See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/335651908

Research Designs

Chapter · September 2019

CITATIONS READS
0 10,009

3 authors, including:

Nitasha Sharma
Postgraduate Institute of Medical Education and Research
31 PUBLICATIONS   144 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

An exploratory study to assess the pathways of care and out of pocket expenditure among patients attending psychiatry care services, PGIMER, Chandigarh 2015-2017 View project

All content following this page was uploaded by Nitasha Sharma on 06 September 2019.

The user has requested enhancement of the downloaded file.

Chapter

11
Dr Jay Prasad, Dr Har Ashish Jindal

Learning Objectives
At the completion of the chapter, the readers should be able to:
  understand concepts of research designs.
  acquire knowledge of different types of experimental research designs.
  discuss the difference between quasi-experimental and true experimental design.
  appraise relevance of internal and external validity of experimental research designs.
  devise an appropriate research design in conducting research.

Chapter Outline
 Introduction
 Characteristics of experimental research designs
 Threats to internal validity
 Threats to external validity
 Classification of experimental research designs
 Advantages and disadvantages of experimental research designs
 Summary table of pre-experimental, quasi-experimental and true experimental research designs
 Summary and key concepts
 Assess yourself
 Nursing Research in 21st Century

There are various methods and criteria’s to classify

Key Terms research designs. However, based upon the criteria of
experiment, we may classify designs into two broad
 Attrition categories: Non experimental and experimental designs.
 Control Non- experimental designs involve the study of given
phenomenon, its description without any kind of
 Cross over intervention or manipulation. Experimental designs
 Differential selection, involve use of some form of experiment or intervention.
The current chapter is going to focus on various types
 Experimental of experimental research designs
 External validity
 Instrumentation bias CHARACTERISTICS OF EXPERIMENTAL
 Interaction effect of testing, RESEARCH DESIGNS
The experimental research designs involve manipulation
 Internal validity
of certain stimuli, or treatments or environmental
 Non equivalent control group design conditions to make an observation on how such
 One group pre test post test design manipulations affect the behaviour of the subject. While
doing such manipulation the researcher must be aware
 One shot case study of other factors which can affect the outcome, which he/
 Post test she either remove or control. Thus, the four basic essential
characteristics of an experimental research design can be
 Pre experimental design explained as:
 Pre-test a. Control: Those variables which are not of interest
to researcher and they can still affect the result also
 Regression effects
called the extraneous variables are either removed
 Research Design or the arrangements are made to minimize their
effects. These arrangements can be either the random
 Static group comparison study
assignment of subjects to the groups, matching the
 Testing effects subjects on extraneous variables and at times keeping
 Time series design groups homogenous or using some form of statistical
technique.
b. Manipulation: In an experimental research,
manipulation refers to a deliberate introduction or
INTRODUCTION operation of an independent variable on the subjects
in experimental group. This operated independent
Section V: Quantitative Research Approaches and Designs

An experimental research design is like a blueprint of the

variable is also called the treatment variable or an
procedures involved in answering a research question. It
experimental variable.
is a detailed plan for conducting a scientific investigation.
A research design is defined as the series of those steps c. Observation: After introducing an independent
taken to ensure that relevant data will be collected in a way variable, the researcher observes its effect on
that permits objective analysis of the different hypotheses dependent variable. This task is called observation in
formulated with respect to research problems. Researchers experimental research.
conceive a research design for any investigation not merely d. Replication: It refers to conducting sub experiments
on premise of convenience or notion; Rather they consider under a large experimental design. The researcher
the specific purpose of research, the types of variables can combine a number of experimental and control
involved and the research settings to evolve a research groups in singe design.
design.1 However, any research design is considered fine The experimental design can be further classified
if it fulfils two basic functions: into three major subtypes: True experimental, quasi
‰‰ Answers the research questions objectively, validly experimental and pre experimental (Figure1). Before
and economically. giving description of different kinds of research designs
‰‰ Enables the researcher to control undesirable variance it seems necessary to have an understanding about the
(extraneous or error variance) while conducting relevance of internal and external validity in various
experiment. research designs. The concepts of internal and external

2
 Part A: Nursing Research

validity pertaining to research designs bear a great due to an historical event wherein one of the persons
significance in any research proposal. So, it would be suffering from lung cancer died. Thus, attribution of
worthwhile to discuss importance of internal and external smoking cessation to intervention would be erroneous
validity in research designs. The detailed description of without taking into account the event that occurred
each design along with its evaluation in terms of internal during the experiment.
and external validity shall be discussed in later section. b. Maturation: The term maturation depicts all
biological or the psychological processes which have
a potential to vary systematically with passage of
INTERNAL AND EXTERNAL VALIDITY IN
time. Therefore, what researcher labels as “effect” of
RESEARCH DESIGN2-4 intervention could be rather one of such maturational
Any experimental research is conducted to see the impact processes. For example, a researcher assesses the
of independent variable upon a dependent variable. Thus, effect of playing video game on cognitive processing
an experimental investigation purports to answer the of children. In such an experiment, the children are
basic question: Did the experimental treatment make any in their growing age wherein there is likelihood of
difference in dependent variable? Internal validity refers natural improvement in cognition. Hence, attributing
to the fact if the researcher has enough evidence to assert, all improvement in cognition to experiment would be
whether or not, experimental treatment/condition has erroneous.
made the difference. If a research design has high internal c. Testing effects: A pre-test may sensitize participant
validity, there are enough reasons to conclude or attribute in unanticipated ways and their performance on
the change in dependent variable to the introduction of the post-test may be due to the pre-test, not to the
independent variable in the form of any intervention or treatment, or, more likely, and interaction of the
treatment. On the contrary, a design with low internal pre-test and treatment. Considering an example, a
validity shall have poor evidence of assuming such study is conducted to assess the impact of structural
causality. There are various threats to internal validity of teaching programme on hand hygiene practices
a research design that might compromise researcher’s in nursing personnel. The improvement in hand
confidence in saying that a relationship is there between hygiene practices may have been due to repeated
the independent and dependent variables. The factors administration of the same test. That is, the nurses
that can confound the effects of experimental variable if simply learned to provide the right answers rather
not controlled by the experimental design act as source of than truly achieving improved hand hygiene practices.
threat to designs’ internal validity. d. Instrumentation bias: This threat which is also
Another important fact that depicts the strength of an referred as “instrument decay” occurs when the
experimental design is its external validity. The external changes in the outcome are due to the change in an
validity refers to the degree to which the results of the instrument used for the assessment. This can be
experiment can be generalized to the entire population. avoided by calibrating the instruments prior to the
Thus, it is also commonly referred as the generalizability. data collection, so that they measure in the same way,
The next section shall offer a detailed description on and to the same degree of accuracy.
factors affecting the internal and external validity of e. Changes related to measurements: This bias may
research designs. also result when two examiners administer the same
test and are not trained adequately prior to the data
THREATS TO INTERNAL VALIDITY collection regarding certain measurements and there
are no written guidelines regarding the procedure.
Internal validity of a research design can be affected by the
following factors:
Let us see an example, a study is conducted to | 11 | Experimental Research Designs
assess the cognitive impairment associated with
a. History: Many a times, the change in scores on Electroconvulsive therapy. While first examiner duly
dependent variable before and after the intervention explains each test item to the subject satisfaction while
are due to the occurrence of some unanticipated the other does not. This difference in the observer’s
event which occur while the experiment is in progress approach to assessment can bring about a change in
and the event has a potential to affect the dependent the outcome. This could be controlled by training all
variable. We can take an example, if a pre experimental the data collectors, and testing them until they achieve
study is conducted to assess the impact of structured the inter rater reliability to a set standard.
teaching programme on tobacco cessation in reducing f. Regression Effects: The regression effect is the
prevalence of smoking. If the researcher finds a statistical effect of a propensity of regressing towards
significant reduction in smoking prevalence after the the mean. The regression effects occur when the
intervention. There is a possibility that this occurred subjects are recruited to an experiment on the basis

