2/25/2022

EnE 202
RESEARCH METHODS IN
ENVIRONMENTAL ENGINEERING

MODULE 2
Environmental Research Design and Statistics

General Types of Research (MULTI-DISCIPLINARY)

Environmental studies may require doing all types of research→ depending on the problem

Descriptive —
survey, historical, content analysis,
qualitative/ethnography

Associational —
correlational, causal-comparative

Interventional —
experimental, quasi-experimental,
action research (sort of)

Del Siegle, 2017

24/02/2022 3

1
 2/25/2022

PREDISPOSITIONS OF QUANTITATIVE & QUALITATIVE MODES OF INQUIRY

e.g.,The significance
of Ifugao Mumbaki
(leader)

Mumbakis & rituals

of the Ifugaos serve
as social control &
agricultural calendars

24/02/2022 https://researchbasics.education.uconn.edu/qualitative_research_paradigm/ 4

TWO MAJOR TYPES OF STUDIES:

• Observational studies (social surveys, time-series field surveys)
- data are collected on individuals without affecting them
(e.g., surveys– questionnaires, polls, phone-in, field observations)
- improperly done surveys result in bogus information
(choice of words/questions, no response, sampling method)
To establish cause → effect
• Experiments (laboratory experiments, “social experiments”)
- data are collected thru (actions/tests) on participants
(e.g., ingestion of drugs, catalysts, reagents, etc.)
- a control set is needed for comparison & conclusion
- researcher bias results in biased results (GIGO)

24/02/2022 5

2
 2/25/2022

IN THE TWO MAJOR TYPES OF STUDIES:

• Both Observational studies & Experiments –
- no matter what the type of study, it has to be designed so that the
research question(s) can be answered in a credible way
- after design of study, subjects are chosen and data are ready to be collected
You want to EXPLAIN /UNDERSTAND a phenomenon;
because understanding a problem leads to its solution(s).
IMPROPER UNDERSTANDING will not solve the problem.
• How to design a study? START WITH A QUESTION.
- “Clear research objectives, framed as questions or hypotheses, is the first
and most critical step in the design and execution of.. research.”
May even aggravate the problem.

(Gatrell, Bierly, and Jensen, 2012 p.25)

24/02/2022 6

RESEARCH DESIGN
“Research design is the critical process that transforms an idea,
interest, or question from “just a thought” into a meaningful (helpful /useful)
and purposeful investigation of social or physical phenomena.”
Clear objective(s)
(Gatrell, Bierly, and Jensen 2012 p.25)

A “research design is a plan for collecting evidence that can be used

to answer a research question.” (Vogt, 2006, p. 8)

- Not the end result, but the process

- Iterative process of negotiation between the researcher,
peers, and the standards of good science.
(as emerging conclusions /theories are continuously subjected to scrutiny)
The beauty of SCIENTIFIC METHOD and advancement of SCIENTIFIC REVOLUTION
(Khun 1962)

24/02/2022 7

3
 2/25/2022

EXPERIMENTATION
“The principle of science, the definition almost, is the following:
The test of all knowledge is experiment. Experiment is the sole
judge of scientific truth… Experiment itself helps to produce the
laws, in the sense that it gives us hints. But also needed is
imagination to create from these hints the great generalizations
– to guess at the wonderful, simple, but very strange patterns
beneath them all, and then to experiment again to check
whether we have made the right guess.”
-- Richard Feynman (1995)
24/02/2022 8

Five Classes of Experimental Problems

1. Determining a subset of important variables from a given
larger set of potentially important variables (Variable Screening)
(Dependent/independent Variables)
2. Determining empirical “effects” of known input variables X
(Empirical Model Building)
3. Determining a local interpolation or approximation function
4. Determining a function based on mechanistic understanding
of the system (Mechanistic Model Building– Just knowing the parts that work)
5. Determining values for the parameters- (Model Fitting; a.k.a. reverse engineering)

24/02/2022 9

4
 2/25/2022

Experimental Design Approaches

Screening Determine a subset of important variables from a given larger set of
variables potentially important variables (Depending on the problem,
(Selecting the most significant which variable are important)
among a multitude of variables)

Empirical model Determine empirical “effects” of known input variables X (Dependent / Independent
Variables)
building
(From direct observation / Determine a local interpolation or approximation function (Linear / Bilinear / Cubic /
or Circular Interpolations)
measurements, predict overall behavior)

Mechanistic Determine a function based on mechanistic understanding of the

model building system e.g., A chemical engineer, can create a mechanistic model of a process based on
(Based on understanding the understanding of components - distilling columns, reactor chambers and particle filters.
component behavior of a system)

Model fitting Determine values for the parameters & the model creating the data
(find the model that is most likely (a.k.a. “reverse engineering”)
to have generated the data)

A mechanistic model predicts the future based on a theory.

e.g., a manufacturer designs a new plane based on what theory predicts will fly well.

