LESSON 7: DESIGNING

QUANTITATIVE STUDIES ⦿ Quantitative experiments are useful for

testing the results gained by a series of
Quantitative research refers to the qualitative experiments, leading to a final
systematic empirical investigation of answer, and a narrowing down of
social phenomena via statistical, possible directions for follow up research
mathematical or computational to take.
techniques. The objective of quantitative
research is to develop and employ DISADVANTAGES
mathematical models, theories and/or ⦿ Quantitative experiments can be
hypotheses pertaining to phenomena. difficult and expensive and require a lot of
time to perform. They must be carefully
The process of measurement is central to planned to ensure that there is complete
quantitative research because it provides randomization and correct designation
the fundamental connection between of control groups.
empirical observation and mathematical
expression of quantitative relationships. ⦿ Quantitative studies usually require
extensive statistical analysis, which can be
Quantitative data is any data that is in difficult, due to most scientists not being
numerical form such as statistics, statisticians. The field of statistical study is
percentages, etc. a whole scientific discipline and can be
difficult for non-mathematicians.
In layman's terms, this means that the
quantitative researcher asks a specific, Examples of Quantitative Research:
narrow question and collects numerical ⦿ Research that consists of the
data from participants to answer the percentage amounts of all the elements
question. The researcher analyzes the that make up Earth's atmosphere.
data with the help of statistics. The
researcher is hoping the numbers will yield ⦿ Survey that concludes that the average
an unbiased result that can be generalized patient has to wait two hours in the waiting
to some larger population. room of a certain doctor before being
selected.
ADVANTAGES
⦿ Quantitative research design is an ⦿ An experiment in which group x was
excellent way of finalizing results and given two tablets of Aspirin a day and
proving or disproving a hypothesis. The Group was given two tablets of a placebo
structure has not changed for centuries, a day where each participant is randomly
so is standard across many scientific assigned to one or other of the groups.
fields and disciplines. The numerical factors such as two tablets,
percent of elements and the time of
⦿ After statistical analysis of the results, a waiting make the situations and results
comprehensive answer is reached, and quantitative.
the results can be legitimately discussed
and published. Quantitative experiments ⦿ In finance, quantitative research into
also filter out external factors, if properly the stock markets is used to develop
designed, and so the results gained can models to price complex trades, and
be seen as real develop algorithms to exploit investment
and unbiased.
hypotheses, as seen in quantitative Hedge ⦿ Alternative versions of independent
Funds and Trading Strategy Indices. variables under study
⦿ Criterion measures evaluating the
WHAT IS RESEARCH DESIGN? effects of the independent variable
⦿ Research design is the researcher’s ⦿ Statistical treatment of data, using
plan of how the study will be conducted, various tools
the type of data that will be collected, the
techniques or the means to be used to Research Designs
obtain these data, the reasons for the
choice of the subjects, the manner of Different types of research designs have
determining sample size, the instruments different advantages and disadvantages.
to be used and their validation , and the
data analysis scheme which includes the ⦿ The design is the structure of
application of the statistical tools for any scientific work. It gives direction and
treatment of data arising from the study. systematizes the research.
These are determined after variables are
identified and quantified. ⦿ The method you choose will affect your
results and how you conclude the findings.
RESEARCH DESIGN HAS: Most scientists are interested in
⦿ Detailed explanation of techniques and getting reliable observations that can help
procedures for data collection and the understanding of a phenomenon.
analysis (Research Design).
DIFFERENT RESEARCH METHODS
⦿ Description of the target population, the
location or setting and the procedure for (1) Descriptive Designs
choice of samples or respondents Aim: Observe and Describe
(Sampling Design or Procedure). ⦿ Descriptive Research
⦿ Case Study
⦿ Clear description of tools and ⦿ Naturalistic Observation
instruments to be used and the statistical ⦿ Survey (The Questionnaire is also a
treatment of data to be applied technique used in many types of research
(Instrumentation and Statistics) designs)

CRITERIA FOR CHOICE OF DESIGN: (2) Correlational Studies

⦿ Logically sound Aim: Predict
⦿ Comprehensive enough to solve ⦿ Case Control Study
problem ⦿ Observational Study
⦿ Reasonably clear and precise ⦿ Cohort Study
⦿ Clearly spelled out ⦿ Longitudinal Study
⦿ Free from weaknesses ⦿ Cross Sectional Study
⦿ Correlational Studies in general
ELEMENTS TO CONSIDER IN THE
CHOICE OF RESEARCH DESIGN (3) Semi-Experimental Designs
⦿ Total population from which to select Aim: Determine Causes
the study sample ⦿ Field Semi-Experimental Designs
⦿ Method of selecting the samples or ⦿ Experiment
study respondents ⦿ Quasi-Experimental Design
⦿ Twin Studies
 (4) Experimental Designs EXPERIMENTAL RESEARCH
Aim: Determine Causes ⦿ This particular design is an inquiry on
⦿ True Experimental Design cause-and-effect relationships, and is
⦿ Double-Blind Experiment conducted in a specialized setting, such
as the laboratory, experimental unit or
(5) Reviewing Other Research research center.
Aim: Explain The researcher controls and manipulates
⦿ Literature Review the independent variable and randomly
⦿ Meta-analysis assigns the subjects to different conditions
⦿ Systematic Reviews or situations.

Test Study Before Conducting a TYPES OF EXPERIMENTAL RESEARCH

Full-Scale Study Aim: Does the Design 1. True Experiment:
Work? An inquiry that considers four (4)
⦿ Pilot Study properties of experimental research:
1. Manipulation
TYPICAL EXPERIMENTAL DESIGNS 2. Control
SIMPLE EXPERIMENTAL TECHNIQUES 3. Randomization
⦿ Pretest-Posttest Design 4. Validity
⦿ Control Group
⦿ Randomization EXAMPLE:
⦿ Randomized Controlled Trials ⦿ “The risk of Post-Portal Infections to
⦿ Between Subjects Design Patients with External Flushing Compared
⦿ Within Subject Design to Patients without External Flushing.”
⦿ “The use of Mask to Prevent Nursery
COMPLEX EXPERIMENTAL DESIGNS Infections”
⦿ Factorial Design
⦿ Solomon Four-Group Design PROPERTIES OF TRUE EXPERIMENTS
⦿ Repeated Measures Design 1. Manipulation- the researcher
⦿ Counterbalanced Measures Design subjects the respondents of the
⦿ Matched Subjects Design study to a particular situation for a
⦿ Bayesian Probability specific purpose. The
experimenter consciously varies
WHICH METHOD TO CHOOSE? the Independent variable and
⦿ What design you choose depends on observes the effect that the
different factors. manipulation has on the dependent
⦿ What information do you want? variable of interest.
2. History- this occurs when some
The aims of the study. event besides the experimental
⦿ The nature of the phenomenon - Is it treatment takes place during the
feasible to collect the data, and if so, course of the study and affects or
would it be valid/reliable? influences the dependent variable.
⦿ How reliable should the information be? 3. Maturation- this takes place when
⦿ Is it ethical to conduct the study? changes within the subjects
⦿ The cost of the design occurs during the experimental
⦿ Is there little or much current scientific study, thus, influencing study
theory and literature on the topic? results.
 the average scores in the post-test of the
EXAMPLE OF MATURATION: experimental group would be
⦿ The subjects may have gained in correspondingly low.
height and weight at a time when the
pre-test and the post-test were being ALL EXPERIMENTAL STUDIES
administered. If the nurse is interested in SHOULD HAVE THE FOLLOWING
gain in weight and height of malnourished CHARACTERISTICS:
children, she will keep in mind that change
in their size may occur during the 1. Internal Validity- this means the
treatment or course of the study. degree to which changes in the
dependent variable (effects) can be
4. Testing- possible testing threat in attributed to the independent variable
studies in which a pre-test is (cause).
requisite. This refers to the
influence of the pre-test, which Threat to Internal Validity:
already projects the results of the Selection Bias - this exists when study
post-test scores. results are attributed to the experimental
treatment, when in fact, results are due to
EXAMPLE OF TESTING: differences among the subjects even
⦿ Test scores of actual study respondents before treatment.
may be altered in the post-test as a result
of their knowledge of the pre-test results. 2. External Validity-this is the degree to
which study results can be influenced or
5. Instrumentation Change- the affected by other populations and
existence of a difference setting.
between pre-test and post-test
results caused by change in the Threat to External Validity:
accuracy of the instrument or the a. The Hawthorne Effect- this
ratings, rather than a result of the occurs when study participants
experimental treatment. respond in a particular manner, or
there is an obvious change of
EXAMPLE OF INSTRUMENTATION behavior because they are aware
CHANGE: that they are being observed.
⦿ A mechanical instrument such as the
sphygmomanometer used in taking the EXAMPLE OF HAWTHORNE EFFECT:
blood pressure affects the accuracy of ⦿ A student nurse applies correct
reading throughout the study. technique in doing nursing procedures
while the researcher observes her.
6. Mortality- this occurs when a
difference exists between the b. The Experimenter Effect - this
subject dropout rates of the refers to a threat to study results
experimental and the when the researcher’s behavior
non-experimental group. influences the behavior of the
subjects such as the researcher’s
EXAMPLE OF MORTALITY facial expression, gender, and
⦿ If a large experimental group scored clothing among others.
very low in a pre-test, some respondents
may have dropped out of the study. Thus,
 c. The Reactive Effect of the ⦿ Thus, a researcher interested in
Pre-Test. This occurs when the quantitative data or planning to perform
subjects have been sensitized to quantitative analysis needs to understand
the treatment by taking the the characteristics of quantitative design.
pre-test.
Independent Variable
d. The Halo Effect- this is the ◼ Every quantitative design must specify
tendency of the researcher to rate an independent variable. The independent
the subject high or low because of variable is the one that the researcher
the impression he/she has on the believes has an effect on the
latter. phenomenon under investigation.

