SAS Base Programming key terms

A short nickname (up to eight characters) that you assign to a SAS library so that programs can refer to the folder or database where data sets are stored without writing the full path each time.

A statement that associates a libref with a physical location, such as a folder of files, so SAS can read and write data sets stored there for the rest of the session.

A file that SAS creates and reads, containing data organized into rows (observations) and columns (variables), plus descriptive information about the file such as variable names, types, and labels.

The area of memory where SAS builds one observation at a time during DATA step execution; it holds all variables, including ones created in the step, and its contents are written out when the step reaches the end of an iteration.

The properties that describe a variable in a data set: its name, type (character or numeric), length in bytes, label, format, informat, and position. Changing them changes how values are stored or displayed.

A statement that tells SAS how to read raw character values and convert them into stored values, for example turning the text 01/15/2020 into a numeric date. Informats can be assigned in a DATA step or with an ATTRIB statement.

A DATA step statement used to set one or more variable attributes, such as label, format, informat, and length, in a single place before the variables are read or created.

Instructions written in parentheses after a data set name that change how SAS reads or writes that data set for one step, such as KEEP=, DROP=, RENAME=, WHERE=, OBS=, and FIRSTOBS=.

A procedure that reads an external file, such as a delimited text file or an Excel workbook, determines its structure, and creates a SAS data set from it, optionally saving the generated code for reuse.

A DATA step statement that reads values from an external file or from instream data lines and assigns them to variables, using list, column, formatted, or named input.

A DATA step statement that reads observations from one or more existing SAS data sets, one observation per iteration, so the step can modify, filter, or create new variables from existing data.

A DATA step statement that combines observations from two or more SAS data sets side by side into one output observation, matching rows when a BY statement is used and requiring the input data sets to be sorted first.

A numeric value equal to the number of days between January 1, 1960, and a given date; SAS stores dates this way, and a date constant such as '01JAN2020'd lets you write one in code.

A statement in the DATA step that tests a condition and, when the condition is false, stops processing the current observation so that it is not written to the output data set.

A statement that selects observations before they enter the DATA step or procedure, so only rows that meet the condition are read and processed.

A statement that tells SAS to process data in groups defined by the values of one or more variables, which must be sorted or indexed first; it enables group-level processing and BY-line output in procedures.

A DATA step statement that keeps a variable's value from one iteration of the DATA step to the next instead of resetting it to missing at the start of each iteration.

A statement of the form variable + expression that adds a value to a variable and automatically retains that variable across iterations, treating missing values as zero and initializing the variable to zero.

A group of statements executed repeatedly a set number of times, controlled by DO index = start TO stop BY increment, often used with an array to process several variables in one pass.

A procedure that orders a data set's observations by one or more variables; the NODUPKEY option removes observations that have duplicate values of the BY variables, keeping the first occurrence.

A custom label mapping created with PROC FORMAT and a VALUE statement that groups or relabels stored values for display or analysis without changing the underlying data values.

Temporary variables that SAS creates during BY-group processing; FIRST.variable equals 1 for the first observation in a group and LAST.variable equals 1 for the last, and both are 0 otherwise.

A function that returns the value of its argument from a previous observation during DATA step execution, so a step can compare the current row with an earlier one.

A group of SAS variables referenced by a single name and an index; arrays let you process many variables with a loop, although they store no data themselves and exist only while the DATA step runs.

Built-in functions such as SUBSTR, SCAN, TRIM, STRIP, UPCASE, and CATX that extract, search, combine, or change the case of character values, often used to clean text before storing or reporting it.

A temporary variable that SAS sets to 1 when an execution-time error occurs during a DATA step iteration and to 0 otherwise; writing it to a data set helps you locate records that caused problems.

A temporary variable that counts how many times the DATA step has iterated, starting at 1 for the first iteration; it is often used to number rows or to limit processing to specific observations.

A problem SAS detects while translating a step before any data is read, such as a misspelled keyword, unbalanced quotes, or an unknown function; the step does not execute and error messages appear in the log.

A problem that occurs while a step is running, such as reading a character value into a numeric variable or dividing by zero; SAS sets ERROR to 1 and handles the problem according to the error and options in effect.

A statement that writes a message, including variable values, to the SAS log during execution, which is useful for tracing program logic and diagnosing why a DATA step produces unexpected results.

A system option that sets the maximum number of observations for which SAS writes error messages to the log; 0 means no limit, and a small number helps you scan a long log.

A procedure that displays the observations of a SAS data set as a simple listing, with optional LABEL, BY, and SUM statements and data set options controlling which rows and columns appear.

A procedure that produces detailed or summary reports with columns you define, supporting grouping, statistics, computed columns, and the ability to send the report to a SAS data set instead of a listing.

A PROC REPORT statement that specifies how a report column is used and displayed, setting options such as usage (group, order, across, analysis), a display label, a format, and a statistic.

A section in PROC REPORT, begun by COMPUTE and ended by ENDCOMP, where you write DATA step statements to calculate new values or apply conditional formatting for a column or for the whole report.

The part of SAS that takes procedure output and routes it to one or more destinations, controlling the form of the output, such as HTML, PDF, RTF, or a SAS data set, and the style applied.

A specific output target opened with an ODS statement, for example ODS HTML, ODS PDF, or ODS LISTING; destinations can be open at the same time and each receives the output sent to it.

Statements that control which procedure output objects are sent to the open destinations during a session, letting you suppress or keep specific tables and reports.

A procedure that builds frequency and cross-tabulation tables for character or numeric variables, showing counts, percentages, and optionally chi-square statistics; the TABLES statement lists the variables to tabulate.

A procedure that produces summary tables in which class variables define rows and columns and analysis variables are summarized by statistics, using a TABLE statement to describe the table layout.