{smcl}
{* 08apr2010}{...}
{hline}
help for {hi:ordvar}{right:(Michael Lacy)}
{hline}

{p2colset 5 16 18 2}{...}
{p2col :{hi:ordvar} {hline 2}} Ordinal variation and consensus statistics {p_end}
{p2colreset}{...}

{title:Syntax}

{p 8 15 2}
{cmd:ordvar} {it:varname}  [if] [in] [,{cmd:low(}{it:#}{cmd:)} {cmd:high(}{it:#}{cmd:)}]  {p_end}

{synoptline}
{cmd:} The {bf:by} prefix is allowed.

{title:Description}

{pstd} {cmd:ordvar} calculates measures of ordinal consensus
and dispersion. These include {it:lsq} and 1-{it:lsq}, which are 0/1 normed ordinal consensus
and dispersion statistics described in Blair and Lacy (2000). The latter statistic is
essentially identical to the IOV measure of Berry and Mielke (1992).  The non-normed version of the 
dispersion statistic, termed d-squared by Lacy (2006), is also calculated, as are delta-method standard
errors for these statistics. These measures do not rely on assumptions about intercategory distances or
distributional form.

{title:Arguments and Options}
{pstd}{it:varname} is the variable for which the dispersion and consensus statistics are desired.
This variable is assumed to be coded with consecutive integers, starting 
with the lowest value occurring in the data up to the highest. As an option, to allow for
situations in which the lowest or highest theoretically possible value 
does not occur in the data, the user may specify these values by using the low(#) and high(#) options.  
Failure to specify this option when necessary will give incorrect results. 

{title:Returned results}

{pstd}{cmd:ordvar} saves the following in {cmd:r()}:

{synoptset 16 tabbed}{...}
{p2col 5 16 20 2: Scalars}{p_end}
{synopt:{cmd:r(onemlsq)}} 1-{it:lsq} normed ordinal dispersion statistic {p_end}
{synopt:{cmd:r(lsq)}} {it:lsq} normed ordinal dispersion statistic {p_end}
{synopt:{cmd:r(dsq)}} {it:d-squared} non-normed ordinal dispersion statistic {p_end}
{synopt:{cmd:r(dsqse)}}  standard error of {it:d-squared} {p_end}
{synopt:{cmd:r(lsqse)}}  standard error of 1-{it:lsq} or {it:lsq} {p_end} 
{p2colreset}{...}

{title:Examples}

{psee}{cmd:}{p_end}
{psee}{cmd:. sysuse auto}{p_end}
{psee}{cmd:. ordvar rep78 }{p_end}
{psee}{cmd:.} // What if some theoretical minimum or maximum value did not occur in the data? {p_end}
{psee}{cmd:. keep if foreign==1}{p_end}
{psee}{cmd:. tab1 rep78}{p_end}
{psee}{cmd:. ordvar rep78, low(1) high(5)}{p_end}


{title:References}

{phang}Blair, J. and M. Lacy. 2000. "Statistics of Ordinal Variation." 
Sociological Methods and Research. 28: 251-280. 

{phang}Lacy, M. 2006.  "An Explained Variation Measure for Ordinal Response 
Models with Comparisons to Other Ordinal R2  Measures." Sociological Methods 
and Research  34:469-520.

{phang}Berry, K. J. and P. W. Mielke, Jr. 1992. "Assessment of Variation in 
Ordinal Data." Perceptual and Motor Skills 74:63-66.

{title:Author}

{phang}Michael G. Lacy{p_end}
{phang}Colorado State University{p_end}
{phang}Fort Collins, Colorado{p_end}
{phang}Michael.Lacy@colostate.edu{p_end}