- For the researcher to minimize For example, if a researcher is planning to

threats to external validity, the investigate how the amount of money a
double blind method may be man invests in the stock market can
used to remove the observer’s predict his number of past romantic
bias, due to the fact that neither partners, he will set "amount of money
the subject nor the observer knows invested in the stock market" as the
the specific research objectives or design's independent variable.
specific subjects who belong to the
experimental or control group. Dependent Variable
Hence, the observer cannot distort ◼ Researchers must state the study's
the data. dependent variable in the quantitative
design.
- If the double blind method is not
feasible, the double observer Quantitative designs make a hypothesis
method may be used to determine about how one variable (the independent
the extent of bias between the two variable) affects another variable.
observers as they observe and
record the subjects’ performance is This other variable is the dependent
a dependent variable. variable; it is the variable in which
researchers wish to observe change.
The Key Characteristics of Quantitative The idea of a quantitative design is to
Design: construct a project in which the
⦿ Quantitative design precedes data independent variable and dependent
collection and analysis. variable are related. Thus, a researcher
must specify the dependent variable in her
⦿ Before collecting data and performing design.
analyses, researchers of quantitative
properties must design an appropriate For the previous example, where the
approach to answering the question of researcher is interested in seeing the
interest. effect of investment size on the number of
romantic partners, "number of romantic
⦿ A researcher employs quantitative partners" is the dependent variable.
design to assure that her research
methods are valid and will yield Sample
scientifically proper results.
◼ The quantitative design also specifies performance of the subjects on
how the researcher will collect the data certain variables.
for study.
2. THE PRETEST AND POST-TEST
This data is called a sample. The DESIGN. This refers to the pre-test
quantitative design should state what and post-test conducted before
sampling methods will and after the experimental
be used (random sampling is often ideal, treatment was done to the control
but not always possible), what population and experimental group.
this sample is to represent and how such
a sample size will influence the resulting Example: Effectiveness of the Use
analysis. of Guava Leaves for External
Flushing to Postpartum Patients to
For example, when researching how Prevent Infection”.
investments affect a man's romantic life, a
quantitative researcher may sample at - The control group will use tap
random 100 males who invest in the stock water for external flushing. The
market. experimental group will use the
guava leaves for external flushing.
Control
◼ Quantitative designs must identify the - Pre-test shall be done prior to
potential confounders, or influences, on treatment of “use of tap water” for
the dependent variable that are unrelated the control group and “use of
to the independent variable. The process guava leaves” for the experimental
of accounting for these influences in group. Then, the post-test shall
the quantitative design is called "control." again be administered to both
groups after their treatment.
Researchers must control the confounders Results of the pre-test and
in a study so that the analytical inferences post-test are then compared and
hold. analyzed for implications.

In the stock-market-and-romantic-partner (2) SOLOMON FOUR GROUP DESIGN

example, one potential confounder that ⦿ This involves two control groups and
researchers must deal with in their two experimental groups.
designs is age -- older men have had
more time for romantic relations, so One control and experimental group takes
researchers must be wary of how age the pre-test before treatment while the
influences the dependent variable. other group does not.

Type of True Experimental Designs However, both groups will get the post-test
(Polit and Beck: 2004): after the treatment, thereby segregating
the effects of pre-test measure and
(1) Basic Experiment intervention before and after treatment.
1. The Post-Test only Design. This
refers to the randomization and Example: “Effects of Genetic Counseling
experimental treatment of the and Prenatal Supervision on the
subjects that were completed, Incidence of Maternal Complication in
followed by a post test on the High Risk Women.”
 (5) CROSSOVER DESIGN:
(3) FACTORIAL DESIGN ⦿ In this study, different people are
- There is simultaneous randomly assigned to different
manipulation of two or more treatments. Subjects are exposed to two
independent variables which different treatments or conditions. The
permit testing of multiple influence of one treatment on the subject
hypotheses in a single experiment. may be carried over to the next
A separate analysis is done on the treatments. Therefore, a difference
effects of the independent existing between each treatment can be
variables on the dependent accurately ascertained by the subjects of
variables and the interaction the study. This Type of within-subjects
effects between these variables. design has a high equivalence advantage
since participants of the study are
Example: exposed to the same treatments or
An experiment to know the performance of conditions.
male and female (Sex-Ind. Variable 1) in
the relief of pain (Dep. Variable) under the EXAMPLE OF CROSSOVER DESIGN:
effects of varying amounts of same drug ⦿ Both Premature and Full Term
dosage measured in milligrams (Ind. Newborn will be exposed to touch and
Variable 2). The researcher is concerned Music Therapy. Results will be analyzed
with differences in pan relief related to: according to how each group responded
a. Sex; to both treatments.
b. The amount of drug taken; and
c. The interaction effect of sex and drug (6) QUASI-EXPERIMENT
● This is an experiment that lacks
(4) RANDOMIZED BLOCK DESIGN one or more of four (4)
- In this design, there are two properties of the true experiment.
independent variables, but one The missing ingredient is either
independent variable is not randomization or the validity
experimentally manipulated due to component. This method ultimately
the heterogeneous randomized depends in part, on human
assignment of subjects. This is judgment rather than on objective
called the blocking variable. The criteria, hence, validity of cause
blocking variable allows the and effect inferences may be
researcher to control the sample challenged.
size if it is perceived to be a
confounding variable. EXAMPLE:
⦿ Administering a drug to a group of
EXAMPLE: infants whose mothers are heroin
⦿ “Effects of touch and Music Therapy on addicted, to see if the treatment will result
the Heart Rates and Respiratory Rates of in weight increase or decrease of low birth
Premature and Full term babies”. The weight infants.
blocking variable could be the birth status
of the newborn. The full term newborn will ⦿ The foregoing study used only an
respond differently to the touch and music experimental group, was used as
therapy. Hence, the premature newborn administering a drug to those “infants
will only be manipulated. whose mothers are not heroin addicted”.
QUESTIONS ASKED IF NO
COMPARABLE OR NON-EQUIVALENT ⦿ Fallacy is prevented by
CONTROL GROUP IS INVOLVED: randomization, the key element in an
a. Is it fair that some external factors experimental design which controls the
caused or influenced the weight gain? influence of extraneous variables, to
b. Is it fair that selection influenced the ensure accuracy, reliability and validity of
weight gain? data. (Polit and Beck:2004)
c. How about the threat of manipulation?
NON-EXPERIMENTAL RESEARCH
If the answer is Yes, then inferences ⦿ The researcher does not interfere
are weakened. Why? Because it is a with nature, and the conditions for
situation in which judgment rather than research are realistic or natural.
data are utilized to reach a decision.
⦿ This research is often known as
ADVANTAGES OF EXPERIMENTAL “surveys”, which have less control over
DESIGN: the study subjects and the setting where it
1. Explains and establishes causal is conducted.
relationships of variables;
⦿ They are research situations in which
2. Increases purity of observations; the researcher cannot control and
manipulate the independent variable.
3. Creates condition in the experimental
setting that approximates the natural ⦿ They are mostly conducted in natural
setting; and settings such as schools, hospitals, and
other public health agencies, and homes
4. Free from the pressures of daily life of patients, among others.
when conducted in a controlled unit.
REASONS FOR UNDERTAKING
DISADVANTAGE OF EXPERIMENTAL NON-EXPERIMENTAL RESEARCH:
DESIGN:
1. Dangerous, particularly if human beings Many research studies involving human
are used as subjects; subjects, including most nursing research
investigations, are non-experimental in
2. Difficult to create conditions; nature.

