SC 215: SCIENTIFIC METHOD (2 UNITS)
SC 215: SCIENTIFIC
METHOD (2 UNIT
A COURSE ON:
SCIENTIC RESEARCH METHODOLOGY
1.
Meaning,
Purpose and Objectives of Science and Scientific Research
1.1 What is Science? What is research?
Science
·
The term
“science” comes from the Latin word “scientia”, which means knowledge.
·
Definition
of Science: (acc. to Webster’s Third International Dictionary, 1986)
o
“accumulated
and accepted knowledge that has been systemized and formulated with reference
(seek 4info) to the discovery of general truths or the operation of general
laws”
o
or “
knowledge obtained and tested through use of the scientific method”
·
3
components to the definition of science:
o
knowledge
§
Observations,
information, facts, theories, principles, solution to a problem, etc.
§
about
biophysical world and human beings
o
knowledge
that has been tested through the scientific method
o
knowledge
that has been systemized, organized and arranged logically
·
Scientific
theories consist of general principles or laws that attempt to explain how and
why something happens or happened” (World Book Encyclopedia)
·
A theory
is only accepted as scientific knowledge when it has been verified by other
scientific studies.
o
In fact, a
theory remains a theory until it has been repeatedly proven or tested
experimentally.
o
E.g., the
Theory of Evolution is still a theory.
·
Technology
is the application of scientific knowledge in the production of tools,
machines, sources of power, materials, techniques, etc. that we use to make our
life or work easier.
Branches
or Disciplines of Science
·
Natural
sciences (biophysical sciences)
o
Mathematics
and logic
o
Physical
sciences (physics, chemistry, geology)
o
Biological
or Life sciences
·
Social
sciences
o
Economics,
Sociology, Anthropology, Political Science, Psychology
o
NOTE: The
humanities (religion, philosophy, fine arts, etc.) are not part of science,
because knowledge of these cannot be scientifically proven or tested.
Scientific Method
= the principles,
processes and procedures used in the systematic pursuit of objective knowledge,
involving the definition of a problem, the formulation of hypotheses, the
collection of data, and finally, the testing of those hypotheses and drawing
conclusions.
Scientific
research
= systematized effort
to gain new knowledge
= a process or a
movement from the unknown to the known.
= the
systematic, creative process or methodology of gaining original knowledge,
either as a solution to problem on a specific topic or as a generalization of
knowledge, i.e., a formulation of a theory.
= an inquiry
into the nature of, the reasons for, and the consequences of any particular set
of circumstances, whether these circumstances are experimentally controlled or
observed as they occur naturally (without manipulation).
1.2 Purpose of scientific research
The purpose of
scientific research is to discover the unknown and to bring into the realm
(sphere, specialty) of systematized knowledge.
1.3 Objectives of
Scientific Research
Objective 1: To gain
new insights into or better understanding about a phenomenon (event, object,
condition or situation).
·
Termed
formulative or exploratory research.
·
Often
comes up with new theories.
Objective 2: To
systematically describe the characteristics of a situation, an object or a
group of objects.
·
Termed
descriptive research
Objective 3: To test
hypotheses about cause-and-effect relationships between variables.
·
Termed
hypothesis-testing research
1.4
Benefits or Significance of Conducting Scientific Research
·
New
contributions to knowledge.
·
Scientific
research as an agent of diversification and evolution of society
o
Enormous
selection of goods, products.
o
People
select the best, which leads to evolution of life styles and society in general
·
Important
role in the advancement or progress of human civilization.
·
Provides
basis for government policies.
·
Facilitates
development and economic advances.
·
Discovers
solutions to social problems
·
Invents
new tools and technologies that makes work easier and save time.
·
Acquisition
of organized or systematic thinking.
o
Science
trains the mind in logic and understanding.
·
The
researcher acquires a new intellectual tool, i.e., ability to look at the world
and every day experiences in life in an objective and analytical way.
·
Man is the
only creature in the Universe (as far as we know) that has advanced
intellectual powers and the capacity for rational thinking.
·
Through
science, man rises above nature and can control his environment
·
A WORD OF
CAUTION:
o
“But
science itself is neither good nor bad” (World Book Encyclopedia, 1991)
o
Depending
upon how science is used, it can be very beneficial or very harmful
o
The use of
science must be controlled by moral, religious and ethical values; otherwise,
it leads to destruction or chaos (e.g., nuclear weapons, computer viruses).
o
It can
cause destruction of the environment and society
o
Thus, it is
important that we aim at only useful, productive or positive uses of science
and technology.
2.
Steps in
the Scientific Method and Its Distinguishing Features
2.1 The Process
·
Steps in
the scientific method are not mutually exclusive; they overlap.
·
The order
may sometimes change.
·
Feedback
from one step back to the previous one
o
Provides
quality control in the research process.
o
E.g.,
analysis and testing of hypotheses sometimes necessitates re-formulation of the
hypotheses.
o
The final
report results in re-defining the same research problem or defining a new one.
·
Feed
forward to a later step
o
Provides
criteria for evaluation.
o
E.g.,
formulating hypotheses provides criteria for drawing conclusions.
Flow Chart of the
Scientific Method or Research Process
|
I
 |
II
 |
III
 |
IV
 |
V
 |
VI
 |
VII
 |
VIII
2.2 Distinguishing features of scientific research
·
Scientific
research must be based on logic.
·
Systematic
(orderly) in methodology and organization of results
·
It
utilizes relevant concepts – don’t try to “re-invent the wheel”
·
It is
based on objective considerations (logical relationships) – not subjective or
based on personal judgment or feelings
·
Requires
ethical neutrality and complete lack of prejudice (this is more for social
sciences)
·
Science
must be rigorous (thorough, following strict procedures, accurate), not
allowing for misinterpretations, ambiguous results or loopholes in the
conclusions. (E.g., if you are conducting an experiment and control light,
salinity, pressure, etc., but you forget to control temperature; that is not
being rigorous.)
·
Procedures
require consistency. (E.g. always use the same thermometer, the same chemicals,
the same procedures, measure at the same time of day, etc.)
·
It
requires empirical evidence (data derived by observation or experiment)
·
Draws
conclusions and makes predictions that are usually probabilistic.
·
Results
should be repeatable and replicable – either by other scientists or by the same
scientist later.
·
Scientific
findings should be communicated to others (journals, books, conferences), so
that they are open to scrutiny by others.
·
Useful
findings are such that they can be generalized or extended to broader
circumstances.
o
They may
result in general axioms or scientific theories.
3.
Characteristics,
Qualities and Attitudes of a Competent Scientist
·
Curiosity,
Inquisitive
·
Observant
·
Self-motivated
·
Open-minded
·
Not
prejudice, not swayed by emotional feelings, without personal bias
·
Does not
jump to conclusions (e.g., You feed chickens on a certain diet for one month,
then you change to a different diet for one month and most of them die. Can you
conclude that the diet killed them?)
·
Creative
·
Hardworking
·
Power of
concentration
·
Ability to
synthesize, interpret, analyze, integrate
·
Collaborative
·
Communicative
·
Practice
professional ethics
·
Not cook
data; report data as it is but discuss why it is different from expected
·
Acknowledge
the contributions of others
·
Etc.
4.
Types of
Science or Scientific Studies
4.1 Observational versus experimental studies
Observational studies
or surveys
·
Researcher
collects data about a particular phenomenon as it occurs in nature.
