Businesses Research Methods 2023 Board Question Paper

Differentiating Applied Research from Fundamental (Basic) Research

Ethical Concerns in Research

Ethical concerns in business and social research govern the moral principles and integrity of the investigative process:

1. Informed Consent & Voluntary Participation: Participants must be fully informed of the study’s scope and retain the right to withdraw at any point without penalty.
2. Confidentiality & Anonymity: Protecting respondent identity and ensuring sensitive corporate data cannot be traced back to individuals.
3. Prevention of Harm: Ensuring participants are protected from physical, psychological, legal, or professional distress.
4. Integrity in Data Reporting: Strict prohibition of data fabrication, data falsification, selective reporting, and plagiarism.

Definition of Theoretical Framework

A theoretical framework is a conceptual foundation and logical model that identifies, defines, and describes the structural network of relationships among the key variables (independent, dependent, moderating, and mediating) that have been identified as central to the problem under investigation. It integrates relevant established theories to provide a rationale for formulating research hypotheses.

Purpose of a Literature Survey

A literature survey serves several vital methodological functions:

1. Identifies Empirical Research Gaps: Uncovers unexplored areas, contradictory findings, and unanswered questions in current scholarship.
2. Prevents Reinvention of the Wheel: Avoids unintentional duplication of studies already completed.
3. Clarifies Variables and Theoretical Frameworks: Helps define core concepts and guides hypothesis formulation.
4. Informs Methodological Design: Reveals validated measurement instruments, scales, and sampling strategies used by prior researchers.

Four Types of Measurement Scales (S.S. Stevens)

In research methodology, data measurement is organized into four hierarchical levels:

1. Nominal Scale: Categorical classification with no inherent order (e.g., Gender, Religion, Department).
2. Ordinal Scale: Ranked or ordered categories where intervals between ranks are unequal (e.g., Customer Satisfaction: Low, Medium, High; Education Level).
3. Interval Scale: Ordered scale with equal, measurable intervals between points, but no true zero point (e.g., Temperature in Celsius, 5-point Likert Scale).
4. Ratio Scale: Highest measurement level featuring equal intervals and an absolute, meaningful zero point (e.g., Sales Revenue, Age, Weight, Profit).

What is a Research Problem?

A research problem is a clear, concise, and focused statement of an unresolved business issue, theoretical contradiction, operational difficulty, or empirical knowledge gap that exists in literature or practice that requires systematic, scientific investigation and data analysis to resolve.

Methods of Observation in Research

Primary observational methods in business research include:

1. Participant vs. Non-Participant Observation: The researcher either immerses themselves as an active group member or observes detachedly from the outside.
2. Structured vs. Unstructured Observation: Using a pre-defined observational coding checklist vs. recording open-ended narrative field notes.
3. Overt vs. Covert Observation: Subjects are either fully aware they are being observed or the observer is hidden/disguised.
4. Naturalistic vs. Controlled (Laboratory) Observation: Observing behaviors in their real-world setting (e.g., retail store aisle) vs. in a controlled experimental environment.

Definition of Descriptive Statistics

Descriptive statistics refers to the mathematical and graphical procedures used to collect, organize, summarize, condense, and present the primary characteristics of a specific sample dataset in an interpretable format without making inferences or drawing generalizations about the larger population.

Key metrics include:

• Measures of Central Tendency: Mean, Median, Mode.
• Measures of Dispersion: Range, Variance, Standard Deviation.
• Visual Summaries: Frequency distribution tables, histograms, bar charts, and pie charts.

Essentials of a Good Research Report

A professional, university-grade research report must fulfill the following criteria:

1. Clarity and Precision: Written in concise, objective, unambiguous academic language without colloquialisms.
2. Logical Organization: Structured coherently (Executive Summary, Introduction, Literature Review, Methodology, Results, Discussion, References).
3. Empirical Grounding: All conclusions and recommendations must be directly supported by analyzed data.
4. Methodological Transparency & Rigor: Clearly details sampling design, measurement reliability/validity, and study limitations.
5. Standardized Formatting & Citations: Strictly adheres to APA citation and layout guidelines.

Example of Referencing an Article in APA (7th Edition) Format

Standard Reference Structure:

Author, A. A., & Author, B. B. (Year). Title of the journal article. Title of Journal, VolumeNumber(IssueNumber), PageRange. https://doi.org/xx.xxx/yyyy

Concrete Business Research Example:

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879

Difficulties Faced in Applying Scientific Methods to Social Science Research

While the scientific method was pioneered in the natural sciences (physics, chemistry) under controlled laboratory conditions, applying it to business, management, and social sciences introduces unique epistemological and methodological hurdles:

                  Challenges in Social Science Research
                                    |
         +-------------+------------+------------+-------------+
         |             |                         |             |
Human Subjectivity &  Inability of Pure      Measurement       Ethical
   Dynamic Nature       Experimental         Imprecision     Constraints
                       Control (Noise)      (Abstractness)   (Human Rights)

---

1. Inherent Complexity and Subjectivity of Human Behavior

• Human thoughts, motivations, and attitudes are volatile and influenced by cultural context, mood, and subconscious biases. Unlike inert physical particles, human subjects change their behavior simply because they know they are being observed (Hawthorne Effect).

