Class 2 Lesson Plan

1
Class 2 Lesson Plan

• Syntax Basics
• Statements
• Comments
• Variables Variable Types
• Constants
• Expressions
• Introduction to Reading Data
• DATA Statement
• INPUT Statement
• SET Statement

2
Class 2 Lesson Plan

• Debugging and Editing
• Error messages
• Correcting errors
• Resolving common errors

3
Syntax Basics - Statements

• Smallest unit of work in SAS.
• SAS is not case sensitive.
• All statements MUST end with a semicolon ().
• SAS ignores spaces, carriage returns, or extra
  lines. All statements end with a semicolon.

4
Syntax Basics
Always end with a semicolon
Usually begins with a SAS keyword
Can begin anywhere on a line
Can continue on several lines (or continue on one
line)
Can skip lines
Use blanks or special characters to separate words
Can use more than one blank space
5
Syntax Basics - Statements

• Examples of different programming styles
• data oneinfile c\new.rawinput case 1-10
  age 11-12
• data one
• infile c\new.raw
• input case 1-10 age 11-12
• data one
• infile c\new.raw
• input case 1-10
• age 11-12

6
Syntax Basics - Statements

• RUN
• Tells SAS that a DATA or PROC step is over, and
  needs to be run.
• Its usually needed at the end of each DATA or
  PROC step, or else youll confuse poor, stupid
  SAS.
• Rule of thumb End every DATA or PROC step with
  RUN.

7
Syntax Basics - Statements

• data one
• infile c\new.raw
• input case 1-10
• age 11-12
• run
• proc print dataone
• run
• proc means dataone
• var age
• run

8
Syntax Basics - Statements

• ENDSAS
• Same as RUN, but also ends the SAS session (i.e.
  quits SAS and sends you back into the operating
  system).
• Ive never used this statement it has limited
  usefulness with SAS for Windows.
• It may be more useful for SAS on mainframes.

9
Syntax Basics - Comments

• Comments are not read by SAS, so you can write
  anything without having SAS tell you that you
  screwed up.
• Its good practice to use comments extensively to
  organize your program for yourself and others.
• There are 2 ways to identify comments.

10
Syntax Basics - Comments

• Use /./
• / text of comment /
• Use .
• text of comment

11
Syntax Basics - Comments

• Examples of comments Headers
• /
  
• Sample SAS Header
• First Created August 31, 2003
• By Han S. Kim
• Located c/sasfiles/class2
• Modification history
• Person Date Changes
• HSK 09/01/2003 added race
• 
  /

12
Syntax Basics - Comments

• Describe lines in the DATA step
• DATA temp
• infile c/sasfiles/class2/exam.txt
• Read variables sampid, case, sex, age
• input sampid case sex age
• run

13
Syntax Basics - Comments

• Or you can
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case sex age Read variables
• run

14
Syntax Basics - Comments

• But dont
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case Read variables
• sex age sampid, case, sex, age
• run

15
Syntax Basics - Comments

• Describe PROC steps
• Read variables sampid, case, sex, age into
• Datafile temp
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case sex age
• run
• Conduct univariate analysis of DATA temp on
  age
• PROC univariate datatemp
• var age
• run

16
Syntax Basics - Comments

• Or comment out statements during program
  development
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case sex age
• run
• /PROC univariate datatemp
• var age
• run/

17
Syntax Basics - Comments

• But be aware of the differences in the 2 comment
  methods
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case sex age
• run
• PROC univariate datatemp
• var age
• run

18
Syntax Basics - Comments

• But be aware of the differences in the 2 comment
  methods
• DATA temp
• infile c/sasfiles/class2/exam.txt
• input sampid case sex age
• run
• PROC univariate datatemp
• var age
• run

19
Syntax Basics - Variables

• Variables are the names of the columns
• Each observation, or row, has a value for each
  variable.
• Examples of variables sample ID, age, sex,
  smoking status.
• Examples of values for variables 232112, 34,
  male, never.