3
 Nursing Research in 21st Century

of their extreme scores. This means the subjects with two groups, hence, it would not precise to attribute
extremely low or extremely high scores on study change in outcome solely as an impact of treatment/
variable are selected. When repeated testing is done intervention.
on such subjects, they reflect tendency to move closer Threats to external validity: Threats to external
to the mean (i.e., regress). For example, if a group validity compromise of the confidence in stating whether
of subjects was recruited on the basis of extremely the study’s results are applicable to other groups.
high anxiety scores and a psychological intervention The common factors which affect the external validity of a
(progressive muscular relaxation) was done, any research design include the following2-4:
post intervention improvement noted could be due a. Reactive or interaction effect of testing: The term
partly, if not entirely, to regression rather than to the interaction depicts that the effect is not merely due
relaxation technique taught. to treatment rather an interaction between treatment
g. Differential Selection: This threat involves bias and pretesting. This effect occurs when the pre-
resulting from differential selection when assigning testing itself sensitizes or changes the behaviour of
participants to groups. If the groups are different at experimental group. A pre-test might increase or
the outset or in the beginning of experiment, then they decrease the sensitivity and response of the subjects
are most likely to yield different outcomes following with respect to the experimental variable. Thus, the
the intervention/treatment to one of the groups,
results drawn are no more representative of the entire
regardless whether or not the treatment has any
population which is un-pretested unlike the study
effect. While randomization is a measure to eliminate
population. Let’s see an example, a study is conducted
such threat, the design in which randomization is
to assess the effectiveness of ‘Structural teaching
not possible, differential selection bias weakens
programme’ on knowledge about infection control
the internal validity. For example, if a researcher is
in nursing personnel. The baseline knowledge is
conducting a study on effect of “structural teaching
assessed at the beginning followed by the intervention
programme” on attitude towards alcoholics. While
and a post-test assessment is done at the end. In such
the experimental group involve caregivers of patients
experiment, the post-test responses are likely to be
undergoing de addiction treatment and control group
affected by exposure to pre-test itself independent of
involves subjects of nearby residential area. In this
the effect of intervention which will affect the external
case, the two groups may be non - equivalent with
respect to their personal experience of staying with an validity of research design.
alcoholic as well as many other variables which can b. Interaction effect of selection biases and the
influence their attitude. experimental variable: This is a compounded effect
h. Experimental Mortality/attrition: The experimental resulting from the interaction between the selection
mortality also called attrition bias which basically bias and the treatment. This means that some selection
refers to the loss or the dropout of subjects. There can factor interacts with the treatment which would not
be multiple reasons for such loss or dropout including have happened if the group was randomly selected.
death, lack of willingness to participate in follow ups, Let’s see this effect in an example: suppose a study
Section V: Quantitative Research Approaches and Designs

non- availability owing to geographical migration or is being conducted on effectiveness of ‘Assertiveness

at times the negative effect of treatment. In conditions training’ on nursing personnel. The researcher has
where the dropout rates are different in two groups, the chosen a group of nurses working in a tertiary care
groups lose their compatibility at the time of post-test, hospital. In this instance, the personal characteristics
giving faulty results. For example, a researcher plans of selected nurses including their knowledge, skill and
a study to see impact of two specific diet regimens involvement in care are likely to make the treatment
on weight loss. While the first regime is very intense more effective which cannot be generalized to all
and hard specific diet regimens on weight loss. While nurses. Since the findings are not generalizable
the first regime is very intense and hard to follow, the across various settings, thus, there is threat to external
second one little flexible and easy to comply. There is validity.
a high likelihood of greater dropout from first group, c. Reactive effects of experimental arrangements:
especially of subjects with less mental hardiness The subjects tend to behave differently just because
and higher weight. Thus, the two groups will not be they are selected for a study. The experimental
equivalent to be compared at end. arrangement of research makes it difficult to generalize
i. Selection- Maturation interaction: This effect the results to non-experimental settings. Let’s see an
occurs when the rate of maturation/growth from example; a study is being done on driving behaviour
pre-test to post-test are different in experimental and of residents of a community. There is a chance that
control group. Since the rate of maturation varies in subjects chosen might behave differently as they know

4
 Part A: Nursing Research

that all their driving related behaviour is under direct and thus a carryover effect persists. This makes
surveillance. The “Hawthorne effect” is also included it difficult to generalize the outcomes to a single
in this which refers to change in behaviour of an treatment. Considering an example of a depressed
individual due to the attention they are receiving from client who was given sessions of mindfulness therapy
researchers rather than because of any manipulation which did not show any apparent results thus he
of independent variables. was started on another alternative therapy following
d. Multiple-treatment interference: This effect is which the client begins to show improved mood.
evident in conditions when subjects are receiving Now it could be possible that the effect of first therapy
multiple treatments and it is at times difficult and could be prodromal while the second therapy was
at other times impossible to eliminate the effect of started. The researcher may erroneously attribute all
previous treatment when switching to new treatment improvement in mood to second therapy which is in
true sense a combined effect of both therapies.