3. Time constraints. There are many different reasons for

choosing a non-experimental:
4. Non-cooperation of subjects;
⦿ A vast number of human
5. Population constraints; and characteristics are inherently not
subject to experimental manipulation
6. A generalization may not be reliable if (e.g. blood type, personality, health
done in an artificial setting. beliefs, medical diagnosis), and thus the
effects of these characteristics on some
⦿ The experimental research design is phenomenon of interest cannot be studied
basically concerned with controlling and experimentally.
eliminating biases which lead to fallacious
data attributed to extraneous variables.
 ● In nursing research, as in other clients and the nurse in the actual work
fields in which human behavior is setting.
of primary interest, numerous L
variables could technically be It is a collaborative process involving other
manipulated for ethical reasons. If members of the health care team that
the nature of an independent provides solutions to practical problems in
variable is such that manipulation the organization.
could cause physical or mental
harm to subjects, then the variable TYPES OF APPLIED RESEARCH:
should not be controlled a. Historical Design.
experimentally. There are many A critical investigation and analysis of
research situations in which it is events, developments, and experiences of
simply not practical to conduct a the past. It describes the past events in
true experiment. Constraints might relation to the present situations and then
involve insufficient time, lack of analyze and interpret the implications of
administrative approval, excessive past events to the present trends or
inconvenience to patients or staff, practices.
or lack of adequate funds.
- Ex Post Facto or Correlation
TYPES OF NON-EXPERIMENTAL Research – the literal translation of
DESIGNS the Latin terms ex post facto “from
1. Basic or Library Research after the fact”. This expression is
meant to indicate that the research
This is intended to increase knowledge in in question has been conducted
science. It is also called pure research after the events. Ex post facto
for purpose of generating new research attempts to understand
knowledge. relationships among phenomena
as they naturally occur, without any
Basic research include reading books and research because basically, a
literature, listening to discussions, journal correlation is an interrelationship or
analysis, browsing articles from the association between two variable,
internet and websites and that is, a tendency for variation in
teleconferencing among others. one variable to be related to
variation in another.
The end product Research is purely
[Link] goals of basic research is Example:
to improve patient care; alleviate the “ The Evolution of Nursing Practice from
problem of nursing shortage; enrich the the time of Florence Nightingale to the
nursing curriculum and strengthen the 21st Century: Significant Indicators of
organization of nursing service, among Change and Transition from Past to
others. Present Practice of Nursing.”

2. Applied Research or “Action 2. Descriptive Design - A study that

Research” describes the nature of the
phenomenon under investigation after a
This is undertaken for practical survey of current trends, practices and
purposes. It seeks solutions to immediate conditions that relate to that phenomenon.
problems that have been identify by the
Descriptive studies involves analysis of an - Do current practices need to be
extremely broad range of phenomena, its modified, or should they be
result is a comprehensive presentation abandoned altogether?
and interpretation of statistical tabulations - Do the cost of implementing a new
of data yielded by a survey. program out weight the benefits?

This type of research answers questions Evaluation has an important role to

and satisfies curiosity about a certain play both in localized setting and in
phenomenon. It also describes and programs at national level.
elaborates the nature and causes of an
existing phenomenon at the time of the EVALUATION RESEARCH MODELS:
study . 1. Behavioral Objective – is the intended
outcome of the program stated in terms of
Moreover, describes studies are the behavior of those people at whom the
concerned with existing conditions, its program is aimed, that is, the behavior of
meaning and significance and then the beneficiaries, rather that the agents, of
making adequate and accurate the program.
interpretations of these data with or
without the aid of statistics (Calderon: 2. Goal-free Approach – proponents of
1993). this model argue that programs may have
a number of consequences besides
ADDITIONAL TYPES OR accomplishing the official objectives of the
QUANTITATIVE RESEARCH: program and that the classic model is
handicapped by its inability to investigate
A. Survey Research - is designed to these other effects.
obtain information from populations
regarding the prevalence, distribution, and TYPES OF EVALUATIONS:
interrelations of variables within those
populations. The samples of individuals, 1. Process or Implementation Analysis
as they usually do, may be referred to as – is a need for descriptive information
samples, they may be referred to as about the process by which a program or
sample surveys. procedure gets implemented and how it
functions in actual operation.
Surveys obtain information from a sample
of people by means of self-report-that is, 2. Outcome Analysis – tends to be
the study participants respond to a series descriptive and does not use rigorous
of questions posed by the investigator. design. Such an analysis simply
documents the extent to which the goals
B. Evaluation Research – is an applied of the program are attained, that is, the
form of research that involves finding out extent to which positive outcomes occur.
how well a program, practice, procedure,
or policy is working. 3. Impact Analysis – attempts to identify
Its goal is to assess or evaluate the the net impacts of an intervention that is,
success of a program. In people who must the impact that can be attributed
make decisions: exclusively to the intervention over and
- should a new program be adopted above the effects of other factors (such as
or an existing one discontinued? the effects of the standard treatment)
4. Cost- Benefit Analysis – to determine
whether the benefits of the program out The Methodological research may, for
weigh the cost, in terms of monetary example concentrate on the development
value. of an instrument that accurately measures
patients’ satisfaction with nursing care.
Cost-Benefit Analysis are typically done in The researches does not focus on the
connection with the impact analysis, that level of patient satisfaction with nursing
is, when there is solid evidence regarding care.
the net impact of an intervention.
Content Analysis Studies – has
5. Needs Assessment – represents an traditionally been used to described the
effort to provide a decision maker with quantification of narrative researchers who
information for action. A study in which a do not quantify their data sometimes refer
researcher collects data for estimating to their analytic work as a content
the needs of a group, community or analysis, the term as its classic sense
organization. refers to “a research technique for the
objective, systematic, and quantitative
Outcomes research – designed to description of the manifest content of
document the effectiveness of health care communication” (Berelson, 1971).
services. It is gaining momentum as a
research enterprise in nursing and health Quantitative Research Examples:
care fields. 1. Nurse-to-Patient Ratios in
Hospital Staffing – Queuing
Outcomes research overlaps in some Perspective by: Francis de
instances with evaluation research, but Verriconnt and Otis B. Jennings
evaluation research more typically focus The immediate motivation of this
review on an appraisal of a specific new paper is California Bill AB 394
intervention, whereas outcomes research which mandates fixed nurse to
represents a more global assessment of patient ratios as a means to
health care services. address the current crisis in the
quality of nursing care in the U.S.
Dephi Survey – were developed as tools with a predictive queuing analytic
for approach to determine whether
short-term forecasting. ratio policies are effective or not
and to provide an alternative
The technique involves a panel of experts staffing paradigm.
who are asked to complete a series of
questionnaires focusing on the experts’ 2. The Case for a Cognitive Theory
opinions, predictions, or judgment of Dreams by: Adam Schneider
concerning a specific topic of interest. and William Domhoffb
Consciousness and Cognition
Methodological Research – refers to describing the many findings that
controlled investigations of the ways of have been developed using
obtaining, organizing, and analyzing data. thousands of dreams and search
engines. The reports come in a
Methodological study addresses the variety of different sources and
development, validation, and evaluation of research studies from people ages
research tools or techniques. 7-74 years old and can be
 analyzed using a search engine AIMS OF EXPERIMENTAL RESEARCH
and statistical program in the • Experiments are conducted to be able to
dream bank which is located in a predict phenomenons.
particular website.
• experiment is constructed to be able to
“Quantitative has quaint explain some kind of causation.
advantages, but Qualitative
delivers the best results for the • Experimental research is important
least money. Furthermore, to society - it helps us to improve our
quantitative studies are often too everyday lives.
narrow to be useful and sometimes
directly misleading” Factors that affect Experimental
Method
• actual design
Lesson 8. METHODS OF RESEARCH • group selections
• feasibility of the study
A. EXPERIMENTAL RESEARCH • Cost
• It is a collection of research designs • Time
which use manipulation and controlled • Ethics
testing to understand causal processes. • How the study can be measured and the
Generally, one or more variables are validity of the study
manipulated to determine their effect on a
dependent variable. B. Non-experimental research
• Research design: the plan and structure
TYPES OF EXPERIMENTAL of research to provide a credible answer to
RESEARCH: a research question

