27.03.2013 Views

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

• The method keywords follow the specifications for break points.<br />

EXAMINE 523<br />

For each of the following methods of percentile calculation, w is the sum of the weights for<br />

all nonmissing cases, p is the specified percentile divided by 100, and X i is the value of the<br />

ith case (cases are assumed to be ranked in ascending order). For details on the specific<br />

formulas used, see the SPSS Statistical Algorithms chapter for EXAMINE.<br />

HAVERAGE Weighted average at ( + )p.<br />

The percentile value is the weighted average<br />

of Xi and Xi + 1 , where i is the integer part of ( w + 1)p.<br />

This is the default<br />

if PERCENTILES is specified without a keyword.<br />

WAVERAGE Weighted average at . The percentile value is the weighted average of<br />

and , where i is the integer part of wp.<br />

ROUND Observation closest to wp. The percentile value is Xi or Xi + 1 , depending<br />

upon whether i or i+1 is “closer” to wp .<br />

EMPIRICAL Empirical distribution function. The percentile value is Xi , where i is equal<br />

to wp rounded up to the next integer.<br />

AEMPIRICAL Empirical distribution with averaging. This is equivalent to EMPIRICAL, except<br />

when i=wp, in which case the percentile value is the average of Xi and<br />

.<br />

NONE Suppress percentile output. This is the default if PERCENTILES is omitted.<br />

Example<br />

EXAMINE VARIABLE=SALARY /PERCENTILES(10,50,90)=EMPIRICAL.<br />

• PERCENTILES produces the 10th, 50th, and 90th percentiles for the dependent variable<br />

SALARY using the EMPIRICAL distribution function.<br />

PLOT Subcommand<br />

Xi +<br />

1<br />

X( i + 1)<br />

X w 1<br />

Xwp X i<br />

PLOT controls plot output. The default is a vertical boxplot and a stem-and-leaf plot for each<br />

dependent variable for each cell in the model.<br />

• Spread-versus-level plots can be produced only if there is at least one factor variable on<br />

the VARIABLES subcommand. If you request a spread-versus-level plot and there are no<br />

factor variables, the program issues a warning and no spread-versus-level plot is produced.<br />

• If you specify the PLOT subcommand, only those plots explicitly requested are produced.<br />

BOXPLOT Vertical boxplot. The boundaries of the box are Tukey’s hinges. The<br />

median is identified by an asterisk. The length of the box is the interquartile<br />

range (IQR) computed from Tukey’s hinges. Values more than<br />

three IQR’s from the end of a box are labeled as extreme (E). Values<br />

more than 1.5 IQR’s but less than 3 IQR’s from the end of the box are<br />

labeled as outliers (O).<br />

STEMLEAF Stem-and-leaf plot. In a stem-and-leaf plot, each observed value is divided<br />

into two components—leading digits (stem) and trailing digits<br />

(leaf).

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!