Theory Section part 3.

Question 1

When is non-parametric techniques ideal for use?

Answer 1

Ideal for use when you have data that are measured on nominal (categorical) or ordninal (ranked) scales. They are also useful when you have very small samples and when your data do not meet the stingent assumptions of the parametric techniques

Question 2

What is one of the disadvantages with non-parametric techniques?

Answer 2

They tends to be less sensitive than the parametric techniques, and may therefore fail to detect (oppdage) differences between groups that actually exist. If you have the “right” sort of data, it is always better to use a parametric technique if you can

Question 3

Kva meiner ein med ordet “parametric”?

Answer 3

The word “parametric” comes from “parameter”, or characteristics of a population

Question 4

What is one of the different assumption between the two techniques?

Answer 4

Parametric tests make assumptions about the population from which the sample has been drawn. This often includes assumptions about the shape of the population distribution (e.g. normally distributed)
Non-parametric techniques, on the other hand, do not have such requirements and do not make assumptions about underlying population distribution (which is why they are somethimes referred to as distribution-free tests)

Question 5

What is the advantages of non-parametric techniques?

Answer 5

Few Assumptions are Required (Normality of the Sampling Distribution doesn’t matter at all)
Can obtain exact P-Values for small sample tests without assumptions about the Population Distribution (contrast with any of the Normal Distribution based tests)
Often easier to understand and less difficult mathematically
Work well on all forms of data, not only interval ratio level data
Insensitive to outliers (typically these are median based and not mean based tests)
Detailed study has shown that although there is a loss in information from the samples in a non-para test, the efficiency is only slightly less than the parametric test when the assumptions are met exactly

Question 6

What is the disadvantages of non-parametric techniques?

Answer 6

The efficiency of statistical tests can be compared by measuring the likelihood of a Type II Error (β) or Beta Error
This term is called Statistical Power and it is the probability of rejecting a false null hypothesis (1-β)
You can increase the power of a test by increasing the sample size or the level of significance (increasing the alpha error)

Question 7

Why do we need parametric tests?

Answer 7

Real world situations are often too precarious to allow a researcher to idly toss aside degrees of freedom
It is important to be able to utilize many tests so that you aren’t limited to forcing a specific test statistic despite violating assumptions
Fortunately for all of us, there are an entire class of statistical tests that make no assumptions about the population parameters

Question 8

What is Type 2 error?

Answer 8

Occurs when we fail to reject a null hypothesis when it is, in fact, false (i.e. believing that the groups do not differ, when they in fact do)

Question 9

Når testar ein for samanhengar?

Answer 9

When you wish to explore the relationship among variables (correlation)

Predict scores on a dependent variable from scores of a independent variable (simple regression)

Predict scores on one variable from scores on another variable (bivirate variable)

Identify the structur underlying a group of related variables (factor analysis)

Question 10

Når testar ein for ulikskapar?

Answer 10

When you wish to test whether there are significantly differences (in the mean scores) between groups

E.x. Is there a significant difference in the mean self-esteem scores for males and females?

Is there a difference in optimism scores for young, middle-aged and old participants?