1. True experimental • Purpose of non-experimental designs:

• Investigates probable cause and effect describe current existing characteristics
relationship by having one or more such as achievement, attitudes,
experimental groups exposed to one or relationships
more treatment conditions and comparing
the results to one or more control groups TYPES OF NON-EXPERIMENTAL
not receiving the treatment. RESEARCH:

2. Quasi-experimental 1. Historical
• Approximates the conditions of true • Objectives and accurate reconstruction
experiment in a setting that does not allow of the past in the relation to the tenability
control and/or manipulations of all relevant of an
variables. hypothesis.

The experimental method 2. Descriptive

• is a systematic and scientific approach to • Describes systematically, factually and
research in which the researcher accurately the nature of a phenomenon
manipulates one or more variables, and under investigation after a survey of
controls and measures any change in trends, practices
other variables. and conditions relevant to that
phenomenon.
 • There are various aspects to remember
3. Developmental when constructing an experiment.
• Investigates patterns and sequence of
growth • Planning ahead ensures that the
and/or change as a function of time on a experiment is carried out properly and that
certain phenomenon. the results reflect the real world, in the
best possible way.
4. Correlational
• Examines the extent to which variations Quasi-Experimental Design
in one variable correspond with variations • Quasi-experimental design is a form of
in one or more other variables based on experimental research used extensively in
correlation coefficient. the social sciences and psychology.

5. Case and field • involves selecting groups, upon which a

• Studies intensively to interpret the variable is tested, without any random
background, current status, and pre-selection processes.
environmental interactions of a given
social unit an individual, Example:
group, community or institution. • to perform an educational experiment, a
class might be arbitrarily divided by
6. Causal-comparative or ex post facto alphabetical selection or by seating
or retrospective arrangement. The division is often
• Investigates possible cause and effect convenient and, especially in an
relationships by observing some existing educational situation, causes as little
consequences or effects and searching disruption as possible.
back
through the data for any plausible cause. • After this selection,
the experiment proceeds in a very similar
7. Action research way to any other experiment, with a
• Develops new skills or new approaches variable being compared between different
to solve problems with direct application groups, or over a period of time.
on nursing intervention or other applied
setting. ADVANTAGES of QED
(Quasi-experimental Design):
IDENTIFYING THE RESEARCH
PROBLEM 1. Especially in social sciences, where
• The research problem is often pre-selection and randomization of groups
operationalization, to define how to is often difficult, they can be very useful in
measure the research problem. generating results for general trends.

• The results will depend on the exact Example:

measurements that the researcher choose If we study the effect of maternal alcohol
and may be operationalized differently in use
another study to test the main conclusions when the mother is pregnant, we know
of the study. that alcohol does harm embryos. A strict
experimental design would include that
CONSTRUCTING THE EXPERIMENT mothers were randomly assigned to drink
alcohol. This would be highly illegal
because of the possible harm the study This is really hard to do properly. One
might do to the embryos. Researchers are group of children may have been slightly
to ask people how much alcohol they used more intelligent or motivated. Without
in their pregnancy and then assign them some form of pre-testing or random
to groups. selection, it is hard to judge the influence
of such factors.
2. Quasi-experimental design is often
integrated with individual case studies; the 4 types of designs of NED
figures and results generated often (non-experimental design)
reinforce the findings in a case study, and
allow some sort of statistical analysis to 1. Studies that describe a phenomena
take place. • Statistical nature of the description (e.g.,
frequency, percentages, averages,
3. Without extensive pre-screening and graphs, etc.)
randomization needing to be undertaken, • Importance of these designs in the early
they do reduce the time and resources stages of the investigation of an area.
needed for experimentation.
Criteria for evaluating descriptive
DISADVANTAGES of QED studies
(Quasi-experimental Design): • Conclusions about relationships should
not be drawn
1. Without proper randomization, statistical • Subjects and instruments should be
tests can be meaningless. described completely

Example: 2. Relationship designs

These experimental designs do not take Designs describing the relationship
into between two or more variables:
account any pre-existing factors (as for the
mothers: what made them drink or not Comparative designs
drink - These studies investigate the
alcohol), or recognize that influences relationship of one variable to another by
outside examining differences on the dependent
the experiment may have affected the variable between two groups of subjects.
results.
If math scores for males are significantly
2. A quasi experiment constructed to higher than those for females, a
analyze the effects of different educational relationship exists between gender and
programs on two groups of children, for math achievement.
example, might generate results that show
that one program is more effective than If the academic self-concept scores for
the other. ninth graders are significantly different
than those for twelfth graders, a
These results will not stand up to rigorous relationship exists between grade level
statistical scrutiny because the researcher and academic self-concept.
also need to control other factors that may
have affected the results. If the third grade achievement scores for
whites are not significantly different that
those for non-whites, no relationship exists QUALITATIVE RESEARCH
between ethnicity and achievement. ⦿ is a type of scientific research which
has its roots in philosophy and human
Criteria for evaluating comparative sciences.
designs
[Link] and instruments are described ⦿ plays an important part in providing
completely. evidence for practice in nursing, and is
[Link] for identifying the different gaining greater
groups is clearly stated. acceptance within medicine. (Bailey C,
[Link] inferences are made about causation 2002)
[Link] and images depict the results
accurately. ⦿ is used to gain insight into people's
attitudes,
Correlational designs behaviours, value systems, concerns,
Simple correlation – studies or examine motivations, aspirations, culture or
the relationship/s between two variables. lifestyles.

Examples: ⦿ is often associated with naturalistic

Math achievement and math attitudes; inquiry.
Teacher effectiveness and teacher efficacy
⦿ aim to gather an in-depth
Caution in interpreting correlations understanding of human behavior and
A relationship between two variables (e.g., the reasons that govern such behavior.
achievement and attitude) does not mean
one causes the other (i.e., positive FOCUS
attitudes do not cause high levels of ⦿ Qualitative studies design
achievement). ⦿ Other types of qualitative researches
⦿ Research with ideological perspectives
Attenuation ⦿ Research examples
- the possibility of low reliability of
the instruments makes it difficult to QUALITATIVE STUDIES DESIGN:
identify relationship. (EMERGENT DESIGN)
⦿ It is a design that emerges as the
Restriction in range researcher
- a lack of variability in scores makes ongoing decisions reflecting what
makes it difficult to identify has already been learned.
relationships.
Everyone scoring very, very low QUALITATIVE DESIGN FEATURES
Everyone scoring very, very high ⦿ Intervention or control over
independent variable
Large sample sizes and/or using many ⦿ Type of group comparisons
variables can Identify significant ⦿ Number of data collection points
relationships for statistical reasons and not ⦿ Occurrence of independent and
because the relationships really exist. dependent variables
⦿ Setting
LESSON 9: QUALITATIVE RESEARCH
DESIGN AND APPROACHES CHARACTERISTICS OF QUALITATIVE
RESEARCH(IN GENERAL)
1. Often involves a merging together of even though the researcher is still striving
various data collection strategies; to understand the full scope of the
phenomenon and to identify its major
2. Is flexible and elastic, capable of dimensions, these various phases might
adjusting to what is being learned during take only a few days to complete, or they
the course of data collection; may take many months.

