data handling 1

HideShow resource information
View mindmap
  • Data handling
    • types of data
      • nominal
        • data that are in separate exclusive categories of the same feature the order of which has no meaning i.e. types of drug
      • ordinal
        • data that are in separate exclusive, categories of the same feature, the order of which does have meaning e.g. growth stages of an insect: 0-5 hours, 6-10 hours etc
      • continuous
        • data that can have any value in a range of allowed values i.e. birefringence values, RI values
      • discrete
        • either nominal or ordinal
    • two statistical tests
      • non-parametric
        • simpler
          • more likely to get errors
      • parametric
        • interprets results and identifies outliers
          • non-parametric
            • simpler
              • more likely to get errors
        • more robust
          • less likely to get errors
    • distribution
      • normal distribuion
        • peak - central
        • parametric tests required
      • skewed distribution
        • negative skew
          • more to the right
        • positive skew
          • more to the left
      • test for distribution using KS test
    • terms to know!
      • dependant variable
        • what you measure
      • independent variable
        • the variable which is being changed
          • example: "the effect of angle of impact on length and width ratio of blood droplets"
            • what you measure
            • angle of impact = dependant variable
            • length to width ratio = independant variable
      • null hypothesis
        • there will be no significatn difference
      • alternative hypothesis
        • there will be a significant difference
      • repeated measures
        • relationship between observations i.e. same person used - no variations
      • independent
        • no relationship between observations i.e. different people used - variations
    • p value
      • 0.05 = 95% confidence
      • 0.01 = 99% confidence
      • greater than 0.05 (p value) there is no significant difference
        • null hypothesis
      • less than 0.05 (p value) there is a significant difference
        • alternative hypothesis
    • mean
      • add all the numbers and divide by how any there are
    • mode
      • most common
    • range
      • biggest number - smallest number
    • median
      • put numbers into numerical order and get the middle number

Comments

No comments have yet been made

Similar Other resources:

See all Other resources »See all FI resources »