The Scientific Method

David Wm. Reed
Department of Horticultural Sciences
TexasA&MUniversity

Empirical Research, and Inductive and Deductive Reasoning
The scientific method has its basis in empirical research.Empirical research generates knowledge derived from observation or experimentation as opposed to theory.Empirical research uses inductive reasoning to draw conclusions about the experimentation and observations. Inductive reasoning is where specific observations or measurements are made in order to develop broader conclusions, generalizations and theories.For example, scientists conduct experiments and collect data to help answer scientific questions and solve problems.Deductive reasoning is where one starts thinking about generalizations, then proceeds toward the specifics of how to prove or implement the generalizations.For example, deductive reasoning is how a landscape architect approaches a project, e.g. they start with a design concept, and then proceed to the specifics needed to implement the project.Of course many researchers use both inductive and deductive reasoning in approaching a problem.

The scientific method conducts empirical research in such a way that it is without bias, is repeatable, and withstands the scrutiny of the scientific community.The opposite of the scientific method would be knowledge gained by testimonials.testimonial would be where observations are made under non-controlled conditions.For example, a person may state that when they started using fertilizer X they produced the “best” garden ever, hence fertilizer X is the best fertilizer to use.The problem with this approach is that there may be many fertilizers that would produce excellent results if they were used properly, or the results may not be due to the fertilizer at all, but rather it may simply have been a very favorable year for gardens.


Types of Research Used in the Scientific Method

There are two basic types of research associated with the scientific method.

1)  Quantitative Research

Quantitative research is based on collecting facts and figures.This type research is common in biology.

2)  Qualitative Research

Qualitative research is based on collecting opinions and attitudes.This type research is common in the social sciences.

Steps in the Scientific Method

1)  Identify a Problem or Question

Identify a problem to be solved or a question to be answered.For example, we know that plants require nitrogen fertilizer.If a crop is not growing very well, we might wonder if the lack of growth is due to inadequate nitrogen.Or, the crop may be growing, but we might wonder if we can stimulate the crop to growth even better if we try a new type of nitrogen fertilizer.

2)  Review Literature and Gather Information

Determine as much information about the topic as possible.Are there published studies that have investigated the same or similar topic?You do not want to conduct a research project that has already been done.You want to add to the current body of knowledge.The best approach is to conduct an exhaustive review of the scientific literature.

3)  Formulate Hypothesis, Null Hypothesis or Research Objective

Develop a hypothesis to be tested.hypothesis is a statement that the experiment will attempt to prove.An example of a hypothesis would be:“Increasing the level of nitrogen fertilizer will increase plant growth.”The purpose of the experimentation would be to prove the hypothesis.

Sometimes one develops a null hypothesis.null hypothesis is a statement that the experiment will attempt to disprove.Sometimes one can never “prove a hypothesis”, so one attempts to “disprove all possible null hypotheses”.An example of a null hypothesis would be:“Nitrogen fertilizer does not effect plant growth.”

More often that not, researchers in biology develop a research objective, such as:
“To determine the effect of increasing nitrogen fertilizer on plant growth.”

4)  Design Experiment

An experiment is designed to test the hypothesis, null hypothesis or satisfy the research objective.This is the critical component of the scientific method.The design of the experiment is what separates the scientific method from testimonials, general observations and assumptions.The scientific method uses the following characteristics to assure creditability.

a)  Unbiased

The experiment must be designed and conducted without bias.The experiment is designed such that one treatment is not favored over another. Sometimes the treatments are blind, and the researcher does not know which experimental units received which treatments.This is very common in human medical research.In addition, the treatments must be random.For example, the experimental units are randomly selected to receive the various treatments, and the treatments are randomly arranged in the area where the experiment is to be conducted.Finally, the experimental units, for example the plants, must be as uniform as possible.

