Greife auf kostenlose Karteikarten, Zusammenfassungen, Übungsaufgaben und Altklausuren für deinen inferential statistics and testing for differences Kurs an der University of Manchester zu.

- 10873 Karteikarten
- 437 Studierende
- 0 Lernmaterialien

Q:

What are inferential statistics?

A:

- Used to make inferences from our data to more general conditions

- they allow us to make probabilistic statements about the truth of hypotheses

Q:

What is hypothesis testing?

A:

- a technique for using data to validate or invalidate a claim about a population

- you set your research questions first then hypothesis for this

Q:

What are the 5 key steps to hypothesis testing?

A:

1. State statistical hypotheses

2. Set significance level

3. Calculate test statistic

4. Evaluate the statistic

5. Interpret your results

Q:

What two statements does state statistical hypotheses have?

A:

They have two opposing statements about a population characteristic:

- null hypothesis

- alternative (research) hypothesis

Q:

What is Null Hypothesis?

A:

- research question is written as a statement that can be rejected

- looking for the relationship between the data

- usually expressed in the form “there is no significant difference“ or “there is no relationship between“

Q:

What is the Alternative (research) Hypothesis?

A:

- if the null hypotheses is rejected by the statistical test used then it is logical to favour the alternative hypothesis

Q:

An example of a statistical hypotheses:

A:

RQ- ‘to what extent is the precipitation in areas over 500m in altitude different to the precipitation in areas below 500m in altitude?’

Null: there is o significant difference in precipitation between areas above and below 500m altitude

Alternative: there is a significant difference in precipitation between areas above and below 500m altitude

Q:

3. How to calculate test statistic?

A:

- each inferential statistical test involves the calculation of a single value called the test statistic

- examples include: t, R, F and Z values

- gives an indication of the strength of the pattern observed in the data

- enables the significance of the result to be calculated

- (more likely to be significant)

- once you have statistic you might have to do another test to see if the test is significant or not

Q:

4. How to evaluate the statistic: automatically using SPSS

A:

- it will calculate significance value for you

- to be significant, the resultant sig value should be smaller than the sig value you have set.

- i.e. If you have set the sig level at 0.05, and the software gave you a sig value of 0.3, this would be significant at the chosen level

- we are usually looking for sig values <0.05 (lower)

Q:

5. How to interpret results

A:

- if the result is significant, we would “Reject null in favour of alternative”

- rejecting the null suggests that the alternative hypothesis may be true

- if the result is not significant: “do not reject null“

- “do not reject the null” does not necessarily mean that the null hypothesis is true, it only suggests that there is not sufficient evidence against null in favour of alternative

- the final conclusion is always given in terms of the null hypothesis

- we always conclude “reject alternative” or even “accept alternative“

Q:

what test do you carry out if the variances of the two datastes are equal?

A:

Levene’s test for Homogeneity of Variances

Q:

What does the data show in SPSS?

A:

- you will get the test statistic (F) and a sig value

- if you get low sig. value > low probability the differences have occurred by chance > the groups have unequal variances

- if the sig value is large > high probability the differences have occurred by chance > the groups have equal variances

Greife kostenlos auf tausende geteilte Karteikarten, Zusammenfassungen, Altklausuren und mehr zu.

Jetzt loslegen