We have a hunger of the mind which asks for knowledge of all around us, and the more we gain, the more is our desire; the more we see, the more we are capable of seeing.
—Maria Mitchell, Astronomer
Each class will complete an experiment designed to quantitatively answer a specific question.
THE IMPORTANCE OF SCIENCE
Science does not and cannot answer all questions we may have about ourselves or about our place in the universe. It does, however, provide powerful tools for understanding the natural world. Part of the strength of science is that it is a self-correcting process. If a procedure cannot be independently repeated, the conclusions based on that procedure are viewed skeptically. Part of the “weakness” of science is that it is done by humans and thus reflects the values of the individuals involved.
To a very large extent, cultural values dictate what science gets done and what science does not get done. Many branches of science now require expensive equipment and that can limit what is done. In the United States, many scientists rely on funding from grants (often from the National Institutes of Health (NIH) or the National Science Foundation (NSF)) to do their science. The process of getting grants is competitive. Every decision to fund one project means that many others are not funded. There are times when people may have great ideas on how to address difficult problems, such as curing certain diseases. Funding agencies must be cautious with their funds (NIH and NSF are funded by our taxes), and so really innovative ideas may not get funding. It is also true that the questions we ask may limit what we learn. That is, if we don’t know enough to ask a relevant question of a particular system, we are not likely to get a relevant answer.
TWO IMPORTANT TOOLS IN DOING SCIENCE
Scientists use a variety of tools in “doing science” and throughout this class we will use two of these tools quite extensively. The first one is modeling and the second is hypothesis testing.
Modeling
In some cases a structure is too small to manipulate in a way a scientist might want. In other cases, such as with global climate change, it is too large. In both of these cases, scientists may use models to investigate some of the properties of a system. In climate change, we use computer models with a large number of variables. In cellular science, we can use things like dialysis tubing as a model of a cell membrane. In this class, we will make use of models throughout the semester to better understand biological systems.
Hypothesis Testing
One tool that is so central it is sometimes called the scientific method is hypothesis testing. We will focus on hypothesis testing through much of this semester.
STEPS IN HYPOTHESIS TESTING
Step 1: Observation
Step 2: Question
Step 3: Hypothesis
Step 4: Prediction
Step 5: Experimentation
Step 6: Data collection
Step 7: Examining and evaluating the data
Step 8: Rejection or support of the hypothesis
FIRST EXAMPLE
You are in a movie theater before other people come in, and you observe that a woman comes in and sits down and then a man comes in and sits down next to her. This is your observation. You wonder, do they know each other? This is your question. You hypothesize that yes they do. If this hypothesis is correct, how could you test it? Perhaps you could move the woman and see if the man moves to sit next to her again. This is your prediction: If they do know each other, then if you moved one, the other would move as well. Note how useful the “If… then” statement is. Next you must do the experiment and actually move them. Collect your data and then examine them, and then make a judgment about your hypothesis.
What is the difference between your hypothesis being supported and the hypothesis being proven? Note that in this method, a hypothesis can never be proven to be correct. Why not?
SECOND EXAMPLE WITH MORE ITEMS: GROWTH OF KUDZU
In this exercise we will start with an example that shows the steps of the scientific method by designing an experiment to determine the effect of light intensity (amount of light) on the growth rate of kudzu. Most of you are probably familiar with kudzu. Kudzu is a plant, native to Japan, that was introduced into the southeastern United States by the Department of Agriculture to help control soil erosion. Unfortunately, someone didn’t do the proper scientific research on kudzu growth in North America, because although kudzu does a good job of covering soils and has a small effect on erosion, it tends to cover everything else as well, including native vegetation. Plants need solar energy to grow, and when plants can’t get enough light, they grow slowly or die. As a result, in areas where there is a lot of kudzu, our native plants have been harmed.
Step 1: Observations
Every scientific investigation begins with observations. The observations may be personal or based on the results of previous research. For example, we know that under proper conditions, kudzu grows very fast. What observations have you made about where kudzu grows? Does it occur within forests or mostly at the edge of forests?
You have probably noted that kudzu occurs along roads and at the edges of woods.
Step 2: Questions
Science is a question-driven process. Observations lead to questions. But for a question to be answerable by a scientific process the observations must be measurable, and the question must be well-defined. You may have noticed that kudzu tends to grow best at the edge of forests or in cleared areas within forests. Why is this? Could it have something to do with the amount of light available at a forest’s edge compared to the light available deeper within a forest? Or might it be the amount of nutrients available? Could it be it was simply planted at the edges of forests? We will investigate just one possible reason for why kudzu occurs where it does. Write down a simple question regarding the effect of light on the growth rate of kudzu.
Question: Why does kudzu grow where it does?
Step 3: The Hypothesis
A hypothesis is an educated guess or “tentative explanation” to answer a question raised after making preliminary observations. A hypothesis is a statement. Scientists often start with a hypothesis of no effect or a null hypothesis. This is because frequently, if we assume a particular cause and effect or if we look hard enough, we can find some evidence to support our ideas. It is important as a first step to show something is going on and we do that by refuting the null hypothesis. Note that in hypothesis testing we support or refute the null hypothesis. We never can prove the alternative hypothesis. Also, our brains are good at finding patterns, and we will sometimes see apparent patterns when things are actually random. We always need to ask ourselves: Am I seeing a pattern because I expect to see a pattern? The alternative hypothesis is the hypothesis that attempts to explain what we are observing. It is always rival to the null hypothesis.
Based on your preliminary observations, make a null hypothesis and an alternative hypothesis about the effect of light intensity on kudzu growth. Remember that a hypothesis should be in the form of a statement.
Null hypothesis: Light intensity has no effect on the growth rate of kudzu.
Alternative hypothesis: Light intensity has an effect on the growth rate of kudzu.
Step 4: The Prediction
Now it’s time for you to devise a test to help you decide if the hypothesis is valid. To help in the design of the proper test, it is useful to make a specific prediction based on your hypothesis that can be tested. Often, the prediction can be summarized in an “if . . . then” format. Write down a prediction regarding light intensity and kudzu growth that you could test if you had the proper equipment. Express your prediction as an “if . . . then” statement.
Prediction: If light intensity has an effect on the growth rate of kudzu, then as light is increased the rate of growth will be greater. Perhaps it is getting more light at the edge of woods compared to in the middle of woods.
Step 5: Data Collection and Experimental Design
Prediction(s) suggested by a hypothesis may be examined by collection of additional data through further observation by the investigator, a careful review and synthesis of data obtained by other investigators, or by an experiment. Experimentation is not always possible or necessary. But, when experimentation is possible, it is usually the most unambiguous method for determining the validity of a hypothesis. In an experiment, scientists manipulate a cause factor and measure the response in whatever they are interested. Validation of the prediction upon which the experiment is based is supportive of the validity of the hypothesis.
Before we can design an experiment to test the effect of light intensity on kudzu growth, we must discuss the proper elements of an experiment.
Elements of an Experiment: Variables, Treatments, and Replication
A. Choosing the Variables
A crucial step in an experiment is identifying the factors, or variables, involved. There are three categories of variables: dependent, independent, and standardized.
The independent variable is the “cause factor” that the investigator manipulates or varies in an experiment.
In this example, the investigator will manipulate the amount of light. Therefore, light is the independent variable.
Most often, an investigator will vary only one independent variable in an experiment. Otherwise, analysis of the results of the experiment can become quite complex.
The dependent variable is the factor that the investigator thinks will respond to, or be affected by, manipulation of the independent variable. It may also be referred to as the response variable. The relationship being investigated between the independent and dependent variable is a cause–effect relationship.
The dependent or response variable in this example will be the growth of the kudzu.
The standardized variable(s) are all variables, besides the one you are investigating, that could potentially affect the response of the dependent variable in your experimental treatments. Since the investigator is normally interested in the effect of only the independent variable, he or she must eliminate the possibility that other factors are affecting the outcome. Otherwise, it may be tough to determine what factor caused a response in the dependent variable. Thus, standardized variables must remain the same in all control and experimental treatments.
With which standardized variables would we have to be concerned within our experiment?
In this example, the standardized variables could include water and nutrients in the soil.
B. The Experimental and Control Treatments
In an experiment of classical design, the individuals under study are divided into an experimental treatment (or treatments) and a control treatment.
