« Previous | Contents | Next »
Listen
SHS Lite - User Guide: A guide to using the Scottish Household Survey simplified dataset
8.4 Example 4 -Crosstab
Figure 42 - rd3 and agerband added as row and column variables
For Crosstab cells, you need to specify which percentages to use. Selecting all percentages would create a very large table so we will use only column percentages.
The Viewer window now opens to display the Crosstab table (see Figure 44).
Figure 43 - Cell Display dialog
Figure 44 - How random adult usually travels to work/education by Banded age of random adult
Again, this table allows us to estimate the true values for all adults. We can also use the table on page to work out if differences between age groups are significant i.e. they are unlikely to have occurred by chance. The table above shows that 25% of 16-24 year olds walk to work/education. Based on 613 cases, this estimate has an approximate sampling error of 5%. We can also see that 13% of 35-44 year olds walk to work/education. This has an approximate sampling error of 4%. Following the rules on page , since the difference between the estimates (12%) is greater than the sum of the errors (9%), we can say that this difference is significant i.e. we can be confident that among the population as a whole 16-24 year olds are more likely to walk to work/education than 35-44 year olds.
8.5 Example 5 - Selecting Specific Cases
You may wish to use specific data only. This example demonstrates how to select cases only in the local authority of Clackmannanshire. First we are going to find out what the current values are for the local authority, or la, variable.
! Note | If you scroll down through the list of values in the variable information box you will find that 'F' is used for Clackmannanshire. |
Figure 45 - la variable information
To select the Clackmannanshire cases
- Select Data, Select Cases... from the menu bar or click
- Select the If condition is satisfied option (see Figure 46)
Figure 46 - Select Cases
! Note | All text entered must be enclosed in double quotes. |
Figure 47 - Local authority equal to Clackmannanshire
! Note | In Data View, all cases where the local authority does not equal  Clackmannanshire are now displayed with a diagonal line through the case number. |
If you now create the crosstab table in Example 4 which starts on page , you will see the following results.
Figure 48 - How random adult usually travels to work/education by Banded age in Clackmannanshire
We can see in this table that all of the percentages are based on small numbers of cases. The table on page shows that even for 100 cases, the sampling errors are large and it would not be possible to say that any of the differences in this table could not have occurred by chance. The variables need to be made into smaller groups - age bands covering broader ranges and broader categories of modes of transport - in order to investigate whether significant differences exist. This can be achieved by recoding the data.
! Note | Compare the results shown in Figure 44 and Figure 48 and you will see a large difference when we look at just the Clackmannanshire local authority. |
! Note | To reset the dataset, choose Data, Select Cases... from the menu bar or click and select the All cases option (see Figure 46). |
« Previous | Contents | Next »