statistics

• distribution: describes how data is spread out across a range, showing where most results sit - represent the average and spread of data
• normal distribution: mean, median and mode are equal - symmetrical distribution
• positively skew - mode < median < mean, long tail on the right
• median is always in the middle of the graph
• negative skew - mean < median < mode, long tail on the left
• mode is always highest point on the graph (most common value)
• when to use wilcoxon T test

  1. hypothesis states difference between two sets of data
  2. sets of data are pairs of scores from one person/matched pair (repeated measures or matched pairs)
  3. data are ordinal or interval
• procedure for wilcoxon

  1. calculate difference between two scores by taking one from the other
  2. rank differences, smallest rank 1 (exclude differences of 0)
  3. add up ranks for + differences
  4. add up ranks for - differences
  5. T is smallest figure of the total differences (either + or -)
  6. N is number of scores excluding 0 difference
• wilcoxon when are results significant

  when observed value T is equal to or less than critical value from table
• if there´s an R in the name of the inferential test

  the calculated value must be more than or equal to the critical value
• if there is no R in the name of the inferential test

  the calculated value must be less than or equal to the critical value
• when to use mann whitney u test

  1. hypothesis states difference between two set of data
  2. independent measures
  3. data is ordinal or interval
• when to use binomial test

  1. hypothesis states difference between two sets of data
  2. repeated measures/matched pairs
  3. data is nominal
• procedure for binomial sign test

  1. find observed value (O)
  2. find N (number of differences found)
  3. compare observed value with critical value in critical values table
• binomial sign test - finding observed value

  1. work out difference between each participant's two scores (subtract in same direction and remove differences of 0) and identify sign of change (+ or -)
  2. add positive and negative signs - smallest total is observed value (O)
  3. use critical values table to find out if results are significant (using number of participants (removing any with difference of 0)
• binomial sign test when are results significant

  when observed value O is less than or equal to the critical value
• when to use spearman's rho

  1. hypothesis states relationship/link between two sets of data
  2. correlational study
  3. data is ordinal or interval
• equation for spearman's rho


  $r=$$1-\frac{6\times\Sigma d^2}{N(N^2-1)}$
• size and sign of Spearman's rho tells you about


  strength and direction of correlation (-1 to 1)
• spearman's rho when are results significant

  when observed value is greater or equal to critical value
• mann whitney u equation


  $U_A=$$N_AN_B+$$\frac{N_A(N_A+1)}{2}-R_A$$U_B=$$N_AN_B+$$\frac{N_B(N_B+1)}{2}-R_B$