Sources of Data

Data can be collected from different sources depending on the system in use. Often data is collected automatically as part of a transaction. This is the case when a bar-coded item is sold at a POS (Point Of Sale) terminal in a supermarket. In other situations data may be collected manually as part of the transaction - a customer paying their electricity bill would be and example of this.

Data may be collected on special forms and entered into a computer using a keyboard. Examples of this might be the entering data collected for a survey. The data might be collected on a special form that could be input directly using OMR (Optical Mark Recognition) or OCR (Optical Character Recognition). In specialised applications the data could be collected using special sensors, an example of this is meteorological data collected from automatic weather monitoring stations that can be seen by the side of many roads.

Quality of Data

The source of the data affects the quality of information the can be produced. Questionnaires may provide unreliable data because people do not answer correctly, only an unrepresentative sample may complete them, the questions may not be clear or may be misunderstood. Data that is not complete may provide misleading information.

The bar code reader will not provide all the data about stock leaving the shop. There may be theft or item may be spilt or broken. The number of bottles of lemonade on the shelf is information that can be obtained from the stock control system. The information may be wrong however if the data collection is incomplete or flawed.

The quality of data usually deteriorates with age. A market research survey may produce data that will be useful over several months or years. A Geological survey will provide data that will be useful and valuable until the resources concerned have been exploited. Detailed weather data used in forecasting is of little value after 24 hours. Data may be date stamped. The computer will automatically attach a date/time field to the data. This will allow the age of the data to be taken into account when it is processed.

© LEV