Choosing a statistical test

Question 1

Why are statistical/inferential tests used?

Answer 1

• to determine whether a significant difference or correlation exists at the p= 0.05 level which assures that the relationship or difference is significant with a 95% confidence rate leaving 5% due to chance
• also allows us to decide whether we can accept our alternative hypothesis & reject the null hypothesis

Question 2

What are the three factors used to decide a statistical tests?

Answer 2

• whether a research is looking for a difference or correlation
• in a case of difference, what experimental design is being used
• the level of measurement

Question 3

What is the first decision that needs to made when choosing a statistical test?

Answer 3

• is the researcher looking for a difference (one tailed or two tailed) between two groups
• or correlation (relationship, association)

Question 4

What is the second decision that needs to made when choosing a statistical test?

Answer 4

• What is the experimental design
• Matched pairs and repeated measures = related > p’s are the same or similar
• independent groups = unrelated > p’s are different
• choose from unrelated or related

Question 5

What is the third decision that needs to be made when choosing a statistical test?

Answer 5

• quantitative data divided into different levels of measurement
• nominal, ordinal and interval

Question 6

Recite table of statistical test

Answer 6

• nominal, ordinal, interval
• chi^2, sign test, chi^2
• Mann-Whitney, Wilcoxon, spearmans Rho
• unrelated t-test, related t-test. pearsons r

Question 7

What is the mnemonic to remember statistical test table?

Answer 7

Carrots should come mashed with salt under roast potatoes

Question 8

What are the characteristics of nominal data?

Answer 8

• no natural order
• categorical data
• discrete, item can only appear in one of the categories
• can be subjective
• can be quantitative or qualitative
  e.g. sex, eye colour, favourite team

Question 9

What are the characteristics of ordinal data?

Answer 9

• has a natural order
• difference between values are not meaningful
• subjective
• can be qualitative or quantitative
  e.g. rating scale, rank, shoe size, behavioural score

Question 10

What are the characteristics of interval data?

Answer 10

• has a natural order
• difference between two values is meaningful
• most scientific > objective
• quantitative only
  e.g. weight, temperature, time

Question 11

What statistical test would you need to calculate the degree of freedom (df)?

Answer 11

• Pearson r
• unrelated t-test
• related t-test
• chi^2

Question 12

How do you calculate the df when using Pearson r?

Answer 12

• N-2
  N= no of p’s

Question 13

How do you calculate the df when using related t-test?

Answer 13

• N-1
  N= no of p’s

Question 14

How do you calculate the df when using unrelated t-test?

Answer 14

• Na + Nb -2
  Na= number of p’s in condition A
  Nb= number of p’s in condition B

Question 15

How do you calculate the df when using chi^2?

Answer 15

• (number of rows - 1) x (number of columns -1)

Question 16

What statistical test should the calculated value be less than or equal to the critical value to be significant??

Answer 16

• sign test
• Mann Whitney
• Wilcoxon

Question 17

What statistical test should the calculated value be more than or equal to the critical value to be significant?

Answer 17

• spearmans Rho
• unrelated t-test
• related t-test
• pearsons r
• chi^2
  RULE OF R