There are four types of data: nominal, ordinal, interval and ratio.
The type of data determines the kind of analysis that you can do. Up until 1946, statisticians disagreed about what kind of analysis to run on various types of data. However, in 1946, psychologist Stanley Stevens of Harvard University classified four types of numeric data: nominal, ordinal, interval and ratio. He listed appropriate statistics that could be used for each data type.
Nominal Data
Numbers used as names or codes only are nominal data. An example of nominal data is to number each of seven different choices of television programs in any given evening. Number one is not greater or lesser than the other numbers on the list, it is simply a designation or short-hand to list names of categories. The mode is the statistic of central tendency to use with nominal data. The mode is a count of which category is most frequently chosen.
Ordinal Data
Numbers that are rank ordered are ordinal data. An example of ordinal data are the top four ranks in a race: 1st, 2nd, 3rd and 4th place. The order of number one is better than number two, which is better than number three, which is better still than number four in ordinal data. The distance between any numbers in ordinal data is not equal; the first place finisher in the race may have completed the race 12 hours before the second place finisher who completed only 20 minutes before the third place finisher. The median is the middle number of a set of data. Either the median or the mode can be used as measures of central tendency with ordinal data.
Interval Data
Number scales that have equal distances between each value and where magnitude can be assessed by numeric value are interval data. The value of zero on an interval scale is arbitrary - it does not mean the absence of what is being measured. An example of interval data is a temperature scale. When it is 0 degrees outside, it is cold but there is not an absence of temperature. Another example is dates on a calendar. April 25th is 24 days later than April 1, but April 0 does not exist. The mode and median can be used on interval data as well as the average, which is a the sum of all of the values divided by the total number of values.
Ratio Data
Number scales with a true zero and equal distances between values are ratio data. Data in the natural sciences is most often ratio data. Examples include centimeters or inches on a ruler, height values and age. The value of zero on a ratio scale means the absence of what is being measured. The mode, median and average can all be calculated on ratio data.
Tags: ordinal data, place finisher, than number, absence what, absence what being