research methodology. 17 september 20152 research? what is it? (definition) should you be doing it?...
TRANSCRIPT
Research Methodology
April 19, 2023 2
Research?
• what is it? (Definition)
• should you be doing it? (Motivation)
• how do you do it? (Steps and Methods)
April 19, 2023 3
Research: Definition (1)
• “Systematic investigation towards increasing the sum of knowledge”(Chambers 20th Century Dictionary)
• “an endeavor to discover new or collate old facts etc. by the scientific study of a subject or by a course of critical investigation.”
(The Concise Oxford Dictionary)
April 19, 2023 4
Research: Definition (2)
• Research: an activity that contributes to the understanding of a phenomenon
– phenomenon: a set of behaviors of some entity(ies) that is found interesting by a research community
– understanding: knowledge that allows prediction of the behavior of some aspect of the phenomenon
– activities considered appropriate to the production of understanding (knowledge) are the research methods and techniques of a research community
April 19, 2023 5
Motivation for Research Methodology
• (qualitatively) control research process• validate research results• compare research approaches• respect rules of good scientific practice
Types of Research
• Applied vs. Fundamental• Descriptive vs. Analytical• Quantitative vs. Qualitative• Empirical vs. Conceptual• Other type of research (e.g.):
– One-time vs. longitudinal (Repeated Observation)– Field-setting research vs. laboratory or simulation
research
Steps in Research• Define research problem• Review the literature• Formulate hypothesis• Research design• Collect data• Analyse data• Test Hypothesis• Interpret
All these process could be iterative. For example, research design may need to be modified depending on data.
Quantitative or Qualitative
Quantitative Qualitative
Inquiry from the outside Inquiry from the inside
Underpinned by a completely different set of epistemological foundations from those in qualitative research
An attempt to take account of differences between people
Are simply different ways to the same end?
Aimed at flexibility and lack of structure in order to allow theory and concepts to proceed in tandem
Involves the following of various states of the scientific research
The results are said to be “hard generalized data”
The results are said to be through theoretical generalization, “deep, rich and meaningful”
Deductive Inductive- where propositions may develop not only from practice, or literature review, but also from
04/19/23 10
Research Methods
• Research methods are generally categorised as being either quantitative or qualitative.
• What matters is that the methods used fit the
intended purposes of the research!
04/19/23 11
Qualitative and Quantitative Paradigms
• The qualitative paradigm concentrates on investigating subjective data, in particular, the perceptions of the people involved. The intention is to illuminate these perceptions and, thus, gain greater insight and knowledge.
• The quantitative paradigm concentrates on what can be measured. It involves collecting and analysing objective (often numerical) data that can be organised into statistics.
04/19/23 12
Qualitative and Quantitative Research
Qualitative Research
Quantitative Research
Also known as
interpretative / responsive
positivist /hypothetico-deductive
Type of reasoning
(usually) deductive (usually) inductive
Link withconcepts
identifies concepts identified concepts and investigates relationships
04/19/23 13
Qualitative Research
Quantitative Research
Action sometimes only describes a situation BUT in action-research openly intervenes
tests relationships between concepts
Outcome illuminates the situation
accepts or rejects proposed theory
Approach to validity
truth seen as context bound (socially constructed)
truth seen as objectiveand universal
Qualitative and Quantitative Research
04/19/23 14
Methodology Comparison
Quantitative
• Explanation, prediction
• Test theories• Known variables• Large sample• Standardized
instruments• Inductive
Qualitative
• Explanation, description
• Build theories• Unknown variables• Small sample• Observations,
interviews• Deductive
19/04/23
Quantitative research • Involves information or data in the form of
numbers• Allows us to measure or to quantify things• Respondents don’t necessarily give numbers as
answers - answers are analysed as numbers• Good example of quantitative research is the
survey
19/04/23Multimedia Training Kit
www.itrainonline.org
Qualitative research
• Helps us flesh out the story and develop a deeper understanding of a topic
• Often contrasted to quantitative research • Together they give us the ‘bigger picture’• Good examples of qualitative research are face-to-
face interviews, focus groups and site visits
04/19/23 17
Other classification of types of research
The purposes of research can be categorised as:
• Description (fact finding) • Exploration (looking for patterns) • Analysis (explaining why or how) • Prediction (forecasting the likelihood of particular events) • Problem Solving (improvement of current practice)
04/19/23 18
Descriptive Research
• Seeks to accurately describe current or past phenomena - to answer such questions as:
a) What is the absentee rate for particular lectures? b) What is the pass rate for particular courses? c) What is the dropout rate on particular degree
programmes? d) What effect does a particularly quality audit process
have on teacher morale?
