Advantages of Non-Parametric methods

Question 1

Advantage 1

Answer 1

Non-parametric tests can be used on all types of data;
data recorded as +- (nominal scaling)
ranked data (ordinal)
Data measured more accurately (ratio scaling)

Question 2

advantage 2

Answer 2

Non-parametric methods make less assumptions about the data then parametric testing procedures do.

More general and hence more relevant to a particular situation, giving a more general conclusion

Question 3

advantage 3

Answer 3

Non-parametric tests are valid even if sample size is very small and can be used regardless of the shape of the distribution

Question 4

advanatge 4

Answer 4

Generally easier than parametric statistical tests and quick to compute with small sample sizes

Question 5

Explain p-value

Answer 5

• Another approach to decide whether to reject null hypothesis
• P-value is the measure of credibility of the null hyposethis
• It is the probability of obtaining a vaue at least as extreme as the one where null hyposthesis is true
• If the p-value is small – reject null hypoethis

Question 6

Discuss Cramers contingency coeffeint

Answer 6

number should lie between 0/1

1 = perfect assosiation

Question 7

what happens when you change alpha?

Answer 7

If the alpha is modified from α = .05 to α = .01, the boundaries expand and move out from the mean of the distribution, which makes it more demanding and difficult to reject the null hypothesis.

The probability of making a Type 1 error decreases from 0.05 to 0.01.