t-test

• If you have two groups and you want to see that they're different you use a t-test.
• There is a signal to noise problem that must be overcome.
• The test calculates signal/noise = (difference between group means)/ (variability of groups)
• Variance is the square of the standard deviation
• I just plugged this data into excel and got the t-test.
• I then used the significance calculator at vassar
• Note that the second column is just the first plus a random number and the second column is the first plus a random number/2.
• Different random numbers give different t-s and significances. I show 4.
• 1 1.442224853 1.427047121
  2 2.301609415 2.135436058
  3 3.317390073 3.2231911
  4 4.299232432 4.06546017
  5 5.973043033 5.489989224
  6 6.466954119 6.289834642
  7 7.167729206 7.071682755
  8 8.479859125 8.101998397
  9 9.907791513 9.331918857
  10 10.63178065 10.38076171
  11 11.92909508 11.06638654
  12 12.13248421 12.16939369
  
  t-test 0.854808487
  significance 0.4131
  t sig
  0.89619 0.3912
  0.9085 0.384
  0.8828 0.3981
  0.8098 0.4369