Stats test

• null-hypothesis

  They’ll be no significant difference or correlation between the things you are measuring and any difference or correlation will be due to chance
• If the thing is looking for an association what test do we do?

  Correlation coefficient also known as spearman rank
• If it’s comparing frequencies, what test do we use?

  Chi squared
• If it’s not looking for an association and not comparing frequencies, what test do we do?

  Students T test
• What is students t test?

  A test for a significant difference between the means of two sets of data
• When is chi squared mainly used

  For inheritance and genetics
• Chi squared test

  Test for the significant difference between observed and expected frequencies
• P<=5

  only if there is a less than a equal to 5% probability that the difference or correlation is due to chance we can see that the difference or correlation is significant
• If the p values less than 0.05

  1. There’s less than 5% probability that the difference/correlation was due to chance
  2. Reject the null hypothesis
  3. Therefore, there is a significant difference between …
• P values greater than 0.05

  1. There is more than 5% probability that the difference/correlation was due to chance
  2. So we accept the null -hypothesis
  3. There’s no significant difference/correlation between…
• Chi squared values greater than the critical value

  Significant difference

  Not due to chance

  So reject the null hypothesis as it’s less than 5% probability that it’s due to change
• Chi squared degree of freedom

  Number of classes minus one
• T test degree of freedom

  Number of samples for both sets of - 2
• Correlation coefficient degree of freedom

  Number of samples