Research I
Research Design
The research design refers to the complete sequence of steps that will be
undertaken in a project. The design should ensure that appropriate data will be
obtained in a way that permits an objective analysis, leading to valid inferences
concerning the stated problem. Preparation of the research design is a step in
problem-solving, consisting of a detailed plan to be followed to obtain the
needed data.
One of the basic elements that you must consider when planning the
sequence of steps that will embody the research design is the identification of
variables. In research, variables are any characteristics that can take on different
values, such as height, age, species, or exam score. In scientific research, we
often want to study the effect of one variable on another one.
Classification of Variables
There are different ways of classifying variables. The method of
classification
will depend on what stage of the research process you are in. When formulating
the experimental design, you classify the variables based on cause and effect
because you need to define the relationships that exist between your variables.
Classes of Variables Based on Cause and Effect:
Independent or Manipulated
Dependent or Responding
Extraneous or Constant
Variables in the Research Design
When you design your experiment, you need to make sure that you have
covered all your bases in terms of the different types of variables used, the
values for the identified variables, and the time that each variable takes on a
value.
Here are guide questions that you may use to determine if you have
covered all the bases:
Independent Variables - What variables will you manipulate or vary?
Extraneous Variables - What will you keep constant during
experimentation or data gathering?
Dependent Variables - What will you observe during and/or after
experimentation/testing?
Here is the example showing the application of the three variables:
Type of variable Definition Example (fertilizer added
to plants experiment)
Independent variables Variables you manipulate The amount of fertilizer added
to affect the outcome of to each plant’s water.
an experiment.
Dependent variables Variables that represent Any measurement of plant
the outcome of the health and growth: in this
experiment. case, plant height.
Extraneous variables Variables that are held The temperature and light in
constant throughout the the room the plants are kept
experiment. in, and the volume of water
given to each plant.
Groups in Experiment Design
An experimental group is a test sample or the group that receives an
experimental procedure. This group is exposed to changes in the independent
variable being tested.
A control group is a group separated from the rest of the experiment such
that the independent variable being tested cannot influence the results.
While all experiments have an experimental group, not all experiments
require a control group. Controls are extremely useful where the experimental
conditions are complex and difficult to isolate. Experiments that use control
groups are called controlled experiments.
Let us Check our New Learning on Variables!
Directions: Read the following experiments and fill in the blanks that follow.
Hint: One problem does not contain a control group
1. The Effect of Screen Time on Students’ Concentration
A psychology student wanted to find out if using electronic devices before studying affects
students’ ability to concentrate. One group of students spent 30 minutes scrolling on their
phones before studying, while another group avoided all screens during that time. Both
groups were then asked to complete the same concentration test, and their scores were
compared.
Independent Variable:
___________________________________________________________________________
Experimental Group:
___________________________________________________________________________
Control Group:
___________________________________________________________________________
Dependent Variable:
___________________________________________________________________________
Extraneous Variables:
___________________________________________________________________________
2. The Effect of Water Type on Plant Growth
A student investigated whether different types of water affect plant growth. She watered
three identical plants using tap water, rainwater, and saltwater for three weeks. She
measured each plant’s height at the end of the experiment.
Independent Variable:
___________________________________________________________________________
Experimental Group:
___________________________________________________________________________
Control Group:
___________________________________________________________________________
(Hint: This study may not have a control group)
Dependent Variable:
___________________________________________________________________________
Extraneous Variables:
___________________________________________________________________________
The research design refers to the overall strategy that you choose to
integrate the different components of the study coherently and logically,
thereby, ensuring you will effectively address the research problem; it
constitutes the blueprint for the collection, measurement, and analysis of data.
The following are the basic elements that you must consider when
planning the sequence of steps that will embody the research design:
1. Identification of variables
2. Application of basic research principles
3. Consideration of the research design structure
I. Variables
A variable is any entity that can take on different measurable or
observable values. The use of variables is an intrinsic part of any investigative
research. Variables are used as the point of comparison between two or more
objects or experimental units, as a means for specifying the focus of the
investigation, and as the basis for defining the scope of a project. (Detailed
discussion of the classification of variables in this module - What's New)
I. Basic Principles of Research Design
For a Research Design to be considered sound, three basic principles are
expected to be present:
Local control refers to the balancing, grouping, and blocking of
experimental units so that the units within each block are relatively
homogeneous. In engineering projects where new products or systems are
created, the product or system itself may be considered as the
experimental unit. Examples are a prototype of a device or a computer
program that is subjected to various treatments (test cases) to determine
its correctness and consistency in performance.
Replication refers to the conduct of an experiment or test using three or
more homogeneous experimental units subjected to the same treatment
under similar conditions. In the case of technical research where the
product of design is the lone experimental unit, replication refers to the
repetition of a test under controlled conditions. Replication makes the test
of significance possible.
Randomization refers to an unbiased method of assignment of
experimental units (or subjects) to the treatment/s or vice versa. It
assures the validity of the statistical tests of significance.
