Unit 1: Foundation of research - Subjective Questions

Research is a systematic, objective, and logical process of enquiry aimed at discovering new facts, verifying existing knowledge, and solving problems. It involves the collection, analysis, and interpretation of data to arrive at valid conclusions.

Significance of Research:

• Solving problems: Research provides scientific solutions to operational and planning problems in business, government, and society.
• Policy formulation: Governments rely on research to frame economic and social policies.
• Advancement of knowledge: It expands the frontiers of human knowledge across disciplines.
• Decision making: Enables evidence-based decisions rather than intuition.
• Career and academic growth: Forms the basis of scholarly work, higher education, and innovation.
• Technological development: Drives innovation in industry and improves quality of life.

In essence, research is the backbone of progress, ensuring that decisions and developments are grounded in reliable evidence.

Basic Research (also called fundamental or pure research) is conducted to expand knowledge and understand underlying principles without any immediate practical application.

Applied Research is directed toward solving specific, practical problems faced by society or industry.

Key point: Basic research often provides the foundation on which applied research builds. For example, fundamental research on the structure of DNA (basic) enabled the development of genetic testing kits (applied).

Quantitative Research deals with numerical data and measurable variables, using statistical tools to test hypotheses.

Qualitative Research deals with non-numerical data such as opinions, behaviours, and experiences, aiming to understand meanings and phenomena.

When preferred:

• Quantitative is preferred when the goal is to quantify relationships, generalize results, or test a theory (e.g., measuring customer satisfaction scores).
• Qualitative is preferred when the goal is to gain deep insight into motivations, feelings, or complex social processes (e.g., understanding why customers abandon a product).

Often a mixed-methods approach combines both for richer findings.

Deductive Approach moves from the general to the specific. The researcher starts with an existing theory, formulates a hypothesis, and tests it through observation.

Inductive Approach moves from the specific to the general. The researcher begins with observations and develops a theory based on patterns found.

Examples:

• Deductive: "All metals expand on heating (theory). Iron is a metal, so iron will expand on heating (hypothesis), tested by experiment."
• Inductive: Observing that many swans are white leads to the tentative generalization "all swans are white."

The research process is a sequence of logically connected steps. Though presented linearly, several steps are interrelated and may overlap.

Steps in the Research Process:

1. Formulating the research problem – Identifying and clearly defining the problem or question.
2. Extensive literature review – Studying existing work to understand what is already known.
3. Developing hypotheses – Formulating tentative, testable statements.
4. Preparing the research design – Planning the blueprint (exploratory, descriptive, experimental).
5. Determining sample design – Deciding the population, sample size, and sampling technique.
6. Collecting data – Gathering data via surveys, experiments, observation, etc.
7. Execution / Data processing – Editing, coding, classification, and tabulation.
8. Analysis of data – Applying statistical/qualitative techniques.
9. Hypothesis testing – Verifying whether data support the hypotheses.
10. Generalization & interpretation – Drawing conclusions and building theory.
11. Preparation of the report – Documenting findings in a structured report.

Flow:

Feedback loops exist, e.g., analysis may reveal the need to redefine the problem.

Good research adheres to certain criteria that ensure its validity, reliability, and usefulness.

Criteria of Good Research:

• Clearly defined purpose: The objective and problem must be stated precisely.
• Systematic procedure: The research design and methods should be well planned and detailed enough to permit replication.
• Careful planning: Thorough planning yields results that are as objective as possible.
• Full disclosure & transparency: The procedure should be reported with sufficient detail; limitations should be frankly revealed.
• Adequate analysis: Data analysis must be adequate, valid, and reliable, using appropriate methods.
• Justified conclusions: Conclusions should be confined to those justified by the data.
• Reliability & validity: Findings must be consistent (reliable) and measure what they intend to (valid).
• Ethical integrity: The researcher should be a person of integrity and maintain ethical standards.

Qualities of a good researcher: systematic, logical, empirical, replicable, unbiased, and ethical.

In summary, good research is systematic, logical, empirical, and replicable.

Longitudinal Study: Data are collected from the same subjects repeatedly over a long period to observe changes over time.

