COMMUNICATION SCIENCE (SBB3802) LECTURE NOTES - Week 5

SBB 3802
SCIENCE COMMUNICATION
Week 5
Dr. Edlic Sathiamurthy
No. 6, Annex. FST
Tel: 668 3192
edlic@umt.edu.my
Week Lecture Activities Hour
1 Introduction 15 minutes 2
-Course
Course aim and objectives essay on course
-Course focus objectives
-Instruction and assessment method etc. Form groups and
*55 minutes speech (5msp) select a contact
person
2 Fundamentals of Communication 5msp 3
-Speech
Speech making tips
-Ethics
3 Practical Aspect of Communication – 1 5msp 3
-Data
Data mining
4 Practical Aspect of Communication – 2 Designing a poster 3
- How to design a poster exercise
P t topic
Poster t i
5 Scientific research methodology – 1 5msp 3
-Qualitative and quantitative
-Inductive
I d i andd deductive
d d i approach h
-Research process
-Research design
Scientific research methodology
• What is research?
• What is the difference between scientific
research and nonnon-scientific
scientific research?
• Is scientific research confined to sciences
only?
l ?HHow about
b t social
i l studies?
t di ?
• What is methodology?
• So… what is scientific research
methodology? gy
What is scientific research?
• It is a search for knowledge or solution
using scientific method
• What is scientific method?
• Read the following article and identify key
points
i t about
b t scientific
i tifi method
th d
Basic Research Principles
• Research
– The search toward the solution of problem
– Calls attention to theories that may be helpful
in predicting future experiences
– Performed in many arenas and may be done
in a varietyy of ways
y
©2006 All rights reserved.

• Two types of research - basic and applied
• Basic research
– The investigator is not concerned with the
immediate applicability of his results but
rather he or she tries to look for
understanding of natural processes
• Applied research
– The investigator has some kind of application
in mind and wishes to solve a problem or in
some way contribute to society

• Research Methodology
– Two main approaches to research
• Quantitative
• Qualitative

– Qualitative Research
• Describes events, persons and so forth without the
use of numerical data
• Consists of interviews, observations and written
documents
• These may be used individually or in combination

– Quantitative Research
• Uses numbers, to describe the study, including
some comparisons of the population and statistical
analysis to describe the results
• Consists of observations, experiments and surveys

Source: Gauch, 2003
Rationalism in methodology
• Rational thinking in science – based on logic or
reason
• Inductive and deductive logic are applied
scientific research approach ((quantitative and
qualitative)
• Reason is based on evidence and proof
• Evidence and proof is gathered from reliable
data
• Reliable data is ggathered from objective
j
measurement using a systematic way or
scientific method
• However….
However Is everything measurable? Is there
a limitation to scientific reasoning
Limitation to rationalism
• Is the belief in God logical?
• Can you proof God exists?
• I science
Is i atheistic
th i ti iintrinsically?
t i i ll ?
• Some fundamentally flawed scientific
rationalism
• There is NO truth in science,, onlyy
probability at varying levels
• Assumption,
Assumption hypothesis
hypothesis, theory and law
Inductive logic
g
• Argues from the real world to the mind
• Make
M k conclusions
l i ffrom observed
b d phenomena
h
• Example:
– Premise 1 – Every donkey that ever has been
observed has a heart
– Conclusion – Every donkey has a heart
• Basis of sciences
• Problems:
– How manyy observations are needed to form a strong
g
conclusion?
– Can we observe every case or phenomenon?
– Can we observe something objectively without any
form of bias?
– How about our perception?
Deductive logic
g
• The opposite of inductive logic – argues from the mind to
the real world
• Make conclusions concerning a phenomenon based on
a tested assumption or hypothesis
• Example:
– Premise 1: Every mammal has a heart
– Premise 2: Every donkey is a mammal
– Conclusion: Every donkey has a heart
• Basis of mathematics
• Problems:
• Can it be purely imagination without real existence?
• How valid is the hypothesis or assumption?
• How many tests, types of parameters and observations
are needed to support a conclusion?
• How sure are we that they are relevant?
Inductive
Deductive
• Research Process
• Six major steps
– Define the problem
– Review the literature
– Design the research
– Collect
C ll t the
th data
d t
– Analyze the data
– Draw
D C
Conclusions
l i

• Define the problem
– Forming a question regarding a topic you would like to study
• The most important thing is to be clear about is what you want to
study
– Often an analysis of historical data, or secondary information,
has gone into the problem definition
• Looking at the past to see what has been done before
• Prevents “reinventing the wheel”
• In previous research, the investigator may have made
recommendations for future studies
– Problems may also be broken down into sub-problems or
smaller problems
– May also define the scope of the study - who or what will be
included in the study

