Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Design experiment with multiple nuisance factors including time

Discussion in 'DOE - Design of Experiments' started by liliput, Mar 8, 2021.

  1. liliput

    liliput New Member

    Joined:
    Mar 8, 2021
    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    1
    this is the case,
    goal : perform an experiment with 4 treatments.
    wish to run each treatment once in the morning, between 9:00 and 12:00. and once in the afternoon, between 1:00 and 4:00. all four treatments in the morning followed by all four treatments in the afternoon. and perform the test for 4 consecutive days.

    Now, days, order of run in mornings or evenings , as well as morning or evening tests seem nuisance factors. so I came to graeco.

    but each cell of graeco table shows the run of experiment for treatment A with blocks b.row , b.col and a greek letter . right.

    in this case, two of nuisances are time, so it can not be a run at like 9 am and simoultanously at 2 pm.

    what do you suggest ? what kind of a design could this be?
     
  2. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    437
    Likes Received:
    332
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    I think your best bet would be to structure this as a repeated measures design. You may also want to try posting this on the Talkstats statistical forum.
     
  3. liliput

    liliput New Member

    Joined:
    Mar 8, 2021
    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    1
    thank you for the suggestion.
    I assume for repeated measurement, same person must test a treatments effect several times under one condition. but here , each treatment is tested although probably by one person but under different conditions like different times in the morning, different day tom ( morning- evening ) and different days.
    I kinda feel confused. I am going to find sample size for this design but I m not sure what design fits perfectly here.
     

Share This Page