Which statement is used to read data in a DATA step?

The statement used to read data in a DATA step is the INPUT statement. The INPUT statement is essential for specifying how the data should be read into SAS variables. It allows you to define the names and formats of the variables directly from the raw data, whether they are contained in an external file or entered directly within the DATA step.

By using the INPUT statement, you can control the parsing of data lines, which is crucial for ensuring that the data are accurately interpreted and stored in the appropriate format in your SAS dataset. This flexibility is vital when working with various types of data sources, as it allows customization based on the data structure.

The other statements mentioned, while important in SAS programming, do not serve the primary role of reading data into a DATA step. The OUTPUT statement, for instance, is used to write observations to a dataset, the SET statement is used for accessing existing datasets (not for reading raw data), and the DATA statement is necessary for initiating a DATA step but does not itself read data.