24/02/2022 10

10

Understanding data & inferring relationships among them using Statistics

Variable Screening

What are the two BEFORE WE DO THAT…

types of
THINK OF THE TOPIC.
“Dictatorship of the problem.”

experiments? (Vogt, 2008)

Modelling Experiments
Understanding processes & modelling them using equations
11

5
 2/25/2022

Types of Experiments
There are many different ways to describe the types of scientific experiments.

In terms of designing experiments:

• Variable screening – analyzing the involved data + statistics to describe
• Modelling experiments – observation + using equations to mimic nature
In terms of the venue:
• Laboratory experiment
• Field experiment
• Thought experiment
In terms of empirical methods employed:
• Experimental
• Observational
• Quasi-experimental (observation + experimenting + action)

12

THE RESEARCH QUESTION May be your discipline, hobby, current problem.

- the essential element of inquiry; RESEARCH

INTEREST
it distills the investigator’s broader
curiosity into a coherent, potentially
RESEARCH
“answerable” thought QUESTION

QUALITIES OF A GOOD RESEARCH QUESTION:

1. Must be answerable (yes, no or maybe) HYPOTHESIS
within the constraints of available data (&time).
2. Must be consistent with the existing
theoretical paradigms (e.g., 2nd Law of Thermodynamics) GENERAL STATEMENT
ON THE
3. Must demonstrate an objective analysis ENVIRONMENTAL
of the findings (i.e., unbiased) PROCESS
4. Must honestly state all assumptions made
24/02/2022 (Gatrell, Bierly, and Jensen 2012 p.31) 13

13

6
 2/25/2022

FROM GENERAL TO SPECIFIC

READ, READ, READ.
• Chronological literature
review and narrowing
concept (FOCUS)
• Finding the gaps &
thinking of modifications
• Aligning your contribution
YOUR
to the body of knowledge
24/02/2022 (Gatrell, Bierly, and Jensen 2012 p.14) 14

14

ITERATIVE PROCESS of RRL & RQ (research question):

RESEARCH
INTEREST

RESEARCH
QUESTION

HYPOTHESIS

GENERAL STATEMENT
ON THE
ENVIRONMENTAL
PROCESS

24/02/2022 (Gatrell, Bierly, and Jensen 2012 p.14) 15

15

7
 2/25/2022

Inter-disciplinary Research Framework

(especially for Environmental research)

• 1st, the research team discusses the conceptual design of their

study which contains the ‘what’ and ‘why’ of the research.
(research problem/question & rationale)

• 2nd, the team discusses the technical design of the study

which contains the ‘how’ of the research.

• Only after the team agrees that the complete research design
is sufficiently crystalized, will the execution of the work
(including fieldwork) starts.
24/02/2022 16

16

How to develop The Research Topic:

- the problem you want to tackle (to help solve), and your motivation(s) for doing so
(may be based on your expertise, your hobby, or today’s pressing problem, etc.)
- your research will help to understand better the problem; possibly, by citing other cases
that helped (do mini-RRL); also how your tools (e.g., GIS/RS, ANN, AI, STAT, etc) can help
→ this becomes your thesis statement
- say something about the expected output /how it will help solve the stated problem
Over-arching problem PROBLEM CONTEXT e.g., MMla TRAFFIC
• Many /complex causes
• High population-more cars
(technical, economic, social..)
• Choose a focus that can be assisted by
Problem Causes • Bad roads (infra)
• Traffic routes
your tools (e.g., RS/GIS, AI, etc) • Driver behavior
• Do review of related literature (RRL) • Etc…
to better understand the topic
• Focus on knowledge gaps YOUR TOOL /EXPERTISE
• Profession
Where your research can help Choose Research Focus • Experiences, hobbies

REVIEW RELATED LITERATURE

RRL (RRL)

17

8
 2/25/2022

Top 10 Environmental Engineering Research Topics

https://guides.libraries.psu.edu/HBGce100/envi

• Climate Change. One thing is certain: We'll always have an environment. ...
• Fire Ecology. Rising temperatures also means more wildfires. ...
• Renewable Energy. Alternative sources to power up, sustain life...
• Urban Ecology. Rising urban sprawl...
• Remote Sensing and Geographic Information Systems. Everything has location...
• Bioremediation. Green solutions...
• Noise Pollution. All pollution...
• Limnology. Aquatic ecosystem…

ALL TO IMPROVE THE QUALITY OF LIFE.