The Experimental Units
Before you start verifying if you have considered the three basic principles
of research design in your project, make sure that you have correctly identified
your experimental units first.
An experimental unit also referred to as the experimental subject, is the
focus of testing or experimentation. It is the organism, material, product, or
device that is subjected to one or more treatments and is, therefore, the source
of observation or data. It could be plants, animals, or inanimate objects.
Examples are lettuce heads, bacterial culture set-ups, mice pieces of wood, and
rabbit human testers.
Checklist of Applied Principles
After identifying your experimental unit, you may now check if the three
basic principles of research design are evident in your project:
1. Local Control
Do all experimental units that receive the same treatment bear
the same characteristics?
2. Replication
For every treatment (experimental or control) or test, do you
have a minimum of three replicates?
3. Randomization
Is there a randomization method used to divide the experimental
units in each block among the various treatments? Note that if
randomization were done, it would not matter what treatment is
assigned to an experimental unit.
Activity:
Instructions: Read each scenario carefully. Then, answer the guide questions
that follow to check whether the three basic principles of research design
are applied.
Scenario 1: Plant Fertilizer Study
A group of students wants to find out which fertilizer helps tomato plants grow
taller. They plant tomato seeds in nine identical pots filled with the same type
of soil. They divide the pots into three groups of three.
Group A gets Fertilizer X
Group B gets Fertilizer Y
Group C gets no fertilizer (control)
The pots are placed randomly on a sunny windowsill, and the plants are watered
equally every day. After four weeks, the students measure the height of each
plant.
Guide Questions:
1. What is the experimental unit in this study
2. How is local control applied in this study
3. How is replication applied
4. How is randomization applied?
5.
Scenario 2: Testing a New Learning App
A researcher wants to know if a new mobile app for math practice improves
student performance compared to traditional worksheets.
He selects 30 students and divides them randomly into two groups:
Group 1 uses the math app for two weeks.
Group 2 uses worksheets for the same amount of time.
At the end of two weeks, both groups take the same math test, and their scores
are compared.
Guide Questions:
1. What is the experimental unit in this study
2. How is local control applied in this study
3. How is replication applied?
4. How is randomization applied?
II. Basic Types of Research Design
A good research design also considers the organizational structure of the
experimental units. There are two basic types: completely randomized design
and randomized complete block design. The other designs are variations of these
two basic types.
Completely Randomized Design (CRD) is used when the set of
experimental units is homogeneous concerning the factors that may
affect the results. Each experimental unit is simply assigned a
treatment randomly. In the figure below, there are a total of nine
experimental units, all of the same type. Any three experimental units
may be assigned for each of the three treatments (T1, T2, T3).
Figure 1: Experimental units assigned randomly in three treatments
Source: Variables - BSICR-Ch-4-Copy
Randomized Complete Block Design (RCBD) is used when the
experimental units have varying characteristics that may affect the
results of the experiment. When this happens, the units are divided into
homogeneous groups or blocks. Within each block, the subjects are
assigned treatment randomly. The result is that all treatments should
be assigned to corresponding experimental units within each block.
Figure 2: Experimental units divided into three blocks and assigned
treatment randomly
Source: Variables - BSICR-Ch-4-Copy
The sample illustration above shows three types of frogs, hence the
division into three blocks. Within each homogeneous block, treatment is
assigned to three frogs. Note that the number of replicates per treatment is still
three. Because the frogs in Block 2 are different from those in Block 1, they are
considered as different set-ups.
Features of a Good Design
A good research design should have a well-defined set of experimental
units. Before execution of the project, it must be verified that the experimental
units have homogeneous characteristics or have been classified accordingly into
homogeneous groups.
If the characterization and blocking of experimental units are done
properly, it should be easy to verify if these basic principles (replication, local
control, and randomization) are applied in the research design.
Once the basic elements of the research design are set, then you are
ready to write the research proposal.
Activity:
Directions: Read the given situation, identify the variables, and illustrate a
research design structure.
Situation: A local pharmaceutical company introduced Moringa D, a new pill
that has a good chance of helping people lower their blood sugar.
Typically, the Hemoglobin A1C test is a way to measure the average
blood sugar for the last three months. The company hopes that the pill
lowers blood sugar as shown with lowered AIC. To experiment, 100
diabetic patients were randomly selected into two groups: patients
who would take the New pill and patients who would take the popular
Brand A pill in the market.
Guide questions:
1. What is the role of Moringa D in the experiment?
____________________________________________________________________________
2. What is the significance of having a lowered AIC in the experiment?
____________________________________________________________________________
3. What are the extraneous variables?
____________________________________________________________________________
4. Which is the experimental unit and what treatment is applied?
___________________________________________________________________________
5. How should the control group in this experiment be treated?
___________________________________________________________________________
6. Illustrate how the samples are assigned using a Randomized Complete Block
Design. Show the repetition of tests in triplicate in each treatment.