stats

• positively skewed graph 


  mode, median and then mean
• continuous data 


  measured = distance, temperature etc
• discrete data/ discontinuous 


  counted = number of students
• parametric test


  there is information about the population parameter (normal distribution)
• non-parametric test


  there is no idea regarding the population parameter - no normal distribution
• example of parametric 


  t-test
• negative skew 


  mean, median and mode
• normal distribution 


  all three mean, median and mode are in middle of curve
• variance 


  a measure of how far each number in the set is spread out from one another
• standard deviation


  a measure of how much a set of values is spread around the mean or average
• standard error of the mean

  indication of how close a sample mean might be to the population mean
• proportion of data included in range of one standard deviation below the mean

  68% (34% one side and 34% the other side on the graph)
• proportion of data included in range of two standard deviation below the mean

  95% (34% (first deviation) + 13.5% on each side of graph)
• 3 deviation


  99.7% (34 + 13.5 + 2.35)
• inferential stats 


  methods which make inferences about the group/ population
• hypothesis 


  precise, testable statement of what the researcher predicts will be the outcome of the study
• null hypothesis 


  statement of no difference between the groups being studied
• alternative hypothesis 


  claim about the groups that is contradictory to what we conclude when we reject
• p value 


  • low = inconsistent with null
  • if p value is smaller than 0.05 where is a difference between groups
• low p value 


  unlikely to happen by chance
• high p value 


  occurred by chance
• t-test 


  used to compare two samples to determine if they came from the same population with a single mean (see if there is a significant difference between two means or if it is due to chance)
• unpaired t-test


  decide whether to accept the assumption that two samples come from populations with the same standard deviations (means are from two independent groups)
• paired t-test


  you expect the differences between paired values to be consistent measure of treatment effect (both means are from same group of individuals at different times)
• two tailed t-test 


  critical area of distribution is two-sided + whether a sample is greater or less than a certain range of values
• one-tailed t-test 


  used to analyse data where critical area is one-sided - greater or less than certain value
• standard curve 


  relationship between concentration or amount of a substance and its measured response
• standard curve also known as...

  calibration curve
• simple linear regression


  used when linear relationship between variables