• Review the literature
– An investigation of all the information about a topic
– Three reasons for Literature Review
• Has research already been done to answer your question?
• Are there any data sources you can use in your study?
• May help make your hypothesis more specific
• Sou
Sources
ces of
o previous
p e ous work
o
– Journals, books, position papers, conference
presentations, videos, interviews and online
subscription
b i ti servicesi

• Design the research and Collect the data
yp of research design
• Several types g
– Exploratory
– Historical
– Conclusive
• Includes descriptive and causal
– Correlational
– Evaluation
– Experimental

– Exploratory research
• Undertaken because a problem is not very clearly defined
• Allows the researcher to study the problem and gather
information about the problem
• May generate a hypothesis
• Generally informal and relies on literature review and
informal discussions with others in order to find out more
about a problem
• This type of research may not help you answer your problem
but rather mayy provide
p yyou with some insights
g to the p
problem

– Historical research
• Involves investigation and analysis of events of the
past
• This type of research does not focus solely on the
past, but it also allows the researcher to apply a
previous researcher
researcher’ss experiences and
conclusions from their study into current
professional practice

– Conclusive research
• Performed in order to come to some sort of
conclusion or help in decision making
• May be done using primary research
– Data collected specifically for your study; or
• Secondary research
– Includes a literature review to see if previous studies can
be used to answer your question
– May
M alsol iinclude
l d summariesi off pastt works
k

• There are two types of conclusive research:
– Descriptive research
– Causal research

• Descriptive research - also called statistical
research
– Provides data about the population you are studying,
including the frequency that something occurs
– The two most common collection techniques of
descriptive research are observations and surveys

• Causal research - tries to answer questions about
what causes certain things to occur
– This type of research is difficult because there may
always be an additional cause to consider
– Uses experimentation and simulation as its data
collection methods

– Correlational research
• Studies that try to discover a relationship between variables
– Variable - anything under study
• The strength
g of the relationship
p is measured
– There may be a positive relationship or a negative relationship
between variables
• This type of research only determines if there is a
relationship between two or more variables
– It does not determine the cause of those relationships
– Uses questionnaires, observations and secondary data as its
data collection methods

– Evaluation research
• A process used to determine what has happened
during a given activity
• The purpose of evaluation is to lead to better
understanding of whether a program is effective,
whether a policy is working, or if something that
was agreed upon is the most cost-effective way of
doing something

– Experimental research
• Entails manipulation of a situation in some way in
order to test a hypothesis
• Certain variables are kept constant and an
independent or experimental variable is
manipulated
• Researchers select both independent and
dependent variables
– Independent variables are the factors that researchers
manipulate directly
– Dependent variables are the measured variables

• Statement of the Hypothesis
– In formal research, a hypothesis is formed
– A statement of the predicted relationship of what the
researcher is studying
– Ap proposed
p solution or explanation
p at which the
researcher has arrived through the review of the
literature
– Simply – it is the tentative answer to a question
– The statement of the hypothesis is important because
it allows the researcher to think about the variables to
incl de in the st
include study
d and ttype
pe of research design to
use in the study

– The research tests the hypothesis proving it to
be positive or negative (correct or incorrect)
• If the hypothesis
yp is rejected,
j that is, incorrect, it
does not necessarily mean that the research is
poor, but only that the results are different from
what was expected
– The formulation of the hypothesis in advance
of the data-gathering
data gathering process is necessary for
an unbiased investigation

– There are two forms of the hypothesis
• The null hypothesis
– This hypothesis states that there is no difference
between the population means or proportions that are
being compared; or that there is no association between
the two variables that are being compared
– For example, in a clinical trial of a new medication, the
null hypothesis is: The new medication is no better than
the current medication

• The alternative hypothesis
yp
– This hypothesis is a statement of what the study is set up
to establish
– For example,
example in our clinical trial of a new medication
medication, the
alternative hypothesis is: The new medication is better
than the current medication

• Collect the Data
– Includes primary research
• Data you get from observations, surveys and
interviews
– Includes secondary research
• Literature review or summaries of original studies
– Irrespective of the data collection method the
research must be valid and reliable

– Validityy
• The degree to which scientific observations
actually measure or record what they purport to
measure
• For example, if you are using a written
questionnaire to collect data, the researcher will
pre test it,
pre-test it that is,
is give it to someone that they may
have included in their population to complete in
order to determine if the instrument is well written,
clear and inclusive of everything he or she is
looking for