04/19/23 19
Analytical Research• Seeking to explain the reasons behind a
particular occurrence by discovering causal relationships. Once causal relationships have been discovered, the search then shifts to factors that can be changed (variables) in order to influence the chain of causality. Typical questions are:
a) Why is there a preponderance of female students on 1st level teacher training programmes?
b) What factors might account for the high drop-our rate on a particular degree programme?
04/19/23 20
Predictive Research
• Seeks to forecast the likelihood of particular phenomena occurring in given circumstances. It seeks to answer such questions as: a) Will changing the start time achieve a higher attendance rate at
our lectures? b) Will introducing anonymous marking reduce the gender
imbalance in the achievement of 1st class degrees? c) Will increasing the weighting for course work encourage students
to adopt deep learning strategies?
04/19/23 21
Problem Solving Research / Action Research
• Action-research is a form of problem solving based on increasing knowledge through observation and reflection, then following this with a deliberate intervention intended to improve practice.
Educational action-research describes a family of activities in curriculum development, professional development, school improvement programmes, and systems planning and policy development.Participants in the action being considered are intricately involved with all of these activities.
04/19/23 22
Typical Methods Descriptive Research
Statistical SurveysSamplingInterviews
Analytical Research
Case Studies Attitude SurveysObservations Statistical SurveysHistorical Analysis
Predictive Research
identifying and / or defining measurable (quantifiable) variables and manipulating them to cause measurable.
Problem Solving/Active Research
action-research spiral: observe reflect plan act observe reflect plan act ………
04/19/23 23
Research Methods Categorised by Activity
Experimental Research
The causal effects of phenomena are tested on one group by comparison with a control group which is otherwise similar but upon which the phenomena is not allowed to act.
Quasi-Experimental Research
Causal effects of phenomena are investigated in a way similar to experimental research BUT full control is not possible
Non-Experimental Research
The investigation of conditions as they really are without an attempt to change any of them - at least while the research is in progress.
19/04/23
Participatory research
• Allows participation of community being researched in research process (e.g. developing research question; choosing methodology; analysing results)
• Good way to ensure research does not simply reinforce prejudices and presumptions of researcher
• Good for raising awareness in community and developing appropriate action plans
Quantitative Research
ContentTheoretical Research: Concepts and MethodsStatistical Research: E.g. Multivariate Analysis (Examples)
– Paradoxes (e.g. Simpson’s Paradox)– Cluster Analysis– SEM (Structural Equation Modeling)
Mathematical Programming– Optimization– Game Theory (and Equilibrium Analysis)
Integrative Analysis (Qualitative & Quantitative Research)– Scenarios Analysis
Theoretical/Fundamental Research: Concepts
• Concepts – Probability (Objective vs. Subjective)– Risk
• Axioms (Assumptions)• Proofs in Mathematics
– Proof by construction– Proof by contradiction
Statistical Research Methods
• Experimental• Non experimental
Multivariate Analysis
• Many statistical techniques focus on just one or two variables
• Multivariate analysis (MVA) techniques allow more than two variables to be analysed at once– Multiple regression is not typically included under
this heading, but can be thought of as a multivariate analysis
Outline of Lectures
•Why MVA is useful and important•Simpson’s Paradox•Some commonly used techniques
– Principal components– Cluster analysis– Correspondence analysis– Others if time permits
•Market segmentation methods•An overview of MVA methods and their niches
Simpson’s Paradox (1)
• Example: 44% of male applicants are admitted by a university, but only 33% of female applicants
• Does this mean there is unfair discrimination?
• University investigates and breaks down figures for Engineering and English programmes
Male Female
Accept 35 (44%) 20 (33%)
Refuse Entry
45(56%) 40 (67%)
Total 80 (100%) 60 (100%)
Simpson’s Paradox (2)
• No relationship between sex and acceptance for either program– So no evidence of
discrimination• Why?
