You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
126 lines
9.0 KiB
Plaintext
126 lines
9.0 KiB
Plaintext
7 months ago
|
{smcl}
|
||
|
|
{* 2013}{...}
|
||
|
|
{hline}
|
||
|
|
help for {hi:valid}
|
||
|
|
{hline}
|
||
|
|
|
||
|
|
{title:Syntax}
|
||
|
|
|
||
|
|
{p 8 14 2}{cmd:valid} {it:varlist}, {bf:partition}({it:numlist}) [{it:options}]
|
||
|
|
|
||
|
|
{p 4 4 2}{it:varlist} contains the variables (items) used to calculate the scores. The first items of {it:varlist} compose the first dimension, the following items define the second dimension, and so on.
|
||
|
|
|
||
|
|
{p 4 4 2}{cmd:partition} permits to define in {it:numlist} the number of items in each dimension.
|
||
|
|
|
||
|
|
|
||
|
|
{synoptset 20 tabbed}{...}
|
||
|
|
{synopthdr}
|
||
|
|
{synoptline}
|
||
|
|
{syntab:Options}
|
||
|
|
{synopt : {opt sco:rename(string)}}define the names of the dimensions{p_end}
|
||
|
|
{synopt : {opt imp:ute}}impute missing item responses{p_end}
|
||
|
|
{synopt : {help valid##impute_options:{it:impute_options}}}options for imputation of missing data {p_end}
|
||
|
|
{synopt : {opt calc:method(method)}}define how scores are calculated{p_end}
|
||
|
|
{synopt : {opt desc:items}}display a description of items and dimensions{p_end}
|
||
|
|
{synopt : {opt graph:s}}display graphs for items description{p_end}
|
||
|
|
{synopt : {opt cfa}}assess structural validity by performing a confirmatory factor analysis{p_end}
|
||
|
|
{synopt : {help valid##cfa_options:{it:cfa_options}}}options for confirmatory factor analysis{p_end}
|
||
|
|
{synopt : {opt conv:div}}assess convergent and divergent validity assessment{p_end}
|
||
|
|
{synopt : {help valid##conv_div_options:{it:conv_div_options}}}options for convergent and divergent validity{p_end}
|
||
|
|
{synopt : {help valid##reliability_options:{it:reliability_options}}}options for reliability assessment{p_end}
|
||
|
|
{synopt : {opt rep:et(varlist)}}assess reproducibility of scores and items{p_end}
|
||
|
|
{synopt : {help valid##repet_options:{it:repet_options}}}options for reproducibility{p_end}
|
||
|
|
{synopt : {opt kgv(varlist)}}assess known-groups validity by using qualitative variable(s){p_end}
|
||
|
|
{synopt : {help valid##kgv_options:{it:kgv_options}}}options for known-groups validity assessment{p_end}
|
||
|
|
{synopt : {opt conc(varlist)}}assess concurrent validity{p_end}
|
||
|
|
{synopt : {help valid##conc_options:{it:conc_options}}}options for concurrent validity assessment{p_end}
|
||
|
|
|
||
|
|
{p2colreset}{...}
|
||
|
|
|
||
|
|
|
||
|
|
{title:Description}
|
||
|
|
|
||
|
|
{phang}{cmd:valid} assesses validity and reliability of a multidimensional scale. Specifically it evaluates
|
||
|
|
structural validity, convergent and divergent validity, reproducibility, known-groups validity, internal consistency, scalability and sensitivity.
|
||
|
|
|
||
|
|
{marker options}{...}
|
||
|
|
{title:Options}
|
||
|
|
|
||
|
|
{dlgtab:Options}
|
||
|
|
|
||
|
|
{phang}{opt sco:rename(string)} allows defining the names of the dimensions. If the option is not used then dimensions are named {it:Dim1}, {it:Dim2},...
|
||
|
|
|
||
|
|
{phang}{opt imp:ute} imputes missing items responses with Person Mean Substitution method applied in each dimension. Missing data are imputed only if the number of
|
||
|
|
missing values in the dimension is less than half the number of items in the dimension.
|
||
|
|
|
||
|
|
{marker impute_options}{...}
|
||
|
|
{phang}{it:impute_options} allow specifying options for imputation of missing. By default, imputed values are rounded to the nearest whole number. If {opt nor:ound} is precised then imputed values
|
||
|
|
are not rounded. If {opt impute} is absent then {opt noround} is ignored.
|
||
|
|
|
||
|
|
{phang}{opt calc:method(method)} defines the method used for calculating scores. {it:method} may be either {bf:mean} (default), {bf:sum} or {bf:stand}(set scores from 0 to 100).
|
||
|
|
|
||
|
|
{phang}{opt desc:items} displays a description of items. This option gives missing data rate per item and distribution of item responses. It also gives Cronbach's alpha for each item, which is the alpha statistic
|
||
|
|
calculated by removing the item from the dimension. Moreover, the option computes Loevinger's Hj coefficients and the number of non significant Hjk. See {help loevh} for details about Loevinger's coefficients.
|
||
|
|
|
||
|
|
{phang}{opt graph:s} displays graphs for items and dimensions description. It provides histograms of scores, a biplot of dimensions and a biplot of items.
|
||
|
|
|
||
|
|
{phang}{opt cfa} performs a confirmatory factor analysis with Stata's {help sem} command. It displays estimations of coefficients and several goodness-of-fit indices.
