iSixSigma

Data

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #51359

    Silva
    Participant

    I have analyzed data and have made a mistake….
    Two Questions regarding data:
    1.  What command do you need to use to analyze the data with?
    2.  How does that command require your data columns to be set up?
     
    Thank you
    Andre

    0
    #177858

    clb1
    Participant

    The first problem is the fact that you have analyzed data.  Data that has been analyzed is no longer raw data.  Analyzing analyzed data is akin to re-cooking a meal and the results will be pretty much what one would expect after such an effort.  (This is much like the problem of what to do with ground water.  Water that has been ground, usually by the blades of a hydroelectric turbine, isn’t good for much of anything except making shaved ice for snow cones and it is the reason the far west has a shortage of water for growing food.) So, before you do anything else you will need to go out and gather fresh (raw) data to analyze.
     
    Once you have a supply of raw data you will be able to give some thought to the commands you might want to issue to it.  Commanding data, like commanding troops, is more of an art than a science but there are a few generalities worth remembering.
     
    1.        Know the personal history of the data. Knowing the background of the data will guide you with respect to appropriate commands.  For example, if the data is nominal you need to remember nominal data rarely does things by the numbers so issuing commands requiring a detailed numeric response will result in nothing but confusion.
    2.        Does your data seem to want to form up into smaller groups or do you have a bunch of militant individualists?  Data that naturally sticks together will be easier to command than data that doesn’t.  In addition, the performance you get from the groups of data will exhibit less variability than that of the individuals.
     
    You can command the data to form up in as many columns as you wish. I usually let the individual raw data point assemble in whatever column it thinks best describes its point of origin or reason for being recruited. Once all of the individual data points have picked their respective columns I’ll order them to line up in descending order or, if they are nominal, alphabetically.

    0
Viewing 2 posts - 1 through 2 (of 2 total)

The forum ‘General’ is closed to new topics and replies.