Choosing a Statistical Test Guide

• What are descriptive statistics used for in quantitative data?

  To summarize and present quantitative data.
• Why do we use statistical tests?

  To determine whether differences or correlations are statistically significant.
• What does it mean if a difference or correlation is found to be significant?

  It means the result is unlikely to have occurred by chance.
• What can be rejected if a significant difference or correlation is found?

  The null hypothesis.
• What are the three questions to determine which statistical test to use?

  1. Are we looking for a difference or correlation? 2. What experimental design has been used? 3. What is the level of measurement?
• What are the types of experimental designs used in statistical tests?

  • Independent groups
  • Repeated measures
  • Matched pairs
• How can you identify if a hypothesis is looking for a difference or a correlation?

  By analyzing the wording of the hypothesis.
• What is an example of a hypothesis looking for a difference?


  "Boys play better scrabble than dogs."
• What is an example of a hypothesis looking for a correlation?

  "There is a relationship between the amount of burgers eaten per day and the size of the waistband."
• What are the levels of measurement for quantitative data?

  • Nominal
  • Ordinal
  • Interval
• What is nominal data?

  Data represented in the form of categories.
• How is nominal data characterized?

  It is discrete and can only appear in one of the groups.
• What is ordinal data?

  Data that is ordered but lacks equal intervals between units.
• Why is ordinal data considered "unsafe" for statistical testing?

  Because it lacks precision and is based on subjective opinions.
• How is ordinal data typically treated in statistical analysis?

  Raw scores are converted into ranks.
• What is interval data?

  Data based on numerical scales with equal, precisely defined units.
• What are examples of interval data?

  Temperature, height, weight, and time.
• What is necessary for using parametric tests?


  Interval data is required.
• What is the mnemonic to remember the order of tests for statistical analysis?

  • URR Carrots Should Come Mashed With Swede Under Roast Potatoes NOI?
• What should you learn to effectively use statistical tests?

  The rules for when to use each test.
• What is the first test mentioned for statistical analysis?

  The Unrelated t-test.
• When should the Unrelated t-test be used?

  When looking for a difference between two sets of interval level data using an independent groups design.
• What type of data is required for the Unrelated t-test?

  Interval level data.
• What is meant by a "safe scale" in statistical testing?

  A public measurement scale made up of equal units.
• What should you do to prepare for statistical tests?

  • Practice as many questions as possible.
  • Check answers against the mark scheme.
  • Understand where marks are awarded.