CLASSIFICATION OF RESEARCH DESIGNS

Fig. 1: Classification of Experimental Research Designs

After getting familiar with what a research design is Each of these types is explained as follows:
and what are the possible threats to external and internal i. One-Shot Case Study: This form of design involves a
validity of given research design, let us see the detailed study of single group of subjects at a single occasion. As
explanation on various research designs depicted in the figure 2, One-Shot Case Study design
has two notations, X & O, which means that there is
While studying the various research designs, the readers are
a single intervention followed by single observation.
expected to be familiar with some symbols originally given by
The assessment of a dependent variable is made
Campbell & Stanley (1963)
subsequent to exposure to some form of treatment.
R: Random selection of subject or random assignment of
The treatment used is presumed to be the cause for the
treatment to experimental groups
change in dependent variable. However, this is done
X: Treatment or the experimental variable or the independent
in absence of any control group. A single instance is
variable. More than one treatment condition is labelled as X1, X2,
carefully compared to general expectations of what
X3 and so on.
the case would have looked like had the treatment not
O: Observation: O1 ,O2,O3 represents subsequent observations.
occurred and to other events casually observed. There
I. PRE EXPERIMENTAL DESIGN is very high likelihood of misplaced precision. This | 11 | Experimental Research Designs
The pre experimental designs are considered as most research design has some inherent limitations. Firstly,
simplest and basic among all research designs. The suffix absence of a comparison group makes it impossible
‘experimental’ affirms to the fact that the experimenter to determine if the outcome scores are any higher
manipulates an independent variable to see its impact than they would have been without the treatment. 
upon a dependent variable. However, the prefix ‘pre’ Secondly, without any pre-test scores, it is impossible
affirms the fact that these designs fail to include a control to determine if any change within the group itself has
group.1 taken place. These factors limit the use of one shot case
The pre experimental designs can be further studies.3 The simplest example could be the subjects
classified into three distinct types: undergo a training to improve emotional intelligence
i. The One-Shot Case Study following which a test of emotional intelligence is
performed on same subjects. The one shot case study
ii. One Group Pretest - Posttest Study.
is diagrammatically depicted in Figure 2.
iii. The Static Group Comparison Study.

5
 Nursing Research in 21st Century

X O Experimental group X O1
Control group No treatment O2
time
time
Fig. 2: One shot case study

Evaluating the design in terms of threats to internal Fig. 4: The Static Group Comparison Study
validity, this design has major weakness owing to history,
As depicted in Figure 4, the experimental group
maturation, selection and mortality. In terms of external
receives a treatment ‘X’ while control group does not
validity, there is threat of interaction of selection and
receive any treatment. However, both groups are similarly
treatment.
tested afterwards. Considering an example, a study is
ii. One Group Pre-test Post-test Study: This form done to assess the effectiveness of expressive emotional
of research design involves the assessment on writing on stress levels. To conduct this study, the two
dependent variable before and after the treatment. groups are chosen. The first group undergoes the sessions
However, like one shot case study, this is done in of emotional writing while second does not undergo any
absence of any control group. The design is superior such sessions. To assess the effectiveness of Emotional
to the One-Shot Case Study as this design enables the writing, both groups are similarly assessed on stress at
comparison of the dependent variable(s) before and the end. The inclusion of one control group removes
after treatment. The One Group Pre-test Post-test the threats of internal validity in terms of history, testing
Study is diagrammatically depicted in Figure 3. In this and instrumentation. The threat related to the selection,
figure O1 is the pre testing or assessment at baseline mortality and interaction of selection and maturation still
which is followed by treatment over time depicted as persists.
X which is further followed by another observation O2
II. QUASI-EXPERIMENTAL DESIGN
i.e. post test assessment.
Like true experimental design, the term experimental
O1 X O2 in this design validates the fact that quasi experimental
designs involve the manipulation of independent variable.
time However the prefix ‘quasi’ which means ‘false’ reflects the
fact that these experiments lack the key feature of random
assignment. Moreover, the researcher lacks full control
Fig. 3: One Group Pre-test Post-test Study over the experimental conditions. The most common of
An example would be, a study conducted to assess all the quasi experimental designs includes the following5
effect of Emotional intelligence training, for which the i. Non–equivalent (Nonrandomized) control group
researcher first assess the baseline EI, followed by training, design:
which is followed by retesting of emotional intelligence. The non-equivalent control group design is a
Owing to absence of control group, the design cannot very commonly employed design. It comprises of an
Section V: Quantitative Research Approaches and Designs

rule out the plausible rival hypotheses (explanations) for experimental group and a control group. The experimental
differences between pre- and post-test (e.g., maturation). group receives treatment and control group do not receive
This very fact explains various threats to the internal any treatment. The assignment to groups is not through
validity of the design including history, maturation, testing, the process of randomization; rather, the researcher use
instrumentation as well as selection and maturation the groups that s/he thinks are similar as the treatment
etc. The threats posed to the external validity are the and control groups. For example choosing two classes, two
interaction of testing and treatment and the interaction of similar schools or may be two similar villages. Although
selection and treatment. the researcher tries to choose the groups as similar as
iii. The Static Group Comparison Study: This design possible so as to have a fair comparison yet, the absolute
takes two groups into account. One group is exposed comparability is not guaranteed. There is a likelihood
to a treatment and the results are tested while a control of non- equivalence of groups and thus the term non-
group is not exposed to the treatment and similarly equivalent control group design. The diagrammatical
tested in order to compare the effects of treatment. depiction of non-equivalent control group design is shown
The experimenter compares the two groups on in Figure5.
dependent variable for the purpose of establishing
the effect of experimental treatment. The design is
diagrammatically depicted in Figure 4.

6
 Part A: Nursing Research

retrospective information on number of educational

Experimental group O1 X O2 programmes attended in previous year, and then compare
Control group O3 O4 it with the similar data in year following launch of such
programme. The Time series design is diagrammatically
time depicted in Figure 6. As shown in figure 6, the group
undergoes periodic observations before and after the
Fig. 5: The non-equivalent control group design intervention.