b)  Control group

There must always be a control.control is a group of experimental units that do not receive the treatment.For example, in a nitrogen fertilizer study, the control group would receive no nitrogen.Sometimes the control must be a standard or normal condition.For example, if plants are growing in the soil where there is natural nitrogen present, then the control group would be the plants grown with the standard or normal amount of nitrogen present in the soil.

c)  Replicates

Each of the treatments is applied to a group of experimental units, for example a group of plants.Single experimental units are never used.Usually, a minimum of 5 uniform experimental units receive each treatment.However, sometimes many more experimental units must be used to collect reliable data.The data collected on the individual experimental units are averaged in step 6) Organize and Analyze data.

d)  Repeat experiment

The experiment must always be repeated to make sure the same, or very similar results, are obtained.

5)  Collect data

The experimental units must be measured in order to determine the effect of the treatments.For plants, measurements may be of growth rate, size, color, flowering, yield, internal physiological factors or constituents, or what ever is needed to determine the response to the treatments.

6)  Organize and Analyze data

The data must be organized and analyzed.The data is averaged and organized into lists, tables, figures and/or graphs.

7)  Interpret data

a)  Identify trend(s)

The data is studied to identify trends, to determine which treatments caused what types of responses, to determine which treatments are better or worse than others, etc.

b)  Determine significant differences

Statistical analysis is used to determine which treatments are different from the others.Oddly enough, in research you can never make a statement that two treatments are “equal”, only that they are “not significantly different”!

c)  Draw conclusions

Finally, conclusions are drawn to support or not support the hypothesis, null hypothesis or research objective.

8)  Communicate results

This is the most important part of the research.The research has no value if the results are not communicated with the scientific community, one’s colleagues, students and/or the public.The research becomes creditable only if it is reviewed by and accepted by one’s scientific peers in the form of a refereed publication.refereed publication is a scientific article that is submitted for publication in a refereed scientific journal.The article is sent to several scientists (the referees) for an anonymous review, and they recommend whether or not the article should be published in the refereed journal.If accepted for publication, the results and conclusions from the study have withstood the highest level of scientific scrutiny and are deemed acceptable by the scientific community.After the results are published in the refereed journal, then it is desirable to publish the findings in popular articles, industry journal, etc.

Structure of a Refereed Publication

Title

The title should be descriptive and succinct

By-Line

The author’s names and affiliation are given.

Additional index words

A list of key words is given.The key words are used by literature search engines.

Abstract

The abstract is a brief synopsis or the article.It may start with a sentence that introduces, justifies or gives the objectives of the research.It presents a very brief outline of the materials and methods, including the plant names, treatments imposed, and procedures followed.Then the major conclusions from the study are summarized.

Introduction

The literature related to the area of study is reviewed.The introduction summarizes the results of previously published studies on the topic, and justifies how this study will add to the body of scientific knowledge.At the end of the introduction, a list of Research Objectives usually is given.

Material and Methods

This section described exactly how the experiment was conducted in enough detail such that the study could be replicated by others.The types of information presented would be: scientific plant names, source and size of plants used, experimental design, treatments imposed, timing of treatments, environmental conditions, physical set-up of the study, the type of data collected and the protocol of measurements made to collect the data, instruments and chemicals used, and statistical analysis used.

Results

The data are presented in tables, lists, figures, graphs and photographs.All of the data is accompanied by the appropriate statistics.For each experiment, the trends in the data are pointed out in order to highlight how the experimental units responded to the treatments.Basically, the Results explain to the reader what the experiment discovered.

Discussion

The results are discussed relative to findings previously published on other scientific research.When a previously published study is discussed, the author and date are cited in the text.Similarities and differences with previous studies are pointed out.Conclusions are drawn as to the scientific meaning and application of the findings.At the end of the discussion, a summary of the major conclusions of the study is presented.Sometimes, the Summary or Conclusions appear as a final section.

Results and Discussion

Sometimes the Results and Discussion are combined into one section.Each experiment is presented then discussed immediately.

Literature Cited

The publications cited in the text are listed.