·
No
manipulation or control of any variables or conditions.
·
Describes
a situation, as it exists.
·
Can
discover correlations between variables.
·
Analyzed
by:
o
Comparing
groups
o
Correlating
variables
·
Cannot
establish a cause-and-effect relationship because there are so many factors
which may affect the variable being measured
·
May
suggest a possible cause-and-effect relationship that can later be tested by
setting up an experiment.
Experimental Studies
·
Researcher
sets up an experiment.
o
Manipulates
certain variables (changes them deliberately)
o
Controls
other variables (makes them constant)
o
Measure
other variables about which he wants to draw conclusions.
·
Can
establish cause-and-effect relationships.
·
Analyzed
by:
o
Comparing
groups
o
Regression
analysis.
4.2 Basic
(Fundamental) vs. Applied Research
Basic, fundamental or
pure research
·
Research
just for the sake of gaining knowledge, regardless of whether it has any useful
application.
·
Basic
research often comes up with knowledge that may be applied much later or lead
to further studies that are applied.
·
Concerned
with generalizations or formulation of theories.
Applied research
·
Aims at
finding a solution to a practical problem facing society (e.g., in agriculture,
health, environmental science, etc.).
Comparative
examples
·
Studying
the various properties of elements and substances is basic research, but
studying how the properties of certain elements can be useful, e.g., in
computer technology, medicine, construction, etc., is applied research.
·
Studying
reproductive behavior in animals is basic research, but using that knowledge to
try to save an endangered species is applied science.
4.3 Qualitative vs.
Quantitative Research
Qualitative research
·
Collects
data that are not measurable; they only describe the object under study by
attributes or qualities.
·
E.g.,
observing color of flowers, rocks, etc.,
Quantitative research
·
Collects
data that are measurable
·
Data are
recorded according to a scale
·
E.g.,
height, force, energy, etc.
·
Usually
provides more information than qualitative research.
Combination of
qualitative and quantitative research
·
Many studies
collect both qualitative and quantitative types of data, which helps to give a
broader range of information about the phenomenon being studied.
4.4 Review Studies
·
Review
documents on a broad topic
·
Review
books, scientific papers, reports, other documents
·
Sometimes
called documentary review
·
Based on
secondary data, not primary data.
·
No
original data collected, but original theories, concepts and trends may be
obtained; thus review studies are also a new contribution to knowledge.
4.5 These types are not mutually exclusive
·
E.g.,
either observation or experimental research may be basic or applied and it may
be qualitative or quantitative.
5.
Defining
the Research Problem (and Setting Specific Objectives)
·
Going from
Step I to Step II of the Scientific Method: Formulation of a general topic into
specific research problem.
·
First, ask
many questions on the general topic
·
Make
general observations and review literature about the general topic.
·
Decide
which questions can feasibly be answered, considering the available resources,
time and knowledge.
o
These will
give you the specific research problem
·
3
sequential components to the definition of the research problem:
o
Statement
of the research problem
§
Outlines
the overall problem which needs a solution
§
Explains
what has already been discovered about that problem
§
States
what still needs to be investigated about that problem
§
States
what you are going to investigate about the problem.
o
General
objective (=Development objective) (Overall aim of the study)
§
States
what will generally be done in the research
§
Briefly
states what will be gained or achieved by doing the study
o
Specific
objectives
§
The
specific questions are stated in the form of specific objectives.
§
E.g.,
Question: Does UV light affect the strength of a certain type of plastic?
§
Objective:
To determine the effect of ultraviolet light on the strength of a certain type
of plastic (chemist)
§
E.g., To
determine the effect of UV light on human skin (biologist)
§
There must
be at least 2 possible outcomes
·
These will
form the hypotheses.
·
E.g., UV
light affects the strength of the plastic
Or UV light does not affect the strength of
the plastic.
·
The
definition of the research problem must be unambiguous.
·
The
specific objectives form the essence or focal point of the whole research plan
o
The
hypotheses are formulated based on the objectives.
o
The
objectives determine the data to be collected
o
The
objectives often determine the techniques to be used in data collection
o
The final
conclusions of the research come directly from the objectives, i.e., the
conclusions are the answers to the specific questions.
6.
Planning
and Preparation for Scientific Research
- Some of these items will go into your
research proposal
6.1 Research problem
and objectives
·
What you
want to do
6.2 Gather literature
on the topic
·
Build up a
library on the topic
·
Find out
what other people have done on the topic
6.3 Methodology
·
How you
are going to conduct the research
o
Research/sampling
design
o
Choose
techniques for data collection
6.4 List resources
required
·
Field
equipment
·
Lab
equipment
·
Materials
·
Transport
·
Human
resources
6.5 Prepare budget
·
Show
example
6.6 Work-plan with
timeframe
·
This is a
schedule of activities.
·
Show
example
6.7 Prepare checklists
·
Checklist
of field equipment and materials
o
This is
very N.B. in order to avoid getting to the field and finding that you have
forgotten one very important item
·
Checklist
of lab equipment and materials
6.8 Data collection
sheets
·
Very N.B.
to prepare in advance to be sure that you don’t forget to record anything.
·
Show
example
6.9 Design database
·
This
should be designed early so that you can enter data as you collect it.
·
Keep
several copies of the data on diskettes or CDs.
·
The
database should be designed such that it can be analyzed directly, without
feeding it into the computer again.
7.
Literature
Survey
7.1 Necessity of
Literature Search
·
In order
to plan research properly, you need to know what are the relevant
o
Relevant
theories and concepts
o
Previous
studies that are relevant to the topic
o
What has
already been discovered
o
Whether
the problem you want to investigate has already been solved
o
What are
the appropriate and accepted techniques for investigating this topic,
especially new techniques
o
Prevents
repeating studies that have already been done by others.
·
Usually you
cannot read everything on the topic
o
There are
millions of scientific publications every year
o
In the
website “google”, there were 496,000 reports on coral reefs last week—this week
there are 502,000 reports
o
On marble,
there are1,910,000 reports.
·
Therefore,
you can only attempt to review papers that are particularly relevant to your
topic
o
E.g.,
coral reefs—predation on coral polyps
o
E.g.,
marble—distribution of marble deposits
7.2 Types of
Scientific Literature to Search
·
Encyclopedias
·
Literature
guides
·
Books
·
Book
reviews
·
Review
Journals, e.g., Reviews of Modern Physics, Chemical Reviews, Annual Review of
Physical Chemistry, Botanical Review, Annual Review of Biochemistry
·
Abstract
and Indexing Journals
o
Available
in libraries, CD rom, and websites
o
E.g.
Science Abstracts
o
Chemical
Abstracts
o
Biological
Abstracts
o
Geophysical
Abstracts
o
Electronic
Engineering Master Index
·
Scientific
journals
·
Theses,
Dissertations
·
Conference
proceedings
·
Websites
o
E.g.,
http://www.google.com
7.3 Literature Cited
in Relevant Publications
·
When you
read a relevant publication, take note of the literature it cites and you may
get many more relevant publications
7.4 Getting Relevant
Information from a Scientific Report or Paper
·
Record
notes on the important points from each report you read, because you will not
remember later what was written
·
Either
quote word for word or paraphrase, but you will have to paraphrase later anyway
if you include those points in your own report.
