STATS lec 6- Is it significant

Question 1

X² (Chi-squared) distribution: Critical Values of X²

Answer 1

• The degree of freedom= Number of measurements - a number of parameters
• Our Chi-squared has to be bigger than the critical value in the critical value table to be more significant
• If our Chi-squared is lower than 5% critical value (12.592 for 6 degrees of freedom) then there is no significant association between the 2 variables
• For this example, there is a significant association found for degrees of freedom = to 4 at P=<0.01
  *

Question 2

what is the significance value

Answer 2

• Chi-squared is below the critical value indicating that there is no significant association found at df= 8 at P=<0.05

Question 3

What is the significance values

Answer 3

*

Question 4

F distribution: critical values of F (5% significance level)

Answer 4

• This is worked out when doing an analysis of variance
• we need to no 2 degrees of freedom
  
  • V1= number of degrees of freedom associated with the factor we are interested in
  • V2= Number of df associated with the error
• If the df for factors interested in was 5 and df associated with error = 3
• Critical F value = 9.01
• if our value is larger then there will be a significant effect

Question 5

Should F be bigger or smaller

Answer 5

• The critical value for F for df=2,2
• p=5% = 19.0
• p=1% = 99.00
• p=0.1% = 999.00
• this shows that the bigger value = more significant

Question 6

Things to do

Answer 6

• Find other statistical tables
  
  • Work out whether the statistics to be BIG or SMALL to achieve significance
• Make up other chi-squared and F statistics: are they significant
• Revise what significant means
• Revise what p values mean
  
  • In terms of whether the null hypothesis can be rejected or not and with what certainty