MPPI session 1: Introduction Research and Scientific Inquiry in IS

 Definition of Scientific Inquiry

o Facts, objective, analytical, measured, deductive/inductive hypothesis
 Definition of Research
o Systematically data collection and interpretation
o Use of facts and theories, data analysis, sampling, experiment
o Problem investigation systematically
 Methods vs Methodology
o Method: techniques and procedure to obtain data
o Methodology: theory of how research should be undertaken
 Research process
o Formulating topic
o Reviewing literature
o Designing the research
o Collecting data
o Analyzing data
o Writing it up
 Objectivism vs Subjectivism
o Objectivism: position that social entities exist in reality external to social actors concerned
to their existence
o Subjectivism: social phenomena are created from the perceptions and consequent actions
of those social actors concerned to their existence
 Deductive vs Inductive
o Deduction: theory and hypothesis are developed and tested
 Theory -> hypothesis -> observation -> confirmation
o Induction: data are collected and theory developed from data analysis
 Observation -> pattern -> tentative hypothesis -> theory
MPPI session 2: Formulating and clarifying research topics
 Applied research
o Research that aimed to create alternatives for decision making and improvements in an
organization/public.
o If it is aimed to policy making or evaluation in an organization, it is named policy research.
 Public policy making steps
o Step 1: policy research
 Output : academic paper
 Process : battle of theories/academic analysis
 Indicator: conceptually accepted
o Step 2: policy design
 Output : policy draft
 Process : battle of interest/political process
 Indicator: politically accepted
o Step 3: finalize policy
 Output : regulation/policy document
 Process : legal process/drafting
 Indicator: legally accepted
 Policy research
o Is an applied research used in social science, aimed to create better policy (products) in an
organizations/society, which improved from time to time (policy analysis)
o Used by: CIO, IS/IT Managers; Consultants and advisors; IS/IT strategist and analyst
 Good research topics attributes
o Capability : is it feasible?
o Appropriateness: is it worthwhile?
o If relevant : does the topic relate clearly to the given idea, possibly by the
organization?
 Definition of Problems
o Reality difference to Expectations
o In organization, expectations are stated in: strategy, policy, SOP, structure, ethics,
regulations, etc.
 Level of Problems
o Observed problems
 Problems are observed and supported by data/information
 Just symptoms and not the real problems
o Hidden problems (causes)
 Hidden aspects, causes of observed problems
 This is the real problems, sometimes called root causes
 Sometimes no data/information available
 Need scientific inquiry/reasoning
 Relevant theories mastery + relevant experience = advantages
 Problem Analysis tools
o Fishbone
o Mind map
o Problem tree
o Logic analysis
 Refining research ideas
o Using Delphi technique (expert panel)
o Conducting a preliminary study
o Continually testing out your ideas
o Integrating ideas from key persons
o Refining topics given by the organization
 Research questions
o Relevant to the problems
o Consistent with expected standards
o Produce clear conclusions
o Not too descriptive
o At the right level, meaning not too difficult
 Research objectives
o Output: direct results of research (ex.: description, correlational analysis, design,
evaluation)
o Outcome: impact to organization or community (ex.: to improve quality, to raise the
efficiency, to give certain direction)
 Research general framework
o Step 1: Background
 Why doing the research
 What problems and facts
 Fishbone mapping
 What are RQ, research output, research benefits, scope and limitations
o Step 2: Literature study
 What are relevant theories
 What are relevant previous research
 Do 3C + 2S
 Develop theoretical framework
o Step 3: Develop research methodology
 What are steps to do the research
 What are output of each steps
 Identify data collection methods, analysis methods, decision making methods
 Qualitative vs quantitative, empirical/positivistic vs interpretive
o Step 4: Data collection
 Primary data (quantitative): survey, observations
 Primary data (qualitative): depth interview, FGD
 Secondary data: from any formal, credible, and relevant documents
o Step 5: Data analysis, develop conclusions, and recommendations
 Quantitative analysis: statistics, operation research, AHP/ANP
 Qualitative analysis: text analysis (hermeneutics)
 Develop specific conclusions and recommendations based on research (not
normative)
 Forum group discussions
o Interactions -> divergent process (opinions and ideas) -> convergent process (pattern) ->
Solution
MPPI session 3: Literature Review (LR)
 Definition of LR
o Surveys of scholarly articles, books, and other sources (dissertations, journal, conference
proceedings) relevant to area of research and theory.
o The purpose is to offer a state-of-the-art overview of significant literatures published.
 Research logical framework
o Literature review -> paper comparison -> methodology -> analysis and interpretation ->
conclusion -> (iteration to LR) …
 Characteristic of good LR
o Logical flow of ideas
o Current and relevant references which consistent
o Appropriate referencing style
o Proper use of terminology
o Unbiased and comprehensive view of previous research on the topic
 Referencing styles
o ACM, APA, Purdue, you name it…
 Urgency of doing LR
o Identify gaps in the literature
o Avoid reinventing the wheel
o Identify methods that could be relevant to research topic
o Carry on from where others have already reached
o Identify other people working in the same fields
o Increase breadth of knowledge of subject area
o Identify similar works
o Provide intellectual context
o Identify opposing views
 Eligible sources for academic
o High
 Scientific articles submitted to academic journals
o Medium
 Research based professional books
 Scientific papers submitted to refereed conference
  Unpublished scientific papers or research report
 University textbooks
o Not eligible
 Non-academic publications: newspapers, magazines
 Popular books or not research based books
 Popular websites: Wikipedia, BSC, etc.
 Personal blogs/websites
 Business white papers
 Consulting reports
 Course assignment papers
o What need to review
 Relevant theories
 Relevant previous research
 Relevant methodology/methods
o Stages of LR
 Problem formulation: which topic is examined and what issues?
 Literature search: finding materials relevant to the subject.
 Data evaluation: which literatures provide significant contribution.
 Analysis and interpretation: discuss findings and conclusions.
o Summary LR
 Put the LR’s into a table to be compared by: problem, theory. Technology,
methodology (method, technique, process), interpretation, finding, limitation.
o Review LR – using 3C + 2S
 Compare : find similarities among literatures