Let’s consider an example: A study to assess effect O1 O2 O3 X O4 O5 O6

of dance therapy on stress levels of nursing students. Time
The researcher chooses the students of first year in
experimental group and the students of second year in
control group. In such an experimental condition, there is Fig. 6: The time series design
a chance that owing to new environment, homesickness
The time series design effectively rules out the threat
and other interpersonal factors, the first year students may
of maturation, testing and regression. Maturation is ruled
have greater levels of stress. This difference in the baseline
out as it is imperative to assume that the maturational
stress levels can largely affect the outcome
changes are gradual, not abrupt and sporadic. Secondly, if
Evaluating the design in terms of threat to internal
there are any testing effects, they would be evident in the
and external validity, the design has a selection bias. For
initial set of observations which are made before the true
instance, if a researcher is interested in assessing the impact
intervention is performed .Thirdly, like the testing effects
of planned peer interaction among first year students
the regression to mean effect shall also come into sight in
in a college on the adjustment to new environment.
initial set of observations9
Such planned interaction is more likely to be successful
The time series non-equivalent control group
in students who are good at expression, have adequate
design: The researcher also has an option of keeping
communication skills and are high on openness. There is
a control group concurrent to an experimental group,
a possibility that the researcher chooses such subjects in
thus employing a Time Series with Non-equivalent
experimental group, while the other subjects who are poor
Control Group Design. Since the control group is not
in such skills end up being in control group.
chosen through random technique, hence, the term
The second important threat in this design is the
non- equivalent is employed. A Time Series with non-
selection –maturation interaction. Many a times, the
equivalent Control Group Design is diagrammatically
groups selected are such that maturation may proceed at
depicted in figure 7.
a faster rate in that group than the other. We might expect,
for instance, that person’s abstraction ability improves Experimental group O1 O2 O3 X 04 O5 O6
with age (maturation) and that more intelligent person Control group O1 O2 O3 04 O5 O6
will display problem solving skills more quickly and
easily. If the treatment group has subjects proportionately
Time
with high I.Q. the group differences are likely to emerge
on such differential rates of maturation rather than any
intervention planned to improve abstract thinking. Such Fig. 7: A Time Series with non- equivalent Control Group Design
an effect can be referred to as a selection–maturation
Let us see this design in line of an example, Lai et
interaction. The history also poses an important threat in
al (2014)10 conducted a study to see the effectiveness of
this type of design.4
ii. Time series Design
a ‘Facebook-assisted teaching method’ on knowledge | 11 | Experimental Research Designs
and attitudes about ‘cervical cancer prevention and HPV
These designs are also referred as the trend designs vaccination intention’ among female adolescent students.
in which repeated observations are made on dependent The quasi experimental time series deign was employed in
variable before and after the treatment on one group the investigation. The researcher did purposive sampling
of subjects.7 The design is used to assess the change of a control group and an experimental group followed
in behaviour over a large scale. For example if an by pre testing three times T1, T2 & T3 over a period of
organisation launches a new policy of annual appraisal in time. Thereafter, both the groups were delivered lecture
terms of number of educational seminars attended by an on cancer prevention. The experimental group received
employee. Now, if we are interested to know if such launch 6 hour discussion sessions which were face book assisted
would bring about any change in employees behaviour while control group had in person discussions after class.
towards attending such programmes, we may effectively The post test 1 was performed at T1, 2 weeks after T0, and
utilize such design. For this purpose, we might obtain post test 2 was performed at T2, 8 weeks after T0. Time

7
 Nursing Research in 21st Century

series is the only design to furnish a continuous record of The true experiment pose a strong control over the
fluctuations in the experimental variables over the entire known confounding variables, hence, they provide more
course of the program. reliable evidence about the causes and effects. A True
The time series with multiple institution of treatment, experimental design is named so by the virtue of three
time series with intensified treatment, and time series with important properties:
withdrawn and reinstituted treatment (Fig.8 )11 are another 1. Manipulation: The basic aim of conducting an
types of times series design, experiment is to see the effect of an independent
O1 O2 X O3 O4 X O5 O6 X O7 O8 variable upon the dependent variable. This is possible
by the manipulation. The term manipulation in
O1 O2 X O3 O4 X+1 O5 O6 X+2 O7 O8
context of experiment basically refers to doing
O1 O2 X O3 O4 (- X) O5 O6 X O7 O8 something to some of the subjects while withholding
Fig: 8: Time series with multiple institution of treatment the same in others. This ‘something’ which is done
to some of the subjects can be a special intervention
In the figure 8, the first row is representing a time or treatment is the independent variable. Thus, an
series design with multiple institution of treatment. Lets independent variable is consciously manipulated
us see an example a researcher is interested in observing by the researcher to see its effect upon dependent
the impact of an educational programme(X) for nurses variable. For example if we are interested in evaluating
on patient satisfaction. Thus educational sessions are the impact of meditation in performance in a task.
provided at specified interval while patient satisfaction Now, to find answer to this problem, the researcher
(O1-O8) is assessed at intervals over a period. shall make a group of subjects perform meditation
In second row of figure 8, time series with intensified before task while others shall not. In this situation,
treatment is depicted. For instance, a researcher intends the meditation acts as an independent variable while
to see the impact of a new drug in diabetes on glycemic performance is a dependent variable. Thus, the
control. In such scenario the two readings of Hba1c at researcher manipulates the independent variable to
taken before initiating a drug, then two readings after test the hypothesis by comparing the performance of
initiating a drug. Now the researcher increase the dose subjects who did meditation to those who did not.
and further makes two set of observations and thereafter a 2. Control: Achieving a control is second important
further dose is increased and observations are made. characteristic of true experimental design. The
In third row, figure 8, time series with withdrawn and experimenter exercise control over various extraneous
reistituted treatment. Lets us understand with an example. variables. The randomization and manipulation
A researcher is interested to see the impact of a mobile are some of the ways to control such variables. The
based personal coach application on weight reduction. other way of control is the use of a control group.
Thus two observations are made prior to the intervention, The term control group refers to a group of subjects
then such phone based application is introduced whose performance on dependent variable is used to
followed by two observations, Thereafter the phone based evaluate the performance of the experimental group.
application is withdrawn and further two observations For example, if we are interested to see the effect of 30
Section V: Quantitative Research Approaches and Designs