– More females apply for the English program, but it is hard to get into
– More males applied to Engineering, which has a higher acceptance rate than English
• Must look deeper than single cross-tab to find this out
Engineering Male Female
Accept 30 (50%) 10 (50%)
Refuse entry
30 (50%) 10 (50%)
Total 60 (100%) 20 (100%)
English Male Female
Accept 5 (33%) 10 (33%)
Refuse entry
15 (67%) 30 (67%)
Total 20 (100%) 40 (100%)
Simpson’s Paradox (3)• In each of these examples, the bivariate
analysis (cross-tabulation or correlation) gave misleading results
• Introducing another variable gave a better understanding of the data– It even reversed the initial conclusions
• The practical significance of Simpson's paradox: – Which data should we consult in choosing an action,
the aggregated or the partitioned?
Treating Many Variables
Many Variables• Commonly have many relevant variables in market
research surveys– E.g. one not atypical survey had ~2000 variables– Typically researchers pore over many crosstabs– However it can be difficult to make sense of these, and the
crosstabs may be misleading• MVA can help summarize the data
– E.g. factor analysis and segmentation based on agreement ratings on 20 attitude statements
• MVA can also reduce the chance of obtaining spurious results
Cluster Analysis• Techniques for identifying separate groups of similar
cases– Similarity of cases is either specified directly in a distance
matrix, or defined in terms of some distance function• Also used to summarize data by defining segments
of similar cases in the data– This use of cluster analysis is known as “dissection”
Clustering Techniques
• Two main types of cluster analysis methods– Hierarchical cluster analysis
• Each cluster (starting with the whole dataset) is divided into two, then divided again, and so on
– Iterative methods• k-means clustering (PROC FASTCLUS)• Analogous non-parametric density estimation method
– Also other methods• Overlapping clusters• Fuzzy clusters
Applications• Market segmentation is usually conducted using
some form of cluster analysis to divide people into segments
• It is also possible to cluster other items such as products/SKUs, image attributes, brands
• Search result grouping: For intelligent grouping of the files and websites, clustering may be used to create a more relevant set of search results compared to normal search engines like Google. There are currently a number of web based clustering tools such as Clusty.
Cluster Analysis Issues• Distance definition
– Weighted Euclidean distance often works well, if weights are chosen intelligently
• Cluster shape– Shape of clusters found is determined by method, so choose method
appropriately• Hierarchical methods usually take more computation time
than k-means• However multiple runs are more important for k-means, since
it can be badly affected by local minima• Adjusting for response styles can also be worthwhile
– Some people give more positive responses overall than others– Clusters may simply reflect these response styles unless this is
adjusted for, e.g. by standardising responses across attributes for each respondent
Structural Equation Modeling (SEM)
• SEM is a statistical technique for testing and estimating causal relations using a combination of statistical data and qualitative causal assumptions.
• General method for fitting and testing path analysis models, based on covariances
• Fits specified causal structures (path models) that usually involve factors or latent variables– Confirmatory analysis
SEM Example:Relationship between
Academic and Job Success
ACT: American College TestingCGPA: Cumulative Grade Point Average
Mathematical Modeling
What use a model?• To reflect complex systems in a simplified form.• Helps to organize large amounts of data.• Examples:
– Economic Policy Models (I-O Analysis) – Environment Policy Models – Product Mix Models– Optimal Pricing Models
Modeling and Simulation
MARKet ALlocation Model:MARKAL
Multi-period linear programming formulation Decision variables in different time periods(e.g.): Energy consumption Electricity generation Capacity utilization Investment in technologies Emissions
MARKAL Model: Overall Architecture
Techno-economic Database
Economic Scenario
EmissionScenarioMARKAL
• Consumption and production of energy• Marginal ‘values’ of energy resources & technologies• Shadow price of external constraints (e.g. emissions)• Introduction and retirement of technologies
Bottom-up 3E (Energy-Economy-Environment) Model System
MINING IMPORT COLLECTION RENEWABLE EXPORT
COALN. GAS
OILBIOMASSNUCLEAR
RENEWABLE