3. Tends to be holistic, striving for an OTHER TYPES OF QUALITATIVE

understanding of the whole; RESEARCH
⦿ Grounded Theory
4. Requires researchers to become ⦿ Ethnography
intensely involved, often remaining in the ⦿ Symbolic Interactionism
field for lengthy periods of time; ⦿ Phenomenology
⦿ Historical research
5. Requires the researcher to become the ⦿ Case studies
research instrument; and ⦿ Qualitative Content Analysis (Narrative
Analysis)
6. Requires ongoing analysis of the data ⦿ Qualitative Outcome Analysis
to formulate subsequent strategies and to ⦿ Qualitative Meta synthesis
determine whether field work is done. ⦿ Descriptive Qualitative Studies

3 PHASES IN QUALITATIVE DESIGN 1. GROUNDED THEORY

1. Orientation & Overview. ⦿Grounded theory is an inductive type of
The researcher enters the study not research, based in the observations or
knowing data from which it was developed.
what is known- that is, not knowing what it
is about the phenomenon that will drive ⦿It uses a variety of data sources,
the inquiry forward. including quantitative data, review of
records, interviews, observation and
2. Focused Exploration. surveys.
The second phase of the study is a more
focused scrutiny and in depth exploration ⦿Concepts are basic units of analysis.
of those aspects of the phenomenon that
are judged to be salient. ⦿Concepts grouped and related to form
abstract categories.
3. Confirmation & Closure.
Undertake efforts to establish that their ⦿Relationships between categories
findings are trustworthy, often by ongoing identified to develop “formal theory”.
back and discussing their understanding
with study participants. ⦿Theory building- concepts, categories,
themes are identified and developed
The three phases are not discreet events. during the study.
Rather, they overlap to a greater or lesser
degree in different projects. 2. ETHNOGRAPHY
⦿Ethnography is a description and
For example, even the first few interviews interpretation of a cultural or social group
or observations are typically used as a or system.
basis for selecting subsequent informants,
⦿The research examines the group’s ⦿Phenomenological research techniques
observable and learned patterns of offer nurses a valuable way to understand
behaviour, customs, and ways of life. the life world of nursing (Walters AJ).

⦿A picture of the “way of life” of some ⦿Analysis /description of everyday life.

identifiable group of people.
Example of a Phenomenologic Study
⦿Ethnography is closely related to Coward (1995) used a phenomenological
Anthropology. approach to studying the lived experience
of self-transcendence as reaching out
⦿Its process include “doing fieldwork” and beyond the boundaries of the self-either
“going native”. outward beyond personal concern or
inward toward increased understanding-to
⦿The focus of ethnography is with achieve broader perspectives and
culture, and how people interact with each behaviors that facilitate the discovery of
other. meaning.

⦿Its is a both a research process and a 4. SYMBOLIC INTERACTIONISM

product. ⦿Making sense of experience via
symbols.
Example of Ethnographic Study ⦿Symbols find meaning through and in
Dreher and Hayes (1993) conducted an interaction.
in-depth ethnographic study of perinatal ⦿Closely related to Grounded theory.
use of marijuana in rural Jamaican
communities. 5. HISTORICAL RESEARCH
⦿The systemic collection and critical
They spent 6 years in the field (starting in evaluation of data relating to past
1983) studying marijuana use among occurrences.
pregnant Jamaican women and It is undertaken to answer questions
subsequent effects on the offspring. The concerning causes, effects, or trends
field researchers lived in a rural parish of relating to past events that may shed light
St. Thomas for most of the on present behaviors or practices.
6-year period.
⦿Nurses have used historical research
This particular parish, in the southeastern methods to examine a wide range of
part of Jamaica, was known for its phenomena in both the recent and more
widespread use of marijuana (called ganja distant past.
in Jamaica). A sample of gravid women Its goal is the discovery of new
was recruited from six different locations knowledge, not the summary of existing
within the parish, constituting a range of knowledge.
rural communities and neighborhoods.
⦿Historical researchers typically need to
3. PHENOMENOLOGY devote considerable effort to identifying
⦿Phenomenology describes the and evaluating data sources on events,
subjective reality of an event, as perceived situations, and human behavior that
by the study population; it is the study of a occurred in the past.
phenomenon.
COLLECTION OF HISTORICAL DATA
⦿ Primary source were involved in a serious illness over a
⦿ Secondary source 2-year period. Data were gathered by
⦿ External criticism observation and in-depth interviewing.
⦿ Internal criticism
CASE STUDY NURSING RESEARCH
EVALUATION OF HISTORICAL DATA SAMPLES
⦿ External criticism
⦿ Internal criticism ⦿ Reese and Murray (1996) conducted
multiple case studies at the individual
6. CASE STUDIES level, examining the issue of how being a
⦿ Are in-depth investigations of a single great-grandmother enhanced possibilities
entity or a small series of entities. for meaning and transcendence.
Typically, the entity is an individual, but
families, groups, institutions, or other ⦿ Perry and Olshansky (1996) engaged
social units may also be the focus. in an in-depth case study of a family in
which one member had alzheimer’s
⦿ It attempts to analyze, and understand disease, studying the ways in which the
the phenomena that are important to the family members made meaning of the
history, development, or care of an situation.
individual or an individual’s problems.
⦿ Twinn and Lee (1997) studied health
⦿ Although the foremost concern of a education in two different settings (a
case study is to generate knowledge of medical ward and a surgical ward) using
the particular case, case studies are case study methods.
sometimes a useful way to explore
phenomena that have not been rigorously 7. QUALITATIVE CONTENT ANALYSIS
researched. (QCA)
- When a studies claim no particular
⦿ The greatest advantage of case studies disciplinary or methodologic roots.
is the depth that is possible when a limited - The researcher may simply say in
number of individuals, institutions, or their reports that they have
groups is being investigated. A common conducted a qualitative study or a
complaint leveled at other types of naturalistic inquiry. Or they may
research is that the data tend to be say that they have done a content
superficial. And the most serious analysis or a qualitative content
disadvantage is the lack of generalizability. analysis.
- The content analysis means that
Example of a Case Study the researcher analyzes the
Price (1992) undertook a series of case content of the narrative to
studies in highland Ecuador to analyze the determine themes or patterns.
processes by which people cope with
serious illness. EXAMPLES OF QCA (QUALITATIVE
CONTENT ANALYSIS)
The case studies were used as a vehicle ⦿ Brink and Ferguson (1998) conducted
for illuminating psychological coping unstructured interviews with 162 men and
mechanism within family and sociocultural women in an exploratory qualitative study
contexts. Price conducted eight (8) concerning the decision to lose weight.
in—depth case studies of families who
⦿ In a study that relied on written diaries competence, and the uplifting effects of
rather than interviews, O’Brien, Reltyea, caring.
and Lidstone (1997) did a qualitative
content anlysis of women’s experiences [Link] QUALITATIVE
with nausea and vomiting during STUDIES (DQS)
pregnancy. - When a research or a qualitative
studies claimed no particular
8. QUALITATIVE OUTCOME ANALYSIS disciplinary or methodological
(QOA) roots, the researchers may simply
- a systematic means of conforming indicate that they have conducted
the applicability clinical strategies a qualitative study or a naturalistic
suggested by a qualitative study inquiry, or they may say they have
and to evaluate clinical outcomes. done a content analysis of their
qualitative data.
Example: - Some studies do not have a formal
In their initial qualitative project, Morse name or do not fit into the typology
and Doberneck (1995) discovered seven and thus we refer to this as DQS.
stages in the process of hoping in four Example:
groups of patients: breast cancer, heart Stubblefield and Murray (2001)
transplantation, spinal cord injury, and interviewed 15 parents whose children
breastfeeding after returning back to work. had undergone lung transplantation to
Next, a hope Assessment Guide was study the effect of the procedure on the
developed to assist nurses in identifying parents interpersonal relationships. The
the stage of hope that a patients was in. authors wrote that they “conducted a
the QOA project was designed to evaluate content analysis to formulate a narrative
the feasibility of using the Hope description of the parents relationships
Assessment Guide in clinical settings. with others.”

