Statistics

Statistics

1. definition :Allows us to build an experiment (model) andanalyse it to draw conclusions

2. Intro to stats

   • Null: No sig difference, any difference being due to chance (sampling or experimental error
     • HIGHER pvalue= H0 correct (>0.05)
   • Alternate (H1) : There IS a sig diff. between measurements, frequencies and there is a relationship between data. (specific thing to blame)
   • P values are always between 0.00 and 1.00
     • 0.00= effect is unlikey due to chance
       • Accept alternative
     • 1.00= Effect is certainly due to chance
       • Accept null
   • Decision making
     • Type 1 error:Reporting a difference when there isn't one
     • Type 2 error: Reporting NO diff when there IS one
   • Normal Distribution
     • Ensures that the data is representative of a "normal" population.
     • Kolmogorov- Smirnov and Shapiro -Wilk tests
       • Used to test data for normal distribution
       • Kolmogorov- Smirnov test
         • Tries to determine if 2 sets of data differ significantly
         • Non- parametric and distribution free
       • Shapiro- Wilk test
         • Normality test, used to determine if a sample or any group of data is normally distributed