SCIENTIFIC PROCESSES

Background Information1

Science is both a body of facts and a way of investigating the natural world. Scientific processes are
those activities related to the systematic collection and interpretation of data, and deriving conclusions
from those data. These activities include: observation, comparison, measurement, testing, predicting
and hypothesizing, creating ideals, constructing conceptual tools and models and formulating
explanation of trends in the data.

Contrary to many science textbooks, there is no universal scientific method. In reality, scientists apply
many methods in their research. There is no single, correct sequence of steps taken by all scientists in all
investigations. The procedures used
by a scientist, and the tools s/he uses
to collect and interpret data, will
depend on the phenomenon being
investigated, the question(s) being
asked, and the feasibility of
conducting research in a particular
setting. For instance, when studying
the behavior of large mammals such
as whales or primates, it is impossible
to control all variables. Instead, an
observational study might be more
appropriate than a strictly controlled
laboratory experiment. In addition,
new technologies can quickly change
the standard protocols used by
scientists in particular sub-fields of a
discipline. For instance, the advent of
DNA sequencing revolutionized the
field of molecular biology and has
made it possible to detect and
predict differences in the composition of DNA of very closely related organisms.

Rather than considering scientific investigation as a single, correct sequence of steps, it is better to
consider scientific investigation as a set of practices, or process skills, used by scientists at different
times and for different purposes suited to the investigation at hand.

Process Skills Used in Scientific Investigation

Formulating testable questions, hypotheses and/or predictions

Scientific questions arise in a variety of ways. They can originate from an individual’s curiosity about the
natural world, inspired by the predictions of a model, theory, or findings from previous investigations.
Scientific questions can also be stimulated by the need to solve a problem. No matter how those
questions arise, they must be testable. In other words, for a question to be scientific, a researcher must
be able to de- sign an investigation that includes the collection and interpretation of empirical data.
Questions are also formulated in the analysis and interpretation of data; for instance, how strong is the
claim based on the data collected by the researchers?

Developing and using models

In science, models are used to represent a system or parts of a system. Models include diagrams,
physical replicas, mathematical representations, analogies and computer simulations. Models never fully
correspond to the real world; while a model might be useful to clarify one aspect of a phenomenon, it
may obscure others. In a scientific investigation, models can be generated at any point in a study, and
they can be evaluated and refined several times during a single study, or across several studies.
Scientists evaluate and refine models through an iterative cycle of comparing their predictions to the
real world and then adjusting them to gain insight into the phenomenon being modeled. When new
evidence is found that can’t be explained by the current model, then the model is refined or modified.

Planning and conducting investigations

Scientific investigations are undertaken to describe

phenomena, or to test a theory or model for how the
natural world works. Scientists plan investigations
based on the question, hypothesis or prediction they’ve
formed. In field studies, planning involves deciding how
to collect different samples of data under different
conditions, even though not all conditions are under
the direct control of the researcher. In controlled
laboratory experiments, scientists must decide which
variable(s) should be intentionally varied from trial to
trial, which variables should be held constant across
trials, and which variable(s) will serve as outputs or
results. Thus in controlled experiments, a consideration
of variables is necessary.

Experimental variable – also known as the independent variable – the variable that is intentionally
manipulated by the researcher; the experimental treatment

Dependent variable – the variable that changes as a result of manipulations to the independent variable;
what is measured in the experiment

Controlled variable(s) – the variable(s) held constant by the researcher during the investigation so that
they do not confound the outcomes or results
Experimental group(s) – the individuals or subjects in an investigation that receive the experimental
treatment, which is manipulation of the experimental variable

Control group – individuals or subjects in an investigation that do NOT receive the experimental
treatment. The control group is important in an experimental study because it provides a means of
comparison for any changes observed in the experimental groups. If all variables are controlled in an
experiment except for the experimental variable, then any observed changes in the experimental group
but not the control group can be inferred as the response to manipulations of the experimental variable.
A control group helps rule out alterative explanations for the experimental results, and thus, heightens
the validity of the conclusions drawn by the researcher.

No matter what type of investigation a scientist conducts, the study should generate sufficient and
appropriate data to serve as evidence for the inferences and conclusions scientists draw.

Analyzing and interpreting data

Once data are collected, scientists must present it in a form that reveals any patterns and relationships,
and allows the results to be communicated to others. Raw data from scientific investigations rarely has
meaning; scientists must be able to organize and interpret raw data through tabulating, graphing and
statistical analysis. Such analysis brings out the meaning of the data set, and its relevance, so that it may
be used as evidence. In the analysis and interpretation of data, a scientist will also identify sources of
error and explain how it affects interpretation of results.

Using mathematics and computational reasoning

Mathematics and computational tools allow scientists to numerically represent variables, symbolically
represent relationships, and predict outcomes of investigations. Math enables ideas to be expressed
precisely and enables identification of new ideas about the physical world. Computational tools enhance
the power of math by enabling calculations that cannot be carried out analytically. They are also potent
tools for visually representing data that allows exploration of patterns.

Engaging in argument from evidence: making claims and constructing explanations

A key characteristic of science that distinguishes it from other ways of knowing is the manner in which
scientists make knowledge claims. In science, the production of knowledge is dependent on the
justification of scientists’ claims. Justification involves explaining how or why the data support a new
idea, but also citing existing theories or concepts as support. In response, other scientists may attempt
to identify the claim’s weaknesses and limitations. Scientists’ explanations and claims can be based on
deductions from premises, on inductive generalizations of existing patterns, or on inferences about the
best possible explanation. Argumentation is also needed to resolve questions about the best
experimental design, the most appropriate techniques of data analysis, or the best interpretation of a
data set.
At any rate, scientific knowledge is never fully proven. Scientific facts, theories, ideas and laws are
always subject to revision in light of new discoveries or data. Thus, hypotheses and predictions are
NEVER proven by an experiment or investigation – they can only be supported or refuted by the data.
Scientific explanation can be enhanced, or further supported, by the reproducibility of results, the re-
liability of a control group, methods taken to reduce experimental error, and illustration of statistical
significance of results.

Obtaining, evaluating and communicating information

Being able to read, interpret, and produce scientific and technical text are fundamental practices in
science. By some estimates, reading and writing scientific texts consumes more than half of a scientist’s
workday. Being a critical consumer of information requires the ability to comprehend scientific reports,
recognize salient ideas, identify sources of error and methodological flaws, distinguish observation from
inference, arguments from evidence, and claims from evidence. Communicating scientific evidence and
ideas is accomplished through many modes: tables, diagrams, graphs, models, equations, and narrative
text.

1. National Research Council. (2012). Framework for K-12 science education: Practices, crosscutting
concepts and core ideas. Washington, D.C.: National Academies Press.