18

DECIDING ON A RESEARCH TOPIC

1. Brainstorm for ideas: General problems in the environment /society
2. Choose a topic that you are interested in /familiar with, and will make
you understand the literature. (e.g., Health care delivery)
3. Narrow down to the use of spatial information /technologies (GE
discipline) to help improve the present situation (e.g., Problems in
accessing health care facilities in the countryside)
4. Make a list of key words associated with the topic & READ about it
for one sem.
5. Ensure that the topic is manageable and that material is available DATA
(on existing location of Health Centers, road maps & census data) & EQUIPMENT (GPS, software, transport)
6. Be flexible (with items 4&5– tweak/REVISE depending on available
materials)
7. FORMULATE Research Objective: To do X using Y in order to Z.
e.g., To combine socio-economic data with geospatial analysis using GIS in order to model accessibility to
Primary Health Care

Modified from-- https://www.umflint.edu/library/how-select-research-topic

19

9
 2/25/2022

DECIDING ON A RESEARCH TOPIC

7. Or FORMULATE Research Objective: To do X using Y in order to Z.
e.g., To combine socio-economic data with geospatial analysis using GIS in order to
model accessibility to Primary Health Care (and propose additional sites for Health
Centers)
8. Define your topic as a focused research question
e.g., How can socio-economic data be combined with geospatial analysis to provide a
better evaluation of access to Primary Health Care?
“State-of-the-art”
9. Research and READ MORE about your topic (RRL - for current status, current methods,
problems with existing methods– how things are currently being done)

10. Formulate a thesis statement (WHAT YOU WANT TO PROVE)

e.g., Access to health care facilities needs to combine analysis of infrastructure and
physical location with socio-economic indicators of the population in order to provide
better health care delivery.
Modified from-- https://www.umflint.edu/library/how-select-research-topic

20

DECIDING ON A RESEARCH TOPIC

Example:
1. General idea /problem: Deforestation→ Monitoring forest reserves
2. With GE discipline /Geomatics Technologies: Using satellite images
3. Read initial RRL: Different image resolutions available, image enhancement
4. Research Problem: On image fusion, On vs. Pixel-based
object-based image analysis,
Image enhancement (from Low-Resolution to High-Resolution)
5. Current approach &problems: Interpolation of imagery, Spatial Statistics Cite DISADVANTAGES
6. New approach: Using Artificial Neural Network (ANN) & Deep Learning in Image
analysis (cite ADVANTAGES)
7. Research Question: How can ANN & Deep Learning be combined in image analysis
techniques to generate High-resolution imagery from Low-resolution imagery?
8. Thesis Statement:
Combining ANN & Deep Learning techniques in image analysis can generate a
better enhanced image for monitoring forests. (to be compared with current
interpolation methods)

21

10
 2/25/2022

What is a review of related literature (RRL)?

RRL is an evaluative report of information found in the literature related to your selected area of study.
The review should describe, summarize, evaluate and clarify this literature.

• an account of what has been published on a topic by authoritative scholars

and researchers

• the purpose is to convey what knowledge and ideas have been established on
a topic, and what their strengths and weaknesses are

• must be defined by a guiding concept – e.g., your research objective or the

problem or issue you are discussing

• RRL is not just a descriptive list of the available materials, or a set summaries;
RRL is an evaluation of all the summaries.

http://advice.writing.utoronto.ca/types-of-writing/literature-review

22

To determine the:
WHAT IS RRL? REVIEW OF RELATED LITERATURE → “State-of-the-Art”

“State-of-the-Art”

RRL is the researcher’s basis for the research strategy (methodology)→ influence of RRL on the current r

23

11
 2/25/2022

MAPPING TREE SPECIES RICHNESS OF TROPICAL FOREST USING AIRBORNE HYPERSPECTRAL (Badola, Anushree, 2019)
REMOTE SENSING
SAMPLE REVIEW OF LITERATURE
(Random Forest)
cite reference
^

(Rotation Random Forest)

(Justification of the research based on RRL)

24

RRL must:
(“state-of-the-art”)
1. be organized around and related directly to the thesis or research
question you are developing
(“knowledge gaps”)
2. synthesize results into a summary of what is and is not known
(different schools of thought)
3. identify areas of controversy in the literature
(basis of your own research)
4. formulate questions that need further research

RRL will help you in:

1. Information seeking: the ability to scan and identify useful literature efficiently (manual or computerized methods);
2. Critical appraisal: the ability to apply principles of analysis to identify unbiased and valid studies.

