* EXAMINING THE CPS DATA

use e:\person.dta

describe

summarize a_grswk
summarize a_grswk, detail

*WHAT SHOULD WE DO WITH THOSE ZEROS?


drop if a_grswk==0
summarize a_grswk

*<DROP> WILL CHANGE THE DATA FOR THE REST OF THE PROGRAM.  TO CHANGE THE SAMPLE 
*ONLY FOR THIS COMMAND, WRITE <summarize a_grswk if a_grswk>0>

clear
use e:\person.dta
summarize a_grswk if a_grswk>0

tabulate a_hga

*GRAPHS AGAIN
graph a_hga
graph a_hga, bin(50)

graph a_grswk a_hga
sort a_grswk
sort a_hga
graph a_grswk a_hga

*OOPS, THERE ARE THOSE ZEROS AGAIN
drop if a_grswk==0
graph a_grswk a_hga

graph a_grswk, hist
graph a_grswk, bin(50)

*USING WEIGHTS
 
*IN STATA YOU MUST SPECIFY THE TYPE OF WEIGHT YOU USE:
*AWEIGHTS: WEIGHTS WHERE THE RECORDED SCALE IS IRRELEVANT,
*    STATA INTERNALLY RE-SCALES THE WEIGHTS TO n
*FWEIGHTS: INDICATES THE NUMBER OF DUPLICATED OBSERVATIONS
*  THESE ARE THE TYPE OF WEIGHTS USED IN THE CPS
*PWEIGHTS: OR SAMPLING WEIGHTS, DENOTE THE INVERSE OF THE
*  PROBABILITY OF INCLUSION IN THE SAMPLE--WILL BE LESS THAN 1.
 
*YOU SHOULD ALWAYS BE ABLE TO USE AWEIGHT OR WEIGHTS
*  WITH THE CPS DATA
 
summarize a_sex
summarize a_sex [weight=marsupwt]
summarize a_sex [aweight=marsupwt]
summarize a_sex[fweight=marsupwt]

tabulate a_sex
tabulate a_sex [aweight=marsupwt]
tabulate a_sex [weight=marsupwt]

gen wagewt=a_ernlwt+marsupwt
*REMEMBER THAT YOU NEED A DIFFERENT WEIGHT FOR A_GRSWK.
summarize a_grswk [weight=wagewt]


*PROGRAMMING:
*OPEN *.DO FILE EDITOR.  PASTE IN PROGSEX.DO FROM 
*THE WEB PAGE
*"DO" THE PROGRAM FROM WITHIN THE .DO FILE EDITOR, THEN
*THIS PROGRAM IS NOW IN MEMORY, AND CAN BE USED THE SAME AS
*ANY OTHER PROGRAM.  SIMPLY TYPE IN THE PROGRAM NAME.

loop

*TO ERASE THE PROGRAM FROM MEMORY, TYPE

program drop loop

*LOOK AT A MORE COMPLEX PROGRAM ON THE WEB PAGE.