Stats

• Midrange-


  max+min/2
• Negative skew-

  • Mean<median<mode
  • Normal distribution tails towards negative
  • Box plot median to the right (more spread out on negative side)
  • M-LQ>UQ-M
• Positive skew-

  • Mean>median>mode
  • Normal distribution tails towards positive
  • Box plot median to the left (more spread out on positive side)
  • M-LQ<UQ-M
• No skew-

  • Mean=median=mode
  • Normal distribution even
  • Box plot median central
  • M-LQ=UQ-M
• Probability of mutually exclusive events-

  • P(A⋂B) = 0
  • P(A⋃B) = P(A) +P(B)
• Probability of non-mutually exclusive events-


  P(A⋃B) = P(A) + P(B) - P(A⋂B)
• Independent events

  • The probability of each event happening doesn't change
  • P(A⋂B) = P(A) x P(B)
• Association-


  General relationship (usually categorical or ordinal data), measured by Spearman's rank correlation coefficiant (SRCC)
• Correlation-


  Linear relationship (continuous or normally distributed data), measured by Pearson's product moment correlation coefficient (PPMCC)
• Null hypothesis-


  H0, original probability
• Alternative hypothesis-


  H1, expected change from the original probability
• Significance level-


  Probability threshold, to accept H0 it must be greater than the SL (need a significant amount for it to be true)
• Probability of incorrectly rejecting a hypothesis = significance level
• Parameter-


  The thing that's being tested
• Hypothesis test concluding statements-

  There is/isn't enough evidence to suggest at a _% significance level to suggest that the probability of (the situation) has increased/decreased from (original probability)
• Carrying out a hypothesis test-

  • State H0 & H1
  • State the parameter & significance level
  • State the values of X, n & p for X~B(n,p) to be used & what X measures
  • Calculate the probability of a more extreme event than the given example occurring
  • Compare the probability to the significance level
  • Accept or reject H0
  • Write the concluding statement