45
ENVIRONMENT
ELECTRICITY PRODUCTION
COAL GAS HYDRO
NUCLEAR SOLAR
ENERGY
FUEL PROCESSING
PETROLEUM REFINERY GAS PROCESSING
75
ENDUSE DEVICES
PUMP TRACTOR
FURNACE MOTOR
BUS TRAIN
CFLTVOVENAC
AGRICULTURE
INDUSTRY
TRANSPORT
COMMERCIAL
RESIDENTIAL
67
235
TECHNOLOGY CAPITAL
EMISSIONS
ENERGY
ECONOMY SECTORS
Agr
Ind
Tran
Bldg
Typical Reference Energy System
MINING
IMPORT
IMPORT
IMPORT
IMPORT
COAL
GAS
REFINERY 2
REFINERY 1
ELECTRIC TRAIN
DIESEL VEHICLE
TRANSPORTEND USE
TECHNOLOGIES
GENERATION+T&D
PETROLEUM PROCESSING
EN
ER
GY
RE
SO
UR
CE
S
D E
M A
N D
ELECTRICITY TECHNOLOGY
EXTRACTION
GAS
COAL
OIL
EXTRACTION
EXTRACTION
OIL
Model Formulation
Objective FunctionTo minimize the discounted sum, over up to 100 years, of investment, operating and maintenance cost of all technologies plus the cost of energy imports and carbon tax
Subject to1. Demand Constraints (one for each end use demand)
Cig(t) >= demandk (t)
i DMD G GRD
V k DM, t T
WhereDMD…end-use demand technologyGRD…set of grades technologies/energy sourcesDM….class of all end use demandsT…..set of time periodsCig(t)…capacity of technology i of grade G in period t
Objective Function (cntd)
1. Capacity transfer constraints (to account for technology vintage carry over time periods)
2. Energy carrier balance constraints (supply >= demand of fuel)
3. Cumulative reserve constraints (fuel extraction <= total reserves)
4. Electricity balance constraints(day and night time modeling for electricity system)
Model Constraints
5. Process technology capacity utilization constraints (process activity <= available capacity)
6. Electricity production capacity constraints (electricity generation <= available capacity)
7. Electricity peaking constraints (extra capacity to meet peak demand)
8. Total emissions constraints (Carbon, SO2 etc)
Model Constraints (cntd.)
Game theory is a study of strategic decision making. More formally, it is "the study of mathematical models of conflict and cooperation between intelligent rational decision-makers.
Game Theory is descriptively also referred to as interactive decision theory.
Game Theory
Prisoners'-Dilemma• A paradox in decision analysis in which two individuals
acting in their own best interest pursue a course of action that does not result in the ideal outcome.
• Dominant strategy: Not to cooperate– If both remain silent, both win– If one betrays the other, than he benefits and the other loses– If both betrays, both loses
Prisoner B stays silent (cooperates)
Prisoner B betrays (defects)
Prisoner A stays silent (cooperates) Each serves 1 year Prisoner A: 3 years
Prisoner B: goes free
Prisoner A betrays (defects)
Prisoner A: goes freePrisoner B: 3 years Each serves 2 years
Prisoners‘ Dilemma
Computational General Equilibrium (CGE) Model
AGRICULTUREgrains and oil seeds
animal productsforestry
food processingother agricultural services
HOUSEHOLDSdemographics
labor supplyENERGY land supply
oil production household savingsPRIMARY FACTORS gas production final product demands
OF PRODUCTION petroleum refiningland surface gas distribution
subsurface resources coke and coal productslabor biomass production
capital uranium productionhydro and solar electric power GOVERNMENT
electricity production general governmentnational defense
education
EVERYTHING ELSEpaper and pulp manufacture
chemical manufacturecement manufacture
primary iron and steelprimary non-ferrous metals
other manufacturingpassenger transport
freight stransportother services
SGM: Second Generation Model is a Global CGE Model (PNL, USA)
Integration of Qualitative and Quantitative Analysis
• Scenarios Analysis• Projections of Future vs. Prediction• Qualitative Storyline• Used for long-term analysis (e.g. economic and
environmental futures)
• Case Study• Projections of Future vs. Prediction• Qualitative Storyline• Used for long-term analysis (e.g. economic and
environmental futures)
Case study
• Definition : an empirical inquiry that investigates a contemporary phenomenon within its real-life context; when the boundaries between phenomenon and context are not clearly evident(Yin, 1984)
• Case study research emphasize detailed contextual analysis of a limited number of conditions and their relationships
• Scenarios: Images of future or alternate futures.
• Scenarios are not Predictions or Forecasts.• Used as a methodology in energy &
environmental analysis: Account for future uncertainties in energy planning and to study likely implications of current policy pathways.
Scenario Analysis
Example
Reference : http://www.ft.com/cms/s/1/4d46cbae-2b4f-11da-995a-00000e2511c8.html
Thank You