Introduction to inferential statistics

Question

What is a Population?

Answer 1

A population is an entire group (or entire set of scores) that is of interest to you as a researcher.

Answer 2

1. Calculate z-score

2. Find z-score in Appendix A

3. find the percentages around the z-score.

Using appendix you can calculate different types of probabilities.

If you that the z-score is from a normal curve, you can work out its probability (using appendix A)

Answer 3

Sampling distribution is a frequency distribution showing means from all possible samples of a given size taken from a population.

Answer 4

The null hypothesis siginificance testing is an inferential statistics technique that produces accurate probabilities about samples when the null hypothesis is true.

You can think of NHST as a battle between the null and alternative hypotheses: if the null hypothesis is incorrect, then we can infer that the alternative hypothesis is correct; if the null hypothesis is correct, than we cannot accept the alternative hypothesis as correct.

Answer 5

1. Null hypothesis ( H0)

The null hypothesis is alwyays a statement about independence or equality in populations. (e.g. there will be no difference between two means)

2. Alternative hypothesis (H1)

Whereas the alternative hypothesis states that there will be an association or differences. (e.g. there will be a difference between two means)

Answer 6

You can test the null hypothesis by asking the following question:

Assuming that the null hypothesis is correct, what is the probability of obtaining our pattern of results?

Other words

"If there is absolutely no real effect, how likely is it that we would have got the results that we have?"

Answer 7

The level of significance (alpha) is the probability value that is the criterion for rejecting the null hypothesis; usually .05.

Answer 8

Because you concluded using the significance that the probability of H0 was very low and concluded that the alternative hypothesis is the better one. This does not proof if it is actually the better hypothesis.

Answer 9

Effect size is a measure of the size of a treatment effect or strength of association unaffected by sample size.

Effect size measures tell us the magnitude of a treatment effect or of the strength of a relationship between two variables.

It indicates the degree of the effect that the independent variable has on the dependent variable.

(effect size provides a useful piece of information that you do not get from just looking at the p-values)

Answer 10

there are two possible outcomes of a null-hypothesis significance test: (1) reject the null hypothesis, or (2) accept the null hypothesis. There are also possible truths about the population: (1) the null hypothesis is true, or (2) the null hypothesis is false. Combining these, we can have four possible situations: two kinds of correct decisions and two kinds of errors.