“Menurut penelitian yang dilakukan oleh Andri (1999), kinerja IRS dengan
menggunakan teknik extended Boolean lebih baik dibanding menggunakan teknik
Boolean saja. Hal ini sejalan dengan hasil-hasil penelitian sebelumnya yang
dilakukan oleh Savoy (1995) dan Salton (1990)”
 Contrast : find differences among literatures
“Hasil penelitian yang dilakukan oleh Santoso (2006) menunjukkan bahwa gaya
belajar konstruktif lebih adaptif terhadap penggunaan ICT. Hal ini bertentangan
 dengan hasil penelitian lainnya yang mengatakan bahwa gaya positivist yang lebih
adaptif terhadap penggunaan ICT (Jones, 1998) dan (Garry, 2010)”.
 Criticize : put own opinion on what is written in the literatures
Sistem Informasi Untuk Eksekutif (EIS) dapat membantu pimpinan mengambil
keputusan lebih akurat sekitar 90% dibanding tidak menggunakan EIS (Hadi, 2005).
Tetapi tidak dijelaskan berapa banyak sample EIS yang disurvei dan kategori
keputusan yang bagaimana yang dijadikan sebagai acuan.
 Synthesize : combine several literatures into an idea
Menurut Hadi (2005) keberhasilan suatu Sistem Informasi Untuk Eksekutif (EIS)
sangat ditentukan oleh tingkat keakuratan menangkap kebutuhan para eksekutif.
Sedangkan menurut Amir (2006), EIS sangat ditentukan oleh kejelasan core bisnis
dari perusahaannya. Dari kedua pendapat tersebut, dapat dikatakan bahwa faktor-
faktor penentu keberhasilan EIS antara lain: keakuratan menangkap kebutuhan
pimpinan, dan kejelasan core bisnis perusahaan
 Summarize : restate article with own words in concise way
o Where to place LR
 At chapter 2 of thesis or dissertation
 May constitute an essential chapter in a thesis or dissertation
 Should logically connected to research problems, methodology, analysis,
conclusions
o To consider in reviewing LR
 Provenance : author’s credentials, arguments supported by evidence.
 Objectivity : author’s perspective even-handed or prejudicial?
 Persuasiveness : which author’s theses are most/least convincing?
 Value : author’s arguments and conclusions convincing enough?
o To discuss in LR
 What’s the problem and why it’s important?
 Is the problem clearly defined?
 Is the research methodology well stated?
 How’s the data being created and manipulated?
 Is the manipulated data sufficiently interpreted?
 How’s contribution of the study?
  Is the conclusions related to the problems?
 Is the evidence sufficient enough to support conclusion?
o Primary Research vs Secondary (Desk) Research
 Primary Research: research that has involved the collection of original data specific
to that particular research project (ex: using research methods like questionnaires)
 Secondary Research: research where no original data is collected, instead using
existing data sources (ex: census, or archive data)
Type of desk research:
 Internal desk research: involves existing organizational resources to
organize collected data so it can be usable and efficient
 External desk research: done outside organizational boundaries, such as
online desk research, government published data, customer desk research
Advantages of desk research:
 Easy availability, from a wide variety of resources
 Faster to collect
 Cheaper to collect
Disadvantages of desk research:
 Update, available information may be outdated and not suitable
 Suitability of UoM, data conversion may be necessary
 Suitability of terminology
 Lack of control, no control over accuracy of the data
MPPI session 4: Research Design
 Research Design in perspective
o Selecting the methodological approach, Operationalizing variables/materials, Crafting
procedures/paradigms, Determining sample, Evaluating the practically of the research
 Research method/design and methodology
o Research method may inclusive in research methodology
o Research method more concise as compare to research methodology
o Research method used to solve straight forward research problem
o Research methodology used to solve more complicated problems requires various steps
o Steps in research methodology may use one or more research method
 Types of research methodologies
o Experimental research
 Allows for the causes of behavior to be determined
 Experiment is a carefully regulated procedure where factors are manipulated
 Ex: Learning effectiveness between with and without e-learning system
o Quasi-experimental research
 Almost but not quite real experiments
 No manipulation of the variables
 Compare groups biased on naturally occurring variables
 Types of natural variables: subject and time variable
 Ex: The impact of e-commerce strategy
o Causal-comparative research
 Explores possible causes and effects
 Independent variable is not manipulated, it has already been applied
 Focus first on the effect, attempts to determine cause of the observed effect
 Seeks to explain differences between 2 groups that have occurred
 Ex: Why IT multinational companies are more innovative than local firms?
o Correlational research
 Measure 2 variables: study methods and grade-point average
 Determine degree of relationship between them: correlation coefficient (r=0.5)
 Allow description and prediction of the relationship
 No attempts to control variables, unlike experiment
  Ex: Is non-technical IT competency will affect the IT department staff performance
o Survey research
 Determine present status of a given phenomenon
 A systematic method of collecting primary data based on a sample
 The purpose is to capture main characteristics of the population at any instant
 Ex: We want to explore the opinion of top executives on IT Manager
o Action research
 Form of collective self-reflective enquiry undertaken by participants in social
situations in order to improve the rationality and justice of their own practices
 Ex: System analysis and design for application development
o Historical research
 Recognition of historical research problem or identification of need for certain
historical knowledge
 Gathering of as much pertinent information about the problem/topic as possible
 The forming of hypotheses that tentatively explain relationships between factors
 The selection, organization, and analysis of conclusion
 The recording of conclusions in a meaningful narrative
 Ex.: Does IT create continuous improvements of firm’s business process?
o Ethnographic research
 Involves field research and requires contextualization – interpretation of result in
data collection
 Focus on organizations, which consist of groups of people who interact in regular
and structured ways
 Ex: The study of leadership style of IS/IT Managers in make strategic decisions
during board meeting
o Case study research
 Empirical investigation of particular phenomenon
 ‘case’ may be an organization, community, country, etc.
 Boundary problems should be determined, Holistic and robust
 Multiple cases are suggested to compare and contrast different cases
 Using qualitative and quantitative data
 Ex: To examine why a particular company success or fail in using IT
MPPI session 6: Qualitative Analysis

