CS Methods in

Research
Topic I – Basic Concepts of Research
Objectives:
 Be able to better formulate a research
question
 Feel more comfortable with the clinical
research process
 Understand the ideas of constructs and
operationalization
 Understand the major differences
between quantitative and qualitative
approaches to doing educational
research
“If we knew what we were
doing, it wouldn’t be called
research, would it?”
Albert Einstein
Developing Your
Question
 Start with a clear purpose
 Know your literature
 Be iterative in your approach
 Try to specify the who, what, where
and when of your purpose
 Ask yourself “What would the answer to
this question add to the literature?”
and…
Developing Your
Question

Don’t let methodology drive

the question!
The Research
Process
1. Identification of general problem/question
2. Literature review
3. Specify questions/hypotheses
4. Determination of design/methodology
5. Data collection
6. Data analysis/presentation
7. Interpretation of findings
The Right and the
Left
Quantitative research - numbers, numbers, numbers

Qualitative research - words, words, words

Quantitative vs. Qualitative:
Assumptions about the World
 Based upon the idea  Based on the notion
of “logical of “constructivism”,
positivism”, that is, which assumes
there is a singular multiple realities that
are socially
reality with stable, constructed through
social facts that are individual and
separate from the collective perceptions
feelings and beliefs of or views of the same
individuals. situation.
Quantitative vs. Qualitative:
Research Purpose
 Seeks to establish  Concern is with the
relationships and understanding of the
explain causes of social phenomenon
changes in measured from the participants’
variables. That is, perspectives. This
the goal of science is requires, to some
to explain and degree, researcher
predict. participation.
Quantitative vs. Qualitative:
Methods and Process
 The scientific  Use of emergent
method, also known design utilizing
as a priori or pre- constant comparison
established design. and revision.
Quantitative vs. Qualitative:
Prototypical Studies
 Experimental or  Use of ethnography,
correlational designs which helps readers
are used to reduce understand the
error, bias and the multiple perspectives
of the situation by the
influence of persons studied.
extraneous variables-- Subjectivity in data
control of bias is analysis and
through design. interpretation is
acknowledged.
Quantitative vs. Qualitative:
Researcher Role
 Detachment from  Immersion in situation
study in order to and the phenomenon
avoid bias. being studied.
Quantitative vs. Qualitative:
Context
 Context-free  Generalizations are
generalizations contextually-bound.
Quantitative vs. Qualitative:
Precision
 Obtained through the  Provided by detailed
use of measurement description of
and statistics phenomenon
Quantitative vs. Qualitative:
Verification
 Results replicated by  Extension of
others. understandings by
others.
Quantitative vs. Qualitative:
Explanation
 Traditionally,  Summary through
parsimonious narrative--importance is
explanations were placed on reducing
sought, but this may be complex realities to
changing due to simple explanations.
technology.
Quantitative vs. Qualitative:
Logical Reasoning
 DEDUCTIVE--What’s the  INDUCTIVE--anyone have
classic example? an example?
Quantitative vs. Qualitative:
Conditional Conclusions
 Statements of statistical  Tentative summary
probability. interpretations.
Types of Quantitative
Studies
 Descriptive
 Descriptive research refers to the methods that describe the characteristics of the
variables under study

 True experimental
 True experimental research relies on statistical analysis to prove or disprove a
hypothesis

 Quasi-experimental
 Quasi-experimental research is similar to experimental research in that there is manipulation of
an independent variable.

 Correlational
 Correlational research refers to a non-experimental research method which studies the
relationship between two variables with the help of statistical analysis

 Predictive
 Predictive research is chiefly concerned with forecasting (predicting) outcomes,
consequences, costs, or effects.
Research Using Primary
Data
 Cross-sectional
 In a cross-sectional study, the investigator measures the outcome and the
exposures in the study participants at the same time

 Case Control
 A study that compares two groups of people: those with the disease or
condition under study (cases) and a very similar group of people who
do not have the disease or condition (controls).

 Cohort
 cohort studies recruit and follow participants who share a common
characteristic, such as a particular occupation or demographic
similarity.

