topic Regrouping Age into a new variable in Statistical Procedures

Regrouping Age into a new variable

madisonahrens — Wed, 15 Jun 2022 19:32:32 GMT

I am trying to regroup the age variable into a new variable called Age_Group. I keep getting errors when I try to run. Here is how I added a new variable:

data work.new;

set work.working_set;

Age_Group=Age;

run;

And here is the code that I am trying to run.

data work.new1;

set new_data;

if Age_Group LE 29 then Age_Group='18-29';

if Age_Group GE 30 AND LE 39 then Age_Group='30-39';

if Age_Group GE 40 AND LE 49 then Age_Group='40-49';

if Age_Group GE 50 AND LE 59 then Age_Group='50-59';

if Age_Group GE 60 AND LE 69 then Age_Group='60-69';

if Age_Group GE 70 AND LE 79 then Age_Group='70-79';

if Age_Group GE 80 then Age_Group='80+';

run;

Can anyone help or give suggestions? Thank you!

Re: Regrouping Age into a new variable

Reeza — Wed, 15 Jun 2022 19:52:21 GMT

The first data step converts the AGE_GROUP variable to AGE, a numeric variable.
The second step attempts to recode the AGE_GROUP variable to a character variable.
AGE should be referenced for each comparison otherwise it compares to the output of the previous comparison
A variable cannot be converted with the same name.
The first data step is not required.

There is also inconsistency in the data sets referenced.

First data step:

input data set: working_set

output data set: new

Second data step:

input data set: new_data (not shown, or typo? No relation to previous step)

output data set: new1

data work.new;
set working_set;

if age LE 29 AND not missing(age) then Age_Group='18-29';
else if age GE 30 AND age LE 39 then Age_Group='30-39';

*simpler way to write the else if;
*else if 30<=age <=39 then Age_Group='30-39';*


else if age GE 40 AND age LE 49 then Age_Group='40-49';
else if age GE 50 AND age LE 59 then Age_Group='50-59';
else if age GE 60 AND age LE 69 then Age_Group='60-69';
else if age GE 70 AND age LE 79 then Age_Group='70-79';
else if age GE 80 then Age_Group='80+';

run;

@madisonahrens wrote:

I am trying to regroup the age variable into a new variable called Age_Group. I keep getting errors when I try to run. Here is how I added a new variable:

data work.new;

set work.working_set;

Age_Group=Age;

run;

And here is the code that I am trying to run.

data work.new1;

set new_data;

if Age_Group LE 29 then Age_Group='18-29';

if Age_Group GE 30 AND LE 39 then Age_Group='30-39';

if Age_Group GE 40 AND LE 49 then Age_Group='40-49';

if Age_Group GE 50 AND LE 59 then Age_Group='50-59';

if Age_Group GE 60 AND LE 69 then Age_Group='60-69';

if Age_Group GE 70 AND LE 79 then Age_Group='70-79';

if Age_Group GE 80 then Age_Group='80+';

run;

Can anyone help or give suggestions? Thank you!

Re: Regrouping Age into a new variable

PaigeMiller — Wed, 15 Jun 2022 19:52:32 GMT

proc format;
    value agef low-29='18-29' 30-39='30-39' /* you type the rest */ ;
run;
data new1;
    set new;
    format age agef.;
run;

Re: Regrouping Age into a new variable

madisonahrens — Wed, 15 Jun 2022 20:55:54 GMT

That worked! Thank you so much @PaigeMiller

Re: Regrouping Age into a new variable

SteveDenham — Thu, 16 Jun 2022 17:59:20 GMT

Why are you categorizing a perfectly good continuous variable? From a statistical point of view, it is the equivalent of taking a Lamborghini into the shop and asking them to cut it up into pieces that might fit into the garage better. See this link:https://discourse.datamethods.org/t/categorizing-continuous-variables/3402

SteveDenham