Policy Analysis Tool Tutorial

Click the following links to download this tutorial:    MS Word     PDF

Basic Analysis Example

Step 1: Select the dataset(s) to search

	Select which datasets that you would like to analyze. Selecting two or more datasets will allow you to compare attention to the selected topic(s) in different venues. Each dataset has various filters which you may select to narrow the search. The Congressional Hearings dataset (1946-2004) allows users to include or exclude: legislative hearings, appropriations hearings, agency or program creation hearings or hearings regarding a specific issue proposed by the President. The default setting is no filters and you may select as many as you prefer. We have selected Hearings (no filters) and U.S. Public Laws (only those introduced by a Democrat). The Congressional Quarterly Almanac dataset does not have any filters at this time.
	The New York Time Index (1946-2002) includes a sample of all New York Times stories. This dataset can be filtered by region (NY City, NY Region, or National), Scope (Domestic, International-US involved, or International-US not mentioned) and by the actors involved (federal agency, President, Congress, courts, campaigns, or interest groups). You can choose to include or exclude these cases.
	Cases in the statutes database can be filtered on the basis of commemorative status, whether the law was vetoed by the President, which party the law was introduced by, and whether the law was discussed in Congressional Quarterly. Users can utilize as many of these filters as necessary for their analysis. There are no filters at this time for the Executive Orders or Gallup's Most Important Problem dataset at this time. For more information about each dataset, click on the link "abstract" located next to the name of each dataset.

Step 2: Select topic(s)

	The Policy Agendas Project's includes 19 major topic categories and over 200 subtopic categories. Users can select any number of these by clicking on the topic of interest. To select multiple topics/subtopics, hold the ctrl key (Apple Command key for Mac users) down while using the mouse to choose the desired topic. To de-select a subtopic, press the ctrl key while clicking the highlighted topics. To select all subtopics within a major topic category, choose the "All subtopics in . . ." Here we have selected the major topics Education and Health.
	If the given topics do not seem to cover an issue that you are interested in, select the link "Search for the best topic" and enter a key word into the search box that appears. This search will identify the major topics most related to the key word. Note: due to the size of the Policy Agendas Project database, the search process will take a few minutes. Please be patient.

Step 3: Using the keyword filter

	The keyword filter allows users to identify those records in a dataset that contain a word of interest. You may choose to search only those topics you have selected or to search the entire database for the search term. The final chart and drill down table will display those records which contain the search term and provide original source information.

Step 4: Select the timeframe

	Decide whether to analyze the data by year or by Congress. It is important to make this selection, otherwise the default is to analyze by year. The database ranges from 1946 through 2004 (or the 79 th through 108 th Congress), though this varies by dataset (for example, Congressional Hearings run from 1946-2004 while CQ Almanac from 1948-2002). See dataset descriptions for details.

Step 5: Select the Unit of Analysis

Select whether you would like to see the data in terms of raw counts of instances or as a percent of all topics per dataset. (For example, this will tell you the percent of Hearings that were about Health issues, or the percent of New York Times stories that were about Environmental issues). In this example we'll view the results as a percent of all topics in each venue.

Step 6: Submit Query and View Results

	To view the basic results, simply select the "Search" button. If you would like to start over, press the "Reset Form" button, which will remove your selections. This will produce a chart displaying the results.
	You can change the features of this chart and customize the display by right-clicking on the chart and selecting properties (see below for an example and some basic instructions).

Customizing Chart Display

This chart was created from the original via the following steps: Right-click in the chart area. In the dialog box that appears, select "properties." In the properties box that appears, select the "General" tab. Click on the "Palette" drop box, select "gray scale" from the options. Click "Apply" from the buttons below. Select the "Series" tab. Click the button next to the word "Gallery" and click on the picture with the two lines (first picture in the upper left corner). Click "Apply." To remove the shapes on the lines, click on the picture next to the word "Shape" and clicking on the blank area. Click "Apply." Go to the drop down box that says "All Series." Click on that box, and select one of the series names (in this case, Health Senate). Click the drop down box next to the word color (in the fill box), to change the color of the line for the selected series. To change the thickness of the line, click on the drop down box next to the word "Weight." To get a dotted line, click on the drop down box next to the word "Style". Click "Apply" after each change. Repeat steps 5 through 7 for each series. When finished, click "OK." To move the legend box, right-click in the legend box and select "Fixed" from the appearing dialog box options. Then click in the now-smaller legend box and drag it (click-and-hold the mouse button while moving the mouse) anywhere on the graph. For information on customizing the chart display see the Chart Options Tutorial.

Drill Down Table & Finding Original Source Information

	Below the chart is a table showing the values for each year or congressional session.
	Clicking on the numbers in this table will display a description of each individual hearing, New York Times story, CQ Story, or Public Law for that year or congressional session. See In addition to the description, there is also information about the original source. Each data set contains original source information to enable users to find out more about a specific case (or record) in the data set. Click here for more information about finding and accessing original sources.

Downloading Filtered Datasets

Users may wish to download specific components of our datasets. Using the policy analysis tool, it is easy to select only those datasets and topics for which you're interested in. Follow the below steps for help downloading the filtered datasets.