Cross-Sectional Study: Data are collected from a population at a single point in time, providing a snapshot.

Examples:

• Longitudinal: Tracking the health of a group of individuals from age 20 to 60 to study aging.
• Cross-Sectional: Surveying people of different age groups today to compare their health status.

Key insight: Longitudinal studies are better for establishing cause-effect over time, while cross-sectional studies are quicker for describing a situation.

Correlational Research is a type of non-experimental research that examines the statistical relationship between two or more variables without manipulating them.

Purpose:

• To determine whether and to what degree a relationship exists between variables.
• To make predictions about one variable based on another.
• Measured using the correlation coefficient , ranging from to .

Types of correlation:

• Positive correlation (): Variables move in the same direction.
• Negative correlation (): Variables move in opposite directions.
• Zero correlation (): No relationship.

Example: Studying the relationship between hours of study and exam scores.

Limitations:

• Correlation does not imply causation — a relationship does not prove one variable causes the other.
• Third-variable problem: An unseen variable may influence both.
• Directionality problem: It is unclear which variable influences the other.

Thus, correlational research is useful for identifying associations and making predictions but cannot establish cause and effect.

Historical Research is the systematic collection and objective evaluation of data related to past events in order to understand causes, effects, or trends that help explain present events and anticipate future ones.

Steps involved:

1. Identification of the research problem.
2. Collection of source material.
3. Criticism/evaluation of sources.
4. Formulation of hypotheses.
5. Interpretation and reporting.

Sources of data:

• Primary sources: Original documents, eyewitness accounts, artifacts, official records, letters, relics.
• Secondary sources: Textbooks, review articles, encyclopedias — information reported by non-witnesses.

Evaluation of sources:

• External criticism: Checks the authenticity (genuineness) of a source.
• Internal criticism: Checks the accuracy and reliability of the content.

Challenges:

• Data cannot be manipulated or repeated (events already occurred).
• Difficulty in verifying authenticity of old documents.
• Incomplete or biased records.
• Researcher's own bias in interpretation.

Example: Studying the economic policies of a nation during the 1930s depression using government archives.

Research Approaches are broad strategies used to conduct research. They are broadly classified into two categories:

1. Quantitative Approach: Involves the generation of data in quantitative form subjected to rigorous statistical analysis. Sub-classified into:

• Inferential approach: Forms a database from a sample to infer characteristics of the population (e.g., survey research).
• Experimental approach: Greater control over the research environment; variables are manipulated.
• Simulation approach: Construction of an artificial environment/model to generate data and observe dynamic behaviour.

2. Qualitative Approach: Concerned with subjective assessment of attitudes, opinions, and behaviour. Uses techniques like focus groups, in-depth interviews, and projective techniques.

Approach vs Method:

Key point: The approach reflects the overall orientation, while methods are the actual tools employed within that approach.

A journal is a scholarly periodical publication that contains articles written by researchers, academicians, and experts.

Types of Journals:

• Scholarly / Academic (Peer-reviewed) journals: Contain original research articles reviewed by experts before publication. High credibility (e.g., Nature, IEEE Transactions).
• Trade / Professional journals: Focus on practical information for a particular industry or profession; not always peer-reviewed.
• Popular magazines: Written for general audiences, articles by journalists, not peer-reviewed (e.g., Scientific American).
• Open Access journals: Freely available online; costs often covered by author fees (APC).
• Subscription-based journals: Require payment or institutional access to read.
• Predatory journals: Charge fees but lack genuine peer review/editorial quality — to be avoided.

Characteristics of a good scholarly journal:

• Rigorous peer review process.
• A reputable editorial board.
• Proper indexing (Scopus, Web of Science).
• A recognizable ISSN.
• Regular publication schedule and good impact metrics.

Journal Indexing refers to the inclusion of a journal in recognized abstracting and indexing databases that catalog, organize, and make research articles discoverable.