20
Syntax Basics - Variables

• Some rules
• Variable names must be 32 or fewer characters in
  length (SAS v8 or higher)
• Names must start with an underscore or letter,
  NOT A NUMBER
• Names must contain only underscores, letters and
  numbers, no !_at_.
• Names are case insensitive

21
Syntax Basics - Variables

• Variable Series
• Be lazy! Save coding time!
• Shorthand way to identify a large number of
  variables
• The order that SAS reads the variables is
  important!
• There are seven List Styles

22
Syntax Basics Variables

• Example program
• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2

23
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• VARn VARx
• x1-x3 refers to the variables x1, x2 and x3
• x1-x4 refers to the variables x1, x2, x3 and x4
• x4-x1 gives an error
• x01-x3 gives an error

24
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• VARn - VARx
• x1--x3 refers to x1, x2 and x3
• x1--x4 refers to x1, x2, x3, name, type, y, z
  and x4
• x3--x4 refers to x3, name, type, y, z and x4

25
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• VARa numeric- VARb
• x1-numeric-x4 refers to x1, x2, x3, y, z and x4
• namenumeric-sum1_3 refers to y, z, x4 and
  sum1_3

26
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• VARa character- VARb
• x3-character-y refers to name and type
• name-character-x4 refers to name and type
• x1-x3 gives an error

27
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• _character_
• Refers to all character variables name and type.

28
Syntax Basics - Variables

• input x1 x2 x3 name type y z x4
• sum1_3 x1 x2 x3
• x4 x4 / 2
• _numeric_
• Refers to all numeric variables x1, x2, x3, y,
  z, x4 and sum1_3.

29
Syntax Basics - Variables

• How do we use variable lists?
• Inputting data
• data temp
• infile c/sasdata/class2/data.txt
• input sampid multi1-x3 nsaid1-y3
• run

30
Syntax Basics - Variables

• How do we use variable lists?
• PROC statements
• proc means datatemp
• var x1-numeric-x4
• run

31
Syntax Basics Data Types

• Two types of data numeric or character (string).
• Numeric variables can only contain numbers.
• Choose numeric if any arithmetic or calculations
  are required.
• Character variables can have anything.
• Choose character if you have names, addresses,
  quotations, descriptions, etc.

32
Syntax Basics Data Types

• Notes on character variables
• Character variables ARE case-sensitive.
• Character variables must be precise
• Keanu Reeves
• Keanu Reeves
• Keanu Reeves
• keanu reeves
• KEANU REEVES

33
Syntax Basics Data Types

• Why use a character variable for numeric data?
• Character variables utilize less computer
  resources.
• Encourages reasoned planning of analyses.

34
Syntax Basics Missing Data

• Missing data can occur because
• Data is lost.
• Respondent did not answer question.
• Structural issues i.e. HRT use in men.
• Remember that there is a difference between a
  missing value and a zero (0) value.

35
Syntax Basics Missing Data

• For numeric variables, the missing datapoint is
  represented by a period (.) or a blank space
  where SAS expects a value.
• For character variables, the missing datapoint is
  represented by a blank space where SAS expects a
  value.
• When referenced within the program, it is
  represented by empty quotes ()

36
Syntax Basics Missing Data

• data patients
• input patient_id 1-5 sex 7 age 9-10
• cards
• 10021 M 24
• 10022 33
• 10023 F .
• 10024 M
• 10025 M 43
• if sex then sex unknown

37
Syntax Basics Missing Data

• How does SAS deal with missing data?
• In calculations, if one of the terms is missing,
  then the target is also missing.
• In PROC steps, SAS does not use missing values in
  its calculations, although with some PROCs, you
  can force SAS to use missing values.
• In DATA steps, variables that you create within
  the program are missing until calculations are
  completed.

38
Syntax Basics Constants

• You can designate constants in SAS, both numeric
  and character.

39
Syntax Basics Expressions

• We can now use variables and/or constants to
  recode, calculate, create new variables..
• Examples..

40
Syntax Basics Expressions

• age 2003 - year_of_birth
• if age gt 18 then agecat Under 18
• bmi weight_kg/(height_m2)
• if sex F and hrt_use 1 then hrt_dose
  dosage freq

41
Syntax Basics Expressions

• General Form
• Variable Variable
• Constant operator Constant
• Expression Expression

42
Syntax Basics Expressions

• Arithmetic Operators

43
Syntax Basics - Expressions

• Some notes on arithmetic operations
• If any of the values are missing, the result of
  the operation will also be missing.
• Division by zero will be indicated by the SAS
  log, but it will not stop the program. It will
  instead create a missing value.

44
Syntax Basics Expressions

• Comparison Operators

45
Syntax Basics - Expressions

• statePA coast. oceanAT tidal89
• Expression Evaluation
• statePA true
• oceanat false
• stateMA true
• tidal ge 89 true
• tidal lt 89 false
• coast lt tidal true

46
Syntax Basics Expressions

• Logical Operators

47
Syntax Basics - Expressions

• statePA coast. oceanAT tidal89
• Expression Evaluation
• oceanAT and statePA true
• oceanAT statePA true
• oceanAT or stateUT true
• tidal gt 100 coast gt 250 false
• state in (PA MA RI) true

48
Syntax Basics - Expressions

• Complex Operations
• Order of operation (arithmetic)
• , , /, , -
• Order of operation (comparison)
• not (), and (), or ()
• Use parenthesis to control order of operation
  SAS will calculate operations inside parentheses
  first, irregardless of the arithmetic or
  comparison order of operation.