 Randomized controlled trial

 A randomized controlled trial (RCT) is an experimental form of impact
evaluation in which the population receiving the programme or policy
intervention is chosen at random from the eligible population, and a
control group is also chosen at random from the same eligible
population.
Cross-sectional Study

 Data gathered at one point in time

 Often used for surveys
 Can not make inferences about
causality
Case Control Study

 Start with the outcome- identify a

sample with the condition of interest
 Identify a similar control group
 Look back to determine exposure
 Calculate the risk in the cases and
controls- odds ratio used
 Can not use to establish prevalence
Cohort Study

 Start with an identified group

 Determine exposure in everyone at the
same time
 Follow the group to determine who
develops the outcome of interest
 Can be used to determine prevalence
 Association measured as relative risk
(rate ratios)
Randomized Controlled Trial

 Gold standard for determining

associations
 Identify a group
 Randomly assign individuals to
exposure
 Only reliable way to control for
confounding
Research Using Secondary
Data
 Literature review
 A literature review is a search and evaluation of the available
literature in your given subject or chosen topic area. It documents
the state of the art with respect to the subject or topic you are
writing about.

 Systematic review
 A systematic review is a review of a clearly formulated question that
uses systematic and reproducible methods to identify, select and
critically appraise all relevant research, and to collect and analyse
data from the studies that are included in the review.

 Metanalysis
 Meta-analysis is a research process used to systematically synthesise
or merge the findings of single, independent studies, using statistical
methods to calculate an overall or 'absolute' effect.2 Meta-analysis
does not simply pool data from smaller studies to achieve a larger
sample size.

 Analysis of existing data collected for another purpose

Analysis of Existing Data
 Use administrative data for research
 Insurance claims data- Medicare
 Central Data Repository (CDR) at Uva

 Use regularly collected survey data

 National Center for Health Statistics performs
multiple surveys periodically- NAMCS, NHIS, NMCES,
NHANES

 Use data collected for another study

RES 313 – CS
Methods of
Research
OBJECTIVES

 Discuss what is meant by and why there

should be ethical standards in Bio Medical
research
 Identify ethical issues in Bio Medical research
which would amount to scientific misconduct
 Explain the role of an ethics review
committee
 Construct a set of guidelines for members of
ethics review committees
Guiding principles

 Autonomy and respect

 Beneficence
 Non-maleficence
 Justice ( free from exploitation)
 Scientific validity
 Honesty
Scenario 1

 Two graduate students have

made some measurements
on a new material. The data
points are as shown. To
prove their hypothesis the
results should lie on the
curve shown. The two
students considered
omitting the two data
points which were off the
theoretical curve.
Answer

 Unethical as it would amount to

falsification of data

 Should include outliers and give probable

reasons or find out statistically
acceptable ways of trimming outliers
Scenario 2

 A group of medical students conducted a

research on the awareness of diabetic
diet in medical clinic participants. Their
research was recognized as the best
undergraduate research and later they
submitted the same research paper to
two different journals to see which
journal publishes it first.
Answer

 Unethical as it would result in "inadvertent

double-counting or inappropriate weighting of
the results of a single study, which distorts
the available evidence
 It would give a false idea of the number of
publications in a given area
 wasting of resources on the review and
publication process
 Should submit to one journal and wait for
response prior to submitting to another
Scenario 3

 Students are required to prepare a

research proposal during their
undergraduate program. Nimal developed
the idea for his project and discussed with
a friend. Several months later, he found
that his idea had been submitted as a
research proposal by his friend without his
knowledge.
Answer

 Unethical as failure to give credit to the

person whose idea it is (intellectual property)
amounts to plagiarism

 Should discuss and include as co-author

Scenario 4

 Four friends decide to work together on a

research project during the vacation. One
of them went abroad during the vacation
and did not contribute to the research.
The friends include all 4 names in a
presentation made at a scientific
congress.
Answer

 Unethical as only those who contributed

intellectually should be cited as authors

 Those who contribute in other ways may

be acknowledged
Scenario 5

 A group of undergraduate students

planned a research project on the
detection of fetal abnormalities in the
second trimester, by ultrasound scanning.
They collected data from the scan room
without informing the mothers
Answer

 Unethical as informed consent was not

taken

 Should have informed mothers of their

intent even though there is no particular
advantage/disadvantage to the mother in
doing so
Scenario 6

 A group of undergraduate students

collected data from a group of bank
officers, with their consent, regarding
their working hours and salary with
regards to the prevalence of high blood
pressure. Subsequently the researchers
gave the same data to another group who
were in need of same data variables.
Answer