·
In your
notes, cite the reference properly, giving all details so that you can find it
later and also include it in the reference section of your report
o
E.g.,
author(s), year, title, Journal, volume number, page nos.
·
Arrange
your notes from different papers systematically
o
Usually
alphabetically by the authors names
o
Write
notes for each report on a separate card
o
In the
computer, you can have a separate file for each paper or arrange them in one
file alphabetically.
8.
Formulating
and Testing Hypotheses
- These are also called “working
hypothesis”, because they dictate how you are going to work on your
research
- These are predictions based on literature,
discussions, personal observations, experience
- Dictate what you are going to measure
- Determine the type of analysis you will do
- Hypotheses are NOT needed in exploratory
or formulative research or descriptive research.
8.1 Formulation of
Hypotheses
·
Come
directly from the objectives
·
Two types
of hypotheses:
o
Difference
between groups, treatments
o
Relationships
between variables
·
Difference
between groups
o
Example 1:
§
Objective:
To determine whether there are greater deposits of marble in Mbeya or Dodoma.
§
Non-directional
or two-tailed hypotheses:
·
Ho: There
is no significant difference in the quantity of marble deposits in Mbeya and
Dodoma.
·
HA: There
is a significant difference in the quantity of marble deposits in Mbeya and
Dodoma.
§
Directional
or one-tailed hypotheses:
·
Ho: The
quantity of marble deposits is not significantly greater in Mbeya than in
Dodoma.
·
HA: The
quantity of marble deposits is significantly greater in Mbeya than in Dodoma.
o
Example 2:
§
Objective:
To determine whether there is a greater density of zebras in Ngorongoro crater
or Mikumi National Park
§
Non-directional
or two-tailed hypotheses:
·
Ho: There
is no significant difference in the density of zebras in Ngorongoro and Mikumi.
·
HA: There
is a significant difference in the density of zebras in Ngorongoro and Mikumi.
§
Directional
or one-tailed hypotheses:
·
Ho: The
density of zebras in not significantly greater in Ngorongoro than in Mikumi.
·
HA: The
density of zebras is significantly greater in Ngorongoro than in Mikumi.
·
Relationship
between variables
o
Example:
§
Objective:
To determine whether there is a correlation between nitrate levels in the soil
and the growth of maize.
§
Non-directional
or two-tailed hypotheses:
·
Ho: There
is no significant correlation between nitrate levels in the soil and the growth
of maize.
·
HA: There
is a significant correlation between nitrate levels in the soil and the growth
of maize.
§
Directional
or one-tailed hypotheses:
·
Ho: There
is no significant positive correlation between nitrate levels in the soil and
the growth of maize.
·
HA: There
is a significant positive correlation between nitrate levels in the soil and
the growth of maize.
8.2 Stating Hypotheses
in Research Proposals and Scientific Reports
·
The common
practice is to state only the hypothesis that you expect to prove to be true.
·
Usually
these will be the alternative hypotheses, in particular, the directional
hypotheses.
8.3 Steps in Testing
Hypotheses Statistically
Step 1: State the null
hypothesis (H0) and alternative hypothesis (HA).
Step 2: Choose an
appropriate statistical test, e.g., t test, analysis of variance, correlation,
etc.
Step 3: Make a rule
about the significance level () and specify the sample size (n).
In biology and many
other sciences, usually = 0.05 or 5%, which means that there should be less
than a 5% chance of making a mistake or you are 95% certain that you are right.
Step 4: Conduct a
study to test the null hypothesis and calculate the test statistic.
·
This gives
the calculated value of the test statistic.
Step 5: Examine the
table showing the theoretical distribution of the test statistic to obtain the
tabulated value of the test statistic, sometimes called the critical value.
·
these
values are given in tables, e.g., t-distribution table, normal distribution
table, etc.
Step 6: Draw a
conclusion, i.e., decide to accept or reject the H0 by comparing the calculated
value of the test statistic and the tabulated value of the test statistic at
the appropriate significance level.
RULES FOR ACCEPTING OR
REJECTING H0
1.
If the
calculated test statistic > tabulated test statistic at , we reject H0.
2.
If the
calculated test statistic < tabulated test statistic at , we accept H0.
(This applies to all
statistical tests except the Wilcoxon paired-sample test.)
Step 7: State the
probability (p) of making a mistake by committing a Type I error.
8.4 Type I and Type II
Errors
|
|
Null Hypothesis
(according to
statistical test)
|
||
|
Accepted
|
Rejected
|
||
|
Null Hypothesis
(in reality)
|
True
|
Correct decision
(no error)
|
Type I error
|
|
False
|
Type II error
|
Correct decision
(no error)
|
|
[Give 2 examples of M
vs. F: intelligence; physical strength]
Type I error =
rejection of H0 when it is in fact true.
·
p (Type I
error) = , i.e., the probability of committing a Type I error is .
·
The Type I
error can be determined by examining the theoretical distribution of the test
statistic.
Type II error =
accepting H0 when it is in fact false.
·
p (Type II
error) =
·
The Type
II error can not be determined
Power of a statistical
test
= 1-
= probability of
rejecting a H0 when it is in fact false
·
The power
of a test can be increased [or the p (Type II error) can be decreased] by:
o
Increasing
sample size (n)
o
Selected
the most powerful statistical test for the situation
9.
Research/Sampling
Design
9.1 Meaning of
Research Design
·
Research
design addresses the following questions:
o
Where will
the study be carried out?
o
What type
of data is required?
o
What should
the arrangement of samples in space be?
o
What
should be the arrangement of samples in time? When? How often?
o
What
techniques will be used to collect data?
o
How will
the data be analyzed?
·
Research
design is advance planning
·
Improves
the reliability of the results.
·
Helps to
avoid mistakes in the research.
·
4
components of research design:
o
Sampling
design – arrangement of samples in space and time
o
Observational
design – techniques for collecting data
o
Statistical
design – the plan for analyzing the data
-
ensures
that samples are collected in a way that they can be analyzed
o
Operational
design – the process and timing of implementing the sampling, observational,
and statistical designs.
·
Research
design must take into consideration:
o
Scientific
reliability - must be scientifically rigorous
o
Skills of
the researcher and his staff
o
Time
available
o
Money
available
o
Equipment
and materials available
9.2 Components of
Research
9.2.1 Independent and
dependent variables
·
Variable =
entity which is being measured.
o
E.g., weight,
force, energy, light, height, color, etc.
o
There is
natural variation in everything, so when you measure several objects or
specimens, you will get different values, thus “variable”.
·
Independent
variables = variables that are not affected by the other variables in the
study, but are expected to affect those other variables.
·
Dependent
variables = variables that are affected by the independent variables.
·
E.g., age
and height – height does not affect age, but age affects height
·
E.g., heat
and reaction rate
·
Independent
and dependent variables are defined by the research problem and are given in
the objectives and hypotheses.
·
In
experimental research, the researcher manipulates the independent variables
(i.e., deliberately changes them) and measures how this affects the dependent
variables.
·
In
observation (non-experimental) research, the researcher does not manipulate the
independent variables, but measures them as they occur in nature.
9.2.2 Extraneous
variables
·
Independent
variables that are not related to the purpose of the experiment, though they
could be of interest in a different experiment
·
These may
affect the dependent variables and lead to “experimental error”.
·
Thus,
extraneous variables must be controlled or made constant throughout the
research.
