Scientific Investigation

This laboratory exercise is intended to introduce you to the methods used by scientists to conduct a scientific inquiry. These methods are applicable regardless of the discipline of science that you decide to pursue. The contents of this laboratory will lay the groundwork for not only the work you will do in this course; it will establish the basics you will use throughout your career as a scientist.

Objectives:

Students should be able to satisfy the following upon completion of this exercise:

  1. Recognize and develop a question that can be addressed through scientific investigation.
  2. Recognize and develop a testable hypothesis.
  3. Identify and describe the components of a good scientific study.
  4. Develop an experimental design for a scientific investigation.
  5. Present the results of a scientific study graphically and defend your findings.
  6. Understand the contents of the introduction, methods, results, and discussion sections of a research paper.

Readings

You should always read the laboratory exercise and related materials before coming to lab so that you will be prepared to carry out the procedures for that day. Reading assignments for this week's lab are:

Introduction

Science consists of two major components.  The first is a body of knowledge that has been built on since humans first started gathering information about the world around them (dating back to to the Greeks and Romans). The second is adding of information to this body of knowledge through continuing scientific investigation.  These, however, are not independent attributes as we need to know and understand what has been contributed earlier to understand where we should proceed today.  

Biology is no exception.  It is interesting to note that the process of science proceeds the same for all disciplines of biology, from studies of cells to studies of ecosystems, despite the obvious differences in the scope of these fields.  The process can be defined in five words: observation, questioning, hypothesis, testing, and explanation.  These words, taken in this order, define the scientific method, the set of procedures biologists follow for sound and rational approach for proper practice of their discipline.  The emphasis here is that they are taken in this order.  A hypothesis cannot be established without knowing what is being observed or what is the nature of the question.  It is through the proper adherence of these procedures and through continuing efforts, that biological theories and concepts are established.

    Observation: Science, no matter the discipline, is rooted in observation.  Think about how many times you have observed the way something looks or the way something moves.  Scientific investigations begin in this manner.  For example, you might be on a bird watching trip and observe that the same species of bird that you have been watching during a walk takes on a different coloration from one area to another.  Noticing in the examination of medical data that the incidence of a disease is much higher in a specific area is another example of an observation.

Question: An observation usually, often at the same time that the observation is made, leads to the formulation of a question.  From our examples above, the questions might be: Why is the color of this bird species changing in a fairly small area? and Why is the incidence of this disease higher in this region when compared to other areas?  Questions come in a number of forms, but for it to applicable in the scientific method, it must be a testable question.  A testable question is a question for which an experiment or investigation can be established and for which a hypothesis can be generated.

Hypothesis: The best definition for a hypothesis is that it is a preliminary answer to your initial question.  This may sound strange and might even sound like cheating - are you not setting things up to get the answer you want?  Neither is try, as the object of later testing is to attempt to find your hypothesis to be false.  Also, you might not have any idea what the answer to the question is (it is not much of a question if you already know the correct answer, anyway).  How is a hypothesis established?  There are three possible sources: (1) your own experiences and knowledge in the field; (2) information that is available from the existing body of knowledge (scientific literature); and (3) a best guess.  If you do not have information on the topic, a guess may be the only option.  It is important to remember that the guess does not have to be correct, as you are trying to establish something that can be tested. The critical thing is that it must be a testable hypothesis.

Testing:  Now it is time to determine the validity of your hypothesis.  This process can proceed through observation, experimentation, or other types of studies.  An important thing to keep in mind for the testing phase is whether or not the results of your efforts are going to be sufficient to provide a strong test of your hypothesis.  It is important that random events do not skew your findings in a manner that does not reflect your experiment.  This is essential because it is often difficult to separate random variation from a response that is related to the nature of your study.  Let's say you are interested in the effect of a pesticide on the growth of a fish.  Do you look at just one concentration and determine how much the fish grow over the course of the experiment?  If you do, to what do you compare your findings so that it is possible to ascertain if the effect is from the pesticide?  A more accurate approach to this question, and the testing of your hypothesis, would be to establish a variable, in this case, pesticide concentration, and test a range of pesticide concentrations from low to high (treatments), in your experiment.  You also do not want to rely on a single aquarium for each concentration - what happens if one group of fish dies or you get a bunch of naturally "slow growers" in a treatment where the is a low concentration.  Each treatment is replicated, that is there is more than one group of fish tested at each concentration.  Replication is applicable to all types of tests, including simple observation and studies in nature.  By using replicates, an average or mean response to a treatment can be determined.  This approach reduces the effects of potential problems or unusual responses in your data.

