Descriptive Statistics

Question 1

LEVELS OF MEASUREMENT: NOMINAL

Answer 1

 Are you bored?
 If I ask you whether this lecture is boring you could reply “yes” or “no”.
 Therefore, people fall into two categories: bored and not bored.
 There is no indication as to exactly how bored the bored people are and therefore the data are merely labels, or categories into which people can be placed.
 Bored/Not Bored

Question 2

LEVELS OF MEASUREMENT: ORDINAL

Answer 2

 We could order people according to how bored they were ‐very bored, moderately or not bored.
 These labels do tell us something about the level of boredom.
 In using ordered categories we now know that the most bored person was more bored than the least bored person.
 very bored/moderately/not bored

Question 3

LEVELS OF MEASUREMENT: INTERVAL

Answer 3

 Interval data are scores that are measured on a scale
along the whole of which intervals are equal.
 For example, rather than asking people if they are bored we could measure boredom along a 10 ‐point scale (0 being very interested and 10 being very bored).
 For data to be interval it should be true that the increase in boredom represented by a change from 3 to
4 along the scale should be the same as the change in boredom represented by a change from 9 to 10.

Question 4

LEVELS OF MEASUREMENT: RATIO

Answer 4

 Ratio data are measured on a scale along the whole of which intervals are equal, but in addition we should be able to say that someone who had a score of 8 was twice as bored as someone who scored only 4. Also, in ratio measurement there is always an absolute zero that is meaningful.

Question 5

EXAMPLES OF DATA TYPES

Answer 5

 Nominal (Unordered categories)
 Gender (male/female); Blood group O, A, B, AB

 Ordinal (Ordered categories)
 Rank: 1st, 2nd, 3rd
 Level Of Agreement: No, Maybe, Yes

 Interval (no natural zero)
 Fahrenheit Temperature Scale

 Ratio (natural zero)
 ruler: inches or centimeters
 weight (continuous)
 number of asthma attacks (discrete)
 Many health scales are treated as ratio (e.g. Short Form
[SF]‐Health Survey Questionnaire 36). The responses to
the questions on each scale are summed to provide a scores between 0 and 100. [a score of 100 is equivalent to no disability].