·
E.g., if
you are testing the effect of fertilizer on plant growth, light, soil type,
water, etc. are extraneous variables and must be kept constant for all
treatments.
9.2.3 Experimental and
control groups
·
Here, a
group refers to several replicates that receive the same treatment
·
Control
group is subjected or exposed to usual conditions
·
Experimental
groups are subjected to new conditions, which the researcher wants to
investigate.
·
E.g., In
an experiment where one is testing the effects of fertilizers on plant growth
o
Control
group is subjected to the usual conditions of soil, water, light, etc., but no
fertilizer.
o
Experimental
groups are subjected to the same conditions + different types of fertilizers or
different amounts of the same fertilizer.
·
Some
studies do not have control groups, but it is best to have control groups when
possible.
9.2 4 Treatments
·
The
different conditions to which experimental and control groups are subjected or
exposed.
·
Sometimes
the term treatments can mean the experimental and control groups
9.2.5 Experimental
units
·
Pre-determined
plots or blocks where different treatments are applied.
9.2.6 Replication
·
Good
experimental results should be repeatable and replicable. One way to replicate
the results is to have several samples or replicates in the same experiment.
·
Replication
is needed to:
o
Check or
confirm the results
o
For
statistical analysis
o
Estimate
the precision (e.g calculate standard deviation) or state the probability that
the conclusion is correct.
·
Statistical
tests are based on replicates.
·
Replication
is particularly important in biological sciences because there is so much
natural variation from one organism to another or from one place to another
within an ecosystem
·
In
physical sciences, where data are collected from the natural environment, e.g.,
geological surveys, replication is also very important due to the great spatial
variation in the environment.
·
When
studying chemical reactions, there is less variability, but replication is
needed to check the accuracy of results.
·
No. of
replicates = no of samples or sample size (n)
·
No. of
replicates depends upon:
o
Degree of
variability of the material or objects being studied
o
Precision
of the techniques used
o
Magnitude
of the difference you expect to find between groups
·
In past
experiments, replication was often insufficient. One experiment was conducted
to see if a certain medicine could cause chickens to recover from a particular
disease. The researcher reported that a chicken recovered from a disease when
treated with a certain drug, another chicken died when treated with a second
drug and no conclusions could be drawn about the third drug because that
chicken ran away.
o
This is
jumping to conclusions.
9.2.6 Factors
·
Independent
variables that are used as treatments in an experiment and which have several
levels
9.3 Types of Sampling
·
Types of
sampling and research design deal with the arrangement of samples in space and
time.
·
Random
sampling (Def’n) = the selection of individuals or units from a population
without bias, such that:
1.
All
individuals have an equal chance of selection
2.
The
selection of individuals is independent, i.e., the selection of one does not
affect the selection of others
·
All
statistical tests assume that samples are obtained randomly from a population
·
Random sampling
can be done by:
o
Drawing
numbers from a box
o
Using a
random number table
o
Using a
computer program
o
Using some
procedure such as throwing a quadrate without looking
·
For
observational science, that means collection of samples randomly
·
For
experimental science, that means allocating the test units or organisms to the
treatments at random.
·
Sampling
with replacement preferred
·
Sampling
without replacement must be done sometimes if the measurements require
destroying the test unit/organism, but it violates the requirement for
independence.
·
3 types of
random sampling
o
Simple
random sampling
§
Every
sample is selected completely randomly and independently
o
Systematic
random sampling
§
The first
sample is selected randomly, then all other samples are selected sequentially,
e.g., every 10 m, every 5 min, every 5th person, etc.
o
Stratified
random sampling
§
The study
area is divided into strata based on a pilot study (preliminary study)
§
Sampling
intensity in each stratum is proportional to the estimated density, etc. of the
items in the stratum
§
Within
each stratum, samples are selected randomly
§
This gives
the most accurate results if there are definite strata in the study area
9.4 Research Designs
For Different Types of Studies
9.4.1 Research design
for exploratory or formulative studies
·
Based on
inductive reasoning
o
i.e.,
making many separate observations and then making a conclusion, which is
usually tentative – a hypothesis.
·
The aim of
exploratory research is to come up with theories, insights or hypotheses that
can later be tested by hypothesis-testing research.
·
Thus, it
is like only the first 4 steps of the scientific method: making observations
about a problem, clearly defining the research problem, review the literature,
making tentative conclusions (hypotheses).
·
Research
design is very simple and does not involve independent and dependent variables,
etc.
·
Methodology
is simple and flexible; it involves making observations in various ways.
·
3 ways of
making observations to formulate conclusions (hypotheses)
o
review of
the literature
o
survey the
experiences of other people
o
analysis
of insight-stimulating examples
·
This type
of research is not so common in modern science.
9.4.2 Research design
in descriptive and diagnostic research
·
Simply
describes a phenomenon as it is, but does not test any hypotheses.
·
The design
involves:
o
Stating
the phenomenon to be studied
o
Designing
or choosing the techniques for data collection
o
Selecting
the samples
o
Collecting
the data
o
Analyzing
the data
o
Writing a
report describing the phenomenon
·
This type
of research is generally not very important by itself
o
It may
lead to information required to plan a full scientific study involving
hypothesis testing
9.4.3 Research design
in hypothesis-testing research
·
Involves
both inductive and deductive reasoning
·
Uses
inductive reasoning to come up with hypotheses
·
Uses
deductive reasoning to test the hypotheses and draw conclusions
·
Thus,
involves all steps of the scientific method
·
Deductive
reasoning
o
Start with
a general statement
o
This is
used to predict specific consequences, i.e., specific hypotheses
o
Then you
make observations to test the hypotheses and draw conclusions that are
definite.
9.4.3.1
Before-and-after without control design
|
|
Time period I
|
Treatment introduced
|
Time period II
|
|
Test group or area:
|
Level of phenomenon
before treatment (X)
|
 |
Level of phenomenon
after treatment (Y)
|
|
Treatment effect = Y
– X
|
|||
Limitations:
o
Extraneous
variables may also affect the phenomenon over time.
9.4.3.2 After-only
with control design
|
|
Treatment introduced
|
|
|
Test group or area:
|
 |
Level of phenomenon
after treatment (Y)
|
|
Control group or
area:
|
 |
Level of phenomenon
after experiment (without treatment) (Z)
|
|
Treatment effect = Y
– Z
|
||
Limitation:
o
Based on
the assumption that the test and control areas are identical
o
However,
often, extraneous variables may affect the two areas in different ways.
9.4.3.3
Before-and-after with control design
|
|
Time period I
|
Treatment introduced
|
Time period II
|
|
Test group or area:
|
Level of phenomenon
before treatment (X)
|
 |
Level of phenomenon
after treatment (Y)
|
|
Control group or
area:
|
Initial level of
phenomenon without treatment (A)
|
 |
Final level of
phenomenon without treatment (z)
|
|
Treatment effect =
(Y – X) – (Z – A)
|
|||
·
This
design is superior to the first two types of design.
·
It eliminates
the effect of extraneous variables, both due to time and space.
·
Takes more
time and more space to conduct the experiment.
9.4.3.4 Completely
randomized design
·
Concerns
the arrangement of replicates in space or the allocation of test units to
various treatments.
·
Test units
are allocated randomly to experimental treatments or plots.
