Evidence from quantitative data

Question 1

If the distribution is symmetrical, where will the mean be?

Answer 1

Near the middle value

Question 2

If the distribution is skewed, where will the mean be?

Answer 2

Pulled in the direction of the tail

Question 3

Is the mean a robust measure of centre?

Answer 3

NO –> it is susceptible to the presence of outliers

Question 4

Is the median a robust measure of centre?

Answer 4

YES –> it does not change with the presence of outliers

Question 5

Advantages of mean

Answer 5

• Takes into account . measures (e.g. 7.5)

Question 6

Disadvantages of mean

Answer 6

• Does not carry meaningful quantitative information for data gathered from nominal or ordinal scales
• Sensitive to extreme values

Question 7

What does variance measure?

Answer 7

the extent to which each observation deviates from the mean. The larger the deviation from the mean, the greater the variability of the observations.
–> deviations can be positive (if the value was above the mean) or negative (below the mean)

Question 8

What is the 68-95-99.7 rule?

Answer 8

In any Normal distribution, approximately 68% of the scores will fall between one standard deviation below and one standard deviation above the mean. Furthermore, approximately 95% of the scores will fall between plus and minus two standard deviations from the mean. Finally, nearly all of the scores in a distribution (approximately 99.7%) will fall within plus and minus three standard deviations of the mean

Question 9

What do we use normal distribution to do?

Answer 9

• Describe the distribution of observations (e.g. height)

- Describe the distribution of statistics (e.g. sample mean)

Question 10

When should statistical hypotheses be stated?

Answer 10

BEFORE the experiment is undertaken

Question 11

What is the null hypothesis?

Answer 11

• A hypothesis of no difference

- The hypothesis to be tested but is usually the opposite of our research hypothesis

Question 12

What does the null hypothesis aim to do?

Answer 12

To decide if we can reject Ho or whether it should be retained

Question 13

What does the p value tell us?

Answer 13

The probability of observing the data by chance if there really was no underlying association

Question 14

What are type 1 errors?

Answer 14

“false positive” –> the error of rejecting a null hypothesis when it is actually true
–> i.e. deciding that there is an effect when really there is no association

Question 15

What are type 2 errors?

Answer 15

“false negative” –> the error of not rejecting the null hypothesis when the alternative hypothesis is the true state of nature
–> i.e. failing to observe an effect when in truth there is one

Question 16

How can the power of a study be improved?

Answer 16

By:

• increasing sample size
• decreasing variability in the response
• increasing the effect size you are trying to detect