Popular indexing databases:

• Scopus (Elsevier)
• Web of Science (Clarivate)
• PubMed / Medline (medical)
• DOAJ (Directory of Open Access Journals)
• Google Scholar

Importance of indexing for researchers:

• Credibility: Indexed journals are considered more reputable and trustworthy.
• Visibility & discoverability: Articles reach a wider audience and are easily searchable.
• Higher citations: Indexed articles are more likely to be cited.
• Academic recognition: Many universities and funding bodies require publications in indexed journals for promotions, PhD, and grants.
• Quality assurance: Indexing databases have selection criteria ensuring quality and legitimate peer review.

Key point: Publishing in an indexed journal increases the reach, impact, and academic value of research, and helps avoid predatory publishers.

Impact Factor (IF) is a metric published by Clarivate Analytics in the Journal Citation Reports (JCR) that reflects the average number of citations received by articles published in a journal.

Calculation: The Impact Factor for a given year is:

Example: If a journal's articles from 2022 and 2023 received 500 citations in 2024, and it published 100 articles in those two years:

Significance: A higher IF generally indicates a more influential and prestigious journal.

Limitations:

• Field bias: Citation patterns differ across disciplines (medicine vs mathematics).
• Skewed by few articles: A few highly cited papers can inflate the IF.
• Only a 2-year window: Ignores long-term impact.
• Can be manipulated through self-citation and editorial policies.
• Journal-level, not article-level: Does not reflect the quality of an individual paper.

The h-index is an author-level metric proposed by Jorge Hirsch (2005) that measures both the productivity and citation impact of a researcher (or journal).

Definition: A researcher has an h-index of if of their papers have each been cited at least times, while the remaining papers have each been cited no more than times.

Example: Suppose a researcher's papers have the following citations (sorted in descending order):

• Paper 1: 25 citations ≥ 1 ✓
• Paper 2: 12 citations ≥ 2 ✓
• Paper 3: 9 citations ≥ 3 ✓
• Paper 4: 6 citations ≥ 4 ✓
• Paper 5: 5 citations ≥ 5 ✓
• Paper 6: 3 citations ≥ 6 ✗

Here, 5 papers have at least 5 citations each, so the h-index = 5.

Advantages:

• Combines quantity (number of papers) and quality (citations) in a single number.
• Not affected by a single highly cited paper or many uncited papers.

Limitations:

• Cannot decrease, so it favours senior researchers.
• Field-dependent (varies across disciplines).
• Ignores author position and self-citations.
• Cannot compare researchers across different fields fairly.

The SCImago Journal Rank (SJR) is a journal-level metric developed by the SCImago research group using data from the Scopus database. It measures the prestige of a journal based on citations, where citations from highly-ranked journals carry more weight.

Key features of SJR:

• Based on a weighted citation approach (like Google's PageRank algorithm).
• Considers the prestige of the citing journal, not just the count of citations.
• Uses a 3-year citation window.
• Normalizes for differences in citation behaviour across subject fields.

SJR vs Impact Factor:

Key insight: SJR gives a more nuanced view of a journal's influence because a citation from a top journal counts more than a citation from a low-ranked one, unlike the Impact Factor where all citations are treated equally.

Journal citation indices are quantitative measures used to evaluate the influence and quality of journals and researchers.

Major citation indices:

• Impact Factor (IF): Average citations per article over a 2-year window (Clarivate/JCR).
• h-index: Balances productivity and citation impact for authors and journals.
• SJR (SCImago Journal Rank): Prestige-weighted metric based on Scopus.
• SNIP (Source Normalized Impact per Paper): Normalizes citations by the citation potential of the subject field, allowing cross-field comparison.
• CiteScore: Elsevier's metric — average citations over a 4-year window using Scopus data.
• Eigenfactor Score: Measures the total influence of a journal, weighting citations by the influence of citing journals over 5 years.
• i10-index: Number of publications with at least 10 citations (used by Google Scholar).
• g-index: Gives more weight to highly cited articles than the h-index.

Purpose of these indices:

• Compare journals within a field.
• Assist researchers in selecting where to publish.
• Help institutions in evaluating research output.

Note: No single index is perfect; they should be used together and interpreted within the context of the discipline.

Reputed / High-valued publishing houses are well-established organizations known for rigorous quality control and wide academic acceptance.

