Sign Test

• what is meant by statistical testing?

  • A way of determining whether hypotheses should be rejected or accepted
  • Whether differences/relationships between variables that have been found during experiments are statistically significant/ or have only occurred by chance
• What can a sign test only be used for?


  • looks for a difference - not correlation
  • uses a repeated measures design
  • collected nominal data
• What are the first 2 steps in conducting a sign test?


  1. State the hypotheses (including the alternative & null hypothesis)
  2. Record the data and work out the sign
• when will the sign be negative?


  If the value has decreased in the 2nd condition
• when will the sign be positive?


  If the value has increased in the 2nd condition
• What if the value has stayed the same?


  It will be ignored and the N adjusted to exclude it
• What is step 3 of conducting a sign test?


  3. Find the calculated value for the sign test , S.
• What is the value of S?


  The number o times the less frequent sign occurs
• What is step 4 of conducting the sign test?


  4. Find the critical value of S by using the calculated N value
• What is the N value?


  The total number of values with the ignored values excluded
• What does p$\le$0.05 mean?


  Theres a less than 5% probability that the results occurred by chance
• What happens if S $\le$ critical value?


  The null hypothesis is rejected - there is a significant difference
• What happens if S$\ge$ critical value?


  The null hypothesis is accepted - there is no significant difference
• What is step 5 of the sign test?


  State the conclusion where you refer back to the hypothesis mentioning the IV and DV and support your conclusion with the exact values of the critical value S,N and what p value was used