Codes that indicate "missing" whenever they appear in the data may be specified with Missing=. They will then be ignored whenever they appear. "Missing" refers to missing observations. It means "observation unknown or observation not collected, so ignore it".

 

There cannot be "missing element numbers". If the element number is unknown (for instance, you don't know which item was being rated), then the observation cannot be included in the analysis - so omit it from the data or comment it out with ";". If the element number does not apply (for instance, an item that does not require a rater), then the element number is null, usually 0, the Keep as null= value.

 

Facets treats as "missing data", any data value (or space or omitted) outside the specified numerical range for a valid observation.

 

Example: Models=?,?,?,R2

Valid data for R2 are 0,1,2 so anything else is treated as missing data.

 

The first time Facets encounters a value outside the valid range (other than blank or omitted), it issues a warning message, just in case you have a data entry error, but proceeds with the analysis:

Check (2)? Invalid datum location: "___" in line: ___. Datum "_" is too big or non-numeric, treated as missing.

 

I recommend being explicit about missing data. Use the same code every time, e.g., ".".

If you do this, then you can specify:

Missing = .

so that Facets will not issue a warning message for ".".

 

So in your file you might have:

Facets=4

Models=?,?,?,?,R6

1,2,3,1-4, 3, 2, ., 4 ; where the response to item 3 is missing

 

Example 1: Missing observations have been entered in the data file with code "9". These are to be ignored and bypassed.

Missing = 9

Facets = 2

Data=

22, 36, 1 ; this observation of "1" is analyzed

23, 37, 9 ; this observation of "9" is ignored as missing data

 

Example 2: Codes "^", "." and "#" are all to be ignored as observation values.

Missing = ^, ., #

Facets = 3

Data=

22, 0, 17, 1 ; this observation of "1" is analyzed. Element "0" means that facet 2 does not apply to this observation.

23, 3, 13, # ; this observation of "#" is ignored as missing data

 

Example 3: To make specific observations into missing data use the missing data model.

Models=

23, 37, M ; all data points with elements 23 and 37 are treated as missing