http://advice.writing.utoronto.ca/types-of-writing/literature-review/

25

12
 2/25/2022

3 Strategies in doing RRL:

(literature related to your research– discussing tools and strategies to be used)

1. use paraphrasing (“in other words..”) –

- Putting an author’s ideas in your own words (rewording)
- Cite the author’s work & year of publication in parenthesis
- example (Lilesand and Keifer, 2007)

2. use precis (summarizing a paragraph) –

- Condensing an author’s ideas to a more succinct statements WITHOUT USING
the original author’s words
- Cite the author’s work & year of publication in parenthesis
- example (McDougal, 2015) (shorter than paraphrase)

3. use direct quotes (“ ”) – (do not abuse direct quotes)

- Using an author’s language word for word (verbatim)
- Use quotation marks and cite the author’s work & page number in parenthesis
- example (Santos and Manansala, 2009:45)
Page No.

26

PARAPHRASING:
(In your own words)
Examples:

Original: Giraffes like Acacia leaves and hay, and they can consume 75 pounds of food a day.
Paraphrased: A giraffe can eat up to 75 pounds of Acacia leaves and hay everyday.

Original: The ceiling of the Sistine Chapel was painted by Michelangelo.

Paraphrased: Michelangelo painted the Sistine Chapel’s ceiling.

Original: Pouring hydroxide peroxide on your wooden cutting board will kill salmonella and other bacteria.
Paraphrased: Bacteria, including salmonella, can be eradicated from your wooden cutting board by pouring
hydrogen peroxide on it.

Original: Ginger shuts down nerve receptors that trigger the vomiting reflex.
Paraphrased: The nerve receptors that cause the reflex for vomiting are stopped by ginger.

27

13
 2/25/2022

PRECIS WRITING:
Example:
It is physically impossible for a well-educated, intellectual, or brave man to make money the chief object of
his thoughts just as it is for him to make his dinner the principal object of them. All healthy people like their
dinners, but their dinner is not the main object of their lives. So all healthy minded people like making money
ought to like it and enjoy the sensation of gaining it; but the gain is something better than money.
The clergyman's object is essentially to baptize and preach, not to be paid for preaching. So of doctors.
They like fees no doubt; yet if they are brave and well-educated, the entire object to their lives is not fees.
They on the whole, desire to cure the sick; and if they are good doctors and the choice were fairly to them,
would rather cure their patient and lose their fee than kill him and get it. And so with all the other brave and
rightly trained men: their work is first, their fee second—very important always; but still second.

1. Write the Main Points:

- Money-making is a common attraction in life.
- But it cannot be the principal aim of well-educated, intellectual, brave persons.

2. Make the Precis Summary:

Money-making is a common attraction in life. But it cannot be the principal aim of well educated, cultured and brave person.
A good clergyman is more interested in the moral welfare of his people than his returns. A good doctor values the care of his
patient far more than his fees. Thus with all the well-educated, intellectual persons, their work is first, money next.

http://englishwithiftikhar.blogspot.com/p/blog-page_26.html

28

PARAPHRASE vs PRECIS
• Paraphrase – a summary -- all in your own words
• Precis – shorter, may use the original author’s keywords (in quotes)
• BOTH are summarizing the original work; we are not going to be
technical about it… just proceed with caution about citing references
to avoid plagiarism.

• There is a tool to check plagiarism – TURNITIN® software license

available in all UP Libraries– you can send to the Engglibrarian,
(email) and they can check for you. Automatically comparing
similarities to a database of references– greater than 25% similarity
will be flagged.

29

14
 2/25/2022

FOR MORE EXERCISES:

https://www.scribd.com/document/219842908/paraphrasing-exercises-with-answers
30

Those 3 Strategies for RRL are used to avoid cheating:

1. Cheating the original author – plagiarism

can lead to lawsuits
2. Cheating the public – loss of reputation
3. Cheating yourself – poor writing ability
“Honesty is the best policy.”
CAUTION:
If wording of the paraphrase is too close to that of the original content, then it is plagiarism.
The main ideas need to come through, but the wording has to be your own.
Use a summary that is much shorter than the original and is an overview of the main points.

31

15
 2/25/2022

32

33

16
 2/25/2022

PHOTO CONTEST WINNER

- The real owner saw the

published winners and
recognized his photo.

- The organizers were

alerted.

