The scientific method is a procedure used by scientists to conduct research which consists of principles and procedures for generating and representing knowledge as accurately as possible. The scientific method generally includes the following main steps
1. OBSERVATION AND ASKING QUESTIONS
In the scientific method always observations would lead to a question and possible answers to the question can be developed into one or more testable hypotheses.
A hypothesis is a testable and falsifiable (True or False) prediction that arrived at logically from a suggested explanation. The tester should test hypotheses using experiments and mathematical analysis and modeling.
Analyzing results will reach a conclusion and it may support or reject the hypothesis. Repeated testing of the hypothesis is essential before the confirmation of validity. The hypothesis that has been consistently validated through additional observations or experimentation can eventually be advanced to the status of theory. A theory is a thoroughly substantiated explanation of some aspect of the observable world.
3. EXPERIMENTATION : TEST, COLLECT, AND ANALYZE DATA
An experiment tests whether your prediction is accurate and thus your hypothesis is supported or not. It is important that the experiment is fair/ unbiased. When testing only one factor is changed at a time while keeping all other conditions the same. Experiments should be repeated several times to make sure that the first result wasn’t an accident.
- IDENTIFYING VARIABLES
When designing an experiment following three aspects must be clearly determined.
What stays the same? , What is measured and changed?
In a properly designed experiment, all factors that could impact the results of the investigation are taken into consideration. These are called variables.
INDEPENDENT VARIABLE :
The variable is changed deliberately so that its effects may be tested (also known as the manipulated variable).
Example: Temperature, pH, Humidity, Light intensity
May be thought of as the results or data of the experiment (also known as the responding variable). The dependent variable changes in response to the independent variable.
Example: Percentage of seed germination, Height of a seedling
CONTROLLED VARIABLES :
Variables that remain constant. Makes the experiment reproducible.
- EXPERIMENTAL (TREATMENT) GROUP AND CONTROL GROUP
There are two groups in the experiment, and they are identical except that one receives a treatment while the other does not. The control group provides a baseline that allows determining if the treatment has an effect (remains unchanged during the experiment). In a comparative experiment, the experimental/treatment group is being tested for its response to a change in the independent variable. There may be more than one experimental group in a study, each testing a different level or amount of the variable.
We should apply each of the treatments generally to a group of similar experimental units called replicates. The use of more replicates increases the reliability of data and data collected on the individual experimental units are averaged during analysis.
- RECORDING OBSERVATIONS AND DATA COLLECTION
Experimental units are being monitored in order to determine the effect of the treatments. In order to do that, the outcomes must be observable or measurable. There are two types of data,
- Qualitative data – Descriptive
Eg: Color, Sound, Acid, or Gas production
2. Quantitative data – Expressed numerically (measurable). Then analyze data using mathematical and statistical methods.
Eg: Height of seedlings, Percentage of germination, Yield per unit area
Qualitative data are useful, but they cannot be statistically analyzed. Thus, no scientific research should be based on qualitative data alone.
ORGANIZING AND ANALYZING DATA
Average the collected data and organize it into lists, tables, figures and/or graphs. Sometimes additional calculations may be necessary for certain experiments (with the use of the collected raw data)
INTERPRETATION OF DATA
Trends should be identified through the responses caused by each treatment and it should be found out which treatments are better or worse than others. Then significant differences can be determined through statistical analysis of quantitative data.
Finally, we can conclude which may support or not support the hypothesis tested.