---|
SIMD (Least deprived) |
|---|
2 | 0.920 | 0.610 | 1.388 |
|---|
3 | 1.388 | 0.946 | 2.038 |
|---|
4 | 1.504* | 1.019 | 2.221 |
|---|
Most deprived | 2.44** | 1.682 | 3.542 |
|---|
Educational attainment (Degree/ FE) |
|---|
Highers/ A'Level | 1.611* | 1.109 | 2.342 |
|---|
Standard grade/ GCSE | 2.009** | 1.437 | 2.808 |
|---|
None | 2.604** | 1.795 | 3.777 |
|---|
Age (18-24) |
|---|
25-34 | 1.110 | 0.696 | 1.769 |
|---|
35-44 | 0.855 | 0.547 | 1.338 |
|---|
45-54 | 0.612* | 0.381 | 0.985 |
|---|
55-64 | 0.613* | 0.378 | 0.993 |
|---|
65+ | 0.492** | 0.302 | 0.800 |
|---|
Contact with 11-24 year olds in local area (knows most well enough to speak too) |
|---|
Knows some | 1.483 | 0.965 | 2.277 |
|---|
Does not know any | 1.998** | 1.260 | 3.166 |
|---|
Number of cases in model: 1451