Scientific Method, Observations and Hypotheses
In Science, Critical Thinking Occurs Throughout the Scientific Method
May 26, 2009
Science is based on logic and reason and the processes of induction and deduction. The scientific method is a simple concept: an observation is made, a hypothesis is proposed, experiments are done and conclusions (scientific facts or theory) are reached. This overview discusses observation and hypotheses.
Observation in Science Is the First Step
In the scientific method the first step is an observation. Scientific observation is any sensory experience that causes the scientist to think and ask a question. The observation leads to a search for an answer. Most observations typically are something seen or read, but sensory experiences such as hearing, olfaction, taste or touch are valid “observations”. Here are a few observations:
- Does diabetes have a single cause, or multiple causes?
- How are sugars used by animal cells?
- What do the varied chickadee, bird calls and sounds mean? How many different calls and signals do they have?
An observation may generate a question that already has an answer, therefore, it is not a true unknown. If the answer exists in the literature or the Internet, it can be found. It is may be that simple, in many cases.
The Hypothesis Moves Observation to Possible Answers
Typically, when confronted with one or more unanswered, unresolved questions or observations, the scientist will propose hypotheses — possible explanations to explain the question or observation. The hypothesis, or hypotheses (plural), may be valid, or invalid.
Hypothesis: Hypo, (Greek, hypo: under, below, beneath); Thesis, (Greek, tithena: to put, place).
The hypothesis is the underlying aspect of the thesis — the main proposition or idea. All theses are final statements of an underlying possibility that was tested and evaluated. A hypothesis is a plausible, probable explanation of some process, event or entity. Not all hypotheses are valid, or correct. Hypotheses may be proven wrong by experiments designed to show them as appropriate.
The Null Hypothesis States the Hypothesis Is Not True for Statistical Purposes
The null hypothesis sets up a premise that states that there no significance or difference between or among compared grouped results or data. The null hypothesis states that the variables are the same and not different. The null hypothesis is valid until is declared null (invalid). Experiments, with proper controls, are designed to validate or invalidate the null hypothesis. The null hypothesis sets up a negative underlying position that must be statistically evaluated. The statistical result is given in a probability value for the hypothesis.
The null hypothesis might be stated like this for a particular disease:
"The presence of these crystals in tissue is not related to the disease."
This null hypothesis is tested experimentally to show if the null hypothesis is statistically correct or incorrect.
In summary, the scientist seeks to know how likely the results are due to chance alone versus, for example, a direct effect of the drug or treatment used for the patients or subjects.
Characteristics of a Good Hypothesis.
A good hypothesis directly addresses the observation. The good hypothesis is:
- relevant and directed to the observation made
- testable qualitatively or quantitatively
- able to be analyzed statistically to indicate the probability of occurrence by chance alone.
A hypothesis may be poor or bad, if it:
- is not easily testable,
- yields unclear, or marginal results,
- is not replicable or reproducible,
- is extremely laborious, or time-consuming, or expensive.
It is useful to propose two or more, different hypotheses, evaluate each pro and con and, finally, determine which may be the best one to use and test.
Observation and hypotheses are the first two important steps in the scientific method. Excellence in science requires at all stages of the method.
To learn more about the scientific method see:
Science is a rewarding career for those who enjoy its processes. Consider a career in the sciences.
Sources
Alters, S and B. Alters. 2006, Biology. John Wiley & Sons Inc., Hoboken, N.J. 755 pp
Glantz, S.A. 1992. Primer of Biostatistics. McGraw-Hill, New York. 440 pp
 |
 |
 |
 |
 |