An additional consideration in an experiment is that if you are using living organisms, they are going to be handled or dealt with in a manner that might induce stress.  It is for this reason that a control is often added to an experiment as an additional treatment.  In our example, the control would be a treatment where no pesticide is added to the aquaria.  The fish placed in the control would be dealt with in the same manner as all of the other fish, except that no pesticide would be added.  This allows you to compare the survival and growth of fish under "normal" conditions to the treatments where pesticide was added.

Explanation: Once you have tested, and probably retested, your initial hypothesis, it is now time to explain or interpret your findings.  This is where the body of knowledge comes in again.  Even if you are looking at something no else has studied directly, it is likely that someone has published a study in the scientific literature that is comparable.  Such information can be used to support your findings, providing additional validity to your hypothesis.

Consistent confirmation of a hypothesis, combined with the lack of credible exceptions, and time, a scientific explanation can reach the status of theory.  To the layperson, a theory is often just an idea or possibility.  Hence, you often hear someone say, "it is just a theory."  This is not the case for a scientific theory, which goes through rigorous examination and testing over time.  It is only when a strong body of evidence has been generated that fails to refute a scientific explanation that it is raised to the status of a theory.  It becomes evident, then, that the establishment of a theory is not something taken lightly in science. 

The following exercises are designed to introduce you to the process of science from observation to explanation.  An answer sheet can be downloaded using the following links:

Answer Sheet for Laboratory 1

You can type your answers in directly on your laptop (my preference).


Exercise 1 - Questions, Hypotheses and Hypothesis Testing

Pick up some 8.5" x 11" and 8.5" x 5.5" paper from the front desk. The first question you want to address is "how many times can you fold a sheet of 8.5" x 11" paper in half until it no longer can be folded?" Develop a testable hypothesis. Remember that a hypothesis is a preliminary answer based on your own experience, the experience of others (acquired through reading scientific literature) or a "best guess" if you have no prior experiences from which to draw upon. Remember, also, that a hypothesis does not have to be the correct answer; it is intended to give you a point from which you can start your experiment. Determine how best to proceed with your experiment so that you reach your conclusions without bias. Did you accept or reject your hypothesis? Where do you go from here?

In part 2, come up with a question, hypothesis, test your hypothesis, and generate a conclusion using the 8.5" x 5.5" paper.


Exercise 2 - Developing an Experimental Design

(Exercise 1.2 and 1.3) Each group will be given some information about a relevant scientific question. Use this information to develop a hypothesis and an experiment that would be suitable for testing your hypothesis. Be sure that your design specifically addresses the original question and will give you a good test of your hypothesis.

Example: you are interested in the differences in weight gain for adding one of three possible supplements to the food of cattle. The question would be "which dietary supplement provides the greatest weight gain when added to the food of cattle." You would next need to develop a hypothesis and an experimental design to test that hypothesis.

A good working hypothesis in this instance might be a null hypothesis. Null hypotheses are commonly used in scientific investigations because they can be easily tested using statistics. "There will be no difference in weight gain among the dietary supplements" would be a good null hypothesis for this example. The experimental design needs to consist of four groups of cattle. Each of three groups would get one of the food supplements, while the fourth group would not receive a supplement and would serve as the control.


Exercise 3 - Presenting Scientific Data

(Exercise 1.4) Each group will be given a set of data that might have been generated by an experiment similar to the one for which they created an experimental design (NOTE: it might not match your design exactly). Present the data in the appropriate graphical format. Write a suitable figure heading and a brief summary of the findings from the study.

Example: The experiment above should result in average change in weight (Final weight - Initial weight) of the cattle. You would have four average weights (the groups of cattle fed each of the three supplements and a control group given no supplement). This would be put in a bar graph because it represents data from a single point in time and is not continuous. If you had measured cattle weight on a regular basis, say weekly, then it would be continuous and would be put in a line graph.

WE WILL STOP HERE FALL 2011


Exercise 4 - Writing in a Scientific Format

You will be given information pertinent to the introduction, methods, results, and discussion sections of a research paper and will need to write 1-2 sentences in the proper format of a research paper. You may create work groups, but everyone must submit their own assignment. Refer to the reading assignments listed above and the web page Writing Scientific Papers for information on writing scientifically.  Use the Scientific Writing Assignment Sheet for completing this assignment.

Exercise 5 - Scientific Literature

The most important information on science is in professional scientific journals.  Journals are peer-reviewed, which means that other scientists review a proposed paper and determine if the contents of the paper are noteworthy and contribute to the body of knowledge of science.  Journal information is more current than textbooks.  The web is a ready source of information, BUT remember that any yahoo (no pun intended) can create a web site and information on a web page almost always has not gone through the scientific scrutiny that journal articles have not.  We will examine, in lab, how to do a literature search on the computer to find journal articles on-line.  The search will be based on a topic of interest relating to a potential impact to Lake Winona, either natural or man-made.