a1. Create a new variable for height in inches, dropping the separate
variables for the feet and inches parts of height from the data set.
2. Recode any unexpected or unusual values you found for CLASS and
HOWMANY to missing values. For all players who indicated "no
burners" last year, recode the value of HOWMANY from missing
to zero.
3. Create an indicator
variable to identify Freshman that takes the value 1 for freshman and zero for upperclassmen (sophmore/junior/senior/5th yr).
4. Label the new variables you created. Assign formats as appropriate.
(Tip: add
to your format program from the previous assignment, as needed,
and rerun it).
5. Print the first 20 observations in the file using the labels, and
run contents on your dataset. Note: the obs=20 option after the data set name will limit the number of observations printed. DO NOT TURN IN A LIST OF ALL 700+ OBSERVATIONS!
(Use: PROC PRINT DATA=libn.dsn(obs=20)
labels; )
For parts 1b and 1c: hw8p1bc.sas
b.
SUBSETTING
Create two temporary datasets from the burner data, one with those
who never had burners, and one with those who have ever had burners.
Keep only subject id, freshman/upperclass indicator, and number
of burners last season in these data sets.
For each data set, get frequency tables of freshman/upperclass status,
and number of burners last year. Use appropriate titles.
c. CONCATENATING
Rejoin the never/ever burner data sets into a single data set. In
the process of rejoining, re-create a variable to indicate ever/never
had a burner in this new data set.
Print frequency tables of freshman/upperclass status and ever/never
had burners using your rejoined data.
