I used the PROC FREQ to count all the specified groups. I am not sure if my code is correct to get in the PROC FREQ. I used the data but the list looks messy. I created to classify all the information into the appropriate name groups but the result of PROC FREQ looks empty but left the title and mentioned Frequency Missing=1.
Obviously, I do not do this code right.
Proc Freq without any modification ( but it looks messy):
American Dialect Frequency
? 1
Creole 256
Craole 1
Californian 603
Chicano 302
Chiano 2
Chiaco 1
Midwest 423
Missing
New York City 409
NYC 35
Other 90
Pacific Northwest 128
Pittsburghese 15
Pittsburhese 2
San Francisco 78
Won't say 12
Modify the data... Here is my code:
PROC SORT data=Speechdata; by Dialect;run;
Data Speechdata2;
set= Speechdata;
If substr(Dialet,1,2)='Cr' then Dialect = 'Creole';
else if substr(Dialect,1,2)='Ca' then Dialect = 'Californian';
else if substr(Dialect,1,2)='Ch' then Dialect ='Chicano';
else if substr(Dialect,1,3)='Mid' then Dialect ='Midwest';
else if substr(Dialect,1,1)='N' then Dialect ='New York City';
else if substr(Dialect,1,2)='Sa' then Dialect ='Californian';
else if substr(Dialect,1,2)='Pi' then Dialect ='Pittsburghese';
else if substr(Dialect,1,2)='Pa' then Dialect ='Pacific Northwest';
else if substr(Dialect,1,1)='O' then Dialect ='Other';
else if substr(Dialect,1,1)='NA' then Dialect ='Missing';
else Dialect ='Missing';
run;
PROC FREQ data=Speechdata2;
table Dialect/nocum nopercent;
title 'Dialect Frequencies';
run;