[Link] METASYNTHESIS RESEARCH WITH IDEOLOGICAL

(QMS) PERSPECTIVES
- Defined as “the theories grand ⦿ Critical Theory
narratives, generalizations, or ⦿ Feminist Research
interpretive translations produced ⦿ Participatory Action Research
from the integration comparison of
findings from qualitative studies.” 1. CRITICAL THEORY
- Originated with a group of Marxist
Example: oriented German scholars in the
Beck (2001) conducted a metasynthesis of 1900’s, collectively refer to as the
14 qualitative studies on the meaning of Frankfurt School.
caring in nursing education. - Variants of critical theory abound in
The metasynthesis revealed five themes the social sciences. Essentially, a
that permeated caring in nursing critical researcher is concerned
education whether it was faculty caring for with a critique of society and with
each other or their students or nursing envisioning new possibilities.
students caring for each other or their - Is concerned with a critique of
clients. These themes centered on existing social structures; critical
reciprocal connecting, which consisted of researchers strive to conduct
presencing, sharing, supporting, inquiries that involve collaborative
 efforts with participants and foster research method and should instead
enlightened self knowledge and employ more than one measurement
transformation. procedure When investigating a research
problem.
2. FEMINIST RESEARCH (FR)
- Like critical research, FR is Triangulation is a scientific technique in
designed to be transformative, but the research process w/c refers to the use
the focus is sharply on how gender of more than one approach to the
domination and discrimination investigation of a research question in
shape women’s lives and their order to enhance confidence in the
consciousness. ensuing findings.

3. PARTICIPATORY ACTION Triangulation offers the prospect of

RESEARCH (PAR) enhanced confidence.
- Produces knowledge through close
collaboration with groups or B. APPLICATION OF MULTIMETHOD
communities that are vulnerable to RESEARCHES
control or oppression by a
dominant culture. 1. INSTRUMENT DEVELOPMENT
- In PAR research, methods take One of the most frequent uses of an
second place to emergent integrated approach in nursing research
processes that can motivate involves the development of instruments,
people and generate community particularly self report instruments that
solidarity. involve questions to be posed to
respondents.

Lesson 10: Multimethod Research 2. EXPLICATING and VALIDATING

CONSTRUCTS
Multimethod research Multimethod research is often used to
- entails the application of two or develop a comprehensive understanding
more sources of data or research of a construct or to validate the construct’s
methods to the investigation of a dimensions.
research question or to different
but highly linked research Such research may be undertaken when a
questions. Such research is also little researched phenomenon has been
frequently referred to as mixed identified as worthy of further scrutiny
methodology. (usually in an in-depth qualitative study) or
when there is a body of existing research
Mixed method design in which some serious gaps have been
- incorporates techniques from identified or doubts have been raised
qualitative to quantitative methods about the prevailing conceptualization.
to answer research questions.
3. ILLUSTRATION
A. The rationale of multimethod research Qualitative data are sometimes combined
is Underpinned by the principle of with quantitative data to illustrate the
TRIANGULATION, which implies that meaning of constructs or relationships.
researchers should seek to ensure that Such illustrations often help to clarify
they are not over reliant on a single important results or to corroborate the
understanding gleaned from the statistical 4. Theoretical or conceptual framework
analysis. development
5. Study assumptions formulation
In this sense, these illustrations often help 6. Limitation of the study specification
to illuminate the analysis and give 7. Hypothesis or research question
guidance to the interpretation of results. formulation
Qualitative materials can be used to 8. Study variables/ terms definition
illustrate specific statistical findings or to 9. Research design selection
provide more global and dynamic views of 10. Population identification
the phenomena under study, often in the 11. Sample selection
form of illustrative case study. 12. Research locale description
13. Pilot study investigation
4. UNDERSTANDING RELATIONSHIPS 14. Data collection
and CAUSAL PROCESSES 15. Data for analysis organization
Quantitative methods often demonstrate 16. Finding interpretation
that variables are systematically related to 17. Results presentation
one another, but they often fail to provide 18. Implications citations
insights about why the variables are 19. Findings dissemination/
related. This situation is especially likely to communication
occur with ex post facto research.
C. MULTIMETHOD RESEARCH DESIGN
In quantitatively oriented studies, the Green and Caracelli (1997) have identified
interpretations are often speculative, several types of research designs that
representing the researcher’s best guest involve a multimethod approach. The
(a guess that may, of course, be built on design cluster into two broad categories
solid theory or prior research) about what that they label component designs and
the findings mean. When a study integrated designs.
integrated both qualitative and quantitative
data, however, the researcher may be in a (1) MULTIMETHOD COMPONENTS
much stronger position to derive meaning DESIGNS
immediately from the statistical findings. In studies that can be described as having
a component design, the qualitative and
5. THEORY BUILDING and quantitative aspects are implemented as
REFINEMENT discrete components of the overall inquiry
The most ambitious application of and remain distinct during data collection
multimethod research is in the area of and analysis.
theory development.. The use of multiple
methods provides greater opportunity for 3 Types of Designs as Component
potential disconfirmation of the theory. If Design
the theory can survive these assaults, it 1. TRIANGULATED DESIGNS- both
can provide a substantially stronger qualitative and quantitative methods are
context for the organization of our clinical used to capture the same phenomenon,
and intellectual work. with a focus on convergence and
increased validity.
TRIANGULATION Scientific manners:
1. Problem identification 2. COMPLEMENTARITY DESIGNS- the
2. Purpose of the study determination results from one dominant method type
3. Literature review are enhanced or clarified by results from
the other type. The use of case studies to integrated simultaneously rather than
illustrate important relationships discerned co-linearly or hierarchically.
in quantitative analyses exemplifies this
type of design. 4. TRANSFORMATIVE DESIGNS
- the emphasis is on the blending the
3. EXPANSION DESIGN - different value commitments of different research
methods are used for distinct inquiry traditions to arrive at a better
components representation of the multiple interests in
- as might be the case in an the larger social context. In general,
evaluation that involved both an integrated designs are better suited to the
implementation and an impact application of theory building and testing
analysis. The results from such that are component designs.
studies are often presented in a
side-by-side fashion, rather than
woven together into a single story. D. STRATEGIES FOR MULTIMETHOD
RESEARCH
(2) MULTIMETHOD INTEGRATED The ways in which a researcher might
DESIGNS choose to combine qualitative and
In studies that Green and Caracelli (1997) quantitative methods in a single study are
refer to as having an integrated design, almost limitless-or rather, are limited only
there is greater integration of the method by the ingenuity of the investigator.
types at all phases of the project, from the Therefore, it is impossible to develop a
development of research questions, catalog of multimethod strategies.
through data collection and analysis, to However, some of the following scenarios
the interpretation of the results. are apt to be especially common.

