Graph > Bar Chart > Values from a table, Two-way table, Cluster > Graph Variables: 'Male' 'Female', Row labels: methods, check Rows are outermost..., Bar Chart Options > Check show y as percent, Within categories at level 1.
| Method | Male | Female |
| Private Swimming Pool | 488 | 219 |
| Bathtub | 115 | 132 |
| Ocean | 231 | 40 |
| Freshwater bodies | 155 | 19 |
| Hottubs | 16 | 15 |
| Reservoirs | 32 | 2 |
| Other Pools | 46 | 14 |
| Pails, basins, toilets | 7 | 4 |
| Other | 40 | 12 |
| Total | 1130 | 457 |
Because the total for men and women are so different it is better to change the table to row percentages:
| Method | Male | Female |
| Private Swimming Pool | 69.0 | 31.0 |
| Bathtub | 46.6 | 53.4 |
| Ocean | 85.2 | 14.8 |
| Freshwater bodies | 89.1 | 10.9 |
| Hottubs | 51.6 | 48.4 |
| Reservoirs | 94.1 | 5.9 |
| Other Pools | 76.7 | 23.3 |
| Pails, basins, toilets | 63.6 | 36.4 |
| Other | 76.9 | 23.1 |
Step 1: Graph
Graph > Bar Chart > Values from a table, Two-way table, Cluster > Graph Variables: 'Male' 'Female', Row labels: methods, check Rows are outermost..., Bar Chart Options > Check show y as percent, Within categories at level 1.
Step 2: Test
Stat > Tables > Chisquare test (Table in Worksheet): 'Male' 'Female'
this gives a warning of one expected cell count less than five. Those are for "Pails,,-Female'. This is no problem
1) α=0.05
2) H0: Classifications are independent (no relationship between the Gender and Method)
3) Ha: Classifications are dependent (some relationship between the Gender and Method)
4) p=0.000 < α
5) Reject H0, There are significant differences between the methods of drowning of men and women