Levels of Measurement (A2 only)
Psychology: AQA A Level, topic 7.3.6
Nominal and Ordinal Data
There are four main types of data: nominal, ordinal, interval and ratio. The types of data will influence how they are statistically analysed.
Nominal data
- Nominal data is essentially data that fits into distinct categories.
- Nominal data is collected by counting the frequency of each category.
- An example of nominal data, in terms of biology, would be hair colour (brown, black, red, etc.)
Nominal data example
- In a study on the bystander effect, a researcher could observe the behaviour of passing individuals.
- Do they A) help or B) not help a person in need.
- This is an example of nominal data because there are two distinct categories.
- The frequency of people in both categories can be counted (out of 30 people, five were helpful and 25 were not helpful).
Ordinal data
- Ordinal data is data that falls along a scale.
- It relates measurements to the same variable.
- Ordinal data is often seen in surveys.
Ordinal data example
- An example is a Likert scale:
- Participants select from a 5-7 point scale (strongly disagree, disagree, neutral, agree, and strongly agree).
- Another example in psychological studies could be a questionnaire on depression:
- Participants could rate their symptoms on scales of 1-to-10.
Interval and Ratio Data
Both are data measured on a scale. They both use objective units and intervals. Ratio data cannot have negative readings.
Interval data
- Interval data is also a measurement taken on a scale.
- But the difference is that the unit is equally sized and is objective.
- Students can get confused between ordinal and interval data.
- For example, 1-to-10 scales can be miscategorised. As these scales can be subjective (a person’s own opinion on their depression symptoms), they are ordinal and not interval.
Objectivity
- For data to be interval, the set units and intervals must be objective.
- Examples of interval data would include time and temperature.
- Seconds are equally sized and objective. No one would argue that seconds are subjective.
- Temperature would also be interval as there is a universal standard of degrees Celsius and standardised measuring equipment.
Ratio data
- Ratio data, like interval data, are measurements taken on a scale.
- But ratio data is set to an absolute zero and there can be no negative values.
- For example, length measured in centimetres by a ruler is set to 0cm. There is no “negative length”.
Examples
- Another example would be percentage grades on exam (0-100%). There would be no negative scores.
- Temperature (measured in Celsius) would be interval and not ratio because you can have negative temperatures.
- An example of ratio data in psychological studies would be IQ tests. There is no negative IQ score.
Tests
- The different type of data is analysed using different statistical tests.
- Nominal data can be analysed using the sign test or the Chi-squared test.
- Ordinal data is analysed using Spearman’s rho, the Wilcoxon test or the Mann-Whitney test.
- Interval and ratio data is analysed using either Pearson’s r test, the unrelated t-test or the related t-test.
Test yourself on Levels of Measurement (A2 only)
Hair colour would be an example of which kind of data?
The Likert scale would be used for what kind of data collection?
Which kind of data CANNOT have negative readings?
Which kind of data is NOT measured on a scale?
Which of the following is NOT an example of ratio data?