Skip to content
alexdean201 edited this page Nov 29, 2018 · 3 revisions

Purpose

This tool tabulates the item metadata included in Smarter Balanced Test Administration packages. It can also be used with Test Scoring packages since they use the same format for the item sections.

Multiple input packages may be specified in which case the output is the aggregation of all input files.

By default the tool processes only test administration packages as determined by the @purpose attribute on the root element in the file. All other .xml files are skipped. The -scoring option will cause it to tabulate scoring packages and ignore all other package types.

The tool must be run on a Windows machine.

Command-line parameters:

-i <input filename>
Specifies an input filename which should be the name of a test administration or test reporting package. The Filename may include wildcards in which case all matching files will be processed. If the filename is a .zip then all .xml files in the .zip will be processed regardless of their path within the .zip file. The -i parameter may be repeated to process multiple input sources.

-o <output filename>
Output filename prefix. The results are stored in the following files:
filename.items.csv Tabulation of item data
filename.stims.csv Tabulation of stimulus data
filename.errors.csv Tabulation of errors (if any)

-ci <file path>
If a content tabulator item output file is specified, input test packages will be cross-tabulated against that csv output file and any discrepencies will be reported.

-cs <file path>
If a content tabulator stimuli output file is specified, input test packages will be cross-tabulated against that csv output file and any discrepencies will be reported.

-sv
If this flag is present and both cs and ci arguments are provided, the tabulator will write cross tabulation errors for stimuli versions"

Sample Command Line Input

Clone this wiki locally