Classify and describe the different types of data, including two examples of each.
Any reasonable classification was awarded marks. Standard textbooks well describe this topic, often in their opening chapter. Broadly, qualitative (defined by some characteristic) vs and quantitative (measured on some numerical scale) data exists. These can also be described as categorical or numerical with subdivisions including ordinal, interval and ratio scales. Numerical data may be described as discrete or continuous. Appropriate examples and descriptions of each were required. For example:
Categorical data -when each individual can only belong to one of a number of distinct categories of the variable.
1. Nominal – categories not ordered but simply have a name e.g. blood group (A,B, AB, O) and marital status (married, single, widowed)
2. Ordinal – categories are ordered in some way e.g. disease staging (advanced, moderate, mild) or degree of pain (severe, moderate, mild, none)
When the variable takes some numerical value
1. Discrete – when the variable can only take certain whole numerical values e.g. the
number of visits to GP in last year, or the number of episodes of illness.
2. Continuous – when there is no limitation on the values that the variable can take e.g.weight or height.
Understanding types of data allows appropriate description and comparison with parametric or non-parametric statistics and better answers highlighted this.
In order to offer a "reasonable" classification, one need to refer to Myles and Gin where (at least in my 2000 edition) "Data Types" is the title of Chapter 1. This is an answer that benefits from a tabulated format
|Qualitative||Described by a characterstic or category||
|Quantitative||Described best by a number||
|Categorical||There is a limited range of possible (qualitative) values||
|Numerical||The variable is expressed as a number||
|Nominal||The range of possible categories is not ordered||
|Ordinal||The range of possible categories is ordered||
|Discrete||Possible categories are restricted to a range||
|Continuous||Possible range of values is on a continuum||
Shades of colour
To be a "better answer", this table would have to be followed by some statement regarding the utility of such classification. In short, if your measurement scale is nominal or ordinal then you use non-parametric statistics, whereas if you are using interval or ratio scales you can use parametric statistical tests. Parametric tests assume that sample data have been taken from a normally distributed population.