·
Example: 4
fertilizer treatments (A, B, C, D) allocated randomly to 20 plots:
|
C
|
A
|
D
|
C
|
|
A
|
B
|
C
|
B
|
|
D
|
C
|
A
|
B
|
|
A
|
D
|
D
|
B
|
|
C
|
B
|
D
|
A
|
·
Suitable
if there is no extraneous variable affecting the plots in a definite pattern.
·
Advantage
is that it allows for different sample size for various groups
o
E.g.: 20
rats were randomly allocated to 4 different diets, but 6 died before the end of
the experiment the final weights of the surviving rats can still be analyzed.
|
Final weights of
rats (g)
|
|||
|
Diet A
|
Diet B
|
Diet C
|
Diet D
|
|
347
|
327
|
265
|
357
|
|
375
|
298
|
247
|
342
|
|
395
|
286
|
Died
|
321
|
|
327
|
Died
|
Died
|
302
|
|
387
|
Died
|
Died
|
Died
|
·
Can be
combined with either:
o
Before-and-after
without control design
o
Or,
after-only with control design
o
Or,
before-and-after with control design
9.4.3.5 Randomized
block design
·
Necessary
when there is a gradient of conditions across the test area.
·
Experimental
area is divided into blocks such that it can be assumed that the conditions
within each block are homogenous, even though conditions vary among blocks.
o
Thus it
eliminates the effect of extraneous variables in space.
·
In each
block, each treatment is represented once.
Gradient
in
moisture
·
Limitation:
o
requires
an equal number of replicates or cells for all treatments
o
If some
treatments die or are lost, either use a special formula to calculate missing
value or delete replicates until there is equal replication.
·
Each
“cell” can either have one replicate or several replicates
·
Can also
be used to eliminate the effect of time on the observations
o
E.g., If a
researcher wants to compare 4 sites and he cannot take measurements in all
sites at the same time, he can take 1 measurement in each site every month
o
This will
eliminate the effect of seasonal variation on the results.
o
Very
useful for marine biology or wildlife ecology.
Gradient
in
Time
·
Can also
be combined with either:
o
o
Before-and-after
without control design
o
Or,
after-only with control design
o
Or,
before-and-after with control design
9.4.3.6 Latin squares
design
·
Very
commonly used in agricultural research
·
Each
treatment must be represented once in each row and once in each column.
·
Like
randomized block design in both directions
Gradient in salinity
 |
Gradient
in
moisture Ocean
River
·
Advantage
is that it eliminates the effect of extraneous variables in two directions at the
same time
·
Limitation
o
There must
be equal columns and rows
o
Again, if
any cell is lost, the missing value has to be calculated or the one row and one
column have to be deleted.
9.4.3.7 Factorial
design
·
Used when
the research wants to test the effects of more than one factor (variable) on
the dependent variable
|
|
|
Factor B
|
|
|
|
|
Level I
|
Level II
|
|
Factor
A
|
Level I
|
Cell 1
|
Cell 3
|
|
Level II
|
Cell 2
|
Cell 4
|
|
·
Each cell
may have one observation or several observations
10.
Proposal
Writing
·
A well-formulated proposal can
form the basis for the first sections of the final project report, i.e.,
Introduction and Methods sections
o You will need only to add more detail.
10.1 Basic Research Proposal
·
For University 3rd year
project, M.Sc. or Ph.D.
Title
·
Should be written very
concisely
·
Should mention the main
independent and dependent variables
·
Mentions study area or study
sites
·
One should be able to have a
good impression of your main objective from reading the title
·
Avoid repetition of words
·
E.g. The effect of UV light on
a specific plastic
·
E.g. The effect of phosphate on
nitrogen fixation by blue-green algae
·
E.G. A comparison of marble
deposits in Mbeya and Dodoma
1.
INTRODUCTION
1.1 General Introduction
·
Gives very general background
information about the research topic
·
Explains the overall research
topic in broad terms
·
Background information about
the variables to be included in the study
1.2 Statement of the Research Problem
·
Describes the research topic in
more specific terms.
·
Clearly defines the research
problem to be investigated
·
Briefly states what is already
known about the topic
·
States what is yet to be
investigated.
·
Concludes by stating the aim or
general objective of the study, i.e., specifically what you are going to
investigate about the problem
1.3 Objectives
·
States the specific objectives
of the study
·
The
specific research questions are usually stated in the form of specific
objectives.
·
May be 2-6
objectives
·
Sometimes
you may have
o
An overall
general or development objective, and
o
2-6
specific objectives
·
Must
mention all the independent and dependent variables
·
E.g.,
Question: Does UV light affect the strength of a certain type of plastic?
·
Objective:
To determine the effect of ultraviolet light on the strength of a certain type
of plastic
·
There must
be (at least) 2 possible outcomes
·
Objectives
must be unambiguous.
·
Form the
focal point of the whole research plan
o
The
hypotheses are formulated based on the objectives.
o
The
objectives determine the data to be collected
o
Determine
what literature to review
·
Usually
formulate these first, before writing anything else.
1.4 Significance of the Study
·
Explain why the study is
important
·
What will be the benefits of
the study to society or to the environment
·
What further research will this
study lead to
1.5 Literature Review
·
Gives details about what is
already known about the overall research topic
·
Presents the findings of other
researchers that are relevant to your study.
·
Does not have to relate very
directly to your specific problem, but can broadly review the literature about
the general topic.
·
Can later be published as a
review paper
·
You must write these findings
in your own words,
·
Cite the author and year of
publication in parentheses [e.g. (Clark, 1991) or (Clark and Roberts, 2003) or
(Hanson et al., 1993)].
·
Should be divided in
subsections
·
Can be considerable flexibility
in the subsections
·
Subsections depend upon the
topic
·
E.g., UV light, Increase in UV
light through ozone depletion, the plastic being investigate (giving its
properties and any previous studies done on the effect of UV light on this
plastic
·
E.g. Properties of marble,
economic importance of marble in Tanzania, Distribution of marble deposits in
Tanzania (what is already known).
·
E.G. The BGA, Nitrogen
fixation, Effects of phosphate on nitrogen fixation
·
Study Area (not Study Site)
§ Only included if yours is a field study
§ The last section of the literature review
§ Describes the general area around the sites where the study will be
conducted
§ Location (may define the boundaries), climate, geography,
environment, geo-physical conditions, biological conditions
§ Map
§ E.G., if you are doing a study in the Dar es Salaam harbour, the
study area in the whole Dar es Salaam area.
1.6 Hypotheses
·
The hypotheses should be stated
concisely based on your objectives
·
Stated according to what you
predict, based on the literature review.
·
Normally, state what you expect
to conclude from your research, i.e., directional, alternative hypotheses.
·
Sometimes you may wish to state
them as null hypotheses, if it is difficult to predict the outcome.
·
Come
directly from the objectives
·
Two type
of hypotheses:
o
Difference
between groups, treatments
o
Relationship
between variables
2. MATERIALS AND METHODS
·
Study sites
o Precisely describes the location and conditions of your sites
o Map
·
Very precisely describes the
research design
·
Explains the techniques to be
used
·
If it is a standard technique
there is no need to explain it in detail
o Just cite reference
·
If it is a new technique,
describe it in detail
·
Often the Methods Section of
the proposal goes into the final report almost without change.
