BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
noha
Calcite | Level 5

Hi

I am using University edition.

I am dealing with complex study design and I am using the output delivery system in proc surveyfreq to output a data set but I am confused between frequency and weighted frequency :

 

proc surveyfreq data = demo5;
cluster v021 ;
strata v023;
weight sweight;
tables elgpop*(mothagec);
run;

 

I have attached the result, there is frequency, weighted frequency and percent.

I want to include frequency and percentage in my result tables, I want to know which numbers I should include?

Do I have to use the frequency then calculate the percentage for each variable by myself? Or I should use the weighted frequency?

 

Hope you could help me.Thank you!

1 ACCEPTED SOLUTION

Accepted Solutions
ballardw
Super User

"Which to use" pretty much always depends on "what question are you answering".

 

If you look at your table you should note that Missing + total under the frequency column = Number of observations. Frequency is number in the data used for calculations. Weighted frequency refers to Frequency with the weight applied which generally can be though of as the estimated count in the population with the combination of values. Though with about 50% missing that might be a very suspect conclusion.

 

Most of the questions I use Surveyfreq for would be predominately using the PERCENT column as in 53% of ELGPOP=1 persons were in MOTHAGEC=1.

 

You should get in the habit of requesting confidence limits as well. That makes it easier to eyeball if the 16 % for Mothagec=0 is likely to be significantly lower than the 30% for Mothagec=2.

 

View solution in original post

2 REPLIES 2
ballardw
Super User

"Which to use" pretty much always depends on "what question are you answering".

 

If you look at your table you should note that Missing + total under the frequency column = Number of observations. Frequency is number in the data used for calculations. Weighted frequency refers to Frequency with the weight applied which generally can be though of as the estimated count in the population with the combination of values. Though with about 50% missing that might be a very suspect conclusion.

 

Most of the questions I use Surveyfreq for would be predominately using the PERCENT column as in 53% of ELGPOP=1 persons were in MOTHAGEC=1.

 

You should get in the habit of requesting confidence limits as well. That makes it easier to eyeball if the 16 % for Mothagec=0 is likely to be significantly lower than the 30% for Mothagec=2.

 

noha
Calcite | Level 5

Thanks for your reply. regarding the missing values, they are not eligible for my study.

 

"Which to use" pretty much always depends on "what question are you answering": your answer is to use the weighted frequency, right? So the sample of my study will be 5799, here is another question that comes up, by summation of the weighted frequencies, it is not equal to 5799. What is the problem here?

sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

Register now!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 8881 views
  • 0 likes
  • 2 in conversation