Home  Back Topics CD Orders  INDEX 
Data encoding
Often data is encoded - each data item is represented by a code.

Examples :

  • Each item in a supermarket is coded by a barcode number.
  • Each book in a library has an ISBN number.
  • Each branch of each High Street Bank has an individual branch sort code
    Eg. 51-81-26
  • Each pupil's gender is recorded on a school's database as 'M' for male or 'F' for female.
  • Country codes for Olympic athletes may be GB, US, F etc..

When the data is input, the codes are used.

 

Reasons for encoding data:
  • The data takes up less space when it is stored.
  • The data is easier and quicker to enter because the codes are shorter.
  • Validation of the data is easier.

 

 Validation is the checking of data to make sure it is sensible data.
There may be problems with encoding data...
  • A code may not be understandable by a human.
    Eg. a barcode 5031366014047....
  • Codes may not be precise.
    Eg. A code may be used on a survey for a person's nationality...E (English), W(Welsh), S(Scottish) etc.. but sometimes a person is Anglo-Welsh...
    Extra codes would be needed...

Codes may be used on questionnaires for value judgements. For example, you may be asked how you liked a certain chocolate bar and to allocate a number - 5 for Excellent, 4 for Good, 3 for Average, 2 for Unpleasant, 1 for Disgusting.

People may interpret the value 'Good' differently or may wish to give a value such as 'Average to Good'.

Extra comments are needed to make the data more useful but it is difficult to code 'It was too tough to chew and kept sticking to my teeth'

It may difficult to interpret the analysis of questionnaire responses where value judgements are used.

 Value judgements can provide broad views of people but will be lacking in details.