·
Someone should be able to read
your methods section and know how to repeat the same experiment
·
Data analysis
o States the exact analysis that you will do
o States the statistical test that you will apply to test each
hypothesis
LITERATURE CITED (OR REFERENCES)
All
literature cited in the text must be given in alphabetical order by
author(s). Following the authors' names,
the year of publication, title of the paper, title of the journal, volume
number and page numbers should be given. In the case of a book, the publisher
and city should be given as well as the total number of pages in the book.
Budget and financial arrangements
-
who your
sponsors are
Work plan (Timeframe)
10.2 Other Components
of Proposals
·
Justification
(Rationale for the study)
·
Beneficiaries
of the study
·
Collaboration
with other research projects
·
Expected
outputs
·
Indicators
o
State
indicators for measuring whether the outputs were attained
·
Research
team to be involved
11.
Data
Collection
11.1
Types of
data
Data = the actual
measurements or observations taken of the variable under study
Datum = a single
measurement or observation
·
Type of
data affects the statistical tests to be used for analysis
11.1.1
Data on a
ratio scale
o
2 main
characteristics
§
Constant
interval size between successive units
§
It has a
true zero point, which makes it possible to establish a ratio
11.1.2
Data on an
interval scale
o
2 main
characteristics
§
Constant
interval size between successive units
§
It has no
true zero point, which makes it impossible to establish a ratio, e.g.
Temperature on Celsius scale, 0 has no real meaning it is arbitrary
11.1.3
Ordinal
scale data
o
Data or
observations which can be put in order from lowest to highest, but which do not
have a constant interval between them, i.e., the data can be ranked
11.1.4
Nominal
scale data
o
Data in
which the variable is classified by some quality or attribute, i.e., the
variable cannot be measured, it can only be put in certain categories
11.2
Precision
and accuracy
·
Accuracy =
the closeness of a measurement to the true value of the variable being measured
·
Precision
= the closeness of repeated measurements of the same quantity
11.3 Sources of Error
in Data Collection
·
Object
being measured – it may be moving, e.g., bird.
·
Environmental
conditions – wind, waves
·
Instrument
– may be biased, not precise, not accurate
·
Researcher
(Observer error)
o
though
scientists try to record data very objectively, sometimes subjectivity may be
involved
o
e.g.,
endpoint in a chemical reaction
o
e.g.,
measuring to the closest 1 mm
12.
Analysis and Interpretation of Data
·
Analysis
and interpretation of data involves two aspects
- Descriptive Statistics = description of the distribution of
data and presentation of data in tables and graphs
- Inferential Statistics = analysis and
interpretation of data by hypothesis testing for the purpose of making
rational scientific conclusions about the phenomena being investigated.
·
Population
= entire collection of measurements of a variable under study, i.e., it
encompasses all the observations about which we wish to draw conclusions.
·
Sample = a
relatively small number of observations from the population being investigated,
i.e., a subset of the population.
- Usually it is impossible or impractical to
measure a variable for an entire population, so a sample is taken
- Sample size = the number of observations
in a sample.
- Inferential statistics allows us to draw
conclusions about an entire population based on data obtained from only a
sample of observations.
A RANDOM SAMPLE
OF SIX
OBSERVATIONS
(6 RANDOM
SAMPLES
|
)
 |
STUDY
SITE |
|||||
 |
|||||
 |
|||||
DRAW CONCLUSION
ABOUT THE ENTIRE
POPULATION
12.1 Descriptive
statistics
12.1.1 Description of
data in terms of statistical characteristics
·
2
important aspects
o
Measures
of central tendency
·
Identifies
the middle of the distribution of observations
·
E.g.,
mean, mode, median, midrange
o
Measures
of dispersion or variability
·
E.g.,
range, variance, standard deviation, standard error, coefficient of variation
(standard deviation relative to the mean), indices of diversity (e.g.,
Shannon-Weaver index of diversity)
12.1.2 Presentation of
data in tables and graphs
·
Usually
raw data are not presented in scientific reports, though they may be included
in the appendices.
·
Tables and
graphs usually present data in summarized form (ratio, interval and ordinal
scale data), particularly showing frequency distribution, means, standard
deviation (or standard error), and sample size.
·
Tables and
graphs showing frequency distribution of nominal scale data
o
(show
Excel file: Tables and graphs—Sheet 1.)
·
Tables and
graphs showing relationship between X and Y (XY graph)
o
(show
Excel file: Tables and graphs—Sheet 2.)
12.1.3 Inferential
Statistics
12.1.3.1 Parametric
versus nonparametric statistical methods
Parametric Statistical
methods:
·
Involve
the estimation of population parameters, e.g., mean, variance, etc.
·
Have
certain underlying assumptions about the populations being tested, e.g.
o
Randomness:
Samples must be randomly selected
o
Normality:
The data must be normally distributed
o
Homogeneity
of variances: when comparing 2 or more samples, all samples must have equal
variances
·
Capabilities/requirements
o
more
powerful than nonparametric tests
o
suitable
for ratio and interval scale data
o
not very
suitable for ordinal scale data
o
cannot be
used for nominal scale data
o
Where the
data fit the assumptions required for parametric methods, these methods are
more powerful than nonparametric methods (usually by 10%), i.e., there is less
chance of making a Type II error
·
Examples:
t test, analysis of variance and regression
Nonparametric
statistical methods
·
Do not use
estimates of population parameters in their calculations and make no hypotheses
about parameters
·
Do not
make any assumptions about the nature of the distribution of the populations
being investigated; sometimes called distribution-free tests
·
Only
assumption or requirement is that the samples must be selected randomly
·
Capabilities
o
can be
used with all types of data (nominal, ordinal, interval and ratio scale data)
·
No
assumptions, except randomness
·
Examples
of nonparametric tests: Chi-square test, Spearman rank correlation,
Mann-Whitney U test
·
Example of
parametric test: two-sample t test
12.1.3.2 Selection of
statistical tests
·
Best to
take a course on statistical analysis where one learns how to calculate the
stats tests – to get an appreciation and understanding of stats.
·
However,
these days scientists just feed data into a computer and get the answer, but
they have to know which is the right test to use for a given problem.
Distinction between
variable and factor
·
Sometimes
the terms “variable” and “factor” are used almost interchangeably, but
sometimes they are used differently, depending upon the context.
·
Variables
– vary together, i.e., as one variable varies, the other varies with it.
·
Factor –
one “variable” is allocated to different treatments or groups and is thus
usually called a “factor”
·
Distinction
between variable and factor depends upon the hypotheses
Hypotheses testing for
difference between groups or treatments (called different treatments of a
factor)
Example: Ho: No
significant difference in growth rate of algae at different temperatures
|
20oC
|
25oC
|
30oC
|
35oC
|
|
7 g/day
|
12
|
15
|
18
|
|
9
|
9
|
16
|
17
|
|
7
|
10
|
13
|
17
|
|
8
|
11
|
13
|
19
|
·
Here, T is
considered as a single factor/ variable, which is independent
·
Growth
rate is the dependent variable
·
We are
testing the hypotheses about difference among groups
·
Analysis
o
If data
fit the assumptions of parametric tests: Single Factor Analysis of Variance
(ANOVA)
o
If data do
not fit the assumptions of parametric tests: Kruskal-Wallis test
Hypotheses testing for
relationship between variables
Ho: There is no
significant relationship between temperature and the growth rate of alga
|
Temperature (x) (C)
|
Growth rate of algae
(g/day)
|
|
20
|
7
|
|
20
|
9
|
|
20
|
7
|
|
20
|
8
|
|
25
|
12
|
|
25
|
9
|
|
25
|
10
|
|
25
|
11
|
|
30
|
15
|
|
30
|
16
|
|
30
|
13
|
|
30
|
13
|
|
35
|
18
|
|
35
|
17
|
|
35
|
17
|
|
35
|
19
|
·
Analysis
o
If data
fit the assumptions of parametric tests: Correlation, Regression analysis
o
If data do
not fit the assumptions of parametric tests: Spearman rank correlation
p = parametric
n = nonparametric
13.
