# Thread: Graphpad or SPSS way to analyze this question

1. ## Graphpad or SPSS way to analyze this question

Dear friends,
I have ~3500 patients. I have ages for everyone of them, and i have a laboratory value for each one as well.
I want to know if the different categories of ages (say, 18 to 30, 30 to 40, 40 to 50, etc) have different means/sds than each other. meaning, if the lab value gets different with age.

what is the best way to go about that? first, i dont know which strata may be more significantly different than other, meaning 18 to 30 vs. 18 to 25 then 25 to 30 then 30 to 35. or 18 to 23, 23 to 27, 27 to 31 etc (i am just bringing random examples).

There is definitely difference as the age progresses, i just dont know how i go about analyzing the data.

Any suggestions? i have graphpad and SPSS

2. ## Re: Graphpad or SPSS way to analyze this question

Well this isn't a formula analysis, but how about 3 scatterplots?

3. ## Re: Graphpad or SPSS way to analyze this question

Dear hlsmith,
thanks for chipping in.

i envisage "discovering" that different age cut-offs will give different "ranges" for that lab values.

I have 3500 participants with their ages and their lab values. This is a normal lab value. my hypothesis is that different age ranges will have different normal range for this lab value.

so for example, if you are 60 and you get X level, this is normal for your age but abnormal if you get it at age of 30 etc

4. ## Re: Graphpad or SPSS way to analyze this question

1 scatter plot, value versus age with dots color coded to better represent categories.

5. ## Re: Graphpad or SPSS way to analyze this question

How can I make a graph that looks like this using SPSS or graphpad?
this is continuous age-dependent percentile. the middle line is the median, the lower and upper are lower range and upper range
http://imgur.com/a/ySPFq

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts