Saturday, August 27, 2016

Science and the Scientific Method

Where are we going with this?
 The information on this page is foundational to science and scientific inquiry. 
Science has, as one of its purposes, the goal of explaining how things work. It looks for relationships between things and attempts to create statements that explain and generalize, that create understanding.
Science is a system of knowledge about the natural world and the methods used to find that knowledge. It is the intellectual and practical activity encompassing the systematic study of the structure and behavior of the physical and natural world through observation and experiment.
Within science are laws. Laws are long-tested and tried and deemed to be nearly unchangeable models of how things work.
law is a statement of fact, deduced from observation, to the effect that a particular natural or scientific phenomenon always occurs if certain conditions are present. 

A law in a statement describing what happens.

Many times, a scientific law can be expressed in terms of some sort of equation.
If there is a possibility for additional information—if there are some parts of the model that are less than 100% certain—the generalization is called a theory.
theory is a pattern or relationship that has been established based on a large amount of experimental data.

theory is an in-depth explanation of how and/or why something happens. It is seem as correct because the pattern or relationship has been tested many times. 


Theories and laws are developed from the analysis of many, many experiments
The purpose of an experiment is to find some explanation of an observed phenomenon. 

Experiments are devised using the scientific method.
Scientific method is a method or procedure consisting of systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses. It is an organized plan used for gathering, organizing, and communicating information that relies on collecting and analyzing data to answer questions or solve problems.
Several steps are commonly described as being part of the scientific method. And… well, there are a lot of ways to organize the steps. Different sources will use different ways to describe the process.
Almost always, though, they essentially reduce to the same process. (Sometimes, steps are combined or skimmed over.)

Observe a phenomenon and ask a question about why it occurs? (Is this one step or two? Depends on who you ask!)
Why are some people taller than other people?
Doing research into previous studies on the topic sometimes can answer the question, but a researcher my have another idea that adds to what is already there. It is important to let the research help focus and refine what is being studied.
Not all problems can be solved and not all questions can be answered using the scientific method. For the use of the scientific method to work, the problem or question needs to be one that can be solved experimentally. 

Ready to see a list? Gotcha! HERE

Focus on a very limited number of factors and speculate how they are related. Then, formulate a guess—construct a hypothesis—that explains the relationship. The hypothesis will be a prediction that answers the research question:
Sample Research Question: What is the effect of gender and age on the height of high school students?
Sample Hypothesis: For students in high school, older students will be taller than younger students and male students will be taller than female students.

There are several things that will make a hypothesis strong and useful (which is good so that researchers don't waste time and resources only to end up with nothing helpful).

hypothesis is based on observation, knowledge, and research. 
hypothesis is a proposal that explains something about the solution to a problem or that answers a research question. They are scientifically testable and can be either found true (supported, accepted) or disproved (rejected).
hypothesis is an educated guess about what will happen in an experiment
If_______ then ___________ .


After developing a hypothesis, the next step is to devise an experiment that tests it.
An experiment is 

  •  a scientific procedure undertaken to make a discovery, test a hypothesis, or demonstrate a known fact. 

  • an organized set of actions that allows, within a controlled setting/situation, the testing the relationship between two or more conditions and responses.

The experiment is designed to answer the research question. There are many ways to set up an experiment. Sometimes, the effect of something is found by having one group experience the something and comparing the outcome to a group that does not experience it. For example, the hypothesis "eating chips at lunch will cause people to read faster in the afternoon" could be tested by measuring the reading speed of two groups: one that ate chips at lunch and one that did not.
The design of this experiment uses two groups: an experimental group that is exposed to the circumstances being tested, and a control group that is not.
An experimental group is a selection of subjects in an experiment that are exposed to the conditions being tested.
control group is a selection of subjects in an experiment that are not exposed to the conditions being tested and to which the experimental group is compared. 
A good control group is (ideally) identical to the experimental group in all ways except for the difference(s) being tested.
Other experiments may be done within a single group in which the frequency or rate of a thing is compared. A research question such as "what percent of white tail deer have antlers with different numbers of prongs on each side?" could be answered by surveying some number of deer and counting the number of occurrences. In the case of our sample hypothesis, a sample of students will be used and, based on their ages, their heights will be compared.
An important part of designing an experiment is determining the variables that will be tested. A researcher will select some conditions that will be compared and will take measurements of other characteristics. Variables will be assigned and named to allow for meaningful comparison. You may compare many things at once, but each additional consideration creates additional complexity in the design of the experiment.
There are two types of variables in an experiment. Those that researchers set and those that are measured. The research question is strongest when it identifies the variables. In the example above, the question is asked, "What is the effect of gender and age the height of high school students?" Almost always, a good research question is in the form of "What is the effect of the independent variable on the dependent variable."