|
||
|
|
|
||
|
|
{marker cfa_options}{...}
|
||
|
|
{phang}{it:cfa_options} allow specifying options for confirmatory factor analysis. {opt cfam:ethod}({it:method}) specifies the method of estimation of parameters. {it:method} may be either {bf:ml} (maximum
|
||
|
|
likelihood), {bf:mlmv} ({bf:ml} with missing values) or {bf:adf} (asymptotic distribution free). The {opt cfas:tand} option displays standardized coefficients.
|
||
|
|
|
||
|
|
{phang}{opt conv:div} assesses convergent and divergent validity. The option displays the matrix of correlations between items and dimensions.
|
||
|
|
|
||
|
|
{marker convdiv_options}{...}
|
||
|
|
{phang}{it:conv_div_options} allow specifying options for convergent and divergent validity. {opt tconv:div(#)} defines a threshold for highlighting some values. # is a real number between 0 and 1 which is equal to 0.4 by
|
||
|
|
default. Correlations between items and their own score are printed
|
||
|
|
in red if it is less than #. Moreover, if an item is less correlated with its own dimension than with another one the correlation is printed
|
||
|
|
in red. The {opt convdivb:oxplots} option displays boxplots for assessing convergent and divergent validity. The boxes represent the correlations between the items of a given dimension and all dimensions. So the box of correlations between items of a given dimension and its score must be higher than other
|
||
|
|
boxes. There is as many boxplots (graphs) as dimensions.
|
||
|
|
|
||
|
|
{marker reliability_options}{...}
|
||
|
|
{phang}{it:reliability_options} allow defining the thresholds for reliability indices. {opt a:lpha(#)} defines a threshold for Cronbach's alpha (see {help alpha}). # is a real number between 0 and 1 which is equal to 0.7
|
||
|
|
by default. Cronbach's alpha coefficients less than # are printed in red. {opt d:elta(#)} defines a threshold for Ferguson's delta coefficient (see {help delta}). # is a real number between 0 and 1 which is equal to 0.9
|
||
|
|
by default. Ferguson's delta coefficients less than # are printed in red. {opt h(#)} defines a threshold for Loevinger's H coefficient (see {help loevh}). # is a real number between 0 and 1 which is equal to
|
||
|
|
0.3 by default. Loevinger's H coefficients less than # are printed in red. {opt hj:min(#)}
|
||
|
|
defines a threshold for Loevinger's Hj coefficients. The displayed value is the minimum Hj coefficient for a item in the dimension. (see {help loevh}). # is a real number between 0 and 1 which is equal to
|
||
|
|
0.3 by default. If the minimum Loevinger's Hj coefficient is less than # then it is printed in red and the corresponding item is displayed.
|
||
|
|
|
||
|
|
{phang}{opt rep:et(varlist)} assesses reproducibility of scores by precising in {it:varlist} the variables corresponding to responses at time 2. Scores are calculated according to the {opt partition()} option. Intraclass
|
||
|
|
Correlation Coefficients (ICC) for scores and their 95% confidence interval are computed with Stata's {help icc} command.
|
||
|
|
|
||
|
|
{marker repet_options}{...}
|
||
|
|
{phang}{it:repet_options} display information about reproducibility of items. The {opt kap:pa} option computes kappa statistic for items with Stata's {help kap} command. The {opt ickap:pa(#)} calculates
|
||
|
|
confidence intervals for
|
||
|
|
kappa statistics. # is the number of replications for bootstrap used to estimate confidence intervals if items are polytomous. See {help kapci} for more details about calculation of confidence intervals for kappa's
|
||
|
|
coefficients. If the {opt kappa} option is absent then {opt ickappa(#)} is ignored.
|
||
|
|
|
||
|
|
{phang}{opt kgv(varlist)} assesses known-groups validity according to the grouping variables precised in {it:varlist}.
|
||
|
|
|
||
|
|
{marker kgv_options}{...}
|
||
|
|
{phang}{it:kgv_options} allow to display graphs for known-groups validity. The {opt kgvb:oxplots} option draws boxplots of scores split into groups of individuals. The {opt kgvg:roupboxplots} option groups
|
||
|
|
all boxplots in one graph. If {opt kgvboxplots} is absent then {opt kgvboxplotsgroup} is ignored.
|
||
|
|
|
||
|
|
{phang}{opt conc(varlist)} assesses concurrent validity with variables precised in {it:varlist}. These variables are scores from one or several other scales.
|
||
|
|
|
||
|
|
{marker conc_options}{...}
|
||
|
|
{phang}{it:conc_options} allow to specify options for concurrent validity. The {opt tc:onc(#)} option defines a threshold for correlations between scores and those of other scales in {it:varlist}. Correlation
|
||
|
|
coefficients greater than # are displayed in bold.
|
||
|
|
|
||
|
|
|
||
|
|
{marker examples}{...}
|
||
|
|
{title:Examples}
|
||
|
|
|
||
|
|
{phang2}{cmd:. valid item1-item20, part(5 4 6 5)}{p_end}
|
||
|
|
|
||
|
|
{phang2}{cmd:. valid item1-item20, part(5 4 6 5) imp graphs cfa cfastand convdiv convdivboxplots kgv(factor_variable) kgvboxplots conc(scoreA-scoreD)}{p_end}
|
||
|
|
|
||
|
|
{phang2}{cmd:. valid item1-item20, part(5 4 6 5) imp rep(item1bis-item20bis) kappa}{p_end}
|
||
|
|
|
||
|
|
|
||
|
|
{marker alsosee}{...}
|
||
|
|
{title:Also see}
|
||
|
|
|
||
|
|
{p 4 13 2}help for {help alpha}, {help delta}, {help loevh}, {help icc}, {help kapci}.{p_end}
|