Writing
Scientific Reports
Different steps in writing a scientific report
1. Generally,scientific reports are the product of slow,careful,
accurate inductive work. The usual steps involved in writing these reports are:
a) Logical analysis of the subject matter
b) Preparation of the final outline
c) Preparation of the rough draft
d) Re-writing and polishing
e) Preparation of the bibliography
f) Writing the final draft
A) Logical analysis of the subject matter
·
It is the
first step which is primarily concerned with the development of the subject
B) Preparation of
the final outline
·
an aid to
logical organization of the material and a reminder of the points of the report
C) Preparation of
the rough draft
·
Very
important
·
write down
the procedures adopted in collecting the material for the study along with various
limitations faced , the technique of analysis adopted , the broad findings and
generalizations and the various suggestions one wants to offer regarding the
problem concerned.
C)Preparation of the
rough draft
D) re-writing and
polishing of the rough draft
·
Time
intensive
·
Cohesion,clarity
, flow and conscisivenes are all checked
E) Preparation of
the final bibliography
·
List of
books in some way pertinent to the research
·
It should
contain all those works which the researcher has consulted
F) Writing the final
draft
·
This
constitutes the last step and should therefore be concise without expressions
like “it seems”,”there may be” and the like
·
Avoid
abstract terminology and technical jargon.
·
Use
diagrams,illustrations, and examples
·
Should be
interesting and show originality
NB: It must be
remembered that every report should be an attempt to solve some intellectual
problem and must continue to the solution of a problem and must add to the
knowledge of both the researcher and the reader
LAYOUT OF THE RESEARCH REPORT
1.
Necessarily
be conveyed enough about the study so that he can place it in its general
scientific context
2.
Judge the
adequacy of its methods thus form an opinion of how seriously.....(blah blah)
13.1
Format of
a Scientific Report
·
The first
sections of a scientific report are usually the same as those of the proposal
Title page
·
Gives a concise title of the
study, the names of the authors, the organization or publishing company
responsible for publishing or producing the report and the year.
·
If the report is for a certain
course at the University, you should put the course code number and your
registration number
·
If it is a third year project,
you should write at the bottom of the page: “A Third Year Project Report
Submitted in Partial Fulfilment of the Requirements for the Degree of Bachelor
of Science at the University of Dar es Salaam”.
·
If it is for a post-graduate
thesis or dissertation, write: ….
Abstract
·
Must be written very concisely,
without wasting extra words
·
State what you did (methodology),
the main results, the important conclusions and any significant
recommendations.
·
Although this is the first part
presented in the report, it is usually the last part written, because it must
synthesize the entire report.
·
The abstract should not exceed
the required number of pages, according to the regulations given by the
particular institution, journal or organization, e.g., 300 words.
Declaration of Copyright
Table of Contents
List of Tables
List of Figures (including maps)
List of Photographs (if any)
List of Appendices (if any)
Dedication
(optional)
Acknowledgements
·
Gives the names of the people
who helped you in any way, briefly mentions the role they played and the
institutions from which they come.
[NOTE ON PAGINATION]
·
Paginate the preceding pages in
lower case Roman numerals (beginning with the title page, though the number is
not shown on the title page)
·
The remaining pages (starting
with the introduction) should be numbered in Arabic numerals.
·
The
following sections are nearly the same as found in the proposal; however, more
detail is added and some aspects may be changed according to experiences gained
while conducting the research.
1.
INTRODUCTION
1.1 General Introduction
1.2 Statement of the Research Problem
1.3 Objectives
1.4 Significance of the Study
1.5 Literature Review
Study Area (not Study Site)
§ The last section of the literature review
§ Sometimes it may be a separate section
1.6 Hypotheses
2. MATERIALS AND METHODS
3. RESULTS
·
Usually divided into
subsections which are identical to, or similar to, the subsections of the
materials and methods section.
·
In this section, you describe
your own results without any mention of the findings of others.
·
Data should be presented in
summarized form in tables or illustrated in graphs. Means and standard deviations should be
shown.
·
Every
graph must be given a number
·
Every
graph that is shown must be cited in the text in parentheses, using the
appropriate number.
·
The text
must comment on each table and graph describing the main trends shown in them.
·
The main trends shown by the
tables and figures should be briefly described.
·
Conclusions inferred from
statistical tests should be stated, giving the test statistic, the probability
of committing a Type I error, and the sample size (or degrees of freedom). If
many statistical tests were performed, it may be best to present the statistics
in tables.
o Example of statistics as part of the text:
“There was a significant difference in
blood pressure between the experimental group, who were given Vitamin E, and
the control group, who were not given the vitamin (two-sample t test: t =
2.386, DF = 22, p = 0.0261).
o Example of many results given in a table:
Table 3.1: Results of the two-sample t test
for the difference in water temperature between cleared and forested areas in
Sites A, B, C and D.
|
|
T
|
DF
|
p
|
Significance
|
|
SITE A
|
7.99
|
22
|
<0.0001
|
***
|
|
SITE B
|
4.827
|
22
|
<0.0001
|
***
|
|
SITE C
|
2.533
|
22
|
0.0189
|
*
|
|
SITE D
|
5.575
|
22
|
<0.0001
|
***
|
4. DISCUSSION
·
Focus on discussing your
results
o in relation to your hypotheses and
o in comparison with the findings of others.
·
Try to give scientific
explanations as to why your findings are the same as, or are different from,
the findings of others and what you hypothesized.
·
Do not repeat the literature
review here; however; simply mention the findings of others (giving the author
and year of publication) in order to compare your results.
·
Do not repeat the results
section, but simple refer to certain results and discuss the reasons for those
findings
·
Do not be alarmed if your
findings are completely different from what you expected. However, try to give reasonable explanations
for the deviations of your results from what was hypothesized.
5. CONCLUSIONS
·
Concisely state the important,
overall conclusions drawn from your study.
·
These conclusions may be
similar to your hypotheses (if they were proven true) or the opposite (if your
findings were to the contrary).
6. RECOMMENDATIONS
·
Give recommendations regarding
possible application of the findings of your study, e.g., if the study was on
environmental issues, recommendations regarding conservations strategies or
resource use management.
·
Give recommendations for
further study or aspects that need to be investigated.
LITERATURE
CITED (OR REFERENCES)
All
literature cited in the text must be given in alphabetical order by
author(s). Following the authors' names,
the year of publication, title of the paper, title of the journal, volume and
page numbers should be given. In the case of a book, the publisher and city
should be given as well as the total number of pages in the book.
APPENDICES
·
May include details of specific
techniques used in data collection, e.g., soil particle size analysis,
spectrophotometry, gas chromatography, methods of chemical analysis, etc.
·
May give formulae or special
statistical procedures used.
