Difference between revisions of "Documentation/How Tos/Calc: DSTDEV function"
(Initial content) |
m |
||
| Line 22: | Line 22: | ||
| − | <tt>'''DSTDEV'''</tt> returns the '''sample standard deviation''', which assumes that the chosen rows of the <tt>'''database_table'''</tt> contain a random sample of a normally distributed population. If the chosen data are the entire population, use '''<tt>[[Documentation/How_Tos/Calc: DSTDEVP function|DSTDEVP]]</tt>''' instead. | + | <tt>'''DSTDEV'''</tt> returns the '''sample standard deviation''', which assumes that the chosen rows of the <tt>'''database_table'''</tt> contain a random sample of a normally distributed population. If the chosen data are the entire population, use '''<tt>[[Documentation/How_Tos/Calc: DSTDEVP function|DSTDEVP]]</tt>''' instead. The <tt>'''DSTDEV'''</tt> calculation uses this formula: |
| + | <center>[[Image:Calc_sample_stddev_formula.png]]</center> | ||
| − | + | where N is the number of values included and x<sub>i</sub> are those values. | |
<tt>'''DSTDEV'''</tt> ignores any cell containing text in the <tt>'''field'''</tt> column. | <tt>'''DSTDEV'''</tt> ignores any cell containing text in the <tt>'''field'''</tt> column. | ||
| + | |||
| + | |||
| + | Simply put, standard deviation is a measure of how widely spread data values are. It is the square root of the <i>variance</i> (see '''<tt>[[Documentation/How_Tos/Calc: DVAR function|DVAR]]</tt>''', '''<tt>[[Documentation/How_Tos/Calc: DVARP function|DVARP]]</tt>'''). Standard deviation is a reliable measure only if there is enough data to examine. | ||
Revision as of 18:22, 4 December 2007
DSTDEV
Returns the sample standard deviation of values in a column of a Calc 'database' table, in rows which meet specified
criteria.
Syntax:
DSTDEV(database_table; field; criteria_table)
where
- database_table is a range defining the data to be examined.
- field is the column to examine. It may be a column number (1 is the first column of the database table, 2 is the second ...) or a column header (enclosed in quotation marks ””) or a cell referring to a column header.
- criteria_table is a range containing criteria, which are used to select which rows of the database_table to examine.
- The Database functions overview describes these parameters in detail.
DSTDEV returns the sample standard deviation, which assumes that the chosen rows of the database_table contain a random sample of a normally distributed population. If the chosen data are the entire population, use DSTDEVP instead. The DSTDEV calculation uses this formula:
where N is the number of values included and xi are those values.
DSTDEV ignores any cell containing text in the field column.
Simply put, standard deviation is a measure of how widely spread data values are. It is the square root of the variance (see DVAR, DVARP). Standard deviation is a reliable measure only if there is enough data to examine.
Example:
In this spreadsheet:
| A | B | C | D | E | |
| 1 | Name | Grade | Age | Distance to School | Weight |
| 2 | Andy | 3 | 9 | 150 | 40 |
| 3 | Betty | 4 | 10 | 1000 | 42 |
| 4 | Charles | 3 | 10 | 300 | 51 |
| 5 | Daniel | 5 | 11 | 1200 | 48 |
| 6 | Eva | 2 | 8 | 650 | 33 |
| 7 | Frank | 2 | 7 | 300 | 42 |
| 8 | Greta | 1 | 7 | 200 | 36 |
| 9 | Harry | 3 | 9 | 1200 | 44 |
| 10 | Irene | 2 | 8 | 1000 | 42 |
| 11 | |||||
| 12 | |||||
| 13 | Name | Grade | Age | Distance to School | Weight |
| 14 | 2 |
DSTDEV(A1:E10; "Weight"; A13:E14)
- returns the sample standard deviation of the weights of children in the second grade. This is not a useful measure, as there are so few children.
See also:
DCOUNT function, DCOUNTA function, DSUM function, DPRODUCT function,
DMAX function, DMIN function, DAVERAGE function,
DSTDEVP function, DVAR function, DVARP function,
Issues
- The OOo2.3 Help implies that 0 as a field will include the entire database table. This seems to be wrong - only the standard deviation of a single column can be found.
- Logical values TRUE and FALSE are interpreted as 1 and 0 in the calculation. This is not compatible with Excel, which ignores logical values. This will very rarely cause difficulty.