4 Types of Integrated designs a. EMBEDDING QUALITATIVE

1. INTERATIVE DESIGNS APPROACHES WITHIN A SURVEY
- involve a dynamic in which the findings
from one method are used as a basis for b. EMBEDDING QUANTITATIVE
moving forward with further research using MEASURES INTO FIELD WORKS
the alternative method (as is typically the
case with instrument development and c. QUALITATIVE AND QUANTITATIVE
refinement). DATA IN EXPERIMENTAL RESEARCH

2. EMBEDDED DESIGNS (NESTED E. OBSTACLES FOR MULTIMETHOD

DESIGNS) RESEARCH
- one methodological approach is Various obstacles that may constrain the
embedded within the other, “interlocking gathering of qualitative and quantitative
contrasting inquiry characteristics in a data in a single investigation, among the
framework of creative tension. most salient are the following:

3. HOLISTIC DESIGN 1. Epistemological biases

- features the essential interdependence Qualitative and quantitative researchers
of alternative methods for gaining a full often operate with a different set of
understanding of complex phenomena. In assumptions about the world and ways of
holistic designs, the methods are learning about it.
For those with a hard-line, purist view, challenges may well be clearer or more
these assumptions may be seen as refined conceptualizations of the
inevitably irreconcilable. According to a phenomenon under study.
survey of nurses with doctorates, however,
extreme positions of this type are atypical 5. Publication biases
among Some journals have a distinct preference
nurse researchers. for studies that are qualitative, and others
learn toward quantitative studies. Because
2. Costs- a major obstacle facing of this, researchers might be concerned
researchers who would like to gather that they would need to write up the
qualitative and quantitative data is that qualitative and quantitative results
such multimethod research is usually separately, foregoing many of the
expensive. advantages of multimethod research.

3. Researcher training However, publication biases are much less

Most researcher obtain graduate-level evident today than they were a decade
training that stresses either qualitative or ago. All of the major nursing journals
(more typically) quantitative research devoted to research publish qualitative
methods. According to a survey by and quantitative studies, and they are
Damrosch and Strasser (1988), only about increasingly publishing reports of
one third of doctoral prepared academic multimethod studies.
nurses have training in both qualitative
and quantitative methods.
Lesson 11. Sampling Designs
Thus, investigator skills may pose an
obstacle to multimethod research. What is Sampling?
However, there is nothing about such ● It is not possible, nor it is
research that suggests that all phases of necessary, to collect information
the investigation must be done by a single from the total population.
researcher. Collaboration among ● Instead, a smaller subgroup of the
researchers might, indeed, be an target population or a sample is
important by products of the decision to selected for the purpose of study.
use a multimethod approach. Such ● Sampling is the strategy of
collaboration provides opportunities for selecting a smaller section of the
triangulation in terms of both methods and population that will accurately
investigator perspectives. represent the patterns of the target
population at large.
4. Analytic challenges
Despite the many advantages of doing Why Take a Sample?
multimethod research, it is nevertheless The main purposes of sampling are:
true that a successful integrated data ● Economies on the resources
analysis is a challenging task. required for collecting and
managing the data from a smaller
The researcher may be confronted with sub-group
issues about how best to combine numeric
and narrative data or about how to resolve ● Improve quality of data by focusing
and interpret inconsistent or contradictory on a smaller group.
findings. However, the outcome of such
Importance of Sampling Example:
● Ensures quality of subjects through - a survey of County employees.
its inclusion and exclusion criteria - An exhaustive list of all County
employees as of a certain date
● Acknowledge and specify the could be obtained from the
scope and limitation of the study Department of Human Resources.
If 100 names were selected from
● Maximize time and effort for better this list using a random number
and accurate collection of data table or a computerized sampling
program, then a simple random
● Ensures quality of data since the sample would be created.
sampling process helps control or
eliminates extraneous variables 2. Stratified Random Sampling
● Divides the population into
● Economic and financial concerns. homogenous subgroups from
which elements are selected at
Sampling Methodologies random.
Sampling methodologies are classified
under two general categories: Example:
1. Probability sampling and - Pick out 200 samples from 5,000
2. Nonprobability sampling. population then group them
according to age, sex, civil status,
1. A probability sample tends to be more and occupation
difficult and costly to conduct.
● However, probability samples are 3. Cluster Sampling/ Multi-stage
the only type of samples where the - The successive selection of
results can be generalized from the random samples from larger to
sample to the population. smaller units by either simple
● In addition, probability samples random or stratified random
allow the researcher to calculate methods. It involves sever stages
the precision of the estimates in drawing the samples from the
obtained from the sample and to population.
specify the sampling error.
Example:
Four basic types of Probability Sample - Province
methodologies - municipality
1. Simple random sampling, - village
2. Stratified random sampling, - individual respondents
3. Cluster Sampling/ multi-stage At each stage, simple random, systematic
Sampling, and stratefied techniques are used.
4. Systematic / sequential sampling
4. Systematic or Sequential Sampling
1. Simple Random Sampling - The selection of every 10th name
- the researcher must first prepare in a list of patients in odd or even
an exhaustive list (sampling frame) numbered rooms;
of all members of the population of - Every 5th house on the block;
interest. - Every 6th baby in the nursery.
 - The sequence of selection can have no evidence that they are
also be done, using odd or even representative of the populations
numbered names on the sampling we're interested in generalizing to
list. -- and in many cases we would
clearly suspect that they are not.
Advantage of Probability Sampling
- There is less bias as every 2. QUOTA SAMPLING
element in the population is given - In quota sampling, you select
an equal (independent) chance to people nonrandomly according to
be selected. some fixed quota.
There are two types of quota sampling:
Disadvantage of Probability Sampling - proportional and non
- It is time consuming, expensive, proportional.
inconvenient and impossible to
obtain. In proportional quota sampling
- you want to represent the major
2. NON-PROBABILITY SAMPLING characteristics of the population by
- Respondents or subjects are sampling a proportional Qamount
selected in a non-random way. The of each.
researcher desires to use available - For instance, if you know the
Subjects at her convenience population has 40% women and
anytime during the period of the 60% men, and that you want a
study. total sample size of 100, you will
continue sampling until you get
TYPES OF NON-PROBABILITY those percentages and then you
SAMPLING will stop.
1. ACCIDENTAL /CONVENIENCE - So, if you've already got the 40
SAMPLING women for your sample, but not
- Uses the most readily available or the sixty men, you will continue to
most convenient group of people sample men but even if legitimate
or objects as study respondents. women respondents come along,
- One of the most common methods you will not sample them because
of sampling goes under the various you have already "met your quota.“
titles listed here. - The problem here (as in much
purposive sampling) is that you
Examples: have to decide the specific
1. I would include in this category the characteristics on which you will
traditional "man on the street“ (of base the quota.
course, now it's probably the - Will it be by gender, age, education
"person on the street") race, religion, etc.?