 Birth of Qualitative method

o First known in studies from Chicago schools in 1910-1940
o At that period, researchers at the Univ. of Chicago conducted research using participant
observation and personal documents
o The qualitative research is rooted in interpretative paradigm
 Qualitative Data
o Refers to all non-numeric data that have not been quantified and can be a product of all
research strategies
 Qualitative vs Quantitative research
o Qualitative Research
 Designed to assist researchers in understanding phenomena in context
 Based on meanings expressed thru words
 Collection results is non-standardized data requiring classification into categories
 Analysis conducted thru the use of conceptualization
 Capture and discover meaning once the researcher becomes immersed in the data
 Concepts are in the form of themes, motifs, generalizations, and taxonomies
 Measures are created in an ad hoc manner and are often specific to the individual
setting or researcher
 Data are in the form of words and images from documents, observations, and
transcripts
 Theory can be causal or non-causal and is often inductive
 Research procedures are particular, and replication is very rare
 Analysis proceeds by extracting themes or generalizations from evidence and
organizing data to present a coherent, consistent picture
o Quantitative Research
 Focuses on measurement. Tends to isolate specific aspects of phenomena by
measuring these thru dedicated instruments
 Based on meanings derived from numbers
 Collection results in numerical and standardized data
 Analysis conducted thru the use of diagrams and statistics
 Test hypothesis that the researcher begin with
  Concepts are in the form of distinct variables
 Concepts are in the form of distinct variables
 Data are in the form of numbers from precise measurement
 Theory is largely causal and is deductive
 Procedures are standard, and replication is frequent
 Analysis proceeds by using statistics, tables, or charts and discussing how what
they show relates to hypotheses
 Deductive and Inductive Research
o Deductive
 Identify problems and develop RQ; then
 Do literature review; then
 Create proposed model (hypothesis); then
 Test with field data and then analyze the result; then
 Recommendations.
o Inductive
 Identify problems and develop RQ; then
 Do data collections; then
 Create proposed model; then
 Validate the model by comparing/contrasting to existing literatures and also
expert panel; then
 Recommendations.
 Positivism and Interpretivism
o Positivism
 (field or lab data) - (field or lab data) - (field or lab data); forming
 Analysis (survey, experimental, etnography); forming
 Model; forming
 Recommendations.
o Positivism
 (human mind data) - (human mind data) - (human mind data); forming
 Analysis (FGD, depth interview); forming
 Model; forming
 Recommendations.
 Qualitative data collection types
o In-depth, open-ended interviews
o Focus group discussions
o Direct observations
o Written documents
 Sampling
o Quantitative methods rely on random sampling
o Qualitative methods rely on purposive sampling
 Type of qualitative analysis
o Summarizing (condensation) of meanings
o Categorizing (grouping) of meanings
o Structuring (ordering) of meanings using narrative
 Qualitative data analysis steps
o Step 1. Selection and Reduction: select only relevant data for the research
o Step 2. Create Pattern: create cause-effect relationship, don’t count the words
o Step 3. Create Conclusions: interpret the findings into verbal display
 Coding = Analysis
o Codes are tags or labels for assigning units of meaning to the information compiled.
o Codes used to retrieve and organize chunks of information, so we can quickly find, pull out,
and cluster the segments relating to particular topic.
o Open Coding (Descriptive): attributing class of phenomena to a segment of text.
o Axial Coding (Interpretive): include more complex, underlying meaning (unsupported
argument), identifying relationships among open codes.
o Selective Coding (Pattern): inferential and explanatory, group codes into smaller number
of themes or constructs, analogous to cluster and factor analysis in statistics.
o Open Coding and Axial Coding are part of thematic coding/analysis.
o Selective Coding mostly used in grounded theory research design.
 Triangulation for Qualitative Research Quality: Respondents – Documents – Theories
 Key success factors for Qualitative Analysis
o Selection of data sources
o The understanding of text/language/meanings
o The coding process
MPPI session 7: Qualitative Research Design
 Qualitative data collections
o Primary Data: Depth interviews, focus group discussions, observations (participant, direct)
o Secondary data: Data published by other researchers, documentations (corporate, org’s)
 World Views (Weltanschauung): framework of ideas and beliefs through which an individual
interprets the world and interacts in it.
 Hermeneutics: science of the interpretation (grammatical, historical, psychological)

 Grounded Theory: set of flexible analytic guidelines that enable researcher to focus their data
collection and to build inductive middle-range theories through successive levels of data analysis
and conceptual development.
o Focus on generating theoretical ideas from the data
o Inductively derived from the study of the phenomena it represents
o Key focus: reflective reading of text and the application of codes
o How to using it?
 Data analysis and data collection proceed together
 Data analysis begins to develop theories that suggest further cases to sample
o Usually used when the research is a new phenomena
 Action Research: generating research knowledge as a scholar and improving “social action” as an
activist (take action to produce something).
o Project takes as its subject matter a social practice
o The project through spiral cycles of planning, acting, observing, and reflecting
o The project involves those responsible for the practice in each moments of the activity
o Data Collection technique:
 Experiencing, through observation and field notes
  Enquiring, when the researcher asks interviews or questionnaires
 Examining, using and making records
o Action research is a good methodology for design research (research to design something)
like enterprise architecture, IT blue print, application design, etc. But it needs a longer time.

 Hard System Methodology (structured systems analysis methods, system engineering, operations
research) assume:
o Objective reality of systems in the world
o Well-defined problem to be solved
o Technical factors foremost
o Scientific approach to problem-solving
o One correct solution
 Soft System Methodology (SSM)
 o Assume:
 Organizational problems are messy, poorly defined
 Stakeholders interpret problems differently
 No objective reality
 Human factors important
 Creative and intuitive approach to problem-solving
o 7 stages of SSM
 Situation considered problematic (real world)
 Problem situation expressed (real world)
 Root definition of relevant systems (system thinking about real world)
 Conceptual models of systems described in root definitions (system thinking about
real world)
 Comparison of models and real world (real world)
 Changes: systematically desirable, culturally feasible (real world)
 Action to improve the problem situation (real world)
 Rich picture diagram: iconic representations – drawn together into a picture which sums up the
important elements of the problem situation

