STATISTICAl TESTS

• what are the three statistical tests in biology?

  • t test
  • correlation coefficient
  • chi squared
• what is the correlation coefficient test for?

  when testing how close the relationship between an independent variable and a dependant variable is
• just because we can see correlation...

  does mean causation
• what to R values in correlation coefficient mean?

  • when its 0 : there's no correlation
  • +1 : there is a strong positive correlation
  • -1 : theres a strong negative correlation
• what type of data do we use correlation coefficient for?

  quantative data : discrete and continuous
• when do we use T test?

  when comparing the means of two independent data sets
• when can't we use the T test?

  when there's outliers or skews
• What does a 0 mean in a t test?

  the means are equal
• when do we use the chi squared?

  to compare observed and expected results
• what does a value of 0 mean in the chi-square test?

  no difference
• what does it mean when the chi-squared value is grater than 0

  there is a difference between, however it doesn't tell how much of a difference
• what type of data doe we use the chi squared for?

  categorical data ( limited range of values)
• What is the null hypothesis?

  claims there is no relationship between the variables
• what is the p value ?

  probability of observing a result by chance
• what happens when the P value is P > 0.05

  • more than a 5% probability the data is due to chance
  • accept the null
• what happens when the p value is P < 0.05

  • there is less than a 5% probability the data is due to chance
  • reject the null
• when finding the p values ourselves, what do we need?

  the degrees of freedom
• how do we find the degrees of freedom for each test ?

  correlation coefficient : (n-2)

  t test : ( n1 + n2 ) -1

  chi squares : (n-1)