are made. And lastly the intervention is re introduced and minute aerobic exercise on weight. Now, if we have
observation are made post intervention. a single group of subjects who take a specific diet for
Advantages of Quasi Experimental Research Design one month period and also show reduction in their
‰‰ Most practical and feasible research design especially weight at the end of one month. However, based on
in nursing. the weight reduction seen in group, can we conclude
‰‰ Less time consuming. that specific diet has caused reduction in weight. There
‰‰ Resources needed for experimentation are also might be a possibility that the subjects were following
reduced. a special exercise regimen as well. Hence, keeping a
‰‰ Enable to compare with other groups. control group who is similar to experimental group in
Disadvantages of Quasi Experimental Research Design all aspects except that they do not take the specific diet
‰‰ It provides comparatively weaker evidence of effect of becomes necessary to conclude the results.
an intervention. 3. Random Assignment12-14: Random assignment is
‰‰ Unexpected factors might affect the results. “the process of assigning individuals at random to
different groups in an experiment”. The purpose of
TRUE EXPERIMENTAL DESIGNS1,5 random assignment is to insure that groups receiving
different treatments are as reasonably equal or similar
True experimental design is regarded as the most in any way that could possibly impact the outcome
accurate and strongest form of experimental research. (or dependent variable). This would mean that the

8
 Part A: Nursing Research

characteristics of the subjects which can have a direct groups: experiment, 25 patients; and control, 25 patients.
or indirect impact upon the outcome of study are All the study subjects are pre tested using Zerit Burden
evenly distributed in two groups. Interview Schedule. While the experimental group shall
The true experimental designs can be employed in get Psycho-education Programme (4 sessions) while
various ways. Some of the most common true experimental control would not receive any treatment. In this particular
designs are being discussed in forthcoming section: example, the author would manipulate the independent
Types of True Experimental Research Designs variable ‘Psycho-education Sessions’ on one group to see
i. Pre test- Post test Control group Design its effect on dependent variable ‘Caregiver’s burden’. At the
This design is considered to be the most effective end of the experiment, both the groups will be tested again
design in testing a cause effect relationship. The design on the ‘Caregiver’s burden.’15
comprise of two groups which are equivalent as they have Evaluating this design in terms of internal validity,
been chosen through randomization. One of the groups the design controls for most of the threats. The threat for
that receive the treatment is the experimental group while history is controlled as if any historical event has affected
the other which does not receive any treatment is the the outcome, it is likely to produce similar effects in
control group. The diagrammatic depiction of design is the control group as well provided both the groups are
shown in Figure 8. assessed simultaneously. The threat of maturation and
testing are also controlled as they manifest equally in both
groups. The threat of regression is too controlled because
the subjects in experimental and the control group have
been randomly picked up from same population. So if
any regression to mean occurs, it is likely to occur in both
groups irrespective of the treatment received by the group.
However, there remain threats to the external validity
Fig. 8: Pre test- Post test Control group Design in this form of design. There is a threat for interaction
As depicted above, the pre- test and post test is given to of testing and treatment, interaction of selection and
both the groups. Such design provides a simple way to test treatment as well as reactive arrangements.2
the effectiveness of treatment on the dependent variable ii. Post-test only control group Design
which can be analysed by making a comparison of gain This is a two group design including an experimental
scores in experimental and control group respectively group and a control group, where assignment to groups
i.e. comparing O2- O1 (mean gain score in experimental occurs through randomization. The experimental group
group) with O4-O3 (mean gain score in control group). is exposed to the manipulation; the control group is not. 
Evaluating this design in terms of external and internal The assessment on dependent variable is made post
validity, the design controls for most of the threats. The intervention only, as the design does not use a pre-test
threat for history is controlled as if, any historical event has measure to establish a baseline.16 This design can be
affected the outcome, it is likely to produce similar effects diagrammatically depicted as shown in Figure 9. Figure 9
in the control group as well provided both the groups are clearly depicts that subjects are assigned to either of the
assessed simultaneously. The threat of maturation and two groups i.e. experimental or control groups through
testing are also controlled as they manifest equally in both random assignment. The experimental group receives
groups. The threat of regression is too controlled because a treatment ‘X’ followed by post- test (O1). The control
the subjects in experimental and the control group have group does not receive any treatment while post test is
been randomly picked up from same population. So if any conducted (O2).
regression to mean occurs, it is likely to occur in both groups | 11 | Experimental Research Designs
irrespective of the treatment received by the group. While
the group controls for most of the sources causing threats
to internal validity except the testing effect. The testing
effect is the gain in post test scores due to the experience
of undergoing pre- test itself. Secondly, the researcher
has no control over the sensitization to treatment i.e. the Fig. 9: Post-test only control group Design
interaction of testing and treatment is not controlled.
Let us consider the Pre test post test Control group Let us consider the post test only group design in
Design in an example: A study is conducted a study to an example: Batra (2014) conducted a study entitled
assess Effectiveness of Psycho educational programme ‘Application of ice cube’ prior to subcutaneous injection of
in reducing the caregiver’s burden. To conduct the heparin on ‘pain perception and ecchymosis’ of patients
experiment participants are randomly allocated to two with cardiovascular problems.’ The independent variable