Experimental treatments are treatments subjected to manipulation of the variable of interest. There may be one or more experimental treatments depending on how many levels of treatment you choose. For example, in our experiment to test the effect of various intensities of light on kudzu growth rate, we would grow kudzu at a range in light intensities. Each of the light intensities would be considered an experimental treatment. The control treatment is a treatment to which the experimental treatments can be compared. The control treatment is subject to all the same conditions as the experimental treatments except for any change of the independent variable. For the control treatment, the independent variable is either eliminated or set at a standard value. For example, in our experiment to test the effect of light intensity on kudzu growth rate, the control treatment would be a group of kudzu plants grown under a standard light intensity to which the various experimental treatments can be compared.
C. Levels of Treatment
The values set for the independent variable are called the levels of treatment. For example, in our experiment we could examine kudzu growth at five different light levels, from low to very high light intensity. The levels of treatment set by the investigator are based on the purpose of the experiment. An investigator would not want to set levels of treatment that are so high or low as to make the experiment unrealistic.
D. Replication
In all biological systems there is variation. In other words, the response of a dependent variable to differences in the independent variable cannot always be expected to be the same even under identical conditions. For this reason, scientists always repeat their experiments using exactly the same conditions to see if the results are consistent. Often, the experiment is repeated several times simultaneously. Replication of an experiment increases our confidence in the results and gives us an idea of how much variation there is in the response of the independent variable.
Step 6: Doing the Experiment—Collecting Data (Results)
After an experimental procedure has been designed, the experiment is conducted. Data (results) collected during the experiment are organized in some way to best show what the results are.
Step 7: Examining and Evaluating Your Data
What do your data show? This very important step of examining and evaluating your data is critical. Is your null hypothesis supported or not supported? Is the alternative hypothesis supported? Hypotheses can never be proven correct, but hypotheses can be rejected. Very frequently at this step you may have a partial answer but more questions.
Be skeptical. Can there be other explanations for your data?
What is the difference between a theory and a hypothesis? The word “theory” is used in a variety of contexts that makes misunderstandings very common. In the context of science, a theory is a body of ideas that has been so thoroughly tested that it is taken as fact. Examples include the Cell Theory or the Theory of Gravity.
For the next part of this period, you will work with your TA on devising an experiment to demonstrate the experimental process. In class, you will test a specific hypothesis by an experiment and as a class you will evaluate your results.
You will complete and turn in the Hand-In by the end of the period.
Reaction time is the time between a stimulus and your response to that stimulus. Many factors can affect your reaction time.
With your TA you should discuss what some situations are in which reaction time is very important.
Does reaction time differ between people who have played video games and those who have not? Does a person’s right hand differ from their left hand in reaction time? What factors may affect a person’s reaction time?
Working in teams of two, we will use a meter stick to measure reaction time. One person will make a “C” with one of their hands and look at the meter stick being held by their lab partner. The partner will drop the stick, and the first person must catch it. How far it drops is a measure of the reaction time.
Each person, one at a time, should record their reaction time for 10 drops of the meter stick. Make a graph of both of your reaction times for each drop on Hand-In 1.
Next consider this question: What are some of the variables that can affect this measurement?
And, for each experiment, what is the independent variable (the cause factor, the variable that the experimenter manipulates)?
For each experiment, what is the dependent variable? This is what you think may be affected by the variable that you manipulate.
What variable can and should be standardized?
Complete the first part of Hand-In 1.
It may be obvious that careful measurements are a key piece of good process, and careful, precise work is the hallmark of a good scientist. It is also true that a measuring device or tool and a measuring process can introduce error and variability. Any time you are making a measurement, you should be aware of how much variability there is in the measuring process.
There are two terms we want you to know about measurements: reliability and validity. Good measures are those that measure a variable both reliably and precisely.
Reliability is the extent to which a measure is repeatable. Validity is the extent to which a measure reflects or assesses what a researcher is trying to measure. In the example of the reaction time, we practiced dropping the meter stick at least 10 times to increase our reliability. How reliable were your measures?
To be concerned about validity would be to ask if this particular technique really measured reaction time. Do you think this is a valid measure?
Complete the last page of Hand-In 1 to practice using metrics and knowing some of your own measures. In addition, all members of the class will do a series of measures on one person.
In the last few minutes of the class, we will introduce another way to measure reaction time, using the Iworx system. We will use this more extensively next time. In addition, before next week, all students need to complete the on-line plagiarism quiz.
The homework for the next lab requires that each TA find a paper that the students in their sections will summarize. This paper must be given to the Lab Coordinator this week.
Elodea and sheep’s blood should be ordered for the diffusion lab.