Docstoc

The Scientific Method History

Document Sample
The Scientific Method History Powered By Docstoc
					Lab 2: Introduction to the Scientific method (1hr, 50 minutes)
Lab Outline:

    1) I am assuming that you have (or will) read the text, which will give you more
       details and background on the history of the scientific method…today we are
       going to do an activity that lets you practice it.

    2) The Scientific Method is a very specific thing…it is a formal method for posing a
       testable question and then answering it, you will work through this today…but,
       the Scientific Method is often confused with…

    3) The Scientific Process which is less specific and encompasses the creative and
       self-correcting aspects of science. It is the nature of this process that people
       sometimes find frustrating because…

             a. It leads to changes in conclusions of science, and we tend to think of
                science as a body of facts…and facts should be unchanging.
             b. It is this very nature of the evolution of thought that makes the scientific
                process so powerful.
                      i. People who don’t understand, naturally tend to distrust it
                     ii. People who do understand it, may tend to accept it more
                         readily…sometimes that is good and sometimes maybe not.


     4) The Scientific Process begins with observations… Darwin was a master of
         observations and spent decades cataloging, describing and observing things.
              a. This is often the part that teachers leave out, but it is the most critical part
                  of the scientific process!
                           Have you ever had a teacher ask you to for a hypothesis only to
                           think…well I have no idea! In order to be able to ask a question,
                           you have to know something about what you are asking, so start
                           with observations.
                                    Observations can be original or, be the process of reading
                                    the research in the field to get an understanding of the
                                    context of the questions in a particular field.
----------------------------------------------------------------------------------------------------------
         Activity 1: Obervations
         We are interested in learning more about earthworm populations (really!)
         1) First, let’s summarize what we think we already know about earthworms…
                  a. Their biology and life histories?
                  b. Where would they likely prefer to live and why?
                  c. What things might control their distributions?

        2) Second, let’s do some direct observations to familiarize ourselves with our
           study area (we aren’t collecting data yet)…
                  a. Describe the habitat types, include things that you might think would
                     be important to earthworms…
                          i. Vegetation
                         ii. Soils
                        iii. Microclimate / habitat conditions (light, moisture, temp., etc.)

         3) Finally, let’s share our observations and devise a Scientific Theory of what
              we think determines earthworm distributions on the LSC campus.
------------------------------------------------------------------------------------------------------------

    5) Scientific Theory -Back to Darwin as an example…from his observations, he
       devised a Scientific Theory for the evolution of species based on natural
       selection.
           a. A Scientific Theory provides an overarching explanation of how we think
               something works FROM WHICH specific testable hypotheses will arise.
           b. For example, the “gene theory” explains how heredity works; “germ
               theory” explained how illness was caused by biological agents; and of
               course… “evolutionary theory” explained how species could evolve
               through natural selection.

    6) From Theory to Hypothesis…enter the Scientific METHOD…

              a. Devising a Hypothesis
                       i. A hypothesis is a testable question that predicts a certain outcome
                           based on our Scientific Theory.
                                1. For example, Darwin’s Scientific theory suggested that new
                                    species could arise by natural selection…natural selection
                                    suggested that some individuals produced more or less
                                    offspring depending on how adaptive their particular traits
                                    were. So “Fred the finch”, has a bill that is a little bit better
                                    cracking a particular nut than does “Sam the finch”. As a
                                    result, more of Fred’s offspring survive to reproduce and
                                    carry that trait with them. So,…
                                         a. The observation is that Fred has a better bill for
                                             cracking nuts than Sam
                                         b. The theory is that this will lead to differential
                                             reproductive success
                                         c. The testable hypothesis is that Fred has more
                                             offspring survive to reproduce than does Sam.
------------------------------------------------------------------------------------------------------------
         Activity 2: Devising an hypothesis
         1) Our observations about earthworm distributions on the LSC campus were…

         2) Our theory about earthworm distributions on the LSC campus were…

         3) Our hypothesis about earthworm distributions on the LSC campus is…
               a. Remember, hypotheses ask testable questions…
                       i. in the form of a prediction,
                      ii. describe the measurable parameter used,
                     iii. define the comparison being made or relationship being
                          described…
                     iv. for example, rather than saying “habitat type affects earthworm
                          abundance on the LSC campus ”, which is untestable.
                          We would say “earthworm biomass will be greater in the
                          forested habitat than the lawn habitat on the LSC campus”,
                          which is testable.

                         OK, now that we have a hypothesis…let’s design a study
----------------------------------------------------------------------------------------------------------

    7) the Scientific METHOD…continued

             a. Observational study vs. Experimental study

             b. Sampling a population
                   i. It is rarely possible to sample an entire population so we have to so
                      something called “sampling” a population.
                  ii. And there’s the trick…we want to sample a population in a manner
                      that we have confidence that our sample is representative of the
                      larger population…otherwise we are just making up stories!