·
May give raw data or semi-raw
data.
EXAMPLES OF APPROPRIATE
PLACEMENT OF TEXT
·
“The fact that there was …
13.2 Scientific Style of Writing
13.2.1 Orderly presentation of ideas
·
Must be a logical sequence or
systematic order in the arrangements of words, phrases, sentences, paragraphs
and sections of the report
·
There must be continuity in the
flow of the arguments, i.e., no gap in the sequence of logic
·
Correct placement of
punctuation marks helps readers to understand the flow of the argument, e.g.,
use of commas
·
Use of transition words:
o Cause-and-effect links: therefore, consequently, subsequently, as a
result
o Strengthening your argument or discussion: in addition, moreover,
furthermore, similarly
o Contrast links: conversely, nevertheless, however, whereas
13.2.2 Smoothness of Expression
·
Distinction between scientific
writing and creative or literary writing
o In literary writing, the author tries to create ambiguity or insert
unexpected ideas in order to arouse attention or interest
o In scientific writing, this would only create confusion.
·
Because you are so familiar
with the work you have done, you may omit some aspects thinking that it is
understood.
o One solution is to put the manuscript aside for awhile and re-read
it later so that you can notice omissions and irrelevancies
o Another solution is to give your manuscript to someone else to read.
·
Be careful not to change verb
tenses
o Literature review: use either past tense [Smith (1994) showed that]
or present perfect tense [Researchers have shown]
o Methodology: use past tense (future tense for proposal)
o Results: use past tense
o Discussion: use present tense [the results of experiment 2 indicate
that…]
o Conclusions: again use present tense
o Recommendations: use present tense [It is recommended that…] or
conditional tense [Steps should be taken to mitigate negative impacts on the
environment such as…]
13.2.3
Economy of Expression
·
Write
only what is necessary or important
·
Particularly
for papers to be submitted to a journal for publication—they will not publish
something that is too wordy.
·
Avoid
overly detailed descriptions of apparatus and procedures.
o
If
it is a new technique, the details can be put in the appendices
o
If
it is a well-known, standard technique just give the name of the technique and
a reference or describe it in one sentence.
·
Wordiness:
avoid unnecessary words or phrases
o
If
you revise your report several times, you can usually cut out a few words each
time.
·
Redundancy:
absolutely essential, in close proximity to, completely unanimous
·
Sentence
length and paragraph length: avoid a chain or very short sentences or a chain
of very long sentences—varying sentence length keeps the interest of the reader
o
The
same applies for paragraph length
13.3 Symbols and Abbreviations
13.3.1 SI Units
·
SI = Systeme Internationale
d’Unites (International System of Units)
·
Based on the metric system
·
Only SI Units should be used in
scientific work
·
Basic SI units
|
Quantity
|
Name of Unit
|
Unit Symbol
|
|
Length
|
Metre
|
M
|
|
Mass
|
Kilogram
|
Kg
|
|
Time
|
Second
|
S
|
|
Electric current
|
Ampere
|
A
|
|
Temperature
|
Kelvin
|
K
|
|
Luminous intensity
|
Candela
|
Cd
|
|
Amount of substance
|
Mole
|
Mol
|
·
Multiples and submultiples of
the basic units
|
1012
|
Tera
|
T
|
|
109
|
Giga
|
G
|
|
106
|
Mega
|
M
|
|
103
|
Kilo
|
K
|
|
102
|
Hecto
|
H
|
|
101
|
Deca
|
Da
|
|
10-1
|
Deci
|
D
|
|
10-2
|
Centi
|
C
|
|
10-3
|
Milli
|
M
|
|
10-6
|
Micro
|
µ
|
|
10-9
|
Nano
|
N
|
|
10-12
|
Pico
|
P
|
|
10-15
|
Femto
|
F
|
|
10-18
|
Atto
|
A
|
·
Derived SI Units with special
names (a few)
|
Physical Quantity
|
Name of Unit
|
Symbol
|
Definition of Unit
|
|
Energy
|
Joule
|
J
|
kg m2 s-2
|
|
Force
|
Newton
|
N
|
kg m s-2
|
|
Power
|
Watt
|
W
|
kg m2 s-3
|
|
Frequency
|
Hertz
|
Hz
|
s-1
|
·
Other derived SI units (a few)
|
Physical Quantity
|
SI Unit
|
Symbol
|
|
Area
|
Square metre
|
m2
|
|
Volume
|
Cubic metre
|
m3
|
|
Velocity
|
Metre per second
|
m s-1
|
|
|
|
|
·
Named units which are
sub-multiples and multiples of SI units
|
Physical Quantity
|
Name
|
Symbol
|
Definition
|
|
Area
|
Hectare
|
Ha
|
104 m2
|
|
Mass
|
Ton
|
T
|
103 kg
|
|
|
|
|
|
·
Do not put a period after
symbols
·
An oblique stroke or negative
index can be used
o E.g., m/s or m s-1 (notice the space between m and s)
·
Other units
o Minutes: min
o Hour: h
o Degrees: 45o
·
Do not abbreviate the
following, unless the particular journal gives instructions:
o Day, week, month, year
13.3.2 Abbreviations
·
Use periods, unlike symbols for
units
·
E.g.,
·
Ed. Or eds.
·
Anon.
·
Cf. = compare
·
Fig. Or figs.
·
Et al. = and others
·
Etc.
·
i.e. = that is
·
Viz. = namely
·
Vs. = versus
13.4 Steps In Writing a Scientific Report
Logical analysis of the subject matter
·
Chronologically
·
logically
Preparation of the outline (table of
contents)
Preparation of the rough draft
·
abstract last
Rewriting and polishing of the rough draft
·
Informal review
·
Re-read later
·
Revise one or two aspects at a
time
o Content/logic
o Grammar
o Spelling
o Consistency in stats, tables and graphs
o Citation of literature
Final draft for submission
Review (formal)/external examiner
Viva voce = oral presentation
Final revision for final submission
Proofs
14. Types of Scientific Publications
14.1 Technical Report
·
Very detailed, not very concise
·
Gives all details of the
methods, results, all data may be included in the appendices
·
Assessments of the environment
14.2 Research
paper/report
·
Original
data
·
Should be
concisely written
·
Substantial
contribution to science
·
Usually
reviewed by referees (Reviewers)
·
Published
in a journal
14.3 Short
communication/ Synopsis
·
Original
data
·
Minor
experiment or study
·
Contribution
to new knowledge on one minor aspect
·
Usually
reviewed by referees
·
Published
in a journal
14.4 Thesis
·
For M.Sc.
or Ph.D.
14.5 Dissertation
·
For M.Sc.
thesis by coursework and research
14.6 Review paper
·
Reviews
the studies of others
·
Usually on
a specific topic
·
No
original data or findings
·
May give
new insights/ concepts/ theories, based on the findings of many others
·
Based on
secondary data
·
If it is
published in a journal, it is usually reviewed by referees
14.7 Conference
proceedings
·
May be
published as a book
·
May or may
not be reviewed by referees
14.8 Book
·
Systematic
collection of information on a broad topic
·
Reviews a
very large number of other studies
·
Usually no
original data
·
May give
new insights/ concepts/ theories, based on the findings of many others
14.9 Book review
·
Reviews, criticises, comments
on a book
Miscellaneous
·
Annotated
bibliography
