BookmarkSubscribe
Choose Language Hide Translation Bar
DataMonkeyMan
New Contributor

Weighting Observations

I'm trying to pull some summary statistics out of a weighted dataset (e.g., proportion of the population that has blue as a favorite colour). However, JMP does not seem to automatically weight counts. Is there a way to get JMP to produce the weighted count?

 

Thanks for your help.

0 Kudos
8 REPLIES 8

Re: Weighting Observations

Which platform are you using, Distribution? Did you put the data column with the Counts in the Freq analysis role in the launch dialog?

Learn it once, use it forever!
0 Kudos
DataMonkeyMan
New Contributor

Re: Weighting Observations

Yes, I am using the distribution command under Analyze. The variable I'm looking at has been assigned to being nominal, and I've included my weight variable. 

0 Kudos

Re: Weighting Observations

Do you have another data column with the weight to be used for each level?

 

Maybe I should ask what you mean by 'weighted' in this case?

Learn it once, use it forever!
0 Kudos
DataMonkeyMan
New Contributor

Re: Weighting Observations

I have two variables; colour preference and weight.  Colour preference are nominal, and weight is continuous. I have a sample of 8,000 that is scaled to a population of 30,000.

0 Kudos

Re: Weighting Observations

Ah, then color preference is the Y variable and weight is the Weight variable. There is not count for the Freq role. Does that clarification help?

Learn it once, use it forever!

Re: Weighting Observations

Here is an example:

 

Capture.PNG

 

Here is the Distribution analysis:

 

Capture.PNG

 

Alternatively, the levels might be pre-summarized:

 

Capture.PNG

 

 

But the Distribution analysis is the same using N Rows in the Freq role.

 

How does weighting enter your analysis?

Learn it once, use it forever!
0 Kudos
DataMonkeyMan
New Contributor

Re: Weighting Observations

I'm wanting to be able to state that X% of the population prefer blue. 

0 Kudos
KarenC
Super User

Re: Weighting Observations

Try using your weight as the frequency.  I am assuming your data tabel has 8000 rows and you have a coulm that "weights" each row to the proprotion of the full population that it represents.  So this "wegithing" column is the "frequency" of that response in the population.  To get your population counts from your sample use your response (color) as your Y and the "weight" as the FREQ column.  Then you will get something like this (my example has N = 28741 rows weighted to a population of N = 955968818)

 

Raw Data

 

Level 

Count

Prob

1

16751

0.58283

2

11990

0.41717

Total

28741

1.00000

 

Population Wise:

 

Level 

Count

Prob

1

572697202

0.59908

2

383271616

0.40092

Total

955968818

1.00000