Data Types

Describe the different types of data

Data are a series of observations or measurements. Can be either qualitative or quantitative.

Qualitative Data

Using words as data rather than numbers, evaluating meaning and process. Common in the social sciences.

Quantitative Data

Uses numbers, or can be coded numerically. Divided into multiple types, each with multiple subtypes.

• Categorical
  Data exist in discrete categories without intrinsic order.
  • e.g. Medical specialty (intensive care, emergency medicine, orthopaedics, cardiology)
  • Descriptive statistics for categorical data can be reported using the absolute number for each category, percentages, or proportions

• Ordinal
  Data exists in discrete categories with an intrinsic order, e.g. age groups (0-5, 6-10, 11-15...)
  • Descriptive statistics for ordinal data are the same for categorical data, but they can also be summarised by the median and the range (e.g. median age group, age group range).

• Numeric
  Data is an actual number. Can be subdivided into discrete or continuous:
  • Discrete
    Can only be recorded as an integer (whole number), e.g. number of hospital admissions.
    • Dichotomous or binary data, which occurs when there are only two categories
  • Continuous
    Where data can assume any value (including fractions), e.g. white cell count.
    • Continuous data can be further subdivided into interval or ratio data:
      • Ratio data
        Are expressed with reference to a rational zero, which is where zero means no measurement.
        • e.g. Temperature in °K is a ratio variable, whilst temperature in °C is not
          This is because 0°K means no temperature, whilst 0°C does not; e.g. 50°K is half the temperature of 100°K, but 50°C is not half the temperature of 100°C.
        • Ratio variables can (unsurprisingly) be expressed as ratios, whilst interval variables can not
      • Interval data
        Do not have a rational 0 - this is just another point on the line (e.g. temperature in °C).

References

1. Myles PS, Gin T. Statistical methods for anaesthesia and intensive care. 1st ed. Oxford: Butterworth-Heinemann, 2001.