[讨论]t test and normality

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

Hi everybody:

I want to make sure I am doing things right. I need to run a t-test to compare age means between 2 samples. I have more than 1000 people in each sample. I checked the normality using proc univariate as follow:

Proc univariate normal; Class sample; Var age; Run;

Am I doing it right? If yes, I found small p-values "0.00" which suggest non normality. Should I use a non-parametric technique like proc npar1way?

Thanks for your help.

hm

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：normality normal Norm test ALI 讨论 test normality

As long as you sample is randomly from populations, with 1000 samples and central limited theory, the sample means are normally distributed. If you want to test mean difference, the proc ttest will give you all results.

HTH

%let randomF=rannor;

data t1; do i = 1,2; do j= 1 to 1000; x_&randomF=i*&randomF(123)+i; output; end; end; run;

proc ttest data=t1; class i; var x:; run;

Thanks so much. But, I used proc univariate to check normality and the results suggested non normality. I am a bit confused. Once I have large samples, by the central limit theoreme, should I assume normality even if the proc univariate suggests non normality?

Thanks