Stats Tests

• Criteria for Mann Whitney.

  • Independent variables.
  • Test of difference.
  • Ordinal/interval data.
• Mann Whitney Method:

  1. Rank the data together,
  2. Find the sum of the ranks for each group,
  3. Use the formula for the smallest sum,
  4. Use the critical values table to compare the the u value.

  
• Criteria for wilcoxon:

  • Repeated measures,
  • Test of difference,
  • Ordinal/interval data.
• Method for Wilcoxon:

  1. Calculate the differences between the scores,
  2. Rank the differences (Ignoring positive/negative),
  3. See how many positive and negative numbers you have,
  4. Find the sum of the ranks for the least frequent differences,
  5. Uce the critical value tables to find the critical value.
• Criteria for Spearman's:

  • Correlation,
  • Looking for a relationship,
  • Ordinal/interval data.
• Spearman's method:

  1. Rank both sets of scores separately,
  2. Find the differences between each score,
  3. Square the difference,
  4. Find the sum of the differences squared,
  5. Apply the formula.
  6. Use the critical values tables where n=number of pairs of scores.

  
• Criteria for Chi squared:

  • Independent measures,
  • Looking for a difference,
  • Nominal data.
• Method of Chi squared:

  1. Add totals for each column,
  2. Calculate observed and expected frequencies,
  3. Apply formula,
  4. Chi squared value is the sum of all cells.
  5. Use the critical values table where degrees of freedom = (number of rows - 1)(number of columns - 1)

  
• Criteria for Binomial:

  • Repeated measures,
  • Looking for a difference,
  • Nominal data.
• Binomial method:

  1. Determine positive and negative values for the data set,
  2. Add each positive and negative signed direction,
  3. Smallest direction score is the observed value,
  4. Use critical values table.
• The observed value is significant when it is gReater than the critical for stats tests that include an R (Spearman's & Chi squared).
• In a significance statement you must have:

  1. Comparison of observed and critical value,
  2. Indicate if the results are significant or not at p=0.05,
  3. Indicate which hypothesis we will accept (and outline this).
• In a conclusion you must have:

  1. If results are significant or not at the significance level,
  2. An inference (this could be because...),
  3. State which hypothesis we will accept/reject.