 Root definitions: short textual statements which define the important elements of the relevant
system being modelled, rather like mission statements.
o Form: a system to do X by (means of) Y in order to Z
 What the system does – X
 How it does it – Y
 Why it’s being done – Z
 o Example:
 A university owned and operated system to award degrees and diplomas to
suitably qualified candidates (X), by means of suitable assessment (Y), (in
conformance with national standards), in order to demonstrate the capabilities of
candidates to potential employers (Z).
 A university owned and operated system to implement a quality service (X), by
devising and operating procedures to delight its customers and control its
suppliers (Y), in order to improve its educational products (Z).
 CATWOE Analysis: a check to ensure that root definitions contain most of what is important.
o C-ustomers : the victims or beneficiaries of T
o A-ctors : those who do T
o T-ransformation: process from input to output
o W-orldview : that makes the T meaningful in context
o O-wners : those with the power to stop T
o E-nvironmental : elements outside the system which constraints are taken as given, but
nevertheless affect its behavior
o Example: (C) candidate students, (A) university staff, (T) candidate students to degree
holders, (W) the belief that awarding degrees and diplomas is a good way of demonstrating
the qualities of candidate to potential employers, (O) the university governing body, (E)
national educational and assessment standards.
 Conceptual Models: necessary activities to ‘do’ the root definition, modelled by verbs.
o Verb + noun phrase: activity – ‘do something’
o A -> B: logical dependency arrow, A must come before B and will impact it results
o Example: A university owned and operated system to award degrees and diplomas to
suitably qualified candidates (X), by means of suitable assessment (Y), (in conformance
with national standards), in order to demonstrate the capabilities of candidates to
potential employers (Z).
 Performance measures
o E1 – efficacy: does the system work, is the transformation effected?
o E2 – efficiency: the relationship between the output achieved and the resources consumed
to achieve it
o E3 – effectiveness: the longer term goal (Z) achieved
 o Example of the complete model:
(E1) are degrees and diplomas awarded, (E2) how many degrees and diplomas, of what
standard, are awarded for the resource consumed, (E3) do employers find the degrees and
diplomas a useful way of assessing the qualities of potential employees.
 Black box/transformational system: specific SSM template for change management