----------------------------------------------------------------------------------------------------------
         Activity 3: Sampling a population
         1) What are we measuring…
                  a. Be sure the parameter you are measuring fits your hypothesis (i.e. are
                       we sampling ALL earthworms or only a certain species? Are we
                       measuring biomass or numbers?)
                  b. Be sure to collect all the needed data for your hypothesis (i.e. habitat
                       type, distance to stream. Etc.)
         2) How are we measuring…
                  a. Use a method that is accepted by the scientific community AND/OR
                       provide evidence that your method is reliable (accurate vs. precise).
         3) Where are we measuring…
                  a. Avoid bias…when in doubt randomize!
                  b. Collect enough samples to reliably estimate your population…how
                       many are enough? (performance curves).
         4) When are we measuring…
              a. Avoid bias and makes sense with the populations being sampled
                 (i.e you would not conduct a study of breeding bird in the fall since
                  they aren’t breeding).

                  OK, let’s go sample…!
---------------------------------------------------------------------------------------------------------

    With data hand, head back to the lab for analysis…

    8) the Scientific METHOD…continued
           a. Testing the Hypothesis… so, now we have data, let’s test our Ho
                   i. Statistical Analysis involved summarizing your data and then
                      using an objective test to determine if your hypothesis is correct.
                      (See the Introduction to Statistics handout for some additional
                      background which you may find useful for writing your paper)
                       i. Evaluate your sampling strategy (Excel page1) by looking at
                          your performance curves.
                              1. Did you collect sample enough to have your mean
                                  stabilize?
                              2. If not what does that mean?
                      ii. Summarize the data (Excel pages 2 and 3), in a meaningful
                          way based on your hypothesis.
                              1. means, medians and modes
                                    a. THE GOAL is to provide an estimate of the real
                                        mean, median or mode for your population if you
                                        COULD sample the whole thing.
                                    b. Will you use the mean or the median? Why?
                                              i. when the distribution is perfectly
                                                  normal the mean=median
                                             ii. depending on how non-normal the
                                                  distribution is the mean and the median
                                                  can be very different, you would need to
                                                  use the most appropriate measure for
                                                  your hypothesis…for example,
                                                  comparing Caucasian vs. African
                                                  American income levels…
                                    c. do you report your data on a per sample basis or
                                        on an area basis? Which is more meaningful to
                                        readers? Why?
                              2. variance, standard deviations
                                    a. all the sample values are not the same…Variance
                                        and S.D. represent the average amount of
                                        variability associated with your mean…
                              3. 95% confidence interval give you the range around the
                                  mean with which you can have “95%” confidence
             contains the mean of your sample (5% chance of being
             wrong)…this can be helpful when you want to compare
             two populations to see if they are really different. For
             example…
               a. if the confidence intervals do not overlap, you can
                  have high confidence (95%) that the means in the
                  two populations really ARE different.
               b. if they do overlap to large degree then that may
                  indicate that the difference in the means is more
                  probably the result of chance

iii. Statistical Analysis (Excel page 4)...this is not a class on
     statistics, but some basic understanding of what this all means
     can be helpful as you evaluate information you see and hear.
     You will be asked to do some basic statistical analysis in a
     future lab so it would be worthwhile to work through the
     template I have provided here in preparation for that lab.
         1. Basic types of statistical tests include…
                a. Regression…describes a relationship between
                    two to factors…for example as a person’s height
                    increases, on average so does their
                    weight…regression gives you a tool to
                    mathematically describe that relationship and to
                    test the probability that the relationship is real or
                    just the result of chance.
                b. T-tests…tests the probability that two means are
                    from different populations (this is what we will
                    use here to day)
                            i. Do the t-test as I demonstrate in class
                           ii. The output of interest is the p-value
                               (written here in scientific notation!)
                                   1. If the p-value is less than or
                                       equal to 0.05…you can be 95%
                                       confident that the difference
                                       between the two means is
                                       “significant”…that the difference
                                       between the mean values really
                                       represents a difference in the real
                                       world…95% confidence is the
                                       minimum standard by which
                                       virtually all scientific
                                       hypothesis are evaluated!
                                   2. If the p-value is greater than
                                       0.05…you cannot be confident
                                       that the difference between the
                                       two means represents a
                                                       difference in the real world, the
                                                       difference may well have
                                                       occurred due to change in your
                                                       sampling… If you do not have
                                                       95% confidence, then your
                                                       hypothesis is REJECTED.

Both intelligence and ethics are involved in the Scientific Process. To ensure this, most
scientific findings go through peer review to ensure that minimum standards are met!

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:116
posted:9/16/2009
language:English
pages:6