The PSMATCH Procedure

ASSESS Statement

ASSESS <LPS> <PS> <VAR=( var-list)> </ assess-options> ;

The ASSESS statement assesses variable differences between the treated and control groups for all observations and for observations in the specified support region. It also assesses variable differences for matched observations if a MATCH statement is specified and assesses variable differences for observations by stratum if a STRATA statement is specified. In addition, the ASSESS statement assesses variable differences for weighted observations provided that the WEIGHT=NONE suboption is not specified.

You can specify variables for assessment by using the following options:

LPS: requests an assessment of differences in the logit of the propensity score.
PS: requests an assessment of differences in the propensity score.
VAR=(var-list): requests an assessment of differences in the specified list of variables. These variables must be binary classification variables or continuous variables in the input data set.

If none of these options are specified, an assessment of differences in the propensity score is produced by default.

In addition, you can specify various assess-options after a slash (/). Table 95.2 summarizes these options:

Table 95.2: ASSESS Statement Options

Option	Description
PLOTS=	Requests variable plots
STDDIFFDIV=	Specifies the divisor for the standardized difference
VARINFO	Displays variable information for the treated and control groups
WEIGHT=	Specifies the weight for the variable distribution

PLOTS <(global-option )> < = plot-request> PLOTS <(global-option )> = (plot-request < …plot-request > )

specifies options that control the plots.

You can specify the following global-options:

ONLY

suppresses the default plots and displays only plots that are specifically requested.

ORIENT=HORIZONTAL | VERTICAL

controls the orientation of the plots:

HORIZONTAL: places the lines and boxes horizontally for variable distribution plots, places the bar lengths horizontally for bar charts, places the variable values horizontally for cloud plots, and places the standardized differences on the horizontal axis for the standardized differences plot.
VERTICAL: places the lines and boxes vertically for variable distribution plots, places the bar lengths vertically for bar charts, places the variable values vertically for cloud plots, and places the standardized differences on the vertically axis for the standardized differences plot.

By default, ORIENT=HORIZONTAL.

You can specify the following plot-requests:

ALL

requests all applicable plots for all variables that are specified in the ASSESS statement. These plots include bar charts for binary classification variables, box plots for continuous variables, cloud plots for all variables, and a combined standardized differences plot for all variables. If you specify a STRATA statement, then PROC PSMATCH also produces the plots by stratum.

BAR <(DISPLAY=ALL | (bar-list ))> BARCHART <(DISPLAY=ALL | (bar-list ))>

requests comparative bar charts for binary classification variables that are specified in the VAR= option. You can specify either of the following options:

DISPLAY=ALL: requests bar charts for all binary classification variables that are specified in the VAR= option.
DISPLAY=(bar-list): specifies a subset of the binary classification variables for which bar charts are to be displayed.

By default, DISPLAY=ALL.

If you specify a STRATA statement, then the bar charts by stratum are also displayed.

BOX <(DISPLAY=ALL | (box-list ))> BOXPLOT <(DISPLAY=ALL | (box-list ))>

requests box plots for LPS, PS, and all continuous variables that are specified in the VAR= option. You can specify either of the following options:

DISPLAY=ALL: requests box plots for LPS, PS, and all continuous variables that are specified in the VAR= option.
DISPLAY=(box-list): specifies a subset of the continuous variables for which box plots are to be displayed.

By default, DISPLAY=ALL.

If you specify a STRATA statement, then the box plots by stratum are also displayed.

CLOUD <(DISPLAY=ALL | (cloud-list ))> CLOUDPLOT <(DISPLAY=ALL | (cloud-list ))>

requests cloud plots for LPS, PS, and all variables that are specified in the VAR= option. The term cloud plot is used here to refer to scatter plots in which the points have been jittered by adding random noise to prevent overplotting, which typically occurs when a continuous variable (such as age) is rounded to some convenient unit (such as years).

You can specify either of the following options:

DISPLAY=ALL: requests cloud plots for LPS, PS, and all variables that are specified in the VAR= option.
DISPLAY=(cloud-list): specifies a subset of the continuous variables for which box plots are to be displayed.

By default, DISPLAY=ALL.

If you specify a STRATA statement, then the cloud plots by stratum are also displayed.

NONE

suppresses all plots.

STDDIFF STDDIFFPLOT

requests a standardized differences plot for PS, LPS, and all variables that are specified in the VAR= option. If you specify a STRATA statement, then a standardized differences by stratum plot is also displayed.

By default, PLOTS=STDDIFF.

STDDIFFDIV=POOLED | TREATED

specifies the divisor for the standardized difference:

POOLED: uses the pooled standard deviation, which is computed as the square root of the average of the sample variance for the treated group and the sample variance for the control group.
TREATED: uses the standard deviation of the variable values in the treated group only.

By default, STDDIFFDIV=POOLED.

VARINFO

requests a variable information table for the treated and control groups.

WEIGHT=ATEWGT | ATTWGT | MATCHWGT | NONE

requests (except when the WEIGHT=NONE is specified) additional variable assessment for weighted matched observations if a MATCH statement is specified, for weighted observations in each stratum if a STRATA statement is specified, and for weighted observations in the support region if neither the MATCH nor the STRATA statement is specified:

ATEWGT: uses inverse probability of treatment weighting (IPTW) to weight the treatment and control groups up to the combined group. These weights are appropriate for estimation of the ATE. This option applies only if the MATCH statement is not specified.
ATTWGT: uses ATT weighting (also referred to as weighting by odds) to weight the control group up to the treatment group. These weights are appropriate for estimation of the ATT. This option applies only if the MATCH statement is not specified. For more information about ATT weighting, see the section ATT Weighting.
MATCHWGT: uses match weighting to weight the control group up to the treatment group. That is, in each matched set, each treated unit has a weight of 1 and each control unit has a weight that equals the number of treated units divided by the number of control units in the matched set. For example, with one-to-one pair matching, each treated unit has a weight of 1 and each control unit has a weight of 1. With this weighting, the total weight of control units is the same as the total number of treated units in each matched set, and the total weight of matched control units is the same as the total number of treated units. This weighting is useful when multiple control units are matched to each treated unit, and is appropriate for estimating the ATT. This option applies only if a MATCH statement is specified.
NONE: does not add weighted variable assessment.

By default, WEIGHT=MATCHWGT if a MATCH statement is specified, WEIGHT=NONE if a STRATA statement is specified, and WEIGHT=ATTWGT if neither the MATCH nor the STRATA statements is specified.

For more information about these propensity score weights, See the section Propensity Score Weighting.