Summary Files published on CD-ROM and DVD by the Bureau of the Census come with powerful extraction software called Allocate. Using the software is often the easiest way to do large extractions of data from the Census 2000 Summary Files.
Allocate is available on the PC network in Electronic Data Service, 215 Lehman Library. You must visit EDS to use these products. Currently it can be used for extracting data from Summary File 1 (entire nation) and Summary File 3 for New York State.
This guide provides details on using the Allocate software in EDS.
- Getting Started in EDS
Use these instructions to find and start the software in EDS. - Workspace
Select a Summary File and optionally save searches or retrieve previous searches. - Pick Geography
Select the specific geographic areas in which you are interested. - Output
Select the format for output. - Variables
Select among all available variables that are arranged on the files in tables. - Retrieving Output
Getting the output you requested.
Getting Started in EDS
If you do not already have work space set up in My Documents, where userid is your Uni, create a work folder on the PC using the EDS WorkSpace icon on the desktop.
Allocate is located in the Start Menu under, Programs -> EDS -> Censu -> Portfolio Allocate or in the Applications folder called Portflio Allocate.
Workspace
When the application window opens, the "Workspace" screen appears. It is one of four work areas (each identified with a tab )used in the selection process. The "Output variables" area, to the right of the tabbed areas, remains visible throughout the selection.
The first dialog box in the "Workspace" screen, labelled "Current Database Set" shows which Summary File that will be used, in this example, Summary File 1. The one that shows is the default. To change the default, pull down the menu under "Current Data Set" and make another selection. A change in the selection takes place by the opening of another Allocate application in a new window.
The "Current Workspace File:" area is used for saving a search or opening one that was previously saved.
Move to the next step by either picking another tab or clicking on the Next button.
Pick Geography
Under the "Pick Geography" tab you are presented with all the available geographic levels in the scroll window on the left. Drill down through this list to find the lowest level of geography for which you want data.
For example, if you want tract data, the level to start with is 140 (tracts within counties) or 080 (tracts within places).
|
|
To choose a few tracts from within a county:
|
Data for Alabama state and Autauga county will not be returned as the shaded check marks indicate only that a sub-geography has been selected. When drilling down in either level 040 or 080, only data at the tract level (lowest level of geography in each of these hierarchies) will be returned.
|
|
To choose all tracts from within a county:
|
To get data for Autauga County, you must also drill down the summary level 050 items to locate and select Autauga County. Hierarchies within a level can be closed by clicking a - (minus sign) without losing your selections and thus enabling you to make further selections in other geographic levels.
Move to the next step by either picking another tab or clicking on the Next button.
Output
You are offered three types of output: File, Report, and Index Base. The File option is the default. Use it when you are making a selection of variables to be retrieved from a Summary File. This option is the only one described in this guide. The Report and Index Base options are not currently supported in EDS. The File option creates a data file. It is important that you correctly set up the path and file name to use since the program always writes out files as the finishing step.
- File option allows for you to:
- pick individual data tables or variables from within a table at any geographic level for which the data is reported;
- pick from among a variety of output file formats: xls .csv .dbf .htm .mdb .db.
To use this option proceed as follows.- Click the radio button next to the word File.
- Double click on the save as button to activate the Windows' file-save-as feature.
- Choose the path My Documents/userid (as specified in the Getting Started step.), and give the file name you want for your output.
- Choose the desired file format using the Save as type pull down menu.
- Clicking save in the Save As window will return you to the "Output" window.
Move to the next step by either picking another tab or clicking on the Next button. - Note, that you do have the option of "CREATE OUTPUT AS SUMMARY," which will report the data as totals for each variable across all the detailed areas selected. In the above illustration where three tracts were selected, choosing this feature would return totals for the area covered by the three tracts, not separate counts for each tract. This would be useful if you wanted data for a neighborhood that encompassed those three tracts.
Variables
Under the "Variables" tab, you are presented with all the available variables in the scroll window on the left. The variables are arranged in folders that reflect the way they are found on the files. Drill down through the folders using the + (plus sign) and - (minus sign) icons in same way they are used to navigate the hierarchy of geographies. If the truncated variable names that appear in the Input Variables window are difficult to understand, hold your mouse over the description and a dialog box with the name for the variable will appear.
To select a variable, highlight it and either double click on it, or click on the Select button. This will cause it to appear in the Output Variables box on the right. In the above example selections were made as follows.
- Opened the folder Geographic Identifiers (the first one in the list), selected Census Tracts. This will return the six digit tract number.
- Closed the Geographic Identifiers folder and opened the P Tables folder.
- Scrolled down to the URBAN AND RURAL folder, and, without opening it, selected the folder thus moving all the variables it contained to the Output Variables box.
- Opened the P3. RACE folder and selected only White Alone.
Move to the next step by either picking another tab or clicking on the Next button.
To remove variables from the Output Variables box, highlight the variable with a single mouse click and, with the variable highlighted, click on the "Delete" button. Clicking the Delete button when no variable is highlight will remove all variables from the output list.
Retrieving the Output
The instructions here apply to the last step which occurs after all your selections and at the point when you click the Finish button. Clicking on the Finish button will write out files. The following window will appear. This window gives you the choice of viewing the files: a data file in the format you selected in the Output step, and a documentation file in ASCII text format that describes the output. You can choose not to view your output by just closing the window.
For the documentation file, the default application
is likely to be a Notepad. The format of the file is comma delimited
and it contains the variable names as they appear on the data file
matched with a textual description for that variable. Using this
example, the text description assigned to variable
P003003 (column J in the table shown below) would be
P Tables 2000 P 3. RACE [71] P003003 White alone.
For the data file, the default application will depend on what format you chose in the Output step. The csv in this example opened in Excel but it can also be read as a text file in SAS or SPSS.
- The column TRACT matches the variable "Census Tract" which was selected from the Geographic Identifiers folder. Values returned from the variables in the Geographic Identifiers folder are the codes assigned within a single geographic level.
- Note that the first two columns, KEY and NAME do not match to any variable that was selected. These are automatically generated.
- KEY is the combination of codes from all levels in the hierarchy of geography that together uniquely defines a selected geographic area. Since we chose level "140--Census Tract by County," the geocode values in this example should be a string of numbers made up of: state code, county code, tract code.
