Results 1 to 7 of 7

Math Help - Averaging discrete variables

  1. #1
    Newbie
    Joined
    Aug 2008
    Posts
    4

    Averaging discrete variables

    I'm doing some data analysis for an experiment that I conducted involving serial dilutions.

    Because of the nature of the experiment, my variable of interest (MIC), can take on discrete values of:
    15, 7.5, 3.75... 15/(2^i) where i = 0, 1, ..., n.

    Now, I'm a little confused. I have done triplicates of the experiment, and I was about to average the data. But I'm not sure if it makes sense to average discrete values such as these (I'm ok averaging discrete values of 1,2,3,4 etc.)

    Let's say I get values of 7.5, 3.75, and 3.75. Does averaging it even make sense?

    Hope someone can help me. I've been trying to figure it out. Thanks!
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Flow Master
    mr fantastic's Avatar
    Joined
    Dec 2007
    From
    Zeitgeist
    Posts
    16,948
    Thanks
    5
    Quote Originally Posted by xtrios View Post
    I'm doing some data analysis for an experiment that I conducted involving serial dilutions.

    Because of the nature of the experiment, my variable of interest (MIC), can take on discrete values of:
    15, 7.5, 3.75... 15/(2^i) where i = 0, 1, ..., n.

    Now, I'm a little confused. I have done triplicates of the experiment, and I was about to average the data. But I'm not sure if it makes sense to average discrete values such as these (I'm ok averaging discrete values of 1,2,3,4 etc.)

    Let's say I get values of 7.5, 3.75, and 3.75. Does averaging it even make sense?

    Hope someone can help me. I've been trying to figure it out. Thanks!
    Is each value equally likely to occur? What are you trying to achieve by finding an average of the measured values?

    The problem as presented is too vague.
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Aug 2008
    Posts
    4
    Hi

    The test I'm doing is an assay with 2x serial dilution. I am trying to determine the minimum inhibitory concentration of an antibiotic against a bacteria strain. The concentration values of the antibiotic across the 10 wells decreases by 2x in each well. Hence with a starting value of 15 ug/mL, the next is 7.5, 3.75 and so on. Due to the nature of the assay, the minimum inhibitory concentration is the smallest value that has no bacteria growth.

    Hence if the 1.875 ug/mL is the first well with bacteria growth (10, 7.5 and 3.75 have no growth), 3.75ug/mL is the experimental minimum inhibitory concentration.

    So if my experiment yields MICs of 7.5, 3.75 and 3.75, does it make sense to average it?

    Hope this is clear. Thanks for the help!
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Flow Master
    mr fantastic's Avatar
    Joined
    Dec 2007
    From
    Zeitgeist
    Posts
    16,948
    Thanks
    5
    Quote Originally Posted by xtrios View Post
    Hi

    The test I'm doing is an assay with 2x serial dilution. I am trying to determine the minimum inhibitory concentration of an antibiotic against a bacteria strain. The concentration values of the antibiotic across the 10 wells decreases by 2x in each well. Hence with a starting value of 15 ug/mL, the next is 7.5, 3.75 and so on. Due to the nature of the assay, the minimum inhibitory concentration is the smallest value that has no bacteria growth.

    Hence if the 1.875 ug/mL is the first well with bacteria growth (10, 7.5 and 3.75 have no growth), 3.75ug/mL is the experimental minimum inhibitory concentration.

    So if my experiment yields MICs of 7.5, 3.75 and 3.75, does it make sense to average it?

    Hope this is clear. Thanks for the help!
    I'd suggest you need to collect much more data before a sample mean can be meaningful.
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Newbie
    Joined
    Aug 2008
    Posts
    4
    My question is not an issue about collecting more data points actually. Let me rephrase the question.

    If the variable can only take on values of, let's say 1, 10, 100, 1000 (instead of a continuous variable from 0 to 1000), can it be meaningfully averaged?
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Flow Master
    mr fantastic's Avatar
    Joined
    Dec 2007
    From
    Zeitgeist
    Posts
    16,948
    Thanks
    5
    Quote Originally Posted by xtrios View Post
    My question is not an issue about collecting more data points actually. Let me rephrase the question.

    If the variable can only take on values of, let's say 1, 10, 100, 1000 (instead of a continuous variable from 0 to 1000), can it be meaningfully averaged?
    It depends on what you want to use the average for.

    If you roll a die (whether it be biased or unbiased) and note the number of spots each roll, you can obviously calculate the average number of spots. But the meaningfullness of this average depends on what you want to use it for ......
    Follow Math Help Forum on Facebook and Google+

  7. #7
    Newbie
    Joined
    Aug 2008
    Posts
    4
    Well, but in a die roll, the numbers are evenly spaced: 1, 2, 3, 4, 5, 6. Let's say we have a die with number 1, 10, 100, 1000, 10000, 100000. Would that average even mean anything? (This may not be a great analogy, since probability governs a die roll, and not really the experiment)
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Discrete random variables
    Posted in the Advanced Statistics Forum
    Replies: 4
    Last Post: October 11th 2010, 11:15 AM
  2. Discrete Random Variables
    Posted in the Statistics Forum
    Replies: 4
    Last Post: April 13th 2010, 03:39 AM
  3. Discrete Random Variables
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: April 27th 2008, 06:55 AM
  4. Discrete Variables
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: March 12th 2008, 12:34 AM
  5. Discrete Random Variables
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: January 29th 2008, 11:48 PM

Search Tags


/mathhelpforum @mathhelpforum