Stats Tests

Question 1

What type of data does chi squared use?
What type of experimental design does it use?
Test of difference or test of relationship?
What Is the significance rule for chi squared?

Answer 1

1. ) Nominal data
2. ) independent measures design
3. ) Test of difference. Used when data is in frequency form counting number of times an event occurs. Could also be test if relationship.
4. ) Observed value needs to be equal too or exceed the critical value of data’s too be significant.

Question 2

Mann Whiteney U test what type of data?
What type of experimental design?
Test of difference or association?
When should it be used?
Significance rule?

Answer 2

1. ) ordinal data
2. ) independent groups design
3. ) test of difference
4. ) Should be used for a two condition unrelated design with different Of each condition.
5. ) observed value needs to be less than the critical value for any significance to be shown

Question 3

Unrelated t-test what type of experimental design?
What type of data?
Test of difference or association?
Significance rule?

Answer 3

1. ) Independent measures design
2. ) Uses interval or ratio data
3. ) test of difference
4. ) Observed value has to be less than the critical value for any significant difference to be shown

You’d use for difference between 2 conditions.

Question 4

Sign test what type of data?
What type of design?
Test of difference or correlation?
Significance level?

Answer 4

1. ) Sign test uses nominal data
2. ) Uses repeated measures design.
3. ) test of difference
4. ) observed value has to be equal to or less than critical value

Question 5

Wilcoxon signed ranks test

What type of data?
What type of experimental design?
Test of difference or test of association ?
Significance rule?

Answer 5

1. ) ordinal data
2. )repeated measures design
3. ) test of difference
4. ) observed value needs to be less than critical value for any significance to be shown.

Question 6

1. )Related t-test what type of data?
2. )Test of difference or of relationship?
3. ) What experimental design?
4. ) significance rule?

Answer 6

Interval or ratio data
Test of difference
Repeated measures design
Observed value has to be less than critical value for any significance to be shown

Question 7

1. )Spearman’s rho correlation what type of data?
2. )Test of relationship or test of difference
3. )Significance rule?

Answer 7

Ordinal data
Test of relationship
Observed value needs to be greater than critical value for significance to be shown

Question 8

1. ) Pearson’s correlation coefficient what type of data?

2. ) test of difference or test of relationship?

Answer 8

Interval data
Test of relationship
Observed value has to be greater than critical value for any significance to be shown

Question 9

1. )Nominal data what is it?
2. ) ordinal data what is it?
3. ) what’s interval data?

Answer 9

Dats that can be put in specific categories and involves counting and data is presented in bar charts that won’t touch.

It involves continuous data where their may be ranking or comparison with others. Data can be presented in histogram where bars touch due to continuous data.

Interval data- also uses continuous data however uses variables that are set in stone that researcher can’t change eg. Minute Is always 60 secs. Shows difference between 1st and 2nd, 2nd and 3rd etc/

Question 10

1. ) Parametric and non- parametric tests what’s the difference ?
2. ) What type of data can you use non-parametric tests on? Why are they limited

When and why should you use parametric tests?

Answer 10

The difference is the method of calculating chance probabilities.

Non-parametric tests can be used on nominal and ordinal data and they aren’t very precise.

Parametric tests can be used on interval data and are more precise. They are more likely to successfully reject null hypothesis so less likely to make a type 2 error

Question 11

1. )Sign test how to work out +’s and -‘s
2. ) How do you work out your s value?
3. ) How do you work out your N value
4. ) How to work out critical value?

Answer 11

1. ) If experiment is less than control you get a minus but if experiment is more than control you get a +.
2. ) S value is the number the less frequent sign occurs
3. ) Your N value is the number of pluses and minuses
4. ) Firstly figure out one tailed or two tailed. Then use your N value at the right significance level generally 0.05 to work out critical. Observed value must be less than or equal to critical value for results to be significant