|
People | Business | Geography | Newsroom | Subjects A to Z | Search@Census |
| data: (da • ta ) n. a collection of facts from which conclusions may be drawn TheDataWeb is a site to support and promote organizational and multi-agency collaborative efforts to enhance the DataFerrett project |
|
INSIDE TheDataWeb:
DataFerrett Page |
|
User Guide home Back Next
Chapter 11 - Step 2: DownloadIn Download we will explore the download capability of DataFerrett using the example databasket from the previous Chapter (9). Remember to save your databasket as a ferrett session file (.fsf).
DataFerrett gives you the option of extracting the data in your databasket or display a number of rows from the databasket on your screen.
To display the data, you have the choice of viewing 50 rows, 100 rows or 200 rows of the raw data. This is an ASCII type display with each variable in a column and the values for each row of data. The view will be offered via a link to an html page:
To save/download this data you will use the right mouse button on the file link. The right mouse click in Internet Explorer it will say "Save Target As: The right mouse click in Netscape will say: Save Link As:" which will download your file(s) to your computer. For this example, the default will always be using Internet Explorer. DO NOT save to favorites or bookmark. The files that save your extraction are periodically cleared out of the cache of the DataFerrett system. Save the file to your own computer.
Choosing the Download Data file formatDataFerrett offers several standard statistical package formats for the user to download. Choosing your download format will be entirely dependent upon your preferences for after-download file manipulation. After making a file format selection, select the Get Extract button to get your data downloaded to your computer. The images above are indicative of the kinds of windows that will come up for you to make choices to before the download process is complete. Below you will see the files that return with your data. Some of them will return with an additional statistical package command file.
Running your Request in Batch Mode This allows you to run the job in batch and get the results later instead of waiting for the results. You will be sent an email message with a link to your data file when the job is complete. You might want to use this option if you are trying to extract a very large number of variables. If you choose this option a window pops up telling you the url to check later for your results. This is where the correct input of your login email address is important. Please do not insert false or incorrect email addresses to login to DataFerrett.
The following may or may not be necessary with the status of you computer's memory. Also, DataFerrett has restricted the download to 50 variables only at one time. But, there are two compression methods available if the user feels more comfortable with this format: zip and gzip.
Copy the link to check later or wait for the email, as shown below.
Working with SIPP Extractions using SASThe instructions included on the DataFerrett/SIPP CD give guidance to the user about how to save a DataFerrett extract as a SAS file. The recommended approach is to save the extract as a SAS file, close DataFerrett, launch SAS, and then import the saved extract into SAS for execution. This approach will minimize the likelihood that the user's computer might exhaust its memory resources. Some users may wish to provide a more automatic invocation of SAS by associating the file suffix of the DataFerrett extract with a particular version of SAS existing on their computers. It is important to note that SIPP extractions from DataFerrett should be processed in SAS version 8.x or higher. Earlier versions of SAS will not properly handle certain features of SIPP data, such as the length of labels. If you wish to create such an association to enable automatic invocation of SAS, please check with your LAN Support specialist regarding how to do so in the most appropriate way in your local environment. |
||
| Back Next
Last update: 3/12/10 |
||||