2. Interviews conducted frequently by Nonproportional quota sampling

television news programs to get a - is a bit less restrictive.
quick (although nonrepresentative) - In this method, you specify the
reading of public opinion. minimum number of sampled units
you want in each category. here,
3. Clearly, the problem with all of you're not concerned with having
these types of samples is that we
 numbers that match the - With a purposive sample, you are
proportions in the population. likely to get the opinions of your
- Instead, you simply want to have target population, but you are also
enough to assure that you will be likely to overweight subgroups in
able to talk about even small your population that are more
groups in the population. readily accessible.
- This method is the nonprobabilistic
analogue of stratified random 4. SNOWBALL/NETWORK SAMPLING
sampling in that it is typically used ● In snowball sampling, you begin by
to assure that smaller groups are identifying someone who meets
adequately represented in your the criteria for inclusion in your
sample. study.
● You then ask them to recommend
[Link] SAMPLING others who they may know who
- In purposive sampling, we sample also meet the criteria.
with a purpose in mind. ● Although this method would hardly
- We usually would have one or lead to representative samples,
more specific predefined groups there are times when it may be the
we are seeking. best method available.
- For instance, have you ever run
into people in a mall or on the ● Snowball sampling is especially
street who are carrying a clipboard useful when you are trying to reach
and who are stopping various populations that are inaccessible
people and asking if they could or hard to find.
interview them? ● For instance, if you are studying
the homeless, you are not likely to
- Most likely they are conducting a be able to find good lists of
purposive sample (and most likely homeless people within a specific
they are engaged in market geographical area.
research). ● However, if you go to that area and
- They might be looking for identify one or two, you may find
Caucasian females between 30-40 that they know very well who the
years old. other homeless people in their
- They size up the people passing vicinity are and how you can find
by and anyone who looks to be in them.
that category they stop to ask if
they will participate. Advantage of Non Probability
- Convenient
- One of the first things they're likely - Economical
to do is verify that the respondent
does in fact meet the criteria for Disadvantage of Non-Probability
being in the sample. Sampling
- Purposive sampling can be very ● It is likely to produced biased
useful for situations where you samples or errors in judgement
need to reach a targeted sample because the researcher cannot
quickly and where sampling for estimate the precise elements of
proportionality is not the primary the population that will be included
concern. in the sample.
 question you are trying to address.
Lesson 12. WRITING AND should be focused on the thesis
PRESENTING RESEARCH FINDINGS question(s).
- All cited work should be directly
GENERAL GUIDELINES relevant to the goals of the thesis.
•Title Page This is not a place to summarize
- Title (including subtitle), author, everything you have ever read on
institution, department, date of a subject.
delivery, research mentor, mentor's
institution Methods
•Abstract - Information to allow the reader to
- Answers to these questions should assess the believability of your
be found in the abstract: What did results. Description of your
you do? materials, procedure, theory.
- Why did you do it? What question Calculations, technique,
were you trying to answer? procedure, equipment, and
- How did you do it? State methods. calibration plots. Limitations,
- What did you learn? State major assumptions, and range of validity.
results. Description of your analytical
- Why does it matter? Point out at methods, including reference to
least one significant implication. any specialized statistical software.
- A good abstract is concise,
readable, and quantitative. Length Results
should be ~ 1-2 paragraphs, • The results are actual statements of
approx. 400 words. observations, including statistics, tables
and graphs.
GENERAL GUIDELINES • Indicate information on range of
● Table of Contents variation.
- list all headings and • Mention negative results as well as
subheadings with page positive. Do not interpret results - save
numbers; indent that for the discussion.
subheadings • Use S.I. units (m, s, kg, W, etc.)
● List of Figures throughout the thesis.
- List page numbers of all • Key results should be stated in clear
figures. The list should sentences at the beginning of paragraphs.
include a short title for each Describe the nature of the findings; do not
figure but not the whole just tell the reader whether or not they are
caption. significant.
● List of Tables
- List page numbers of all tables. Conclusions
The list should include a short title - Refer back to problem posed, and
for each table but not the whole describe the conclusions that you
caption. reached from carrying out this
investigation, summarize new
Introduction observations, new interpretations,
- Sufficient background information and new insights that have
to allow the reader to understand resulted from the present work.
the context and significance of the Include the broader implications of
 your results. Do not repeat word ● Look pleasant, enthusiastic,
for word the abstract, introduction confident, proud, but not arrogant.
or discussion. Remain calm. Appear relaxed,
even if you feel nervous. Speak
Recommendations slowly, enunciate clearly, and show
• Include when appropriate (most of the appropriate emotion and feeling
time) relating to your topic.
• Remedial action to solve the problem. ● Establish rapport with your
• Further research to fill in gaps in our audience. Speak to the person
understanding. farthest away from you to ensure
• Directions for future investigations on your voice is loud enough to
this or related topics. project to the back of the room.

Acknowledgments ● Vary the tone of your voice and

- Advisor(s) and anyone who helped dramatize if necessary.
you: technically (including ● If a microphone is available, adjust
materials, supplies) intellectually and adapt your voice accordingly.
(assistance, advice) financially (for ● Body language is important.
example, departmental support, Standing, walking or moving about
travel grants) with appropriate hand gesture or
- References cite all ideas, facial expression is preferred to
concepts, text, data that are not sitting down or standing still with
your own. head down and reading from a
prepared speech.
Appendices
• Reference data/materials not easily ● Using audio-visual aids or props
available (theses are used as a resource for enhancement if appropriate and
by the department and other students). necessary, be sure all necessary
• Tables (where more than 1-2 pages). equipment is set up and in good
• Calculations (where more than 1-2 working order prior to the
pages). presentation
• You may include a key article as ● Speak with conviction as if you
appendix. really believe in what you are
• List of equipment used for an experiment saying.
or details ● Do not read from notes for any
of complicated procedures. extended length of time although it
• Note: Figures and tables, including is quite acceptable to glance at
captions, should be embedded in the text your notes infrequently.
and not in an appendix, unless they are
more than 1-2 pages and are not critical to ● Maintain sincere eye contact with
your argument. your audience.
● Speak to your audience, listen to
PRESENTING YOUR RESEARCH IN AN their questions, respond to their
ORAL PRESENTATION reactions, adjust and adapt.
● Dress appropriately for the ● Pause. Allow yourself and your
occasion. Be solemn if your topic is audience a little time to reflect and
serious. think. Add humor whenever
appropriate and possible.
 terms I, me, and my, as well as the
● Have handouts ready and give phrase personally speaking...
them out at the appropriate time.
● Know when to STOP talking. ● Use past tense in the abstract,
● To end your presentation, introduction, and method. Results
summarize your main points in the and discussion sections can be in
same way as you normally do in the present tense.
the CONCLUSION of a written
paper. ● When abbreviating any terms, spell
them out the first time (in both the
GENERAL RULES FOR WRITING abstract and again in the body of
● You must use complete sentences. the manuscript, if need be). For
example, The Palawan State
● The first sentence of a paragraph University (PSU) was used to…
must be independent (able to
stand on its own). For example All measurement reporting is done
consider While these studies are in metric units. In other words, use
important, there is... This sentence centimeters and meters rather than
would be correct in the middle of a inches and feet.
paragraph, but as the first
sentence, it should more ● The numbers zero through nine
appropriately read, While studies are spelled out (except when it is a
of the effects of whatever on table or figure number, or a metric
whatever else are important, there measurement, etc.). The numbers
is... 10 and above are written as
numbers.
● Try not to use slang (e.g., ...put a
damper on...). ● If you use someone's words or
ideas, you must give them credit
● Do not use contractions. That is, with a citation. This is particularly
instead of it's, use it is. important, since the penalties for
plagiarism are severe.
● If you are doubtful about the
spelling of a word, do not guess. ● The first time the reference is cited
Look up the correct spelling in an in the text, spell out all of the
appropriate reference sourceIf you authors last names. For example,
are doubtful about the spelling of a Miller, Rosellini, and Seligman
word, do not guess. Look up the (1975) suggested that... With
correct spelling in an appropriate articles that have three or more
reference source (e.g., authors use the Latin abbreviation
[Link]). for "and others" when the
reference is cited a second (or
● Proofread the copy that you submit third) time. For example, Miller et
and do correct minor typographical al. (1975) suggested that... or ...
errors, formatting, spelling, or even some fact (Miller et al., 1975).
the wording, with a pencil. These
corrections are inevitable and will Multiple citations in parentheses
communicate. Avoid use of the are placed alphabetically and are
 separated by a semicolon and a • Dividing a table or figure - confine each
space. For example, Some fact figure/table to a single page
(Carlson, 1972; Moon, 1968; • Submitting a paper with pages out of
Partin, 1980). order

● Avoid the use of first person FORMAT AND STYLE

pronouns. Refer to yourself or the In all sections of your paper
research team in third person. • Use normal prose including articles ("a",
Instead of saying "I will ..." or "We "the," etc.)
will ...", say something like "The • Stay focused on the research topic of the
researcher will ..." or "The research paper
team will ...". • Use paragraphs to separate each
important point (except for the abstract)
REDUCING BIAS IN RESEARCH • Indent the first line of each paragraph
REPORTING • Present your points in logical order
• Avoid informal wording, don't address
Research bias, also called experimenter the reader directly, and don't use jargon,
bias, is a process where the scientists slang terms, or superlatives
performing the research influence the • Avoid use of superfluous pictures -
results, in order to portray a certain include only those figures necessary to
outcome. presenting results.

MINIMIZING BIAS
• Use sampling techniques that result in
choosing
groups from the same population
• Use methods that result in high
recruitment rates
• Use valid reliable tools to measure all
study subjects

FORMAT AND STYLE

GENERAL STYLE
• Print or type using a 12 point standard
font, such as Times, Geneva, Bookman,
Helvetica, etc.
• Text should be double spaced on 8 1/2" x
11" paper
with 1 inch margins, single sided
• Number pages consecutively
• Start each new section on a new page
• Adhere to recommended page limits

MISTAKES TO AVOID
• Placing a heading at the bottom of a
page with the following text on the next
page (insert a page break!)