34

PLAGIARIZING ILLUSTRATIONS

ORIGINAL WORK COPY

35

17
 2/25/2022

RRL EXERCISE#1: Paraphrase & cite the reference

In order to understand complex, dynamical societies, we need appropriate data

to base analyses on. Unfortunately, acquiring such data is very hard.
The traditional methods of analysis in sociology have been to gather
qualitative data from interviews, observation or from documents and records,
and to carry out surveys of samples of people. While qualitative data can
illustrate very effectively the emergence of institutions from individual action,
because of the nature of the data most analyses inevitably remain
somewhat impressionistic.

Agent-based social simulation: dealing with complexity

By Nigel Gilbert
Published by the University of Surrey, 2004. page 14

36

RRL EXERCISE#2: Use precis writing in this example

When we survey our lives and efforts we soon observe that almost the whole of
our actions and desires are bound up with the existence of other human beings.
We notice that whole nature resembles that of the social animals. We eat food that
others have produced, wear clothes that others have made, live in houses that
others have built. The greater part of our knowledge and beliefs has been passed
on to us by other people though the medium of a language which others have
created. Without language and mental capacities, we would have been poor
indeed comparable to higher animals.

We have, therefore, to admit that we owe our principal knowledge over the least to
the fact of living in human society. The individual if left alone from birth would
remain primitive and beast like in his thoughts and feelings to a degree that we
can hardly imagine. The individual is what he is and has the significance that he
has, not much in virtue of the individuality, but rather as a member of a great
human community, which directs his material and spiritual existence from the
cradle to grave.

Alder, K. (2002). The Measure of All Things. New York: Free Press.

37

18
 2/25/2022

RRL EXERCISE#3: Use the 3 RRL strategies in this example

RFID is a non-contact automatic identification technology that uses

radio frequency signals to automatically recognize targets and access to relevant
data. The identification work does not require human interference and can work
in a variety of harsh environments. But if there is no network to transmit data,
it will be difficult to play its advantage. Under the influence of environmental
conditions, the traditional wired network may not be a better way to achieve the
desired effect. The feature of wireless sensor network is no center and
self-organized; it is a powerful complement of RFID, and can solve the drawback
of poor anti- interference, when the effective transmission distance is short.

Research on Wireless Sensor Network Technology

By Aamir Shaikh and Siraj Pathan
International Journal of Information and Education Technology, Vol. 2, No. 5, October 2012, page 477

38

“USING DIRECT QUOTES FROM THE AUTHORS”

Example: See how many references were read and summarized in two paragraphs:

The strategy I am putting forward to address the complexity of Natural Resource Management
(NRM) is a learning-based approach— a “joint learning” (Pretty 1994:25). Joint learning is the interaction of
knowledge from different perspectives in understanding and improving a problematic situation (ibid.). In the
same vein, Daniels & Walker (1996:72) argue for “collaborative learning” in order to address the failure of
NRM organizations in accommodating the “interdependence among good science, good civic dialogue, good
local knowledge, and good learning.” Woodhill and Röling (1998:47) call it “social learning” and underscore
the “need to focus on integrating the creative capacities of people, whether they be land users, lay people,
natural scientists, social scientists, policy makers or politicians” through cultural transformation and
institutional development. Emphasis is given to “understanding” because in a complex situation involving
multiple perspectives, defining the problem is itself problematic.
In arguing for collaborative endeavors, Checkland (1993:65) refers to the German word
“Weltanschauung,” which describes the existence of meanings that different people attribute to what they
perceive according to the “human activity system” they are engaged in, and which shapes the way they
handle issues they face in this world. Zaltmann (1982, cited in Baraba 1990:45) calls it the “law of the lens”
in using metaphor to illustrate the differences in people’s perspectives likened to the use of different eye
glasses. Warren (1975:20) calls it “world view” in recognizing that although we live in the same world, we live
differently, and so we live different worlds. Schütz (1945:33) refers to the existence of “multiple realities” to
better understand the differences in our actions or reactions to stimuli; hence, the need to learn from each
other’s ways.

39

19
 2/25/2022

DIRECT QUOTES

“All models are wrong, but some are useful.” - George Box, 1976
Because models are simplifications of reality

“It is easy to lie with statistics. It is hard to tell the truth without it” – Andrejs Dunkel

“No problem can be solved from the same level of consciousness that created it.” -- Albert Einstein

“If I had an hour to solve a problem, I'd spend 55 minutes thinking about the problem and five minutes
thinking about solutions.” – Einstein

- "Statistics is, or should be, about scientific investigation and how to do it better, but many statisticians believe
it is a branch of mathematics. Now I agree that the physicist, the chemist, the engineer, and the statistician can
never know too much mathematics, but their objectives should be better physics, better chemistry, better engineering,
and in the case of statistics, better scientific investigation. Whether in any given study this implies more or less
mathematics is incidental." -- George E. P. Box

40

METHODOLOGY vs. METHOD

METHODOLOGY – the research design (and justification)
- a systematic theoretical analysis of the methods
applied to a field of study (as a result of RRL).
- the general research strategy that outlines the way
in which research is to be undertaken and
identifies /describes the methods to be used in it.