– Reliability
y
• The repeatability, including interperson
replicability, of scientific observations
• With reliability
reliability, the major question is: Can another
researcher reproduce the study using a similar
instrument and get similar results?
– A study
d may b be reliable
li bl b
but not valid
lid
– That is, the study may be able to be
replicated but still doesn’t
replicated, doesn t answer the
research question

– Depends
p on the type
yp of research the
investigator wishes to conduct
– For example, if the researcher want to
establish
t bli h a causall relationship,
l ti hi ththey should
h ld
conduct one of the experimental studies
– If they are breaking new ground in a poorly
understood area of practice, they may want to
consider an exploratory study in a qualitative
design

– Surveys
y
• Gathers data from a relatively large number of
cases at a particular time
• Can include interviews and questionnaire surveys
– The questions should be well thought out to be sure to
answer the question of the research study
– The questions can be restricted - closed ended
» The investigator only wants certain answers
» Provide for unanticipated response by providing an
“other”
other category that permits respondents to indicate
other thoughts

– Surveys
• Questionnaire surveys
– The questions can be unrestricted - open ended
» This allows the participant to express a freer
response in his own words
» These questionnaires can be difficult to tabulate,
however, they are easier to write

– Observation
• Instead of asking a participant questions, the
investigator observes the participant
– Non-participant observation, the examiner is a neutral
observer who does not interact with the participants
– Participant observation, the researcher may y also
participant in the actions being observed, however they
try to maintain their objectivity

– Observation
• Ethnography or Naturalistic Inquiry
– The researcher observes, listens to, and sometimes
converses with the subjects in as free and natural an
atmosphere as possible
– The assumption is that the most important behavior of
individuals in groups is a dynamic process of complex
interactions and consists of more than a set of facts,
statistics, or even discrete incidents
– A position of neutrality is important in this type of
research

– Experimental study
• This provides a logical, systematic way to answer
the question, “If this is done under carefully
controlled conditions, what will happen?”
• Experimentation is a sophisticated technique for
collection of data

• Selecting an Instrument
– The instrument, also called a tool, is a consistent way
to collect data
– There are many different types of instruments to use
– Should
Sh ld fit th
the purpose off th
the researchh
– Researchers should not develop an instrument until
they have established that one does not already exist
– If you decide to develop your own instrument take
your time in developing the questions
• It would be too costly and time consuming to have to repeat a
survey because you neglected a question or two

• Samples
– The selection of subjects for the study
– For example, if your study concerns health
information departments in the United States, you
could study every health information department in
the United States, (this, by the way is referred to as a
census, as you are including all of the population)
• That would not be very practical and could be very costly and
time consuming
• The next best thing is a sample
– When chosen correctly samples are considered to be
representative of the population

• Samples
– Types of Samples
• Two types of sampling techniques
– Probability sampling
– Non-probability sampling

• Probability Samples
– Simple Random Sample (SRS)
• Consists of individuals from the population chosen in such a way
that every individual has an equal chance of being selected
• The simplest way to select a sample is to place names in a hat (the
population) and draw out a handful (the sample)
• Statistics
St ti ti b books
k also
l iinclude
l d a ttable
bl off random
d numbers
b th
thatt can b
be
used for the selection of random samples
• Systematic random sampling
– A syste
systematic
at c patte
pattern is
s used with
t random
a do sa sampling
p g
– For example, let’s say you are choosing a list of discharged patients
from the last month, you choose the first patient randomly, then choose
every fifth patient (or whatever number you choose) thereafter
– In this sample
p the choice of the first p
patient determines the others

– Stratified Sampling
• First divide the population into groups of similar
individuals, called strata
• Then choose a separate SRS in each stratum and
combine these SRSs to form the full sample
• For example,
example you may want to choose a selection
by gender, by patients with private insurance and
those without or separating hospitals by bed size

– Cluster Sampling
• The population is selected from groups or clusters
p let’s say
• For example, y yyour study
y includes health
information practitioners working in large cities
– We would choose the cities first – the cluster
– Then we would randomly choose the health information
practitioners
– This is called Two-stage cluster sampling

• Non-probability Samples - Almost all qualitative
research methods rely on this type of sampling
– Judgment Sampling
• The researcher relies on his or her own judgment to select
the subjects
• For example, let’s say you are performing a study on
transcription departments in acute care hospitals that use an
incentive plan to pay employees
– There are a number of large hospitals as well as small and
medium sized hospitals using an incentive plan
– In a judgment sampling, the researcher decides which hospitals
to use in the groups