Select datasets and topics of interest

	To begin, select the datasets you would like to use. Here we have selected Congressional Hearings, CQ Almanac, NY Times Index and US Public Laws.
	Next, select the major topics, the years or congresses of interest, and the unit of analysis. We have selected Social Welfare policy and the default years, 1946-2004.

Downloading the filtered datasets

	A chart based on the criteria you selected will appear. At the top of the page you will see the heading "Download Filtered Datasets" followed by a listing of the filtered datasets available for download. These datasets will only include the major topics and the years or congressional sessions you selected.
	To download one of these datasets into an Excel spreadsheet, click on one of the links. The text file should appear in a new browser window. Go to File, Save As and enter a name for the file. Then click Save. The text file is now saved to your computer. These files are tag delimited and can be imported into most statistical software packages. See your software manual's help section for more information on this feature.

	There are two ways to import data into Excel. The following steps will give you instructions for both First, go to the directory where you saved the text file, probably My Documents or the desktop. Right click on the file and select Open With and Choose Program.

	A dialogue box will appear. Select Microsoft Office Excel, then click OK. An Excel Spreadsheet will appear with the data automatically imported into the proper columns and rows.
	Next, go to File, Save As. A dialogue box will appear. Name the file and select "XML Spreadsheet (*.xml)" then click Save. Now the tab delimited text file has been converted into an Excel spreadsheet and is ready for use.
	The second method is demonstrated in the following steps. Open Microsoft Office Excel by going to Start, Programs, Microsoft Office, and then clicking on Excel.

	Once Excel is open, go to File, Open. The Open File dialogue box appears. Under the drop down box "Files of Type" select "Text Files (*.pm*.txt*.csv)". Now select the text file you saved earlier and then click Open.
	The Text Import Wizard dialogue box will appear. Select the delimited option under "Choose the file type that best describes you data", then click Next.

	Choose tab as the delimiter. Excel will preview the data the box below. Then click Finish.
	The text data file has now been successfully converted into an Excel spreadsheet. Congratulations!

Finding Original Source Information

A Note on Finding Original Sources: These sources can be found in the government reference section of your local library. Another useful resource to consult is Thomas, the Library of Congress's online database of congressional information.   Here you can find information on bills (93rd - 108th) and public laws (93rd - 108th).

Step 1: Selecting the options for query

Choose the dataset(s), policy areas, unit of time and analysis. Then click the Search As Is button. We have chosen the Congressional Hearings, CQ Almanac, NY Times Index, and US Public Laws; the policy topic Energy; years (1970-1999); and activity graphed as the number of cases per year. For help with any of these features see the Basic Analysis Example.

Step 2: The Drilldown Table

A list of the filtered datasets available for download will appear, below which is the chart created based upon our criteria. The drilldown table is located just below the chart and displays the values for each year or congress.

Step 3: Finding the Original Source Information

To find the original source information for one selection, click on a value. A window will appear that displays all the items for the year or congress and the dataset you selected. This table displays a description of each individual hearing, New York Times story, CQ story or Public Law for that year or congressional session. Each description is attached to a source code. The source code is different for each dataset. Read on to decode the source information for each of our datasets.

Step 4: Interpreting the source code

	The hearings source code will look different depending upon the year(s) you selected. For hearings prior to 1970, the source code will look something like this: CIS years6769 91 S2045-3. The first component, "CIS" signifies the source, which is the Congressional Information Service. The second component,, "years6769", tells you the years contained within the CIS book in question. The third component, "91" is the congressional session. The fourth component, "S2045" signifies the chamber (S for Senate, H for House, J for Joint) and the CIS committee number. The last component is the hearing sequence number. For hearings after 1970, the source code will look similar to the following: CIS Abstracts 1970-J801-4. The first part, "CIS Abstracts" signifies the source. The second part, "1970" tells you the year of the hearings. The third part, "J801" displays the chamber and the CIS committee number. Finally , the last part signifies the hearing sequence number. Full text of congressional hearings are available online at GPO Access from the 104th to the 109th Congress. For hearings in past congressional sessions, you may either visit your local library's government publications section or consult Lexis-Nexis. To find full text information from Lexis-Nexis, click on the link "Search by Number" under Congressional Publications. Once there, select the option "Bibliographic Number" in the search window and change the default "SuDoc" to "Accession Number." Then copy and paste the source information from the drill down table to the search box. Due to formatting changes in Lexis-Nexis, users will now have to edit the hearing number slightly. For example, if you cut and paste "2002-H-161-1" from the drilldown table, change it to "2002-H161-1." If you cut and paste "CIS Abstracts 1999-H181-58" change it to "99-H181-58." Note: Years before 2000 only have two digits.
	Congressional Quarterly stories source code will look something like this: CQ Almanac 91 p. 410. This conveys the source, which is Congressional Quarterly Almanac for the 91st congressional session, page 410. To access the full text of Congressional Quarterly stories, visit the government publications section of your local library.

	New York Times stories source code will appear similar to this: NYTIndex 1968 p. 1269. This reveals the source, which is NY Time Index for 1969, page 1269.

Public Laws source code will look similar to the following: P.L. 92307. This is the public law number, the first two or three digits of which are the congressional session. Public Law information is available from the GPO Access from the 104th (1995-1996) Congress to the 109th (2005-2006). Prior to that, final bill summary and status is available through THOMAS. We are currently working on a project to dynamically link these external resources to the datasets available on this website. Check back soon!