 Unethical as violating principles of consent and

confidentiality
 Data can be used for a secondary purpose which
was not first considered as long as
 informed consent for sharing has been given
 identities anonymised
 due consideration to access restrictions
 Develop ethical guidelines for data sharing?
Why should there be
research ethics?
 To protect participants /patients /society
/resources /researcher?
 To ensure accuracy of scientific
knowledge
 To protect intellectual and property rights
To protect
participants/patients/society/re
sources
 Protect from harm
 Show respect
 privacy /confidentiality
 Informed consent
 Refrain from Coercion and undue
inducement When do incentives amount
to coersion/undue inducement?
To protect
participants/patients/society/re
sources
 Refrain from exploitation of vulnerable
participants Who are the vulnerable groups?
 Refrain from indiscriminate use of resources
 Ensure Favorable Risk-Benefit Ratio: risk
should be minimized & potential benefit to
society must outweigh risks
To protect
participants/patients/society/re
sources
 Must be of social value: improvement of
health/knowledge for the benefit of
society/science
 high social value: use of stem cells to improve
quality of life for Huntington's disease patients
 less social value: drug studies conducted to
obtain data that allows a new drug to compete
in the healthcare marketplace even though
existing effective and often cheaper
therapeutics are already available
To ensure accuracy of scientific
knowledge

 Should be methodically rigorous - Scientific

validity
 Fair subject selection: with inclusion / exclusion
criteria & a valid number of subjects in order to
project results to the population
 State research method clearly so that another
person can conduct advanced study in future by
using publication
 Is it ethical to copy the methodology from a
published paper?
To ensure accuracy of scientific
knowledge

 Do not gloss research method

 Should not falsify/modify/omit data
 Use actual data for analysis/cannot include someone else's
data
 Report errors
 Be aware of conflict of interest
 Should not withhold and/or ‘vaguing up’ information
 Keep data and material for 5 years
 Data and material should be available to others
To ensure accuracy of scientific
knowledge

 Do not present/publish paper from

incomplete research or from anticipated
outcomes
 Should not duplicate publications and
submissions
 Avoid piecemeal publication
 Should be reviewed Independently by
unaffiliated individuals
To protect intellectual and
property rights
 Citation and authorship
 Inclusion
 writing and significant scientific
contribution
 order
 order of contribution
 -actual researchers
 -approval must be sought to include a
name
To protect intellectual and
property rights
 Whenever somebody else’s work is quoted
reference should be made to the original
author (Piracy vs plagiarism)

 Acknowledgement should include the

names of person who helped
 Breach of ethics in
research would
amount to scientific
misconduct
Scientific Misconduct

 Fraud : invention/fabrication of data

 Plagiarism : copying data, ideas, text
without acknowledgement of source
 Piracy : infringement of a copyright
 Submitting/Publishing the same paper to
different journals
Scientific Misconduct

 Not informing a collaborator of your

intent to file a patent in order to make
sure that you are the sole inventor

 Including a colleague as an author on a

paper in return for a favor even though
the colleague did not make a serious
contribution to the paper
Scientific Misconduct

 Trimming outliers from a data set without

discussing your reasons in paper
 Using an inappropriate statistical
technique in order to enhance the
significance of your research
 Bypassing the peer review process and
announcing your results through a press
conference without giving peers adequate
information to review your work
Scientific Misconduct

 Conducting a review of the literature that

fails to acknowledge contributions of others
 Stretching the truth on a grant application
in order to convince reviewers that your
project will make a significant contribution
to the field
 Giving the same research project to two
graduate students in order to see who can
do it the fastest
Scientific Misconduct

 Overworking, neglecting, or exploiting

research students
 Making derogatory comments and
personal attacks in your review of author's
submission
 Making significant deviations from the
research protocol approved by the Review
Board without informing the committee
Scientific Misconduct

 Not reporting an adverse event in a

human research experiment
 Wasting animals in research
 Exposing students and staff to biological
risks
 Rejecting a manuscript for publication
without even reading it
Scientific Misconduct

 Sabotaging someone's work

 Rigging an experiment so you know how it
will turn out
 Deliberately overestimating the clinical
significance of a new drug in order to
obtain economic benefits
Ethical review committee
(Institutional Review Boards)

Convened to
 maintain ethical standards of practice
in research
 ensure protection of subjects/research
workers from harm or exploitation
 to provide reassurance to the public
 protect researchers from unjustified
criticism
Procedure for ethical
clearance
 Fill ethical clearance form and attach
proposal, questionnaires, informed
consent forms, information leaflets etc.,
and submit
 When ethical clearance is granted, data
collection can commence according to the
approved methodology