{smcl}
{* 26 April07}{...}
{hi:help mavmb}{right: ({browse "http://www.stata-journal.com/article.html?article=st0140":SJ8-1: st0140})}
{hline}

{title:Title}

{p2colset 5 14 16 2}{...}
{p2col :{hi:mavmb} {hline 2}}Testing configuration weighted means on outcome for two groups{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 14 2} {cmd:mavmb} [{it:outcomevar} {it:configurations}]
[{it:{help if}}] 
[{cmd:,}
 {opt group(varname)}
 {cmdab:sig:only}
 {opt slev:el(#)}
 {bind:{cmdab:great:er(col1} | {cmd:col2)}}
 {cmdab:last}
 {cmdab:usered:uction}]

		
{title:Description}

{pstd}
{cmd:mavmb} performs a test of the means of each configuration entered in
{it:configurations} with the outcome versus a 2-category group variable (e.g.,
male/female). Each mean is weighted by the membership in that configuration.
The configurations do not need to exist in the dataset, but their individual
single, capital letter versions must exist.


{title:Options}

{p 4 8 2}
{opt group(varname)} determines the groups to test the means against.
{it:varname} must be a 2-category variable.  A group variable must
be specified if {cmdab:mavmb} is entered, unless one was specified in the last
run call to {cmdab:fuzzy} or {cmdab:mavmb}. 

{p 4 8 2}{cmd:sigonly} displays only those configurations that have
significantly different means.	

{p 4 8 2}{opt slevel(#)} changes the significance level to be used in 
determining {cmd:sigonly}.  The default is {cmd:slevel(.05)}.

{p 4 8 2}
{cmd:greater(col1} | {cmd:col2)} displays only those configurations in which
the specified column value is greater than the other column value. 
{cmd:col1} is always the mean with the outcome for the first group in the group
variable, while {cmd:col2} is the mean for the second group.

{p 4 8 2}
{cmd:last} uses the outcome variable and configurations (or sets) from
the last run {cmd:fuzzy} or {cmd:settest()} option call as {it:outcomevar}
and {it:configurations}.  When this option is specified, {it:outcomevar} and
{it:configurations} should not be entered. Specifically, it uses the last
displayed set of configurations.  Thus, when using this option, a
{cmd:settest()} option must have been invoked on the last call to {cmd:fuzzy}
or just run as a stand-alone command.  When {cmd:last} is done, only those last
displayed configurations will be used in the test. 

{p 4 8 2}{cmd:usereduction} uses the outcome variable and reduced
configurations from the last run {cmd:reduce} command for the means test.	       


{title:Examples}

{p 4 8 2}{cmd:. mavmb Y Abc aBc AC, group(female)}

{p 4 8 2}{cmd:. fuzzy drink age grades friends, label(D A G F)}{p_end}
{phang}{cmd:. mavmb, group(female) greater(col1)}

{p 4 8 2}{cmd:. fuzzy drink age grades friends, label(D A G F) settest(yvn) sigonly}{p_end}
{phang}{cmd:. mavmb, group(female) greater(col1) last}

{p 4 8 2}{cmd:. fuzzy D A G F, settest(yvn yavyb) group(female) sigonly greater(col1) reduce}{p_end}
{phang}{cmd:. mavmb, usereduction sigonly}

{p 4 8 2}{cmd:. fuzzy drink age grades friends, label(D A G F) settest(yvn)}{p_end}
{phang}{cmd:. yvv, last sigonly}{p_end}
{phang}{cmd:. mavmb, group(female) last sigonly}
{p_end}


{title:Also see}

{psee}
Article: {it:Stata Journal}, volume 8, number 1: {browse "http://www.stata-journal.com/article.html?article=st0140":st0140}
{p_end}