MISSINGSUMMARIES Subcommand (MULTIPLE IMPUTATION command)
The MISSINGSUMMARIES
subcommand
controls the display of missing values of analysis variables.
OVERALL Keyword
The OVERALL
keyword displays
a graph that summarizes missingness for cases, variables, and individual
data (cell) values:
- Number and percent of analysis variables that have one or more missing values.
- Number and percent of cases that have one or more missing values on analysis variables.
- Number and percent of individual data values that are missing among all analysis variables and cases.
Overall summaries are displayed by default.
VARIABLES Keyword
The VARIABLES
keyword displays
a table of analysis variables sorted by percent of missing values.
The table includes descriptive statistics (mean and standard deviation)
for scale variables. The table is not shown by default.
You
can optionally control the maximum number of variables to display
(default=25) and minimum percentage missing (default=10) for a variable
to be included in the display. Specify one or both in parentheses.
For example, VARIABLES(MAXVARS=50 MINPCTMISSING=25)
requests that the table display up to 50 variables that have at
least 25% missing values. The set of variables that meet both criteria
are displayed. For example, by default, if there are 25 analysis variables
but only 5 have 10% or more missing values, the output includes only
5 variables. If there are more analysis variables than MAXVARS
and all meet the MINPCTMISSING
criteria, then variables with
the most missing values are displayed.
-
MAXVARS
must be a positive integer. -
MINPCTMISSING
must be a nonnegative number that is less than 100.
PATTERNS Keyword
The PATTERNS
keyword displays
tabulated patterns of missing values. Each pattern corresponds to
a group of cases with the same pattern of incomplete and complete
data on analysis variables. You can use PATTERNS
output to determine whether monotone imputation method can be used
for your data, or if not, how closely your data approximate a monotone
pattern.
To display patterns, specify the PATTERNS
keyword. The chart orders analysis
variables and patterns to reveal monotonicity where it exists. Specifically,
variables are ordered from left to right in increasing order of missingness.
Rows (patterns) are the sorted first by the last variable (nonmissing
values first, then missing values), then by the second to last variable,
and so on, working from right to left.
You can use PATTERNS
output to determine whether monotone
imputation method can be used for your data, or, if not, how closely
your data approximate a monotone pattern. If no nonmonotone pattern
exists after reordering you can conclude that the data have a monotonic
pattern when analysis variables are ordered as such. If data are not
monotone, the chart shows how to achieve reach monotonicity.
NONE Keyword
NONE
suppresses all MISSINGSUMMARIES
output.
NONE
generates an error when used with any
other keyword.