| Topic:
Phases of Research:
Steps in Gathering and Evaluating Evidence
The following are general steps for
conducting a research
study.
Step
1. Generate Ideas
Step
2. Formulate a Research Question
Step
3. Develop a Hypothesis
Step
4. Devise a study to test the hypothesis
Step
5. Run the Study & Get Results
Step
6. Analyze & Interpret the Results
Step
7. Communicate the Results
Step 1. Generate Ideas
The starting place of all research studies is a
single idea. Ideas come to us through our observations of the
world,
from conversations with other people, from sources such as books or
television,
or from reading other journal articles. Ideas represent the
interests
of a person. The idea does not have to be a full fledge research
project. All that is needed at this point is a thought, curiosity
about the topic, and interest in pursuing the topic further.
Let's take an
example that we will use for all of the steps. You notice that
you
are able to memorize items such as phone numbers better when you eat
candy
and wonder if there is a connection between candy and memory.
Step 2. Formulate a Research Question
After you have your idea, it is now appropriate
to examine the literature on the topic in order to ascertain what other
researchers have done and how they have conceptualized and tested the
topic.
Afterwards, convert your idea into a clearly posed research question or
set of research questions.
Continuing with
our example, let's say that you researched the topic of memory
enhancement
and saw that other researchers have looked at how certain herbs and a
healthy
diet can improve memory. However, no one* has examined if sugar
can
improve memory. Therefore, your research question is:
Can sugar improve memory?
(*Keep in mind this is just an example
and does not accurately
reflect the research literature on memory enhancement.)
Step 3. Develop a Hypothesis
A hypothesis is a testable prediction/belief of
what will happen in some circumstance. Recall from the review of
theories
that a good theory is one that is testable. In developing a
hypothesis,
you take your research question and restate it in specific terms that
make
a prediction. (You may also choose to keep the hypothesis more
general
if you are conducting an exploratory study.)
For example,
if the research question is to see if sugar affects memory, the
hypothesis
would be more specific and state:
"College students
who ingest 2 grams of sugar before studying for a memorization task
will
perform better than college students who do not ingest sugar."
Step 4. Devise a study to test the hypothesis
In this step, you (a) define all important terms
and variables in your study, (b) decide on the best research method to
test your hypothesis, and (c) select what statistical methods you will
use for analyzing the data.
(a) The reseacher should define, or create an operational
definition of, all important terms. An operational definition
is a definition that specifies the operations or procedures used to
measure
some variable.
In our example,
"sugar" and "memory" are terms that need to be defined in order for a
reader
to understand exactly what the researcher means by those terms (and for
others to replicate the study). Sugar could be defined as white
granulated
sugar while memory could be described as a score from a memorization
test
of nonsense syllables.
A variable is a characteristic that can have more
than 1 value. Examples of variables include intelligence, sex,
status
in a family, type of behavior, etc. In a typical research study,
there are 2 main types of variables:
1) Independent Variable
The independent variable (IV) is generally defined as the variable that
influences the dependent variable. It is any variable that
defines
different groups of participants who are measured via the dependent
variable.
It can also be a variable that describes the qualities of the
participants.
(An independent variable is also defined in some textbooks as the
variable
researchers manipulate so that its effects may be observed. But
this
definition can be limiting, as explained below*.)
In our above example, 2 grams of sugar would be the independent
variable.
Why? First, it is the sugar that is hypothesized to influence (in
this case, improve) memory; not the other way around. Second, the
IV is often the variable that can be changed or altered by the
experimenter
such as 1 gram of sugar or 2 or 3.
Another example would be if we were comparing spelling ability for
children
with mental retardation and children without mental retardation.
The IV for this example would be: mental retardation
condition (students with it and without it). Why?
First, whether or not one has mental retardation is hypothesized to influence
spelling ability, not the other way around. Second, it is the
variable
that defines the groups of participants who are to have their spelling
abilities measured. Third, it describes the quality of the
participants
(either having or not having mental retardation).
(*The experimenter cannot directly manipulate the degree of mental
retardation
in the participants; however, the experimenter could indirectly
manipulate
this through the recruitment procedure (i.e., recruit children with
mild
and moderate levels of mental retardation and children who are of
normal
intelligence). For this reason, defining IV's as the variable
researchers
manipulate can be confusing and limiting because it does not include
pre-existing
independent variables such as the sex or age of a person.) |
2) Dependent Variable
The
dependent variable (DV) is the variable that is hypothesized to change
in response to the IV. It is the variable that is hypothesized to
be influenced by the IV. The DV is often a score
of
some sort but NOT ALWAYS.
In our sugar experiment, "scores on the memorization task" would be the
dependent variable because that is what is hypothesized to change
because
of or be influenced by (in this case, improve) the independent variable
of sugar.
In the other example above, "score on a spelling test" would be the DV
because it is hypothesized to change in response to whether or not one
has mental retardation.
A
study can have more than one IV and more than one DV.
Click here for a quiz on IV's and DV's. |
(b) After you have these variables defined,
you need to choose the best research method. Click here
to review the different methods.
For
our sugar study, we would pick the experimental method. We would
have 2 groups: college students who ingest 2 grams of sugar and
college
students who ingest 2 grams of a harmless placebo-powder (no
sugar).
After both groups ingest their respective powder, they would have 3
minutes
to memorize a list of nonsense syllables. After a 5 minute rest,
both groups would recall as many of the syllables as possible.
The
number of correct syllables recalled would represent their memory
score.
(c)
You also need to select your statistical procedures, but that is beyond
the scope of this tutorial.
Step 5. Run the Study & Get Results
Now that everything has been planned, it is time
to run your study.
Step 6. Analyze & Interpret the Results
Statistical procedures are now employed to analyze
the collected data. After analysis, you need to examine the
results
to see if they do or do not support your hypothesis.
Step 7. Communicate the Results
The communication of research results is a vital
step in the overall process. Researchers typically communicate
their
results through publication of journal articles. They may also
choose
to present a poster or conduct a discussion at a psychological meeting
or research conference.
|
