Clicking the name of the category sends the user to a description of the specified category.

The Data Dictionary provides a short two to three sentence description of each category. However, the user can access a longer description by clicking the ‘Full category documentation’ link at the end of each passage.

The full category documentation provides:

• A description of the policies captured in the category
• A description of the evolution of these policies under AFDC and TANF.
• "Overlap issues" between categories. This information is helpful in two ways. First, this section indicates which, if any, additional categories a user may need in order to understand a policy. For instance, the variables contained in the Income Eligibility Tests category do not, on their own, completely describe a state’s eligibility tests. Variables in the Dollar Amounts category are also required in order to determine which standards are used for each test. Second, in some cases, a user may assume that a policy is captured in one category, but that policy may actually be captured in another category. These types of situations are discussed in the overlap issues section.
• "Special issues" regarding how the category is coded. These details may help the user understand category constructs.

The Data Dictionary also provides a list of variables contained in each category and descriptions of these variables. Below each category description, the variables are listed in the order in which they are coded. In some cases, the order of the variables does not matter, because the variables are independent of each other (for example, In-kind Income variables). However, in other categories, several variables explain each policy (for example, five separate variables are used to describe each earned income disregard). In these cases, a user must know the ordering of the variables to understand the policy.

Documentation, in Appendix B of the Coding Manual, must be read closely in order for the user to understand which variables are need for a complete query. Often, one variable is not sufficient to explain a policy.

The Data Dictionary also provides a response distribution table for each variable. Each table lists the frequency of each response for a selected variable. The response distributions represent the number of times the particular response occurs for all records recorded in the WRD. For a variable response distribution in a specific year, the query feature must be used.

Obtain the counts by clicking the variable name of interest. A table with the responses and counts will appear.

After becoming familiar with the categories and variables, a query can be constructed.

To begin a query:

• Click the ‘Query the Database’ link on the Introduction page. The interface which accesses the data will appear.
• A query is constructed using the following steps.

After choosing a category, the variables contained in the selected category will appear in the ‘Variables’ window.

For category definitions, click on the blue ‘Dictionary’ button above the ‘Categories’ window. A short description of the category will appear.

For a more comprehensive description, click on the ‘Full category documentation’ link and the longer category documentation will appear. (See section IIA for a description of category documentation).

The variables contained in the selected category will appear in the ‘Variables’ window. Using the variable descriptions in the Data Dictionary, as well as the category and variable documentation in the Coding Manual, determine which variable(s) to include in the query.

Up to 40 variables can be selected at a time.

To select a block of consecutive variables, click on the first variable, hold down the [Shift] key, and click on the last. Alternatively, click the first variable, hold down the mouse button and drag the mouse to the last variable.

To select multiple variables that are not in a block, hold down the [Ctrl] key and click on each variable.

The Data Dictionary contains descriptions of the each variable. Click on the blue ‘Dictionary’ button above the ‘Variables’ window and the variables with their descriptions will appear. (See section IIB for more information on variable descriptions ).

In addition to providing definitions, the ‘Variables’ window dictionary also provides distribution counts for the responses of each variable.

Click on the individual variables within the dictionary and the responses with distribution counts will appear. (See section IIC for a description of distribution counts).

Determine which state(s) to include in the query. A query can search for data in one state, a selected group of states or all states.

To select a block of consecutive states, click on the first state, hold down the [Shift] key, and click on the last. Alternatively, click the first state, hold down the mouse button and drag the mouse to the last state.

To select multiple states that are not in a block, hold down the [Ctrl] key and click on each state.

To select all states, click the ‘All’ box above the ‘States’ window.

Determine which year(s) to include in the query. A query can search for data in one year, multiple years or all years.

To select a block of consecutive years, click on the first year, hold down the [Shift] key, and click on the last. Alternatively, click the first year, hold down the mouse button and drag the mouse to the last year.

To select multiple years which are not in a block, hold down the [Ctrl] key and click on each item.

To select all years, click the ‘All’ box above the ‘Years’ window.

Determine whether to query all records or only the majority rule records. The interface can search the database by ‘Majority only’ or ‘All records’.

To retrieve only the information that affected a majority of the caseload for most of the year, click the ‘Majority only’ box.

If, however, the user wants to retrieve data on all policies in effect in a state, click the ‘All records’ box. An ‘All records’ search can produce several records for each year and state, because some states have several different policies in effect at any one point in time depending on the unit type or geographic area in which the unit resides, or have different policies in effect at different times during the year.

For more information on coverage, see the Coding Manual. There is an extensive discussion of majority rule and coverage variation.

Data can be retrieved in one of the following five formats:

1. Short Table

The short table displays data in the browser window as a state/year crosstab. Only majority rule records are displayed and no header variables are included. For more information on header variables, see the Coding Manual.

Note: Without the header variables, it is impossible to determine whether the rules apply to all assistance unit types or only a selected group of units, such as recipients only or two-parent families. In addition, the short table does not provide any information regarding the geographic area or percentage of the caseload covered by the given rules. Finally, majority rule records are not necessarily consistent across categories. The majority rule record in one category may not describe the same population or time period as the majority rule record in another category. For example, the majority rule record in a given year for Income Eligibility Tests may apply to recipients from July to December. However, in the same year, the Dollar Amounts majority rule record may apply to all units in only 20 counties from January to December.

2. Full Table

The full table displays all selected variables as well as the header variables in the browser window. This format can be used for ‘Majority only’ or ‘All records’ queries. Displaying the header records is often necessary in order to fully understand which assistance units are affected by policies in the state. For more information on header variables, see the Coding Manual.

3. Distribution Counts

The distribution counts table calculates and displays the frequency of each response to a given variable. Distributions by state, by year, and by state/year combinations are calculated. If more than one variable is selected, distribution counts will be shown only for the first.

4. Tab-delimited

The tab-delimited format saves the data to a text file, with each value separated from other values by a tab. This format can be downloaded directly into a spreadsheet, such as Microsoft Excel or Corel Quattro Pro. Notes associated with a particular variable, if any, will be included with that variable's value, separated by a space.

5. Comma-delimited

The comma-delimited format saves the data to a text file, with each value separated from other values by a comma. This is the most common file format, and can be imported into almost any application. Notes associated with a particular variable, if any, will be included with that variable's value, separated by a space.

After constructing a query, click the ‘Display Data’ button. The query will begin to process. Depending on the amount of data requested, this process may take a few seconds to several minutes.

When the data appears, all variable notes, if they exist, will appear next to the variable. In addition, if the record contains any Additional Notes, they will appear at the end of the variable list under the heading Notes.

If the tab-delimited or comma-delimited options are chosen, a dialog box entitled ‘File Download’ will appear. The user is given the option to open the file with a program or save it to disk. Select the option ‘Save this file to disk’ and click ‘ Ok’. Select the preferred spreadsheet package. Follow the file import procedures for the selected software.

If questions or problems arise while constructing a query, the help buttons above each window provide a quick guide.

If the tables are too wide (i.e. more variables are selected than are visible on the screen), they will not print directly from a web browser. Only the portion of the table that is visible on the ‘Selected Rules’ page (without scrolling across the page) will print. This is an internet software problem that cannot be avoided. Therefore, if the user wants data for several variables, constructing more than one query with fewer variables will produce printable tables. Alternatively, the data can be retrieved in a comma or tab-delimited file and displayed in an spreadsheet package.

To print the interface screens, go to ‘File’ then ‘Print’. A ‘Print’ dialog box appears. At the bottom of the box, a section entitled ‘Print frames’ appears. Click the ‘As laid out on screen’ option and click ‘Ok’. This prints all of the information as seen on the screen.