somebody Posted April 6, 2008 Report Posted April 6, 2008 Hey guys, my microsoft excel is acting very weird. I am performing paired ttest on 2 set of data, each with 2 samples. SO i tried type in =TTEST(array1, array2, 2, 1) where array 1 displays my first data (1.06 and 0.73), array 2 displays my second set of data (0.87 and 0.54), 2 represents 2-tails and 1 for paired ttest. And what i get is this error #DIV/0!. Now, i did some research on it and found that it says that because we divide by zero but in my case i am not even using 0 or using an empty cell. I though, the problem is that i only have 2 cases but i tried with my other data which worked out fine but its only giving me trouble on this set of data. I wanted to contact microsoft about this but they want to charge me like $45 so Some one here please help me solve this problem.(my paper is due tomorrow, help by the morning would be greatly appriciated) Quote
Tormod Posted April 6, 2008 Report Posted April 6, 2008 Funky. I don't know what to do but maybe you can get some hints here: How to use the IF worksheet function to suppress the #DIV/0! error value in Excel Quote
somebody Posted April 7, 2008 Author Report Posted April 7, 2008 hah, i already read that but that just suppresses the problem, i want to fix and and the thing is that i do not have any 0 for demominator so doesn't even make sense.I think its because i have 2002 version and i think i should invest in office 2007. thanks for you reply, i appreciate it Quote
Tormod Posted April 8, 2008 Report Posted April 8, 2008 you know, Pyrotex here at Hypography is proficient in Excel. Maybe you should pop him a PM. Quote
C1ay Posted April 8, 2008 Report Posted April 8, 2008 Do you have any missing or incomplete data? See You may obtain incorrect results and misleading labels when you use the Analysis ToolPak t-Test in Excel for some possible causes... Quote
alexander Posted April 9, 2008 Report Posted April 9, 2008 it would be nice if you posted the file here too, or at least some test data that does not seem to work. it's hard saying what the problem is if we have only a brief understanding of what you are doing.... 2002 or 2007 matters little, if you have a problem with your data, you have a problem on both. Quote
somebody Posted April 11, 2008 Author Report Posted April 11, 2008 Hey guys not to worry, i think i "figure" out the problem. Ok, my data was in only 2 decimal so now i try to have 3 decimal data which solved my problem so for instance if i had: Group A GroupB2.45 3.461.45 0.78 I was getting error for some reason but then when i change these values just a little bit (in terms of adding decimal values), i was getting a p-value from my ttest. so when i change my 2.45 to 2.450000001, i got my p-value. But this is not the real data and i only needed 3 decimal value. But thanks for your reply and i will keep in mind that Pyrotex is the "excel go to guy" along with the whole hypography. I just turned in my paper in time (with in matter of 1 min :beer: ) Thanks guys Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.