49
Syntax Basics - Expressions

• a4 b-2 c2 d1 e3 f0
• Expression _
• a2 / b c2 d - 2
• 16 / (-2) 4 1 - 2
• -8 4 - 2
• -6
• a(2/b) c (2d) - 2
• 4(-1) 2(2) - 2
• .25 4 - 2
• 2.25

50
Syntax Basics - Expressions

• a4 b-2 c2 d1 e3 f0
• Expression _
• a(2 / (b c2)) d - 2
• 4(2 / (-2 4)) 1 -2
• 4(2 / (-2)) 1 - 2
• 4(-1) 1 - 2
• .25 1 - 2
• .25 - 2
• -1.75

51
Syntax Basics - Expressions

• Another way to control complex operations break
  them down into sub-statements
• F a(2 / (b c2)) d - 2
• OR
• x b c2
• y 2/x
• F ay d - 2

52
Data Step Reading the Data

• SAS can read data from a number of data types and
  sources.
• Data read can be fixed or list.
• Fixed data is data in which the location is fixed
  and consistent
• List data is data separated by commas, tabs or
  spaces

53
Data Step Reading the Data

• Fixed data is data in which the location is fixed
  and consistent
• ----5----0----5----0----5----0----5----0----5----0
  
• 12210 male case 8410309061968ma
• 12211 female control 8410803051968mi

54
Data Step Reading the Data

• List data is data separated by commas, tabs or
  spaces
• 12210, male, case, 84103, 09061968, ma
• 12211, female, control, 84108, 03051968, mi

55
Data Step Reading the Data

• DATA Statement
• Tells SAS the DATA step is starting.
• Names the SAS dataset.
• Sets variables used in the dataset to missing
  values.

56
Data Step Reading the Data

• Format
• DATA dataset_name options...
• Two elements of the DATA statement
• The dataset_name is the name of the dataset.
  Naming restrictions are the same as variables.
• Options on controlling data.

57
Data Step Reading the Data

• Identifying the data source
• The INFILE statement
• INFILE fileref options
• The fileref is the pathway and filename of the
  external file that you want to import and convert
  into a SAS datafile.

58
Data Step Reading the Data

• Identifying the data source
• The FILENAME statement
• FILENAME fileref pathnamefilename
• The fileref is a name you want to call this
  external dataset and filename is the location and
  name of the external file that you want to import
  and convert into a SAS datafile.

59
Data Step Reading the Data

• Identifying the data source
• The CARDS statement
• cards
• 123 case female 32
• 234 control male 45
• Use this to enter small datasets directly into
  the data step.

60
Data Step Reading the Data

• Describing the data - INPUT
• INPUT tells SAS how to read the data from the
  dataset identified, and what to name it in SAS.
• You can specify location of the variable within
  the dataset, the name of the variable, and
  whether its a character or numeric
• You can also specify decimal place if necessary

61
Data Step Reading the Data

• Format of INPUT statement
• INPUT var start -end ...
• INPUT var modifier start -end .dec...
• INPUT var modifier...
• INPUT var ...
• Examples of INPUT statements
• input name 1-20 grade 22 status 24
• input city state zip
• input month day year low_temp high_temp
• input loc 1-4 span 8-17 .1 rank

62
Data Step Reading the Data

• Notes on the INPUT Statement
• You can input using both fixed and list formats
  in the same statement.
• Decimal spaces can be specified in the input
  statement using DEC if the decimal place is
  already in the dataset, it supercedes the DEC
  specification in the INPUT statement. If DEC is
  not used, the decimal places in the data are kept.

63
Data Step Reading the Data

• Errors in data
• Bad data happens. O is entered as 0, columns
  dont line up, caps are not consistent, different
  formats are used to enter numbers (use of commas
  for thousands, for currency, etc.)
• How do we deal with this?

64
Exercise in Reading Data

• Create a SAS program that will
• Input the following data
• Calculate age as of 2004
• Calculate height in inches (1 inch 2.54 cm)

65
Error Types

• Syntax Error
• Statements do not conform to SAS language rules
• Data Error
• Logic errors

66
Some Common Errors

• Sample SAS Code 1

67
Some Common Errors

• Sample SAS Code 2

68
Some Common Errors

• Sample SAS Code 3

69
Some Common Errors

• Sample SAS Code 4