Summarising Data

Question 1

What are the X2 types of data that exist?

Answer 1

1) quantitative

2) categorical

Question 2

What is quantitative data?

Answer 2

Any data that can be measured (e.g. height).

Question 3

What is categorical data?

Answer 3

Data which can not be measured but instead falls into one of a number of given categories (e.g. male or female).

Question 4

What are the X2 subcategories of quantitative data?

Explain what each of these mean.

Answer 4

Continuous
= data values which are on a continuum (e.g. height)

Discrete
= data which can only take certain values (e.g. number of children in a family)

Question 5

What is categorical data called when it has only X2 categories?

Answer 5

Dichotomous or binary data.

Question 6

What is categorical data that has greater than X2 categories split into?

Explain what each of these means.

Answer 6

Ordered or ordinal
= follows a logical order (e.g. stages of cancer = 1, 2, 3…)

Unordered or nominal
= follows no set order (e.g. nationality = English, French, Spanish…)

Question 7

When finding the centre of quantitative data, what is meant by the mean?

Answer 7

This is the arithmetic average found by adding all of the values and dividing the total by the number of data pieces.

Question 8

When finding the centre of quantitative data, what is meant by the median?

Answer 8

This is the middle data value when the data pieces are ranked (lined up).

Question 9

What is standard deviation?

Answer 9

Is is a value which expresses how far an individual piece of data lies from the mean value of the data set.

Question 10

What is variance and how is it calculated?

Answer 10

It tells you how spread out the data in a set is.

It is the standard deviation value squared.

Question 11

What would a positive skew in a histogram look like?

Answer 11

It would have the highest peak far to the left and a long tail leading to the right.

Question 12

What would a negative skew in a histogram look like?

Answer 12

It would have the highest peak far to the right and a long tail leading to the left.

Question 13

Which type of skew is most common in medicine?

Answer 13

A positive skew.

Question 14

What percentage of data lies within:

(a) +/- 1SD
(b) +/- 2SD

…of the mean in a normal distribution?

Answer 14

(a) 68%

(b) 95%

Question 15

What is 2SD more specifically?

Answer 15

1.96

Question 16

How can you deduce a positive skew from mean and standard deviation values alone?

Answer 16

If the standard deviation > half the mean then then the lower limit of the95% CI must be below zero.

For data which MUST be positive (e.g. number of days of stay on a ward) then this MUST infer a positive skew!