# Thread: Using The Correct Statistical Test

1. ## Using The Correct Statistical Test

Hi there, I'm new to the forum. I'm off to uni in September to study Maths and am trying to do a bit of work on my weakest area, Statistics. I am struggling to see how you know which test is appropriate to which situation.

I have a table containing 30 students coursework marks, overall marks and their sexes, and am trying to answer the following question:

Select the correct test for each situation:
1. a test to see if the male and female overall marks have equal variances.
2. a test to see if the mean overall mark is different for
males and females.
3. a test to see if there is a significant difference between the proportions of males and females that pass.
4. a test to see whether the second coursework mark is on average higher than the first.

Probably very basic for someone on here, but I just don't understand how you know what test to use in each case. Any advice would be much appreciated.

2. Originally Posted by stanger2005

I have a table containing 30 students coursework marks, overall marks and their sexes, and am trying to answer the following question:

Select the correct test for each situation:

What is the actual question?

3. Probably very basic for someone on here, but I just don't understand how you know what test to use in each case.
Z, Student-T, Chi-Squared and F tests are all specific to present information. Have a look here for common tests we run and the conditions under which we run them: Common test statistics

4. Originally Posted by stanger2005
1. a test to see if the male and female overall marks have equal variances.
Try the ANOVA test

Originally Posted by stanger2005
1. a test to see if the mean overall mark is different for
males and females

Hypothesis test for difference in means (two sided)

Originally Posted by stanger2005
1. a test to see if there is a significant difference between the proportions of males and females that pass.

Hypothesis test for difference in proportions (two sided)

Originally Posted by stanger2005
1. a test to see whether the second coursework mark is on average higher than the first.
Hypothesis test for difference in means (one sided)

5. ## Some clarification

I think the last statement needs to be clarified. For finding out whether the average second coursework grade (for men and women) is higher than the first one paired test should be used instead because we average not the marks themselves but the differences like: (man or woman) kth observation second mark - kth observation first one. Paired T-test seems to be a good choice.

6. Thank you all for your suggestions, most appreciated! No doubt I'll have more questions as I explore statistics further.