Independent Variables 

The values of the independent variable are picked and changed by the researcher. 
They may be categories like age group, gender, number of coin tosses, type of tire tread, eye color, presence or absence of a calculator, presence or absence of <any treatment condition>, and many other things. 
The independent variable is also called the manipulated variable because it is purposely manipulated.

Dependent Variables 

The values of the dependent variable are measured or counted. 
They are the ones being observed. They may be measures like height, breaking distance, number of times heads occurs on a coin toss, speed of solving math problems, etc. 
The dependent variable is also called the responding variable.
An experiment can have as many independent variables as the researcher wishes. However, with each additional variable, the complexities of the experiment goes up. 
Sometimes a concept called a constant is useful. A constant is a quantity that does not change. While it is possible to measure a constant, you cannot change it. It is something that the experiment should not alter.
Even though you can measure it, it is not really an experimental variable. 
The idea of keeping certain values in an experiment constant leads right into the idea of experimental controls. 
In order to make sure that only the factors being tested are those considered by the hypothesis, it is absolutely necessary to put in place controls. 
Experimental controls are the measures taken to make sure that the variables present in the hypothesis are free from other influence. 
Comparing braking distance as a result of the type of tire tread would only be appropriate if the measures were done under exactly the same conditions—same car, same track surface, same speed, etc.
In the example of comparing height based on age of high school students, numerous controls need to be in place. These factors can be thought of as the controlled variables. The idea is that these variables are kept exactly the same for each measure of each person in each age group. By controlling everything else, the relationship of age and height can be isolated.
Sample Control Variables/Conditions (for an experiment answering the question about relationship of age on height):
No shoes—controls the variable "heel height"
Measure subjects in same place—controls the condition of an uneven floor
Push hair down and measure from top of skull—controls for poofy hair.
An experiment that is well-designed needs to have all the controls necessary to assure that only the independent variable is having an effect on the dependent variable. This is as important as any other part of the experiment.
From the above emerges elements of a well-designed experiment.
1. Based on the hypothesis, identify the independent and dependent variables.
2. Develop the control measures needed to keep eliminate everything that might affect the outcome other than the independent variable.
3. Create a list of needed materials and conditions.
4. Develop a step-by-step procedure that assures that the dependent variable will be measured accurately.
Once ALL this is done, all that remains is to conduct the experiment that was designed. It is possible that the outcome will not be as expected because of problems with the experiment. Researcher may need to revise the procedure or even go back and rethink the experiment all together. Once the experiment is properly designed, results should be recorded.

The next step in scientific method is to analyze the data and draw a conclusion.
All of the data collected must be reported. Sometimes, data will not fit what was expected, but it needs to be reported anyway.
It is possible that the data does not support the hypothesis. If this is the case, it does not mean that the experiment failed. The first course is to review the procedures and make sure they were appropriate, then repeat the experiment and confirm the outcome. There is always a possible that something was wrong with the experimental design and not the hypothesis, so repeating the experiment—possibly with revised procedures—is important.
When the results clearly do not support the prediction, it means that the hypothesis was incorrect. Finding that the hypothesis is false is just as important as finding out it is right. 
When the results do not support the prediction, researchers need to explore the possible reasons for the unexpected outcomes.
For instance, if the hypothesis that eating apples causes cavities is tested and is proven incorrect, then the practice of eating apples can continue without people worrying.
When a hypothesis is proved false, often, new ideas for additional research are formed. 
Examining the outcomes of one experiment often can lead to new research questions, different hypothesis, and new experiments.

Using the scientific method dates back centuries to the dawn of modern science. Its rigor leads to humanity's ability to better understand how things in creation work together. Its use has led to advances in technology and medicine that have changed the lives of everyone.

Contributions from Troy Smigielski