Stats Tests

Question 1

What is a null hypothesis?

Answer 1

A statement that assumes there is no sognificant difference or correlation between data

Question 2

What are the three types of stat tests?

Answer 2

Chi-squared, correlation coefficient (spearmans rank), students T- Test

Question 3

What does chi-squared investigate?w

Answer 3

The significant difference between expected and observed frequencies.

Question 4

What does a T-Test investigate?

Answer 4

The significant difference between the means of two sets of data

Question 5

What do correlation coefficient tests investigate?

Answer 5

Used to sumarise the correlation of a relationship between two variables

Question 6

What is the spearmans rank correlation? And how is it interpreted?

Answer 6

A technique that measures the strength and direction of a relationship between two variables
Rs is a number between -1 and +1.
If Rs is negative, correlation is negative.
If Rs is positive, correlation is positive.
The closer the Rs is to -1 or 1, the more ‘perfect’ the correlation

Question 7

In the spearmans rank test, once Rs is calculated how do you test the significance of the relationship?

Answer 7

Use the spearmans rank table of critical values
Compare number of pairs of measurements with the CV

If Rs is greater than the critical value, null hypothesis is rejected and there IS significant correlation.
There is less than (p) probability that the correlation is due to chance

If Rs is less than the CV, accept null hypothesis and there IS NO significant correlation between. There is a more than (p) probability that the correlation is due to chance

Question 8

When you have a value for Chi squared (X^2), how do you interperate the value?

Answer 8

Calculate the Degrees lf freedom (number of categories-1)
Compare degrees of freedom to critical value in a CV table.

If Chi squared is greater than the CV, reject null hypothesis. There IS significant difference between the observed and expected frequencies.
There is a less than (p) % probability that the difference is due to chance

If Chi squared is less than the CV, accept null hypothesis. There IS NO significant differemce between the expected and observed frequencies.
There is a more than (p)% probability that the difference is due to chance

Question 9

How do you calculate the degrees of freedom in Chi Squared tests?

Answer 9

Number of categories-1

Question 10

How many repeats of the t-test should you carry out?

Answer 10

20

Question 11

Once a t value is generated in the t-test, how do you interpret it?

Answer 11

Calculate the degrees of freedom using n1+n2 -2
Compare degrees of freedom to CV on a CV table.

If T is greater than the CV, reject null hypothesis. There IS significant difference between the means
There is a less than (p)% probability that the difference is due to chance

If T is less than the CV, accept null hypothesis. There IS NO significant difference between the means.
There is a more than (p)% probability that the difference is due to chance

Question 12

What is standard deviation?

Answer 12

A value that indicates the spread around the mean

Question 13

In normally distributed results, what lercentage pf values are found within 1 standard deviation from the mean?

Answer 13

68%

Question 14

In normally distributed results, what percentage of values are found within 2 standard deviations of the mean?

Answer 14

95%

Question 15

What is implied when standard deviations of two sets of data do or do not overlap?

Answer 15

Overlap implies there is NO significant difference between data
Not overlap implies there IS significant difference between the data

Question 16

What are confidence limits- how are they calculated?

Answer 16

Take the mean value for data and subtract the standard deviation and then add the standard deviation. These two values paired together create confidence limits for the mean.