Hi guys:   I have  many ANOVAs  to be run and each of them has its respective p value and R squares. All ANOVAs have the same amount of observations,  this means thar ERROR I and III could be taken to make part of the table i need to get There we go, we have 3 ANOVA, everyone is a different answer on each environment:   data have; input order environment country$ observation; cards; 1 1 a 11.7062885827733 2 1 a 12.5326186073177 3 1 a 12.1121866473611 4 1 a 10.6770770786762 5 1 b 12.7272303160372 6 1 b 12.5069020572854 7 1 b 10.3801953337563 8 1 b 11.3699075949466 9 1 b 14.7913170706891 10 2 c 11.7957215280586 11 2 c 12.7967928055036 12 2 c 12.3156597291194 13 2 c 11.4903775240095 14 2 c 12.2868489869184 15 2 d 12.0244581514243 16 2 d 11.2738963155379 17 2 d 11.2017213686765 18 2 d 12.6084387827896 19 2 d 13.2818434423956 20 3 e 12.7756560519949 21 3 e 11.0392289428102 22 3 e 12.0000826320338 23 3 e 11.139270842614 24 3 e 13.5663343765743 25 3 f 11.7650627921355 26 3 f 12.6102709811389 27 3 f 11.5515913330262 28 3 f 14.0639297127507 29 3 f 12.3158309468017   proc glm data = have; where environment = 1; class country order; model observation = country order; random order; means country / duncan tukey lines; run; quit;   proc glm data = have; where environment = 2; class country order; model observation = country order; random order; means country / duncan tukey lines; run; quit;   proc glm data = have; where environment = 3; class country order; model observation = country order; random order; means country / duncan tukey lines; run; quit;   now, we have to mount a new table containing: Environment rsquared P value Model P value Error I p value ERROR III 1         2         3           IMPORTANT INFORMATION IN BIG CAPS DATA PRESENTED HERE IS ENOUGH TO ACCOMPLISH THE PUZZLE, NO NEED TO GET MORE DETAILS.   Thanks for the help   Bets regards, always ... View more

Good night guys:   Last post i tried to explain an example, but for some logistic reasons i couldn't finish the explanation there so here i go again:   i have this data set:   data have; input id South North east west ; cards; 1 2 4 12 9 2 3 . 11 10 3 2 4 12 10 4 6 4 13 12 ; proc sql; create table want as select *, sum(south + north)/(count(south + north)) as mean_south_north, std(south and north) as sd_south_north from have; quit; proc print data = want; run;, but unfortunately i got this:   Obs id south north west east sum_south_north mean_south_north sd_south_north sum_west_east mean_west_east 1 1 10 . 15 9 116 25.2 8.48528 74 12.3333 2 2 40 12 10 14 116 25.2 8.48528 74 12.3333 3 3 50 14 13 13 116 25.2 8.48528 74 12.3333   The RED BOLD text in the table above represents the standard deviation wrongly found, where the correct value must be  18.4715998224301, as the real value of the standard deviation of 10, 40, 50, 12 and 14. Must be noted that it is a missing value in the data set, so it is very important to consider also.   Other important issue, it would be grate make the program to find the standard deviation from other columns and put them as other variable, for example the standard deviation of west and east observations , as std_west_east, considering GREEN DATA in the table above   i appreciate any help to complete this task correcting the code i provided to you above.   thanks in advance ... View more