AM Statistical Software > Manual > Getting Started > Retrieving Data

Table Of Contents

Manual
- Getting Started
  - Starting the Program
  - Retrieving Data
  - Manipulating Data
  - Saving and Re-running Actions
    - Reviewing and accessing the code
    - Saving and re-running statistical models
- Sampling
- Procedures
  - Measurement Models
  - MML Models for Test Data
  - Other Available Procedures
- Graphics
- Tools
  - Estimation Methods
  - Optimization Techniques
  - Variance Estimation
- Post-hoc Procedures
- More user input instructions
  - The User Interface
    - Completed Run Queue
    - The Variable List
  - Input Instructions
  - Options
  - Output Precision
- Glossary of Terms and Symbols

General Import/Export

AM can directly import/export data from/to a large number of databases directly as well as via ODBC. This functionality requires AM's DFTransfer component.

If DFTransfer is not installed, you can get it from http://am.air.org.

You can access the general import/export component from the file menu (File|Import|General Import.). If the component is installed, this selection will bring up a dialog box that will enable you to import your data.

If you are going to use the MML procedures, you may still require a dictionary file after you import data. The MML procedures for test data require that tests and item parameters be specified (see Importing AM text for more details). In these cases, your dictionary need only include the data that is not included in the dataset header.Specifically, the dictionary must contain information about the tests, such as item parameters and the arrangement of items into subtests. After the initial import, select File|Update Metadata from the File menu to read in the dictionary.

The DFTransfer component is built on the import/export engine from DataFlux(tm), which we have licensed for distribution and distribute with AM. The component comes with it's own help system, so press the help button on the dialog box. A few important things to keep in mind:

Data files must have their standard extensions, or the system gets confused
Many different file types have the same extension (the extension is the set of letters following the last dot in the file name). It is your responsibility to make sure that you specify the right type of file for the data you are trying to import. You specify the type of file using the drop-down list on the file dialog.
ODBC is a generic access method for databases. If you don't know what it is, you probably won't be using it. It provides a software-independent way to access local or remote data.
Because AM does not currently support character-type variables, character variables in the imported dataset will be skipped.
If you are importing or exporting to/from SAS, you need to read the paragraphs below.
The statistical package names are all the property of the companies who make them

Attention SAS(tm) users:

If you want to preserve your SAS value labels (formats), you have to save them as a SAS dataset first. You can do this by running the following line in SAS:

proc format data=XXX cntlout=yyy;

where xxx is your SAS dataset and yyy is the new dataset to be created. When importing, you will be prompted to specify which dataset contains the formats.

When exporting data to SAS, the value labels (formats) will be output in a sas program named zzz.pgm, where zzz is the name of the file to which you exported. You should use this program to create the formats prior to reading the SAS file. if you choose not to use the formats, you will have to add the following system option to your SAS program:

options nofmterr;