Grouping results in variables

[Monaghan]

Hello!  I am trying to perform a one way ANOVA comparison of merged data.  I have been able to merge the data, but I am having trouble with combining some variable data.  I am wanting to look at the variable of age (RIDAGEYR) with the variable of days of physical activity per week (PAQ670) .  For my project, I would like to group the days into 3 categories (1-2 days, 3-5 days, 6-7 days) and run a ANOVA test to get the mean age for each of the three categories.  I have been able to run the ANOVA, but not able to separate the wanted categories.  Here is my code and output--

data work. agephysicalhealth;
merge SASFILE.DEMO_H SASFILE.PAQ_H;
by SEQN;
run;

proc anova data=WORK.AGEPHYSICALHEALTH;
class PAQ670;
model RIDAGEYR = PAQ670;
means PAQ670 / scheffe;
title;
run;

The output lists the *class PAQ670 and values 1 2 3 4 5 6 7 99*

I have not figured out how to separate these - not sure if I should sort the data prior to running the procedure or sort after.  I would like to use the same code for all of my procedures because I am running multiple analysis on the variable of days of physical activity.

Thank you for any help!!

Carrie

[Kurt_Bremser]

class needs no sorting, but you should try to create a format that converts your days into categories and use that for the class variable.

[Reeza]

Here's an example of how you can create a format and apply it. You can run the code and examine the output, it should work in any version of SAS. 

/*this is an example of creating a custom format and then applying it to a data set*/

*create the format;
proc format;
value age_group
low - 13 = 'Pre-Teen'
13 - 15 = 'Teen'
16 - high = 'Adult';
run;

title 'Example of an applied format';
proc print data=sashelp.class;
format age age_group.; *applies the format;
run;


data class;
set sashelp.class;
age_category = put(age, age_group.); *creates a character variable with the age category;
label age_category = 'Age Category'; *adds a nice label for the printed output;
run;

title 'Example of creating a new variable with the format';
proc print data=class label;
run;

*show format used directly;

proc freq data=sashelp.class;
table age / out= formatted_age;
format age age_group.;
run;

For your specific group your format would look like:

proc format;
value days_category
1-2 = "1 to 2 days"
3-5 = "3 to 5 days"
6-7 = "6 to 7 days"
99 = "Missing";
run;

Then apply it in your code:

proc anova data=WORK.AGEPHYSICALHEALTH;
class PAQ670;
format paq670 days_category.;

model RIDAGEYR = PAQ670;
means PAQ670 / scheffe;

title;
run;
 

---

@Monaghan wrote:

Hello!  I am trying to perform a one way ANOVA comparison of merged data.  I have been able to merge the data, but I am having trouble with combining some variable data.  I am wanting to look at the variable of age (RIDAGEYR) with the variable of days of physical activity per week (PAQ670) .  For my project, I would like to group the days into 3 categories (1-2 days, 3-5 days, 6-7 days) and run a ANOVA test to get the mean age for each of the three categories.  I have been able to run the ANOVA, but not able to separate the wanted categories.  Here is my code and output--

data work. agephysicalhealth;
merge SASFILE.DEMO_H SASFILE.PAQ_H;
by SEQN;
run;

proc anova data=WORK.AGEPHYSICALHEALTH;
class PAQ670;
model RIDAGEYR = PAQ670;
means PAQ670 / scheffe;
title;
run;

The output lists the *class PAQ670 and values 1 2 3 4 5 6 7 99*

I have not figured out how to separate these - not sure if I should sort the data prior to running the procedure or sort after.  I would like to use the same code for all of my procedures because I am running multiple analysis on the variable of days of physical activity.

Thank you for any help!!

Carrie

---