 Summary of SSM (learning process of F-M-A

o Framework(s) of ideas -> Methodology -> Area of application
o SSM in real life:
 IS/IT strategy priority analysis and decision
 Business process efficiency analysis
 IS/IT resource optimization analysis
 IS/IT investment analysis
o SSM is a good methodology specially for design research, and it’s time flexi.
 Case Study Research: a method research or a research strategy
o Advantages depending on:
 Type of research question
 Control over behavioral events
 Focus on contemporary as opposed to historical phenomena
 o Preferred when:
 How and why questions
 Little control over events
 Focus on contemporary phenomena within real-life context
o Type of case study research
 Exploratory: to develop pertinent hypothesis and propositions for further inquiry
 Descriptive: to describe the incidence or prevalence of a phenomenon
 Explanatory: to trace operational links over time
 Principles of Data Collection
o Principle 1: use multiple sources of evidence
 Single source: problems of accuracy and trustworthiness
 Triangulation: rationale for using multiple sources of evidence
 Construct validity -> valid sources
 More expensive/time consuming/need different skills
o Principle 2: create a case study database
 Need to separate between collected evidence and final report
 Increases reliability
 Contents: notes, documents, quantitative data, narratives
 Other people should be able to access data
o Principle 3: maintain a chain of evidence
 To allow an external observer to follow the derivation to any evidence
 Trace steps:
 From conclusions to research questions
 From research questions to conclusions
 Final report <-> database <-> evidence and circumstances <-> procedures and
questions in protocol <-> initial research questions
 It is not mandatory to open your data to external people. But, it is mandatory to open them to your
supervisor. Explain your research protocol, especially how you combine with others like SSM,
hermeneutics, action research, or grounded theory.
 You can use either or combinations of: HERMENEUTICS, GROUNDED THEORY, ACTION RESEARCH,
SOFT SYSTEMS METHODOLOGY
MPPI session 9: Quantitative Research: Collecting, Presenting, and Analyzing Quantitative Data
 Quantitative Research
o Refers to systematic empirical investigation of social phenomena via statistical,
mathematical, or computational techniques.
o The objective is to develop and employ mathematical models, theories, and/or hypotheses
pertaining to phenomena.
o Quantitative data is any data that is in numerical form such as statistics, percentages, etc.
o Quantitative methods can be used to verify which of such hypotheses are true.
 Source of Data
o Quantitative (numerical): containing discrete (only certain values are possible and having
gaps) and continuous (any value within an interval is possible).
o Qualitative (categorical): containing discrete.
 Types of Data
o Primary: data observed and recorded/collected directly from respondents, through
observation, questionnaire, or experimentation.
o Secondary: data compiled both inside and outside organization for some purpose, through
printed or electronically.
 Categorical data
o Ratio data: differences between measurements, true zero exist. Ex.: height, age, weekly.
o Interval data: differences between measurements but no true zero. Ex.: temperature,
exam score.
o Ordinal data: ordered categories (ranking, order, scaling). Ex.: service quality rating,
standard & poor bond rating, student letter grades.
o Nominal data: categories (no ordering or direction). Ex.: marital status, type of car owned.
 Collecting quantitative data
o Identify unit analysis
o Specify the population and sample
o Information that will be collected: observations, interviews and questionnaire, documents,
and audiovisual materials.
 Presenting data
o Techniques: tables and graphs/charts
o Summary: centralized and deployment measurements
MPPI session 9: Survey Research in IS and IT
 Survey is a system for collecting information to describe, compare, or explain attitudes, and
behavior. Types of survey: polling, questionnaires, observation, FGD, consensus panels, interviews.
 Planning a survey:
o Deciding on research question
o Choosing the format of your questions
o Choosing the format of your interview
o Editing your questions
o Refining your survey instrument
o Choosing a sampling strategy
 Sampling strategy
o Probability sampling: provides statistical basis for saying that sample is representative of
the study or target population.
 Simple random sampling
 Stratified random sampling
 Systematic sampling
 Cluster sampling
o Non-probability sampling: chosen based on judgement regarding the characteristics of the
target population and the needs of the survey.
 Questionnaire
 How to do the survey
o Step 1. Determine the purpose
o Step 2. Target population
o Step 3. How will the survey be evaluated
o Step 4. How the survey be administered
o Step 5. Determine the question format
 Open-ended question
 Advantages: freedom of response, Sensitive to respondents desire for
expression, Well suited for exploratory studies, Suitable for certain types
of questions
 Disadvantages: Interpretation of the responses can bias the findings,
Researchers must form representative categories that encompass all
 responses, May discourage respondents from completing the
questionnaire, Coding open-ended questionnaires can be expensive and
time-consuming
 Closes-ended question
 Advantages: easy to administer, reduce interpreter bias, tend to motivate
respondents to complete surveys, simplify coding and tabulating efforts,
