Good Data

Good Data.

Qualitative data: Qualitative data is information about qualities; information that can't actually be measured. Some examples of qualitative data are the softness of your skin, or the colour of your eyes, (opinion based/subjected ideas). however, try telling Photoshop you can't measure colour with numbers. 

Quantitative data: Quantitative data are measures of values or counts and are expressed as numbers. Quantitative data are about numeric variables (e.g how many; how much; or how often). (measurable, and number based deals in facts and figures). 

Both Quantitative data and Quantitative can be good and can be bad, for the data to be good if must have:

  • Valid: Should be unbiased, representative and verifiable.
  • Reliable: How well the information fits with other facts the user already knows how well can the source be trusted.  
  • Timely: The information should be available when its needed for decision making and not some time afterwards. 
  • Fit-for-purpose: Was the information provided for the purpose of being used? (E.g: monthly budget prepared six months ago be relevant to the information of today? 
  • Accessible: Must be able to do calculations with the data. For example, a printed report may be valuable, but if it contains a lot of data you would not want to have to key it all in again in order to do a calculation.
  • Cost-effective: The cost of capturing and producing the data should be very much less than the value of the decisions made on that data. 
  • Accurate: Information needs to be accurate enough, but not always 100% exact. 
  • Relevant: There is no point in capturing  information if it is not relevant to the decision you what to make from it.
  • Having the right level of detail: The user needs to have captured enough detail for the purpose that is required, but no more.  
  • From a source in which the user has confidence: The user must need to know how believable the data is. (e.g: The user may accept a story form multiple reports from different national newspapers than just one web page. 
  • Understandable by the user: The information must be at the user's level of understanding. (For example: Share-buying advice in a weekend newspaper for the general public might have one paragraph for each share. 
https://www.youtube.com/watch?v=sXYeaNLgt6Y


Bias: Sometimes information can have a bias even if it wasn't intended. (e.g: facial recondition doesn't seen black peoples faces because the software may have been trained by people with white faces or lighter toned skin).

Comparable: The information must be able to be compared to other/similar information and get the same result. (e.g: sales figures in a certain reason on a type of product). 

Comments

Popular posts from this blog

The Digital Divide.