(e.g., random sampling, case study, RS/GIS, linear regression, etc)

METHOD – defines the means or modes of data collection or
how a specific result is to be calculated.

Sources: http://www.uq.edu.au/student-services/learning/sections-thesis
https://www.bcps.org/offices/lis/researchcourse/develop_writing_method_quantitative.html

41

20
 2/25/2022

METHODOLOGY
Using Flowcharts

42

METHODOLOGY

43

21
 2/25/2022

METHODOLOGY
Using Flowcharts

https://research-methodology.net/research-methodology/research-structure/

44

RESEARCH/PROJECT PROPOSAL SECTIONS

Significance /Context
Statement of Purpose
Research Objective
Hypothesis

Past Work
Current Work
RESEARCH/ Gap /Modifications
PROJECT SUMMARY Study Area Data
Experiment Design Field Techniques
+ABSTRACT Work Plan – Flowchart Analyses/Software
+GANTT CHART Timeline – Gantt Chart

+Anticipated Problems

References

24/02/2022 (Gatrell, Bierly, and Jensen, 2012) 45

45

22
 2/25/2022

WHAT YOU CAN DO WITH

(To gain new knowledge) RESEARCH
1. Categorize - forming a typology of objects, events or concepts,
i.e. a set of names (categories) into which these can be sorted.
2. Describe - observation as a means of collecting data to examine situations in order
to establish what is the norm, and what can be predicted to happen again under
the same circumstances.
3. Explain - moving beyond ‘just getting the facts’ in order to make sense of the
myriad other elements involved, such as human, political, social, cultural and
contextual.
4. Evaluate - making judgements about the quality of objects or events. Quality can
be measured either in an absolute sense or on a comparative basis.

Walliman, N. (2011). Research Methods: The Basics. New York: Routledge.

46

WHAT YOU CAN DO WITH

(To gain new knowledge) RESEARCH
5. Compare - contrasting cases are examined to highlight differences and
similarities, for a better understanding of phenomena.
6. Correlate - the relationships between two phenomena are investigated to
see whether and how they influence each other (e.g., cause & effect).
7. Predict - if correlations are already known, possible future behavior or
events are made in similar circumstances
8. Control - after understanding an event or situation, find ways to control it.
The cause and effect relationships are determined to prevent undesirable
effects or promote the desired effects.
e.g., CATEGORIZE,
DESCRIBE, EXPLAIN,
CORRELATE, PREDICT
Walliman, N. (2011). Research Methods: The Basics. New York: Routledge.

47

23
 2/25/2022

CAREFULLY DESIGNING THE RESEARCH METHODOLOGY

Research Methodology : (step-by-step process--- overall plan)

- explains the research process and addresses the
issues of the research philosophy. It contains the
explanation of the research design, and the choice
and implementation of data collection methods.

- Sampling aspect of the study and discussions of

ethical considerations are also included.
(e.g., data privacy, guidelines in using participatory methods, acknowledging sources)

Research methods: are the strategies, processes or techniques utilized

in the collection of data or evidence for analysis in order to uncover new information
or create better understanding of a topic. (e.g., random sampling, regression analysis, PCA-principal component analysis)
(e.g., FGD- focus group discussion, survey questionnaire, interpolation, triangulation, etc.)

https://research-methodology.net/research-methodology/research-structure

48

SAMPLING METHODS (1)

We take samples from the whole population in order to bring the study to a manageable number;
To reduce cost (time and money) in studying the whole population.

• Probability Sampling – Uses randomization and takes steps to ensure that all
members of a population have a chance of being selected. There are several
variations on this type of sampling and following is a list of ways probability
sampling may occur:
– Random sampling – every member has an equal chance
– Stratified sampling – population divided into subgroups (strata) and the members are
randomly selected from each group (e.g., by age-group, by economic class, by slope class, by region,
by elevation class, etc.)
– Systematic sampling – uses a specific system to select members such as every 10th
person on an alphabetized list; or every other grid cell, etc. (e.g., exit poll)
– Cluster random sampling – divides the population into clusters, clusters are randomly
selected and all members of the cluster selected are sampled (not defined grouping /strata)
– Multi-stage random sampling – a combination of one or more of the above methods

Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods approaches. Sage Publications, Incorporated.