prompts choices and respondents to ponder alternatives they might not
otherwise consider
 Disadvantages: extensive effort required in questionnaire construction to
create an exhaustive list of choices or a short list of the most likely choices,
the list can have an inhibiting effect on respondents, relying on the
respondent to know whether to select one or check all that apply
 Dichotomous question
 Closes-ended questions that only offer two response choices
 Used to understand respondents demographic composition or behavioral
inclination
 Multiple choice question
 Closes-ended questions that give respondents several choices
o Step 6. Question Construction
o Step 7. Develop response scale
o Step 8. Pilot test
o Step 9. Distribute the survey
 Questionnaire problem
o Double-barreled questions
o Ambiguous questions
o Ambiguous word meanings
o Leading questions or phrases
o Level of question difficulty
o Unbalanced response categories
o Missing questions
 Interviews as a survey methods
o Advantage: comprehensive, ensure participant, understand the question, minimizes
missing data, enables clarification of unclear responses
o Disadvantage: expensive, people more like to refuse participation, can be risky for
interviewer, interviewer may bias the responses
o It is record time, place, name (optional), and role of the interviewees (mandatory). The
interviews record then will be transcribed. 1 hour interview equal to 5 hour transcription!
o Type of interviews survey methods:
 Face-to-face: expensive and time consuming
 Telephone: need to use random-digit dialing to reach listed and unlisted numbers
 Mail: return rate is usually low (20-30%)
o Type of interview:
 Structured: Questions are pre-planned in advance to elicit responses in a specific
area. Benefits: consistency, even coverage, content validity. Issues: limits coverage
 Unstructured: Questions are unplanned. The interviewer can ask questions at
his/her discretion. Benefits: cover a wide range of issues, flexibility. Issues:
reliability, validity.
MPPI session 10: Structural Equation Modelling (SEM)
 What is SEM?
o SEM is a statistical model explaining relationships among multiple variables (Hair, 2010).
o The structure of interrelationships is expressed in a series of multiple regression equations.
o SEM integrates the following multivariate techniques into a general modelling framework:
 Measurement theory from psychology
 Factor analysis from psychology and statistics
 Path analysis from epidemiology and biology
 Regression modelling from statistics
 Simultaneous equation from econometric
 PLS-SEM vs CB-SEM
Criteria PLS-SEM CB-SEM
Orientation Prediction Parameter
Approach Variance Covariance
Assumption Predictor specification (non- Normal multivariate distribution,
parametric) independence observation
(parametric)
Measurement model Indicator can be formative or Should have error terms and only
reflective, no need error terms reflective indicator allowed
Model complexity Big complexity, 100 or 1000 laten Small to medium complexity
(less than 100 laten)
Sample size 30-100 200-800
Model evaluation No goodness-of-fit criteria Should be goodness-of-fit
Implication Optimum for prediction Parameter oriented
Parameter estimate Consistency at large Consistency
Laten variable score Predictor specification (non- Indeterminate
parametric)
MPPI session 10: Statistics for IS/IT Research
 Descriptive and Inferential statistics
o Descriptive: use to describing sample
o Inferential: using sample to create inference for the population (generalization)
 Conduct hypothesis tests
o Identify a null and alternative hypothesis
o Set level of significance (alpha level) for rejecting the null hypothesis
o Collect data
o Compute the sample statistic
o Make decision about rejecting/failing to reject
o H0 ditolak: penelitian terbukti secara nyata (empiris)
o H0 diterima: penelitian tidak terbukti secara empiris
 Contoh Hipotesis
o Rumusan masalah deskriptif: seberapa baik layanan TI di perusahaan X?
o Hipotesis: Pelayanan TI di X paling tinggi 80% dari yang diharapkan
o HIpotesis Operasional:
 H0: μ <= 80%
 H1: μ > 80%
o Rumusan masalah asosiatif: adakah hubungan positif dan signifikan antara kualitas
pelayanan TI dengan kinerja perusahaan?
o Hipotesis: diduga terdapat hubungan yang positif dan signifikan antara kualitas pelayanan
TI dengan kinerja perusahaan
o Hipotesis nol: tidak ada hubungan yang positif dan signifikan antara kualitas pelayanan TI
dengan kinerja perusahaan
o Rumusan masalah komparatif: adakah perbedaan yang signifikan pada layanan TI antara
cabang A, B, dan C?
o Hipotesis: terdapat perbedaan yang signifikan pada layanan TI antara cabang A, B, dan C
o Hipotesis nol: tidak ada perbedaan yang signifikan pada layanan TI antara cabang A, B, C
 Level of significance
o The significance testing is a blend of Fisher’s idea of using the probability value p as an
index of the weight of evidence against a null hypothesis
o Fisher suggested that 95% is a useful threshold for confidence
 o Report summaries with
 One star * attached to indicate p <= 0.05 (for IS/IT research)
 Two stars ** to indicate p <= 0.01
 Three stars *** were used to indicate p <= 0.001 (for research in medicine)
 Type 1 Error (alpha level)
o The incorrect rejection of a true null hypothesis
o H0 ditolak, padahal H0 benar
 Type 2 Error (beta level)
o The failure to reject a false null hypothesis
o H0 diterima, padahal H0 salah
 Variance
o Measures how far a set of number is spread out
o A variance of zero indicates that all the values are identical

