3. Old stuff
          3.1. Old pharm stuff (pre 2009)
              3.1.2. Statistics
 3.1.2.1. Data types 

Data types

[MG1:Chp 1, p1-p6]

Types of data

  • Qualitative data
    * Data which describes some quality
    * e.g. gender, types of admission
  • Quantitative data
    * Data which are measured on a numerical scale
    * e.g. pain scale, temperature

Categorical data

  • Qualitative data are best summarised by grouping observations into categories
    --> Often referred to as categorical data or nominal data
  • When there are only two categories, the data can also be called dichotomous data or binary data
    * e.g. gender

Ordinal data

  • When there is an order among categories
    --> Ordinal data
    * Example: Pain score, ASA score
  • There is no direct mathematical relationship with ordinal data
    * e.g. pain score of 6 is not really double that of a pain score of 3

Quantitative data

  • Commonly referred to as numerical data
  • Can be:
    * Discrete
    * Continuous
  • Discrete numerical data can only be recorded as whole numbers
    * Is usually counted
    * e.g. episodes of MI, post-tetanic count
  • Continuous data can be any value
    * Is usually measured
    * e.g. Temperature, body weight, respiratory rate

NB:

  • Respiratory rate is a special case
    * Normally counted
    * But non-whole number values (e.g. 5.5 breath/min) is still meaningful

Interval and ratio scale

  • Continuous data can be further divided into
    * Interval
    * Ratio scale
  • Ratio scale is when the data has a true zero point and any two values can be numerically related
    * e.g. Temperature in Kelvin is ratio data --> 100 degrees Kelvin is twice that of 50 degrees Kelvin
    * e.g. Temperature in Celcius is NOT ratio data --> 100 degrees Celcius is NOT twice that of 50 degrees Celcius
    * e.g. Age is ratio scale