49

24
 2/25/2022

SAMPLING METHODS (2)

• Non-probability Sampling – Does not rely on randomization techniques to select
members. This is typically done in studies where randomization is not possible in order to
obtain a representative sample. Bias is more of a concern with this type of sampling.
The different types of non-probability sampling are as follows:
•

– Convenience or accidental sampling – members or units are selected based on availability

– Purposive sampling – members of a particular group are purposefully sought after (e.g., most senior)
e.g., factory workers
– Modal instance sampling – members or units are the most common within a defined group and
therefore are sought after
e.g., all with honors
– Expert sampling – members considered to be of high quality are chosen for participation
e.g., 50% M, 50%F..
– Proportional and non-proportional quota sampling – members are sampled until exact proportions of
certain types of data are obtained or until sufficient data in different categories is collected
– Diversity sampling – members are selected intentionally across the possible types of responses to
capture all possibilities e.g., different year levels
e.g., recommended persons /locations, etc.
– Snowball sampling – members are sampled and then asked to help identify other members to sample
and this process continues until enough samples are collected
Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods approaches. Sage Publications, Incorporated.

50

Sampling is a serious matter… GIGO.

Harry Truman was

elected President
of the USA in 1948.
(Though he also
served the remaining
term of Pres. Roosevelt
who died in 1945).

https://cirt.gcu.edu/research/developmentresources/research_ready/quantresearch/sample_meth

51

25
 2/25/2022

Too small a size might include “outliers” that render anomalous results.
SAMPLE SIZE Too big a size will be very costly & complex, although results are better.
(not the whole population)
VARIABLES INVOLVED IN DETERMINING THE DESIRED SAMPLE SIZE:
(Can just be an estimate)
1. Population size – depends on your purpose (e.g., if about dog owners, only how many own dogs)
2. Margin of error – or confidence interval=how much error is allowed (e.g., + 5%)
3. Confidence level – different from #2. How confident you want to be that the actual mean of the data falls
within your margin of error (e.g., 90% or 95% or 99% confident) (Most common confidence level)
4. Standard deviation – how much the sample responses you receive vary from each other
and from the mean number (how far the data is from the mean value)
Maximum Variability – when set to 0.5 (half-half) is better than e.g., 60-40%

(Safe choice before running a survey)

Usually set to 0.5 (to make sure sample Empirical rules in statistics:
68% of the x values lie within the Std.Dev. of the
size is enough)
mean
95% of the x values lie within 2 Std.Dev. of the
Low Std. Dev. = all the values will be
34% 34% mean
clustered around the mean 99.7% of the x values lie within 3 Std.Dev. of the
2.35% 2.35% mean
High Std. Dev. = values are spread out 0.15% 13.5% 13.5% 0.15%
TOTAL Area under the curve = 1
across a much wider range Z= -3 Z= -2 Z= -1 Z=1 Z=2 Z=3 www.sakowskimath.com

52

68 + 27 =95%
Std. Dev. -- a measure of how dispersed
the data is in relation to the mean. Empirical rules in statistics:
68 + 27 + 4.7 =99.7%
68% of the x values lie within the Std.Dev. of the mean

95% of the x values lie within 2 Std.Dev. of the mean

99.7% of the x values lie within 3 Std.Dev. of the mean

TOTAL Area under the curve = 1

www.sakowskimath.com

of the sample pts.

Confidence level

Origins of statistics rules:

Since thousands of yrs ago..
- Probability theory ~17th Century census
- Inferential Statistics ~18th century
12 12 12 12 12 - Sociology & education ~19th century
12
- Regression & correlation – computer analysis 20th c.

53

26
 2/25/2022

(Margin of error e.g., + 5%)

(90%, 95%, 99% - most common level of confidence)

(Standard Deviation)

(The more heterogenous the population is, the larger sample is required
to obtain a required level of precision)

54

Miaoulis and Michener, 1976

William Cochran (1909-1980)

(unknown population)

e.g., wearing a mask

Confidence Z-value
=0.05 Level
90 1.645
95 1.960
From Z-table 99 2.576

55

27
 2/25/2022

Miaoulis and Michener, 1976

(known population)

56

Taro Yamane (1923-1979)

(known population)

= 371 (From Cochran)

Use the bigger value

Miaoulis and Michener, 1976

57

28
 2/25/2022

Or follow from RRL of more recent researchers

58

DEFINITIONS: Median

e.g., Sample data: 1,000,000 50,000 30,000 20,000 10,000

222,000 - AVERAGE or MEAN – sum of the measures divided by the number
Denoted by “xbar”: x of measures: x =x Average =
1,110,000
= 222,000
n 5
30,000 - MEDIAN – the middle value when the numbers are arranged decreasing/increasing
None - MODE – the value that occurs most often in a set of numbers
505,000 - MIDRANGE – the mean of the minimum and maximum of the values in a data set
Midrange = 1,00,000 + 10,000 = 505,000
2
Z-table
Confidence Z-
Level value
90 1.645
95 1.960
Data are close to the mean = Data are far from the mean =
99 2.576 Less dispersed. More dispersed.

