Scientific Method (SM)
The SM is simply a way to gain an understanding of things around us. The SM is a
step-by-step procedure for studying things--it is a systematic approach to learning
about phenomena in education and other disciplines. In education, as in other disciplines,
the SM is used to test, evaluate, and refine theories.
Theories are systematic explanations for observed facts or phenomena. That is, a theory
provides a consistent, systematic view of variables (variables) and relationships among
these variables. [To learn about variables, see this page on variables.]
The purpose of a theory is to provide explanation and perhaps prediction regarding some
phenomena. A theory is a general explanation for different kinds of behavior found in
education, like Piagets notions about the stages of development which have
substantially impacted the teaching of children.
For example, the figure below displays one possible explanation for student attrition
(dropping out) from high school. Various constructs (variables) are displayed as well as
the possible linkages (the arrows) between them. For example, in this theoretical model,
academic performance is expected to have a direct effort on whether a student drops out of
Note that theories make assumptions about causal linkages (causality)
between constructs, between variables. Causality essentially means that for a given
action, there will be a reaction. Stated differently, if A occurs, then B will result. Use
the model above, the theory holds that if academic performance in school for a student is
poor, then that student is more likely to drop out of school.
How are theories tested, evaluated, and refined? The application of the SM enables
researchers to test and refine theories to some extent. When one applies the SM to testing
theories (or hypotheses), five general steps are followed:
- Recognition or identification of the problem (something that needs to be
understood better or tested);
- formulation of hypothesis (expectations about what ought or will
happenusually hypotheses are formulated from theories, so if there is no theory, it
is often difficult to determine proper hypotheses or expectations);
- collect data that will enable one to test the hypotheses formulated above (data
are the information one uses to make judgments about the problems and hypotheses);
- analyze the data (one analyzes data in some systematic fashion in order to better
understand whether the data do or do not support the hypotheses).
- statement of findings or conclusions regarding the hypotheses are presented.
Implications about the theory, based upon the tenability of the hypotheses, are
drawni.e., do the findings support the hypotheses which in turn support the theory?
Example Application of the Scientific Method to the
Analysis High School Dropouts and Dropout Prevention Programs
Suppose a local high school receives a $1,000,000 grant to develop a dropout prevention
program for Bulloch county. How should the money be spent; what factors should
be targeted to help prevent student dropouts?
Using a systematic approach to gaining information to answer this question is the first
step, so we will use the SM to answer this question.
- problem: Students leave high school before graduation, and society has determined
that this is a problem.
- hypothesis or expectations: Some prior research suggests a number
of reasons students leave school before graduation: poor grades, lack of
interest, peer pressure, financial reasons, etc. We build our expectations of the causes we have noted from
our experience dealing with high school students, and from the research
literature. We will want to identify factors causing
students to drop out so we can better serve them.
- data collection: Several approaches exist. We could interview
students, both those who remained in school and those who left, to ask why
they stayed or dropped out, then compare responses in hopes differences will
emerge. Differences could help us identify key reasons for leaving school.
Another approach is to assess various educational, psychological, and
financial reasons via questionnaires and scales, and perform statistical
analyses from a large sample to learn which of these factors best predicts
- analysis of the data: For the interview data we code responses then
assess common themes and differences among participants. We may find, for
example, that 70% of all dropouts have grades that are very low, and that
50% of all dropouts were retained at least once in a grade. So we may see that two primary
problems for these students is academic performance which leads to grade retention.
For the questionnaire/scale data, we could build a statistical similar to the
school persistence model shown above and statistically test which variables
are most important for prediction dropouts. Results from these analyses may
help us identify factors that can be addressed to help prevent large numbers
of school dropouts.
- conclusions: Given our findings, we may conclude that money for the dropout
prevention program would best be spent on elementary school children since that is where
poor school performance is first noticed and since that is where students are most likely to
be retained a grade.