Examples of reputed publishers:

• Elsevier (ScienceDirect, Scopus)
• Springer Nature (Springer, Nature)
• Wiley (Wiley Online Library)
• Taylor & Francis (Routledge)
• IEEE (engineering & technology)
• ACM (Association for Computing Machinery)
• Oxford University Press and Cambridge University Press
• SAGE Publications
• American Chemical Society (ACS)

Characteristics of a reputable publisher:

• Rigorous peer review process ensuring quality.
• Strong editorial board of recognized experts.
• Journals are indexed in Scopus/Web of Science.
• Transparent publication and ethical policies (e.g., COPE membership).
• Wide readership and high citation impact.
• No hidden or deceptive charges (unlike predatory publishers).
• Long-standing reputation and history.

Key insight: Publishing with a reputed house enhances the visibility, credibility, and impact of research, whereas predatory publishers damage academic reputation.

Researchers can disseminate their work through several types of publications, each serving different purposes.

Types of Publications:

• Research/Original articles: Present original empirical findings; the most common form of scholarly output.
• Review articles: Summarize and critically evaluate existing literature on a topic; do not contain new experiments.
• Conference papers/proceedings: Presented at academic conferences; often shorter and time-sensitive.
• Short communications / Letters: Brief reports of significant, novel findings requiring rapid publication.
• Case studies: In-depth analysis of a particular case, common in medicine, management, and law.
• Book chapters: Contributions to edited scholarly books.
• Monographs / Books: Detailed scholarly treatment of a single subject by one or more authors.
• Editorials / Commentaries: Opinion pieces written by experts or editors.
• Technical reports / White papers: Detailed documents on specific technical work, often by organizations.
• Theses & Dissertations: Academic works submitted for degrees.

Key point: The choice of publication type depends on the nature of the work, target audience, and the need for speed vs depth.

Peer Review is the process by which a scholarly manuscript is evaluated by independent experts (peers) in the same field to assess its quality, validity, and originality before publication.

Steps in the Publication Review Process:

1. Submission: Author submits the manuscript to a journal.
2. Editorial screening: The editor checks scope, format, and basic quality (may desk-reject).
3. Assignment to reviewers: Suitable expert reviewers are selected.
4. Peer review: Reviewers evaluate methodology, novelty, and significance.
5. Reviewer recommendation: Accept, Minor revision, Major revision, or Reject.
6. Revision: Author revises based on feedback and resubmits.
7. Final decision: Editor makes the final decision.
8. Proofreading & Publication: The accepted paper is copy-edited and published.

Types of Peer Review:

• Single-blind: Reviewers know the author's identity, but authors do not know reviewers.
• Double-blind: Both authors and reviewers remain anonymous — reduces bias.
• Open review: Identities of both are known; reviews may be published.
• Post-publication review: Article is reviewed/commented on after publication.

Importance: Peer review ensures quality, credibility, and integrity of published research, filtering out flawed or fraudulent work.

Research can be classified in several ways depending on the purpose, objective, approach, and time dimension.

1. Based on Application:

• Basic (Fundamental) Research: Aims to expand knowledge; theory-oriented.
• Applied Research: Aims to solve immediate practical problems.

2. Based on Objective:

• Descriptive Research: Describes characteristics of a phenomenon (surveys, fact-finding).
• Analytical Research: Uses existing facts to make a critical evaluation.
• Exploratory Research: Investigates a problem not clearly defined.
• Explanatory Research: Establishes cause-effect relationships.

3. Based on Data/Approach:

• Quantitative Research: Based on numerical, measurable data.
• Qualitative Research: Based on non-numerical, subjective data.

4. Based on Logic:

• Inductive Research: Specific observations → general theory.
• Deductive Research: General theory → specific testing.

5. Based on Time:

• Longitudinal: Data collected over a long period.
• Cross-sectional: Data collected at a single point in time.

6. Other types:

• Conceptual Research: Related to abstract ideas/theory.
• Empirical Research: Based on observation and experiment.
• Historical Research: Study of past events.
• Correlational Research: Examines relationships between variables.

Key point: These categories are not mutually exclusive — a single study may combine several types (e.g., an applied, quantitative, cross-sectional study).