59

29
 2/25/2022

STATISTICAL METHODS

• Statistics is a branch of science that deals with the collection,

organization, analysis of data and drawing of inferences from the samples
to the whole population.
• Statistics helps us turn data into information-- data that have been
interpreted, understood and are useful to the recipient.
• Involves identification of variables -- characteristics that vary from one
individual member of population to another individual.
• Relationships among variables are obtained through statistics – mean,
average, standard deviation, regression, correlation etc. to give meanings
to the numbers.

https://research-methodology.net/research-methodology/research-structure/

60

TYPES OF STATISTICS:
1. DESCRIPTIVE STATISTICS – numerical data that describe an object or phenomena
(population, total area, temperature, etc.)

2. NUMERICAL STATISTICS – statistical parameters presented as numbers

(standard deviation, average, variance, mean, etc.)

3. PICTORIAL STATISTICS – presenting numerical data in terms of pictures or

graphs (bar graphs, pie charts, icons, etc.)

4. INFERENTIAL STATISTICS – a way of working with numbers to answer /draw

conclusions about a phenomena.

61

30
 2/25/2022

• DESCRIPTIVE STATISTICS -
descriptive statistics describe a big chunk of data
with summary charts and tables, but do not attempt
to draw conclusions about the population from which
the sample was taken. (e.g., PSA, NEDA stats)

62

• NUMERICAL STATISTICS - A variety of numerical measures

used to summarize data.
The primary numerical measure for qualitative data is the proportion
or percentage. For quantitative data: The mean, median, mode,
percentiles, range, variance, and standard deviation are the most
commonly used numerical measures.

63

31
 2/25/2022

• PICTORIAL STATISTICS -
pictorial statistics describe a big chunk of data
with picture, graphs, pie charts, etc. with each
unit representing a quantity.

64

65

32
 2/25/2022

• INFERENTIAL STATISTICS –
Not just to describe data, but to
make inferences about the collected data to solve a problem.
The statistician must do the following to make an educated guess:
(BETTER THAN CHANCE)
1. Gather data (numerical data)
2. Organize the data (sometimes graphically)
3. Analyze the data (using tests for significance, etc..)

May look simple, but requires to be performed in certain prescribed ways.

(USING BOTH DESCRIPTIVE & INFERENTIAL STATISTICS)

66

WORKING WITH VARIABLES /DATA

Characteristics that are measurable

or describable

e.g., color, shape, e.g., height, weight, length

taste, old/new

e.g., Small, Medium, Large e.g., 30, 29, 355 e.g., y= f(x)
High, Low

e.g., 20-100, 40-60 e.g., 50% , 20ppm.

Zulfiqar Ali and S Bala Bhaskar, 2016. Basic statistical tools in research and data analysis. Indian Journal of Anaesthesia. Vol.60(9)

67

33
 2/25/2022

a.k.a. METHODOLOGY (e.g., urban greenery)

SAMPLE
SAMPLE ONLY ONLY

Expound RRL here..

To do X using Y in order to Z.

using RS/GIS techniques to aide in urban planning. X

Y Z

HYPOTHESES (To be tested):

24/02/2022 68
https://research-methodology.net/research-methodology/research-structure/

68

HYPOTHESIS TESTING
• Is a four-step procedure:
1. Stating the hypothesis (null and alternative)
2. Setting the criteria for decision (confidence level, acceptable error, etc)
3. Collecting data (sampling methods)
4. Evaluate the null hypothesis (accept or reject according to the set criteria)
(WELL DESIGNED)
→Done by experiments. (Next module – Experimental Design)
WATCH OUT: Two types of mistakes are possible while testing the hypotheses.
Depending on
Type I : occurs when the null hypothesis (H0) is wrongly rejected. the experiment
Type II: occurs when the null hypothesis (H0), is wrongly accepted. carried out.

(e.g., testing the efficacy of two drugs– which one is better; or no difference)69
24/02/2022

69

34
 2/25/2022

- END -

70

35