 Hubungan vs Pengaruh
o Pengaruh lamanya penayangan iklan di TV terhadap nilai penjualan barang: X -> Y
o Rumusan masalah:
 Berapakah rata-rata waktu penayangan iklan di TV?
 Berapakah nilai penjualan barang yang telah diiklankan?
 Apakah ada hubungan positif dan signifikan antara lamanya penayangan iklan di
TV dengan nilai penjualn barang?
 Bagaimana pengaruh lama penayangan iklan TV terhadap nilai penjualan barang?
 Assumptions for using parametric tests
o Independent observations
o Normal distribution
o Equal variances
o Interval of ratio scales
 Advantages of non-parametric tests
o Easy to understand and use
o Usable with nominal data
 o Appropriate for ordinal data
o Appropriate for non-normal population distributions
 Univariate techniques
o Testing hypothesis on a single mean
 One sample t-test: to test hypothesis that the mean of population from which a
sample is drawn is equal to a comparison standard
 Co.: seberapa tinggi produktivitas pegawai di PT XYZ?
o Testing hypothesis about two related means
 Paired samples t-test: a parametric test that examines differences in the same
group before and after a treatment.
 The Wilcoxon signed-rank test: a non-parametric test for examining significant
differences between two related samples or repeated measurement on a single
sample. Used when population cannot be assumed to be normally distributed. As
an ordinal data.
 Co.: apakah ada perbedaan IP mahasiswa sebelum dan sesudah menggunakan e-
learning?
 McNemar’s test: non parametric test on nominal data. It assesses the significance
of the difference between two dependent samples when the variable of interest
is dichotomous. It is used primarily in before-after studies to test for an
experimental effect.
o Testing hypothesis about two unrelated means
 Independent samples t-test: to see if there are any significant differences in the
means for two groups in the variable of interest.
 Co.: apakah ada perbedaan IPK antara pria dan wanita?
 How to calculate sample size
 G power: tool to calculate sample size
 Compute required sample size: Compute required sample size: given α ,
power (1−β), and effect
o Testing hypothesis about several means
 Analysis of Variance (ANOVA): examine the significant mean differences among
more than two groups on an integral or ratio-scaled dependent variable.
o Regression analysis
 Simple regression: one metric independent variable is hypothesized to affect one
more dependent variable.
 Multiple regression analysis: more than one (metric or non-metric) independent
variable to explain variance in a (metric) dependent variable.
o Multivariate analysis: metode analisis statistic melibatkan multi-variable secara simultan.
 Contoh: structural equation modelling (SEM)
 How big is the sample for SEM? Minimum sample size 100 to 500, depends on the
number of constructs (unobserved variables) and communalities (how good the
unobserved constructs explaining variance in the measured variables).
 Example of SEM
MPPI session 11: Mixed Methods Research Design
 Mixed methods research design is a procedure for collecting, analyzing, and mixing both
quantitative and qualitative data in a single study to understand a research problem.
o Pre-determined and emerging methods
o Open and close-ended questions
o Multiple forms of data drawing on all possibilities
o Statistical and text analysis
o Across databases interpretation
 Why mixed methods?
o Strength of confirmatory results derived from the quantitative analyses
o Deep-structure explanatory descriptions from analysis of the qualitative text narratives
o Offers a concurrent and integrative methodology for rigorous analyses that aim to attain
“the best of both methodologies, an integrative qualitative + quantitative design.
 When use mixed methods?
o Strength of confirmatory results derived from the quantitative analyses
o Deep-structure explanatory descriptions from analysis of the qualitative text narratives
o Offers a concurrent and integrative methodology for rigorous analyses that aim to attain,
“ the best of both methodologies, an integrative QUAL+QUANT design”
o When both QUAL+QUANT data provides a better understanding of a research problem
than either type by itself
o To provide a complete picture of the research problem
 Quantitative: trends and describing population
 Qualitative: participants’ words
o When you want to build from one phase of a study to another
 Explore qualitatively then develop an instrument
 Follow up a quantitative study to obtain more detailed information
o Community of practice argument
o To compare results from QUAL and QUANT research
o To use QUAL research to help explain QUANT findings
o To explore using QUAL research and then to generalize findings to a large population using
QUANT research
o Multiple angles argument. The more evidence the better argument.
 How to identify mixed methods?
o Title for words such as mixed-methods or multi-methods
o Data collection section indicates both qualitative and quantitative data were collected
o Purpose statement and or research questions indicate that the researcher intends to
collect both quantitative and qualitative data during the study
o Priority or weight: qualitative or quantitative
o Sequence of collecting quantitative and qualitative data is indicated
o Analyze both data sets: combined in one analysis, or separate analysis
 Types of mixed methods
o Convergent Parallel Mixed Methods