• Non-probability Samples
– Quota Sampling
• The population is first divided as in stratified sampling
• Then the researcher uses his or her own judgment to select
the number of subjects
j based on a specified
p p
proportion
p
• Using our example of transcription departments that use an
incentive plan to pay employees
– If there are 100 hospitals in your state, and 50 are using an
incentive program and 50 are not, you may choose to perform
your study on 20 of the hospitals as that would be 20% of the
total population of hospitals in your state

• Non-probability Samples
– Convenience Sampling
p g
• The selection is based on availability of the
subjects
• Using our example above above, you might decide to use
only the hospitals in your city or within a certain
driving distance in order to expedite the study
• Not
N t clear
l if thi
this ttype off sample
l iis really
ll
representative of the population
– However, you may be able to get some insights about
th question
the ti you are ttrying
i tto answer

• Institutional Review Board (IRB)
– Responsible for reviewing all research
projects for approval
– IRB is the common name, but can be called
by whatever name an organization chooses
– Purpose
p is to ensure that steps
p are being
g
taken by the researcher to protect the rights
and welfare of anyone participating in the
research study

• Institutional Review Board
– Responsible for reviewing the research
procedures before the study is begun
– May require periodic reviews during the study
– May approve, revise or deny requests for
research
– Research on humans must be done careful to
ensure that subjects are not abused in any
way

– Researcher must obtain permission from the
IRB before any research study is started and
prepare a plan for the research
• This includes
– Preparing the instrument
– Deciding how to select a sample
– Deciding how to collect date
– Getting informed consent forms for the subjects to read
and sign

– Organizations may also have a Monitoring
Committee to ensure that there are no averse
effects on participants
– May recommend closure of research that is
not meeting safety standards, does not have
scientific merit or is not meeting the goals of
the research

• Sample Size
– The ideal sample is large enough to serve as an
adequate representation of the population about
which the researcher wishes to generalize and small
enough to be selected economically
• In terms of subject availability, expense in both time and
money, and complexity of data analysis
– Th
The larger
l the
th sample,l the
th smaller
ll ththe magnitude
it d off
sampling error
– Surveyy studies ordinarilyy have a larger
g sample
p size
then experimental studies

• Sample Size
– If yyou mail questionnaires,
q , your
y response
p rate
could be a low as 20 to 30 percent so a large
initial sample is recommended
– If you are planning
l i tto have
h subgroups
b ffrom
your population, start out with a large group to
makea e su
sure
e you have
a ee
enough
oug paparticipants
t c pa ts for
o
the subgroups
– Subject availability and costs are legitimate
considerations in determining a sample size

• Analyze the Data
– The investigator tries to determine what the
data disclose
• There are a varietyy of statistics used to analyze
y
data
• Most investigators use a variety of techniques to
describe the data
– Two types of statistical applications are
relevant to most studies – Descriptive and
Inferential statistics

– Descriptive statistics
• These are statistics that describe the data
• Includes measures of central tendencyy and
measures of variation
– Inferential statistics
• These statistics allow the researchers to make
inferences about the population characteristics
(parameters) from the sample
sample’s
s characteristics
• T tests and analysis of variance are examples

– Statistics Software Packages
• Help the researcher produce descriptive statistics,
graphs and charts

• Draw Conclusions
– Any conclusions should be related to the hypothesis,
or if using the qualitative approach
approach, then the problem
identified
– Results for each hypothesis should be described
– Any limitations that the researcher discovered during
data analysis should be reported
• For example, our study of transcription departments studied
y – this may
one city y be a limiting
g factor
– New hypotheses may be proposed if the data do not
support the original hypotheses
– Researchers usually include tables and graphs in this
section of the research report to clarify the data

• How to develop research skills
– Take at least one course in statistics and
research methodologies
– Begin to read research studies in professional
j
journals
l tto see h
how others
th h
have performed
f d
their research
– Agree to work with a skilled health information
researcher on a study to gain experience
– Learn how to p present data effectively
y both in
written form and verbally

• Draw Conclusions
– Research always raises new questions about
what to study
• These become suggestions
gg for future research
– The researcher should always include
conclusions as to whether or not this problem
is better understood or perhaps even resolved
from the research

You
Again!

COMMUNICATION SCIENCE (SBB3802) LECTURE NOTES - Week 5

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

COMMUNICATION SCIENCE (SBB3802) LECTURE NOTES - Week 5

Загружено:

Авторское право:

Доступные форматы

SBB 3802

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

©2006 All rights reserved.

Вам также может понравиться