9
 Nursing Research in 21st Century

for the study was the ice cube application for 3 min and the researcher can presume that the pre-test has had some
the dependent variables were pain perception and effect upon the results.19-20
ecchymosis. To answer the research question of whether Let us consider an example : Chang et al (2014)
application of ice cube prior to subcutaneous injection conducted a RCT to assess the effect of an asynchronous
of heparin shall bring any change in pain perception e-learning curriculum on the knowledge level of medical
and ecchymosis, a total of 60 subjects were randomized residents.16 The Solomon four group deign was employed
to experimental and control groups. The researcher for the same. Block randomization was done to assign
manipulated the independent variable in experimental each participant to either of the four groups. The Group
group while assessment on pain score and ecchymosis A received the pre test followed by access to the e learning
was performed on both groups.17 modules which was followed by the post test. While group
Evaluating this design in terms of internal and B was pre tested at beginning of rotation and post tested
external validity, since there is no pre testing so this design at end without receiving any access to e learning module.
eliminates the threats associated with pretesting effect and The group C was given access to e learning and post test
pre-test-manipulation interaction bias. This type of design was done, however, pretesting was not done. Lastly, group
also minimizes the experimental mortality by reducing the D was post tested only, there was neither pre-test nor
overall duration of study. intervention. The results of the study revealed no pre-test
iii. Solomon Four group Design18 sensitization while use of e learning modules improved
Also referred as the hybrid experimental design, this the knowledge significantly.21
design makes use of four groups. It is a combination of Despite of strengths in terms of internal and internal
standard pre-test-post-test two-group design and the post validity, the Solomon four group design is not often
test only controls design. To have a better understanding, used owing to requirement of large sample, along with
consider the diagrammatic depiction of the Solomon four the difficulty faced by researcher in introducing all the
group design as shown in figure 10. treatments to four groups simultaneously. The researchers
also find it cumbersome to perform randomization in four
groups and lastly the burden of performing the intricate
statistical analysis. 22
iv. Factorial Design: There are conditions in which the
research question is not limited to see the effect of only
one independent variable. The researcher might be
interested in manipulating two or more independent
variables labelled as factors. Hence, a factorial
Fig. 10: Solomon four group design research design is utilized in such investigations
As depicted in Figure 10, the random assignment is where researcher aims to assess the effect of two or
used to make four separate groups. The group 1 undergoes more independent variables upon a single dependent
a pre-test followed by treatment, further followed by a post variable. In conventional terms, the independent
test. Group 2 is a control group that undergoes both pre- variables are labelled as factors. The number of ways
Section V: Quantitative Research Approaches and Designs

test as well as post, but there is no treatment. The group in which a factor varies is labelled as the levels. An
3 is also an experimental group that receives a treatment experimental involving two independent variables
followed by post -test only. The group 4 is a control group (factors) with two levels for each factor would be
that has post- test only, in absence of either pre -test or represented as: 2X2 factorial design. The experiment
treatment. This particular design enables the researcher to is planned in such a way that all levels of independent
make multiple comparisons. The researcher can measure variable are combined with all levels of another. This
the direct effect of treatment by comparing group 1 and enables the researcher in observing the direct effect
group 2. The comparison of post test results of group 1 of independent variable on dependent variable as
with the post test results of group 3 enables to researcher well as the interaction effects of the various levels
to conclude the testing effects i.e. the condition where of independent variable on dependent variable.
the act of taking a test itself affects the performance on a Lets consider an example: A study on the effects
retest or post-test. The comparison of pre- test post- test of nicotine gum and counselling among African
scores of group 2 with that of group 4 helps to infer if there American light smokers using a 2 × 2 factorial design.
are any external factors beyond treatment itself that are In this particular example, the independent variables
responsible for change in post test scores. Lastly, if the included the use of 2mg nicotine gum and the use of
difference between the post-test results of Group 3 and motivational interviewing and health education while
Group 4 is different from the group 1 and group 2, then the dependent variable was 7 day quit rates. Here, there
are two factors i.e type of gum and type of counselling

10
 Part A: Nursing Research

and each factor has two levels. Thus factor refers to v. Randomised Block Design24-26 While conducting an
the number of independent variables and term level experiment, if the researcher feels that any known
refers to the subdivision of a factor. So if it is a 3X 3 variable is likely to produce some degree of variation in
factorial design, this shall mean that there are three the findings, then, in order to minimize that variance,
independent variables and each independent variable randomised block design can be employed. With a
has further three levels. Such a factorial design shall randomized block design, the experimenter divides
generate a total of 9 groups. (3X3=9) subjects into subgroups called blocks, such that the
variability within blocks is less than the variability
Type of counselling
between blocks. This implies that each block
Health Motivational introduced is a homogenous subgroup. The researcher
Education Interviewing controls the variability by blocking. Thereafter
B1 B2 subjects within each block are randomly assigned to
2mg Nicotine gum A1B1 A1B2 all treatments. Each block contains a complete set of
A1 treatments, therefore differences among blocks are not
Placebo gum A2B1 A2B2 due to treatments. This variability can be estimated as
Type of gum A2 a separate source of variation.
See this example
Fig. 11: Factorial designs
A study was carried out to evaluate the effect of
The figure above depicts a 2x2 factorial design which exercises on shoulder dysfunction and lymphedema
can help the researcher to test multiple hypotheses, by amongst the patients undergoing breast surgery. The
manipulating more than one independent variable. In patients were randomized into three groups. Group I
this study the subjects shall be assigned to one of the four received the intervention Presurgery; Group II received
arms: the intervention postsurgery, and Group III served as
a. The first group will receive 2mg nicotine gum and a control group. Block randomization was done. Each
health education block containing one A, B and C. Single sequence of block
b. The second group will receive 2mg nicotine gum with randomization was used. Order of these was randomly
motivational interviewing. permuted (rearranged) so that all possible permutations
c. The third group will receive placebo gum with health are created. Thus, in total we had 6 blocks.
education 1. “ABC
d. The fourth group will receive placebo gum with 2. ACB
motivational interviewing. 3. BCA
The researcher shall test various research questions with 4. BAC
this design : 5. CAB
a. Is 2mg nicotine gum beneficial in tobacco cessation 6. CBA”
than placebo gum A random number sequence was used to choose a
b. Is the health education with placebo gum effective particular block, which sets the allocation order for the
in tobacco cessation compared to 2mg nicotine gum first three subjects. Similarly, new number is selected-
with heath education corresponding block was allocated to the next three
c. Is motivational interviewing more effective than subjects. The process was then repeated. From the
health education in tobacco cessation Random number Table a number was chosen (e.g. first no.
which we got was: 4-BAC, then 3- BCA, then 6- CBA and
| 11 | Experimental Research Designs
d. Is motivational counselling with 2mg nicotine gum
more beneficial than motivational counselling with so on…. till we reach at our targeted 330). Permuted block
placebo gum. randomization ensured intervention groups (A and B) and
Thus, the design shall test the main effect of using a control group (C) numbers are evenly balanced.
nicotine gum and placebo gum, the main effect of the type vi. Crossover designs 27
of counselling used. The superiority of the design lies in The crosses over designs are also commonly referred
the fact that it shall assess the interaction effects of type as the repeated measures design. There is more than
of gum and type of counselling used. An interaction effect one independent variable to be manipulated. However,
exists when differences on one factor depend on the level the subjects who are exposed to different treatment
you are on another factor. The factorial designs are highly conditions remain the same. Such designs are thus also
flexible designs with great utility in examining treatment called the within subject designs. The most distinct feature
variations. of crossover design is that each patient serves his/her own
control. This also ensures the equivalence among the

