statistical tests

Question 1

What are the two branches of statistical test?

Answer 1

parametric and non-parametric

Question 2

What is the observed value?

Answer 2

the result from a statistical test

Question 3

What is the critical value?

Answer 3

the number taken from a statistical table

Question 4

What are the types of parametric test?

Answer 4

tests used in core studies e.g., t-tests

Question 5

What are the types of non-parametric test?

Answer 5

• Binomial Sign
• Chi-Squared
• Wilcoxon Signed
• Mann Whitney U
• Spearman’s Rho

Question 6

When are parametric tests used?

Answer 6

• Normal Distribution
• Interval/Ratio data

Question 7

When are non-parametric tests used?

Answer 7

• Non-normal Distribution
• When the median (or mode) provides a better average
• Nominal, Ordinal, and Interval / Ratio data

Question 8

What are the 3 factors than can be used to identify which statistical test should be used?

Answer 8

• difference or correlation
• experimental design (independent or repeated measures)
• Level of Data (Nominal, Ordinal, or Interval0

Question 9

How do you know when to use a binomial Sign test?

Answer 9

• DV is nominal data
• Repeated Measures Design

Question 10

How do you know if there is a significant difference by the result of the binomial test?

Answer 10

Observed value ≤ Critical value

Question 11

What are the steps to the binomial tests?

Answer 11

1- assign a sign to conditions (+ or -)
2- ignore the same results across conditions
3- add up frequency of each sign
4- use least frequent sign (observed value)
5- n= total participants aside from those ignored
6- O ≤ C

Question 12

How do you know when to use a Wilcoxon test?

Answer 12

• ordinal/interval data
• repeated measures design

Question 13

How do you know there is a significant difference in a Wilcoxon test?

Answer 13

Observed Value ≤ Critical Value

Question 14

What are the steps for a Wilcoxon test?

Answer 14

1- calculate difference (condition A - condition B)
2- rank (smallest to largest)
- ignore signs (for now)
- ignore any 0 value
3- use least frequent sign (positive or negative)
4- add up ranks of least frequent sign = observed value
5- compare to critical value
6- N= participants (do not use ignored)
7- O ≤ C

Question 15

How do you know when to use a chi squared test?

Answer 15

• nominal data
• independent measures

Question 16

How do you know there is significance in a chi squared test?

Answer 16

Observed value ≥ Critical value

Question 17

What are the steps to a chi squared?

Answer 17

1- calculate row and column totals
2- calculate expected value for each condition
3- use formula for each condition
4- add up formula results for an observed value
5- compare to the degrees of freedom

Question 18

What is the formula to find the expected value in a chi squared?

Answer 18

row total x column total / overall total

Question 19

How do you find the observed value in a chi squared?

Answer 19

(observed-expected)²/ expected

Question 20

What is the degrees of freedom equation in a chi squared?

Answer 20

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

Question 21

How do you know when to use a spearman’s rho test?

Answer 21

if its a correlation

Question 22

How can you tell if a relationship is significant when using a spearman’s rho test?

Answer 22

Observed value ≥ Critical value

Question 23

What are the steps to a spearman’s rho test?

Answer 23

1- rank each variable (lowest to highest)
2- calculate difference between the ranks of the two variables (variable 1 - variable 2)
3- square the difference
4- total up all of the squared differences
5- use formula = observed value
6- compare observed value to critical value table

Question 24

How do you know when to use a Mann Whitney U test?

Answer 24

• independent measures
• interval data

Question 25

What are the steps to a Mann Whitney U?

Answer 25

1. rank all participant results (lowest - highest)
2. put the ranks of participants into the og table
3. add up all the ranks for each condition
4. use the smallest total
5. formula
6. O ≤ C