Percentiles based on numeric variables
Quantum can calculate percentiles based on the results of numeric variables. An example is where the data contains the actual family income rather than a code representing a range of incomes. This enables you to obtain a more accurate figure by using inc= to read the values from the data. The format of percentiles with inc= is:
n01Text ;inc=variable ;percentile=factor; c=condition
For example:
n01First Quarter;inc=c(123,127);percentile=25;c=c(123,127)u$ $
This calculation uses an n01 for both the elements and the percentile: it does not use an n30.
An example of using percentile could be where you want to produce a table showing the actual family income, by region, along with the median and quartile figures. Thus, the example might read:
tab totalincome sex
l totalincome
val c(123,127);Base;hd=Household Income (in Pounds);=; 0; i;
+1-3000; 3001-6000; 6001-9000; 9001-12000; 12001-15000; 150001+
n03
n01Lower Quartile;inc=c(123,127);percentile=25;c=c(123,127)u$ $
n01Median Income;inc=c(123,127);percentile=50;c=c(123,127)u$ $
n01Upper Quartile;inc=c(123,127);percentile=75;c=c(123,127)u$ $
l sex
col 110;Base;Male;Female
The table it produces is:
Base Male Female |
Base 464 232 232 |
Household Income (in pounds) |
0 76 28 48 |
1-3000 22 13 9 |
3001-6000 44 11 33 |
6001-9000 102 38 64 |
9001-12000 144 96 48 |
12001-15000 18 15 3 |
15001+ 58 31 27 |
|
Lower Quartile 4951 7364 4451 |
Median Income 8418 10002 6874 |
Upper Quartile 11077 11098 9591 |
Note You can use the keyword median instead of percentile=50.
To request percentiles on a table, follow the rules below:
Data contains | Calculate percentiles with | Request percentiles with |
|---|
Codes | Factors | fac= on elements and n30 |
Numeric values | Values from data | n01 and inc= |
However, when you are preparing a study for Windows-based Quanvert, always specify percentiles by using fac= on the elements and the n30 statement, because Windows-based Quanvert does not produce reliable figures for percentiles that have been specified using n01 statements with inc=.
As with the other statistical statements, the percentile and median keywords suppress percentages for the row or column of percentile values.
When Quantum calculates percentiles from numeric variables, it writes information to temporary files on the disk which it then reads later in the calculation. If you see error messages about Quantum being unable to locate or open files at this point it is possible that the disk is full and the file was not created. If this happens, try to make some extra space on the disk before rerunning your job. Alternatively, remove some of the percentile requests or split the job into two smaller ones.
See also