11
 Nursing Research in 21st Century

groups exposed to treatment conditions. And hence, the Advantages of Experimental Research Design
problems of comparability of experimental versus control ‰‰ Experimental research designs are the most powerful
group with regard to various confounding variables are methods to establish the causal relationship between
resolved. The two terms used in cross over design are the variables.
sequence and period. The sequence refers to the order ‰‰ These designs provide the highest quality evidence
of treatment administration and the period is the time of regarding the effects of specific interventions.
treatment administration. The design can be represented ‰‰ There is better extent of purity in observation as the
diagrammatically as shown in Figure 12. study is conducted under controlled environment.
‰‰ Conditions that are not found in natural setting can
be created in experimental setting in a short period of
time that may take years to naturally occur (therefore
very useful in genetic studies).
‰‰ Because the study is carried out in experimental
setting the problems of real life situations and the
personal problems of the researcher is eliminated.
‰‰ The environment in which the research takes place
can often be carefully controlled. Thus, it becomes
Fig. 12: Cross over design easier to assess the true effect of the variable of interest
Let us consider an example of crossover design: on the outcome of interest.
Khalkhali et al (2014) conducted a study on effect of Disadvantages of Experimental Research Design
applying ‘cold gel pack’ on the pain associated with ‘deep ‰‰ Though there are number of benefits of experimental

breathing and coughing’ after open heart surgery. To research design, it has certain limitations also. These
assess the effect of using ‘cold gel therapy’ on postoperative are
pain associated with deep breathing a coughing, the ‰‰ Sometimes, there could be a problem if the subjects

researcher used the cross over design ad recruited 50 themselves have the discretion about participation in
subjects28. The subjects were divided into two groups the treatment.
using random allocation. As depicted, the subjects in both ‰‰ If the experiment is conducted in the lab, and the

groups were pre assessed for pain. In group 1, the cold gel intervention is being carried out by the person other
application was done for 15 minutes following which the than the researcher himself, then it may be difficult
deep breathing and coughing was performed and pain to determine that if the subjects in the experimental
assessment was done. While in group 2, the subjects were group have actually received the intervention.
prepared for deep breathing and coughing without any ‰‰ In experimental studies conducted in natural settings

cold gel application. Group 2 subjects were also assessed like a hospital or community, it may not be possible to
for pain while deep breathing and coughing exercises impose control over extraneous variables.
were performed. This was followed by a washout period ‰‰ Experiments are often more impractical when the

of 2 hours. After the washout period, there was a cross over effect of independent variable may require a lengthy
Section V: Quantitative Research Approaches and Designs

which means that in period 2 , the group 1 performed deep period of time before it can emerge as a response on
breathing and coughing without cold gel application while the criterion measures.
group 2 performed it with cold gel application , and pain ‰‰ It is very difficult to obtain permission from the

was assessment was done. participants.

‰‰ Because the size of the sample is kept small especially
Thus, crossover designs are highly beneficial designs
provided utilized rationally. It is very essential that the studies involving humans, the representativeness of
researcher ensures a washout period between two trials the findings of such study is questionable.
‰‰ Experimental situation may not relate to the real
being compared to rule out the carryover effect in such
design. The carryover effect refers to the impact of first world.
‰‰ It may be unethical or impossible to randomly assign
treatment condition while subject is being assessed for
the second. In many situations, it might be tedious to people to group e.g. deliberately depriving the children
test the assumption of carryover effects or it might not off the sleep.
be possible to determine the precise length of washout Summary table of pre experimental, quasi
period. Such situations limit the use of crossover designs. experimental and true experimental research designs
Also mentioned the another limitation of crossover design After a detailed view on pre experimental, quasi
revealing the “period” effects in addition to carryover experimental and true experimental research designs, let
effects viz. Progression of disease and dropouts. us have a brief summary table on unique characteristics of
each type of design.

12
 Part A: Nursing Research

Characteristics Pre experimental design Quasi experimental design True experimental design
1. Introduction of independent variable
(Manipulation)
2. Inclusion of a control group (Control)
3. Participants are randomly assigned
to either the treatment or the control
group.
(Random Assignment)
*Green indicates presence while red indicates absence.

SUMMARY AND KEY CONCEPTS

The chapter gave a detailed description on various types of research designs on the basis of type of experiment. The
chapter entailed upon the most basic forms of research beginning from the pre experimental to quasi experimental to
the true experimental designs. While each research design has its own inherent strengths and weaknesses, however,
choosing the right design for any investigation is largely affected by the nature of research question and the resources
available to conduct study. It is imperative to critically appraise a design opted for any study in terms of its external and
internal validity to get most precise results.

| 11 | Experimental Research Designs

13
 Nursing Research in 21st Century

A ssess Y ourself
5. Which of the following is not a property of Pretest-
Long and short answer type questions Post test Control group Design :
a. The threat for history is controlled.
1. Describe the classification of research designs. Discuss
b. The threat of maturation and testing are also
in detail the true experimental research designs.
controlled.
2. What are the various threats to external and internal c. Threat of regression towards mean is controlled.
validity of a research design? d. Threat of testing effects is controlled.
3. What are the characteristics of true experimental 6. A 2X3 factorial design depicts that:
research designs? Describe the cross over design with a. There are two independent variables and each
appropriate example. independent variable has further three levels.
4. Define internal validity of research design. b. There are three independent variables and each
5. Define external validity of research design. independent variable has further two levels.
c. There are total five groups.
6. Enlist the different types of quasi experimental types.
d. There are total two groups.
7. Describe briefly a randomised block design.
7. A research design in which each patient serves his/her
8. Explain a time series design. own control are commonly referred as
Multiple choice questions a. Cross over design
1. The evidence to assert, whether or not, experimental b. Factorial design
treatment/condition has made the difference refers c. Non equivalent control group design
to: d. Solomon four group Design
a. External validity 8. The bias resulting from difference in the rate of
b. Internal Validity maturation/growth from pre-test to post-test are in
c. Criterion validity experimental and control group is called:
d. Reliability a. Selection- Maturation interaction
2. The degree to which the results of the experiment can b. Maturation
be generalized to the entire population refers to: c. History
a. External validity d. Interaction effect of testing
b. Internal Validity 9. The given icon represents which research design:
c. Criterion validity
d. Reliability Experimental group O1 O2 O3 X 04 O5 O6
Control group O1 O2 O3 04 O5 O6
3. The designs in which repeated observations are made Time
on dependent variable before and after the treatment
on one group of subjects are referred as:
Section V: Quantitative Research Approaches and Designs

a. Time Series with non- equivalent Control Group

a. Time series design Design
b. The non- equivalent control group design b. Factorial design
c. Pretest- Post test Control group Design c. The time series design
d. Post-test only control group Design d. The non- equivalent control group design
4. Which of the following is not a property of true 10. Which of the following are the threats to internal
experimental design: validity of research designs:
a. Manipulation a. Multiple-treatment interference
b. Control b. History
c. Random Assignment c. Maturation
d. Stratification d. Experimental mortality

