biol139-lecture20-2011

Class number observed number expected2 number

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 44 142 yyR_ yyR_ 124 142 yyrr 146 142 (1 2 ( ( 1 5 4 −1 4 2 ) 2 4 4 −1 4 2 ) ( 1 2 4 −1 4 2 ) 2 1 4 6 −1 4 2 ) 2 + 142 + + 142 142 142 χ = 1 .0 1 + 0 .0 2 8 + 2 .2 8 + 0 .1 1 3 = 3 .4 3 5. Use the chi-square value and number of degrees of freedom to determine a p value : the probability that a deviation from the predicted numbers at least as large as that observed in the experiment will occur by chance (Table 5.1 in text) chance Remember: Χ 2 = 3.43 and df = 3 Remember: 3.43 p value ≈ 0.3 null hypothesis (no linkage) is accepted value is If observed values deviate from values predicted by our hypothesis, is this significant (reflecting a real difference between observation + theory) or insignificant insignificant (reflecting random sampling error or chance)? P value lets us know this…. 6. Evaluate the significance of the p value. The 6. convention is that a 0.05 p value is the boundary convention between accepting and rejecting the null hypothesis. P value – the probability that a deviation from the predicted numbers at least value as large as that observed in the experiment has occurred due solely to chance Very small p values indicate a high degree of significant difference significant between observed and expected data i.e. between 0.05 = 5% chance that the observation is due to chance alone (in other words, a 95% probability that the observation is not due to chance and so is a REAL difference) REAL If very low, deviation of observed from expected results becomes significant, very significant, unlikely hypothesis being tested explains data (i.e. if p value = 0.01, only 1 in 100 times will that observation be due to chance alone) If high, likely hypothesis being tested is correct, deviation from expected is high insignificant (i.e. p value = 0.90 -90 out of 100 times the observation would occur by chance if the null hypothesis is correct) occur Are genes A and B linked in testcross AB/ab x ab/ab ? Are AB/ab ab/ab ab ab ab ab χ 2= (18-12.5)2 + (15-12.5)2 12.5 12.5 χ 2= 2.42 + 0.5 + 2.42 + 0.5 χ 2= 5.84 Df =...
View Full Document

Ask a homework question - tutors are online