Stats testing

Question 1

What is a type 1 error?

Answer 1

A false positive that occurs when a significance level is too lenient

Question 2

What is a type 2 error?

Answer 2

A false negative that occurs when a significance level is too strict

Question 3

What are descriptive statistics?

Answer 3

Mathematical conclusions such as averages, graphs, dispersion

Question 4

What are inferential statistics?

Answer 4

When a small group is tested and then the results are generalised to the entire population

Question 5

What is a parametric test? When can you use them?

Answer 5

Calculations with the mean and standard deviation, which can only be used with normally distributed interval data

Question 6

What are the criteria for each stats test?

Answer 6

• Chi-Squared (nominal, independent, differences)
• Sign test (nominal, repeated, differences)
• Mann-Whitney (ordinal, independent, differences)
• Wilcoxon (ordinal, repeated, differences)
• Unrelated t (interval, independent, differences)
• Related t (interval, repeated, differences)
• Spearmans (ordinal, repeated, correlation)
• Pearsons (interval, repeated, correlation)

Question 7

What are the steps for sign testing?

Answer 7

1. Record data and find differences
2. Count the signs and label the lowest value as ‘S’
3. Find critical value alongside significance level
4. If the S value is less than or equal to given value, reject H(0)

Question 8

What are the steps for a chi-squared test?

Answer 8

1. Create a contingency table
2. Calculate degrees of freedom with (c-1)(r-1) formula
3. Identify critical values
4. Write conclusions

Question 9

How do unrelated and related t tests calculate degrees of freedom?

Answer 9

n-1

Question 10

How does the Pearsons test calculate degrees of freedom?

Answer 10

N-2