🚀 go-pugleaf

From:   NAME: DON SCHAFFNER
        FUNC: FOOD SCIENCE
        TEL:                                  <DON AT A1 AT CAFT1>
To:     inet"STAT-L@MCGILL1.BITNET"@mrgate


        In our laboratory we develop mathematical models which
        descibe the growth of bacteria as a function of environmental
        conditions (i.e. pH, temperature, salt content, etc.).  Up until
        now we have only studied one variable (temperature) at a
        time.  We are now beginning to think about
        experimental designs with two dependent variables.

        From what I know about experimental design, most designs
        are set up around what the experimenters think will be the
        optimum conditions.  In other words, if the organism grows
        best at 37 degrees and a pH of 7, an experimental design
        might look like this:

                pH= 6     7     8

        30 degrees  2     2     2
        37 degrees  2     4     2
        42 degrees  2     2     2

        where the numbers 2 and 4 represent the number of
        replicates.

        This is fine if the 37 degrees, pH 7 condition is the most
        interesting one!

        Since we study microbes which grow in food and are either
        pathogenic, or which spoil the food, the conditions which are
        most interesting to us are the ones far away from the optimum,
        where growth either doesn't occur, or occurs very slowly.

        If we change the design so that the center of the design is
        where growth barely occurs, then we have another problem.
        Many of the conditions in this new design will show no
        growth.  It will cost a lot of time and money to produce data
        that is "0" (no growth).

        My question is this:  Is there is special group of experimental
        designs that can be used to study problems like the one outlined
        above?

        Since I'm not a statistician, and my knowledge of
        experimental design is limited, I thought I'd pose it to the
        list.  (If I'm way off base, or if my question is a silly one,
        please let me know too!)

        If you would like to respond to me directly (see email addresses
        below), I'll summarize and post to the list.

        =================================================================
        Don Schaffner                        Schaffner@Daemon.Rutgers.Edu
        Food Science Building, POB 231       Schaffner@Zodiac.Rutgers.Edu
        Extension Specialist in Food Science        Don@CAFT1.Rutgers.Edu
        Rutgers - The State University of New Jersey       (908) 932-9661
        New Brunswick, NJ 08903                        FAX (908) 932-6776
        =================================================================