Thanks. I appreciate your information and the SAS/IML function. I tried the function for different sample sizes to check the effect on the results. ... View more

Thank you very much. Both methods work. ... View more

Thanks. Yes, I want to compare only 2 categories, but by considering the total of all 8 categories, i.e 35 in this case and not just the total of those 2 categories. The proc freq with chisq considers total of only 2 categories.   Basically, I want to test if there is a significant difference in the frequencies/counts of the 2 selected categories by considering the total observations. So, for the “general” and “central” categories, I want to consider the proportions 4/35 and 3/35 (and not 4/7, 3/7). Is there any way to test it?   (Sorry, the error message in my program was old.)   ... View more

Thank you. ... View more

Thank you very much.  I can do that way by removing the gaps between the bars of the products for each year.  So they will look more like clustrered for each year.  Thanks. ... View more

Thanks for a quick reply.  But that example gives only a stack bar chart, not a stack and cluster bar chart.  I would like to have at least 2 such stack bars (in a cluster) for each x-category.  Fo example, if you have that type of data for another year then we would like to show such stack bars for 2 years for each product in the same chart. How to do it? ... View more

You are right.  I also did study and understood the results correctly.  I switched to  lsmeans option as those results are more clear to me. I used same options as you mentioned in addition to cl just to check CL values.  My problem is now solved. By the way, (Pr>F) = 0.0393 was from another example in which I had the similar problem. Thanks for all the explanation. ... View more

P-val = 0.0393 from my first Eample and p-val =.0007 from my second Ex (both using MEANS option) are also the same as the adjusted (from second set) p-values obtained using LSMEANS options. So, are they supposed to be significant  at .05 and .001 levels, respectively, even though CLs don't mark those results for significance in MEANS option? ... View more

Thanks for the information. But I didn't mention F-value, I gave values from pair-wise comparisons.  I am checking only main effect.  So, I used proc glm data=&DataName2;    class grpcode2;    model BMI = grpcode2 /ss3 ;    means grpcode2 / tukey alpha=.001 ;      contrast 'Obese(1) vs OverWeight(2)' grpcode2 -1 1 0 ; .... run;   (I tried with lsmeans/adjust=tukey and had the same results with no confidence limits.)   The partial results are:           Difference                    grpcode2    Between Simultaneous 99.9%                   Comparison     Means Confidence Limits                     1 - 2      12.2686   7.9828   16.5543  ***                     1 - 3      16.6226  12.3368   20.9083  ***                     2 - 1     -12.2686 -16.5543   -7.9828  ***                     2 - 3       4.3540  -0.2751 8.9831                     3 - 1     -16.6226 -20.9083  -12.3368  ***                     3 - 2      -4.3540  -8.9831 0.2751 Contrast                       DF Contrast SS Mean Square F Value Pr > F    Obese(1) vs OverWeight(2)       1  878.020762  878.020762 131.98 <.0001    Obese(1) vs Control(3)          1 1611.807639 1611.807639 242.28 <.0001    OverWeight(2) vs Control(3)     1   94.786580   94.786580  14.25 0.0007 So, I thought 2 vs 3 should be sig at .001 level. ... View more

Thanks, but it doesn't fit to my situation.  Without replication is within a sample, not among samples.   I again tried, but I couldn't get it. ... View more

Linlin, your code is short, efficient, and easy and It worked. Thanks. ... View more