Ans.
1. b 2. a 3. a 4. d 5. d 6. a 7. a 8. a 9. a
10. a

14
 Part A: Nursing Research

R eferences
illness attending psychiatry OPD and admitted in psychiatry
1. Singh AK. Tests, Measurements and Research Methods in
Ward. [M.Sc. Nursing Thesis]. National Institute of Nursing
Behavioural sciences. 5th edition .Patna: Bharti Bhavan,
Education :Post graduate Institute of Medical Education &
2010.
Research, Chandigarh,India;2015
2. Ohlund B & Yu Chong-Ho. Threats to validity of Research
16. Nancy D & Miller – Albers. Post test only control Group . Cited
Design. Cited [ 2015 March1].AvailableFromhttp://web.pdx.
[ 2015 March 3]Available from http://facultyweb.berry.edu/
edu/~stipakb/download/PA555/ResearchDesign.html
nmiller/classinfo/323/2.4/true_control_after_only.htm
3. Campbell, D. & Stanley, J. (1963). Experimental and quasi-
17. Batra G. Application of ice cube prior to subcutaneous
experimental designs for research. Chicago, IL: Rand-
injection of heparin in pain perception and ecchymosis
McNally.
of patients with cardiovascular problems. Nurs J India.
4. Fife-Schaw C. Quasi experimental Designs 2012 Cited
2014;105(4):155-9.
[2015 March 4]Available from http://www.sagepub.in/upm-
18. William MK. Hybrid Experimental Designs. Cited [March
data/46877_Breakwell_Ch04.pdf
2015]. Available from http://www.socialresearchmethods.
5. Polit DF & Hunglar BP. Nursing research: Principles and
net/kb/exphybrd.php.
methods. 5th edition. Philadelphia: Lippincott. 1997
19. Mchae TW & Tingen MS. The Use Of The Solomon Four-
6. Yoo MS, Park JH. Effect of case-based learning on the
Group Design In Nursing Research. Southern online journal
development of graduate nurses’ problem-solving ability.
of nursing research .9(1) http://www.resourcenter.net/
Nurse Educ Today. 2014;34(1):47-51. doi: 10.1016/j.
images/snrs/files/sojnr_articles2/Vol09Num01Art14.pdf
nedt.2013.02.014
20. Lavanyakumari P. Significance of Solomon four group
7. Gottman, JM, Mcfall RM & Barnett JT. Design and analysis of
pretest-posttest method in True Experimental Research- A
research using Time series. Psychological Bulletin 1969, 72(
Study. IOSR-JAVS 2013; 5(2):51-58.
4) , 299-306
21. Chang et al. Paediatric Emergency medicine acynchronous
8. Lai CY, Wu WW, Tsai SY, Cehng SF , Lin KC & Liang SY. The
E- Learning. A multicenter randomized controlled
Effectiveness of a Facebook-Assisted Teaching Method
Solomon four group study. Academic Emergency
on Knowledge and Attitudes About Cervical Cancer
Medicine.2014;21(8);912-919.
Prevention and HPV Vaccination Intention Among Female
22. Mcghaee TW and Tingen MS. The Use Of The Solomon Four-
Adolescent Students in Taiwan. Health Educ Behav.2014; pii:
Group Design In Nursing Research. SOJNR 2009; 9(1): 1-7.
1090198114558591. [Epub ahead of print]
Available from http://www.resourcenter.net/images/snrs/
9. PsychoMetrics. Threats to Internal Validity. Cited [2015
files/sojnr_articles2/Vol09Num01Art14.htm
March 4]Available from http://www.psychmet.com/id12.
23. Alhuwalia JS. The effects of nicotine gum and counseling
html
among African American light smokers: a 2 x 2 factorial
10. Nancy D & Miller- Albers. Posttest Only, Control Group. Cited
design.Addiction. 2006 Jun;101(6):883-91
[2015 March 3] Available from http://facultyweb.berry.edu/
24. Randomised Block Design. Cited [April 2017]
nmiller/classinfo/323/2.4/true_control_after_only.htm
Available at http://stattrek.com/statistics/dictionary.
11. Polit DF, Beck CT, Nursing Research: Generating and assessing
aspx?definition=Randomized%20block%20design
evidence for nursing research. Wolters Kluwer (India) Pvt.
25. Dallal GE. Randomized (Complete) Block Designs. Available
Ltd., New Delhi. Eighth Edition. Page 267-271.
at http://www.jerrydallal.com/lhsp/ranblock.htm
12. Creswell J. Educational research: Planning, conducting,
26. Willia, MK. Randomized Block Designs. Available at https://
and evaluating quantitative and qualitative research. New
www.socialresearchmethods.net/kb/expblock.php
Jersey: Pearson: Merrill Prentice Hall. 2008
27. Blettner M & Welleck S. On the Proper Use of the Crossover
13. Slavin R. Educational research in an age of accountability.
Design in Clinical Trials: Part 18 of a Series on Evaluation of | 11 | Experimental Research Designs
Boston: Pearson Education. 2007
Scientific Publications. Dtsch Arztebl Int 2012; 109(15): 276-
14. Characteristics of Experimental Research. Cited [2015 June
81; DOI: 10.3238/arztebl.2012.0276
10]. Available from https://www.aea267.k12.ia.us/system/
28. Khalkhali H R, Tanha, ZER, Feizi A & Ardabili SS. Effect of
assets/uploads/files/1468/experimental_research.pdf
applying cold gel pack on the pain associated with deep
15. Kaur M, An experimental study to assess the effect of
breathing and coughing after open heart surgery. Iran J Nurs
caregiver support interventions on the subjective burden
Midwifery Res. 2014 Nov-Dec; 19(6): 545–549
of caregivers of patients suffering from severe mental

15

View publication stats