o Explanatory Sequential Mixed Methods

o Exploratory Sequential Mixed Methods

o Triangulation Mixed Method Design Explanatory Mixed Method Design

o Exploratory Mixed Method Design Embedded Mixed Method Design

o Sequential Embedded Mixed Method Design

 o Complex Research Design

 Mixing qualitative and quantitative

 Mixed methods research steps

o 1. Determine if a mixed methods research is feasible
o 2. Identify a rationale for a mixed methods research
o 3. Identify the data collection strategy and types of design
o 4. Develop QUANT+QUAL research questions
o 5. Collect QUANT+QUAL data
o 6. Analyze data separately or concurrent
o 7. Write the report as a one-or-two phase research
 The procedure
o Sequence: concurrent or sequential or both
o Emphasis: emphasis on QUAL or QUANT
o Sometimes both concurrent and sequential phases are used
o Designs may include more than two phases
o Think about using simple and elegant design
 Mixed methods in IS research
o Technology acceptance model (TAM), User experience (UX), Software QA, IT Governance
Maturity Analysis, Impact Analysis of IT investment, Strategic Fit Analysis of Business IT
Alignment, Business Process Reengineering with IT, Impact Analysis of Social Media
 Example
o TAM
 Step 1. Identify the model (QUANT survey research using SEM)
 follow up with
 Step 2. Develop recommendations for improvements (QUAL using depth
interviews or FGD and text analysis/coding)
o The impact of social media to workplaces
 Step 1. Survey the impact (QUANT survey research using questionnaires/stats)
 Interpretation and merging/concurrent
 Step 2. Survey the impact (QUAL using depth interviews and text analysis/coding)
 Notes!
o If you do mixed methods research design, you should draw the building blocks and explain
the research procedures and protocols.