turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Using mean and standard deviation as data input

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 09:18 AM

Hi!

I have calculated standard deviation of mean values via Gaussian error propagation because I used a rather complex formula. After this i wanted to compare means (with LSD or HSD). I was searching through the net and all I could find was the input of raw data (e.g. 4.5g,4.7g, and 5.1 g for sample A and 8.5g, 9.1g, and 10.8g for sample B).

How can I import the data into SAS if I have the mean value and the standard deviation (e.g. 4.8g (±0,3) for sample A and 9,5g(±1,2) for sample B?

Thanks!

All best regards

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 09:34 AM

You could do it brute force in a DATA step with these instructions

https://www.utdallas.edu/~herve/abdi-**LSD**2010-pretty.pdf

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 09:43 AM

Hi Doc@Duke,

Thanks for your reply! If I understand you correctly you are suggesting to calculate my post hoc testing manually. I was also considering this but I thought importing data in the questioned format must be such a basic function...

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 10:34 AM

The PROCs that compute the LSD (like ANOVA and GLM) expect to get raw data. It's not the importing of the summary data that is the issue, it is the computation of the subsequent statistics.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 10:58 AM

Yeah, you'd either have to simulate the data or program the test yourself, SAS expects raw data not summarized data.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 11:23 AM

There is this 25020 - One-way ANOVA on summary data

I think you will also need N.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

11-13-2013 01:11 PM

and this