Metadata Model to Quantum name change. MDM2Quantum has been renamed to Metadata Model to Quantum.

Categoricals. You can now allocate categoricals using different data formats (including fill zero on the left). Valid data formats for single categoricals are Numeric, Literal, and Punch. Valid data formats for multiple categoricals are Numeric, BitString, and Punch. A new global option on the Preferences dialog allows you to allocate all categoricals as numerics. Updates to the Edit Variable dialog allow you to set CodeWidths and Values. This also impacts the Data Map and Quantum specifications. For more information, see Allocating card, column, and punch Codes manually.

Enhanced grid support. The data map now uses FullLabel instead of Label for better grid support.

Precision and scale. Precision and scale is now used in determining the column count.

Column range. The column range has been increased to support additional formats beyond Quantum.

Triple-S export support. Support has been added for allocating columns for multipliers to support Triple-S export. For more information, see Support for Triple-S export.

Shared lists. You can now allocate shared lists separately for each question.

You can now export categorical variables as numerics through the DataFormat custom property. When DataFormat is set to "numeric," the variable is exported with the value set in the Value custom property. If the Value property is not set, the value is determined by combining the column offset and punch.

Differences in data types when importing a data map. If a variable has the same name in both the currently open document and the data map file being imported, but the data types are different, Metadata Model to Quantum issues an error message. Errors of this type do not immediately halt the import process; instead, Metadata Model to Quantum continues attempting to import the data map and displays the errors at the end of the procedure. For example, if the currently open document shows the age variable as being numeric but the import data map shows it as categoric, the error message will be "age expected to be type Lng. Is of type Cat."

Support for UNICOM Intelligence Interviewer 3 (IOM) question features. Metadata Model to Quantum supports the additional question and response features available with UNICOM Intelligence Interviewer 3, including the option

Enhanced facilities for reading and writing files. The Metadata Model to Quantum component now supports the following methods:

Events. The new methods for reading from files can raise the following events:

Error codes. Three new error codes have been added to ErrorConstants. These codes are returned by the new methods described earlier.

The Metadata Model to Quantum component has been ported to Visual Basic .NET in order to provide better interaction with browser-based applications. This change should be invisible to users but has the advantage that developers who use .NET can now use the assembly instead of the COM interop.

Multiple response variables. Changes in UNICOM Intelligence Data Model 2.7 led to a problem when the ExplodeMultiPunch option was used in a document that had one or more shared lists that were used in both single and multiple response questions. To resolve this problem, the AllocateCardColPunch and AllocateVariables methods now allocate the punch codes on the ElementInstance objects of multiple response variables rather than the Element objects.

Multiplier variables. The AllocateCardColPunch and AllocateVariables methods now allocate multiplier variables (these are sometimes found in data sets that originated in Quanvert). In previous versions of the UNICOM Intelligence Data Model, multiplier variables were ignored.

Creating the Quantum specification. The new MDSC feature of the Quantum DSC means that it is now easy to create a basic Quantum specification programmatically. The Metadata Model to Quantum sample mrScriptBasic files have been updated to provide examples of doing this. For more information, see Creating an Quantum specification using a script.

Working with hierarchical data. In UNICOM Intelligence Data Model 2.8, the Metadata Model to Quantum component is not designed for use in data sets that cannot be represented in a flattened form. For example, the Metadata Model to Quantum component is not designed for use with Quanvert levels projects or other metadata that contains unbounded loops (Array objects of type mlLevel).

Improved Quantum specifications. Metadata Model to Quantum now uses the new MDSC feature of the Quantum DSC to create the Quantum specification. As a result, a number of problems have now been resolved and you may notice a few minor differences compared to previous versions. For example:

In addition, you can now select the user context and language to be used for the texts in the specification. For more information, see General preferences.

Using Metadata Model to Quantum with UNICOM Intelligence Interviewer - Server Admin. Changes in UNICOM Intelligence Interviewer - Server Admin 2.1 mean that the recommendations for how to use Metadata Model to Quantum with UNICOM Intelligence Interviewer - Server Admin have changed. For more information, see Using Metadata Model to Quantum with UNICOM Intelligence Interviewer - Server Admin.

Working with hierarchical data. In UNICOM Intelligence Data Model 2.8, Metadata Model to Quantum is not designed for use in data sets that cannot be represented in a flattened form. For example, Metadata Model to Quantum is not designed for use with Quanvert levels projects or other metadata that contains unbounded loops.

Punch definitions. Metadata Model to Quantum now writes punch definitions to the DataSourceProperties collection on the Element rather than the ElementInstance object. This change is largely invisible because the MDM inheritance rules mean that DataSourceProperties on the ElementInstance object inherits from the Element object. However, this change has affected the handling of variable instances that share a category list with other variable instances (for example, when a variable instance is part of a grid or loop or uses a shared category list). If you clear or allocate an individual variable instance that shares a category list with any other variable instances, the punch and column offset details for those other variable instances will be silently updated too.

However, this change has the advantage that the categories in category lists that are shared between questions have the same punch codes in each question and has led to a significant reduction in .mdd file size in projects that have large numbers or grids, loops, or defined lists.

Punch errors. Metadata Model to Quantum continues to write the punch error codes to the DataSourceProperties collection on the ElementInstance objects. This means that CheckCardColPunch can mark individual element instances in a grid whose punch definitions are inconsistent. However, CheckCardColPunch does not do a similar consistency check on the element instances that are in a loop or part of a shared list.

System variables. In Data Model 2.6 and earlier, the elements in the categorical system variables had built-in punch and column offset definitions. These have been removed in UNICOM Intelligence Data Model 2.7 because they resulted in different punch allocations being made depending on whether or not they were cleared before allocation. However, the card and column definitions that are built into the system variables themselves are unchanged, except that the column count built-in to the DataCollection.Status system variable has been increased to 10 to make it compatible with the ExplodeMultipunch option.

Improved version handling. When you open a file that has more than one version, Metadata Model to Quantum now lists the available versions and lets you choose the version or versions you want to work on. By default, the version or versions you selected last time are automatically selected for you. For more information, see Version handling in Metadata Model to Quantum.

Improved handling of shared category lists. Category lists are frequently shared between questions and variables. For example, all of the subquestions in a categorical grid question share the same category list and when the same category list is required by several questions, it is often set up as a shared list (sometimes called a defined list). In UNICOM Intelligence Data Model 2.6 and earlier, the punch and column offset details were allocated and stored separately for each category in each variable. In UNICOM Intelligence Data Model 2.7, they are now allocated and stored only once for each category in a category list that is shared. This has the advantage that the categories will always have the same punch codes and has led to a significant reduction in .mdd file size in projects that have large numbers or grids, loops, or defined lists.

Improved handling of system variables. The system variables have the column count definition built-in. In UNICOM Intelligence Data Model 2.6 and earlier, they also had punch and column offset definitions built-in. However, this lead to an anomaly whereby Allocate Cols and Punches defined different punch codes depending on whether you had run a clear in the session or not. Therefore in UNICOM Intelligence Data Model 2.7, the system variables no longer have punch and column offset definitions built-in. In addition, the column count built-in to the DataCollection.Status system variable has been increased to 10 to make it compatible with the Explode Multipunched Questions option.

Order of variables. Changes to the MDM in UNICOM Intelligence Data Model 2.7 mean that you may see differences in the order in which variables that relate to the same question or loop are displayed in the Metadata Model to Quantum window. For example, in the museum_qq.mdd sample there is a question called School that has an Other Specify category. In the metadata the School question has three helper variables: School.Other, which is a text variable for storing the open-ended responses to the Other Specify category, School.Other.Codes, which is a categorical variable for storing the coded responses to the Other Specify category, and School.Other.SourceFile, which is used to store the name of the file that contains an image of the scanned response. In UNICOM Intelligence Data Model 2.6, Metadata Model to Quantum listed the variables in this order:

Whereas in UNICOM Intelligence Data Model 2.7, they are listed as follows:

You may notice similar changes in the order in which the variables that belong to a grid or a loop are displayed. This change also affects the order in which the card and column allocations will be made. However, Metadata Model to Quantum will not overwrite any existing allocations if you choose the option to keep previous allocations. For more information, see Allocate Cols and Punches.

Opening other types of data. You can now open in Metadata Model to Quantum any type of data for which you have a read-enabled MDSC. For example, you can now open a .sav file in Metadata Model to Quantum. Metadata Model to Quantum uses the to create an MDM document from the .sav file and you can save the MDM document as an .mdd file. For more information, see Opening a file or database of another type.

Specifying the serial number variable. When you open a file that does not include the Respondent.Serial system variable, you can now specify which variable is to be used as the serial number variable. This is useful when you are using Metadata Model to Quantum with non-UNICOM Intelligence data. For more information, see Select Serial Variable.

Clearing individual variables. You can now clear the card, column, and punch allocations on one or more individual variables. For more information, see Clearing one or more individual variables.

Improved version handling. Metadata Model to Quantum no longer creates a new version when you open a questionnaire definition file in which the most recent version is locked. Now Metadata Model to Quantum gives you the option of creating a new version prior to clearing old-style card, column, and punch definitions from the most recent version of the questionnaire definition.

Improved handling of errors by Allocate Cols and Punches. When there is not enough space for a variable (for example, because it requires more columns than are available on the card) Allocate Cols and Punches now sets the column count to the number of columns that the variable requires. This makes it easier to resolve the problem.

Allocate separate card for Codes variables. This option no longer affects the allocation of the coding variables associated with Other Specify categories. For more information, see Card and Column preferences.

New Close command. There is now a Close command on the File menu. You can use this command to close a file without closing the Metadata Model to Quantum window.

The main changes since UNICOM Intelligence Data Model 2.4 are:

Internal changes. The part of the program that writes the card, column, and punch information to the questionnaire definition is now contained within a separate COM component, called the Metadata Model to Quantum Component. This has been done to make it easier to allocate card, column, and punch definitions programmatically: for example, in a DataManagementScript (DMS) file.

Improved error checking. Metadata Model to Quantum now performs better error checking and there are now more possible error codes. For more information, see Metadata Model to Quantum error codes.

Punch code errors. The element grid in the Manual Allocation dialog now has an Error Flag column. When an element has a punch code error, it is now displayed in this column. (You need to use the scroll bar to be able to see the Error Flag column.) For more information, see Allocating card, column, and punch Codes manually.

New shortcut menus. You now get a shortcut menu when you right-click in the Metadata Model to Quantum window or the grid in the Manual Allocation dialog. These new shortcut menus provide handy commands for changing the allocations for several variables or elements at once. For more information, see Metadata Model to Quantum shortcut menu.

Card and Column Preferences. The CardColumns tab in the Preferences dialog has some new options. For more information, see Card and Column preferences.

Quantum Specification Preferences. The Quantum tab in the Preferences dialog has two new options. For more information, see Quantum Specification preferences.

Version Selection. The Select Version option in the Tools menu lets you choose which version of the questionnaire definition you want to work on if you do not want to allocate cards and punches based on the latest version. For more information, see Version Selection. This option replaces the View Version History option which has been removed.

In addition, the documentation of Metadata Model to Quantum has been improved:

Metadata Model to Quantum 2.3 comes with UNICOM Intelligence Data Model 2.4. The main changes since Metadata Model to Quantum 2.0 (which came with UNICOM Intelligence Data Model 2.2 and 2.3) are:

Version handling. Metadata Model to Quantum has been changed to improve its handling of questionnaire definitions that have multiple versions and is now integrated with the Metadata Model Version Utility. For more information, see Version handling in Metadata Model to Quantum.

Allocating punch codes. You can now allocate punch codes to individual categories manually. For more information, see Allocating card, column, and punch Codes manually.

Automatic allocation. The Overwrite Existing Columns and Codes option has been removed from the Preferences dialog and you now specify whether you want to overwrite the existing allocations when you run the Allocate Cols and Punches command. This change has been made to give increased flexibility and to reduce the risk of overwriting allocations inadvertently. In addition the way the command allocates card, column, and punch codes has been improved. For more information, see Allocate Cols and Punches.

Quantum specification. You now define all of the options for the Quantum specification on the new Quantum tab in the Preferences dialog. You can also choose which variables are to be included in the export and the export itself has been improved. For more information, see Export Quantum Specification.

Log file. Metadata Model to Quantum now writes an entry to a log file after each operation that changes the card, column, and punch definitions in the questionnaire definition. For more information, see Metadata Model to Quantum log file.

Data map. The data map feature has been improved and now includes punch code information. For more information, see Using the Data Map.

View menu. Metadata Model to Quantum now has a new View menu, which gives you convenient access to the the log file, the Quantum specs, the data map file, and the questionnaire definition version history. For more information, see Metadata Model to Quantum View menu.

Multilingual studies. In multilingual studies you can now choose the language in which to display the variable texts. If texts are available for use in more than one context, such as interviewing and analysis, you can also choose the context to use. For more information, see General preferences.

Integrity checking. The Check Cols and Punches command has been improved and it now checks the integrity of all punch code definitions for each categorical variable as well as the card and column definitions for all of the variables in the questionnaire definition.

Serial number. The serial number system variable (Respondent.Serial) is no longer shown in the Metadata Model to Quantum window because the serial number is automatically recorded in the serial field on every card. For more information, see Metadata Model to Quantum window.