2. Impossibility of Pure Experimental Control

• In business settings, a researcher cannot hold all external variables constant. Economic shocks, competitor price cuts, government regulations, and organizational politics act as uncontrolled confounding variables that obscure cause-and-effect relationships.

3. Measurement Imprecision of Latent Constructs

• Physical sciences measure tangible constants (mass, velocity, temperature). Business research measures abstract, latent psychological constructs (e.g., job satisfaction, brand loyalty, organizational commitment). These cannot be measured directly and rely on proxy indicators prone to measurement error.

4. Ethical and Practical Constraints

• Researchers cannot arbitrarily manipulate variables that cause severe psychological distress, workplace terminations, or financial losses to test corporate hypotheses.

5. Researcher Bias and Value Judgments

• Social researchers are themselves active participants in society, making absolute detachment and value-free positivism exceptionally difficult to maintain.

Types of Research Interviews Used for Data Collection and Analysis

An interview is a purposeful, guided two-way conversation between an investigator and an interviewee designed to gather valid and reliable research data.

+--------------------------------------------------------------------------+
|                       Types of Research Interviews                       |
+-------------------+--------------------+---------------------------------+
| Structured        | Semi-Structured    | Unstructured / In-Depth         |
| Rigid questions,  | Flexible guide,    | Open-ended, conversational,     |
| quantitative data | balanced probes    | deep exploratory insights       |
+-------------------+--------------------+---------------------------------+

---

1. Structured Interviews

• Characteristics: The interviewer reads an identical, predetermined list of standardized, predominantly closed-ended questions in the exact same order to every participant without deviation.
• Best Used For: Large-scale descriptive and quantitative studies where high comparability across respondents and minimal interviewer bias are paramount.
• Analysis: Easily codified and analyzed quantitatively using statistical software (SPSS).

2. Semi-Structured Interviews

• Characteristics: Guided by a thematic interview guide outlining broad topics, but the interviewer has the freedom to alter question order, rephrase wording, and ask clarifying probing questions based on respondent answers.
• Best Used For: Explanatory business research where the researcher wants to understand the underlying rationale behind organizational decisions.
• Analysis: Analyzed via qualitative thematic analysis and content analysis.

3. Unstructured (In-Depth) Interviews

• Characteristics: Non-directive, open-ended conversation where the researcher introduces a general topic and allows the informant to narrate their experiences freely without a fixed questionnaire.
• Best Used For: Exploratory research, grounded theory, and understanding complex organizational crises or leadership transitions.

---

Interview Administration Modes:

• Face-to-Face Interviews: Allows observation of non-verbal cues and high rapport building.
• Telephone Interviews: Cost-effective, geographically wide reach, faster turnaround.
• Computer-Assisted Personal Interviewing (CAPI / Video Zoom): Combines digital recording and immediate data logging.

Comparative Analysis: Quantitative vs. Qualitative Research Design

Both quantitative and qualitative paradigms provide rigorous, complementary frameworks for business inquiry:

+--------------------------------------------------------------------------+
|                       Research Paradigm Comparison                       |
+-------------------+------------------------------------------------------+
| Quantitative      | Positivist, deductive, numerical, statistical tests  |
| Qualitative       | Interpretivist, inductive, textual, deep context     |
+-------------------+------------------------------------------------------+

---

Comprehensive Comparison Matrix:

Similarities:

• Both adhere to rigorous, systematic inquiry processes.
• Both aim to understand empirical phenomena and solve managerial problems.
• Modern business studies frequently combine both via mixed-methods research.

Applicable Research Designs in Business Research

A research design is the master blueprint specifying the methods and procedures for collecting and analyzing needed information. In business and management research, the choice of research design depends directly on the clarity of the research problem:

Problem Uncertainty                Applicable Research Design
-----------------------------------------------------------------------------
Highly Ambiguous           -->     Exploratory Research (Qualitative/Grounded)
Partially Defined          -->     Descriptive Research (Surveys/Correlational)
Clearly Defined / Precise  -->     Causal / Experimental Research (A/B Testing)

---

1. Exploratory Research Design

• When Applicable: Used when the research problem is ambiguous, ill-defined, or novel, and the management team lacks preliminary insights.
• Techniques: In-depth interviews with industry experts, pilot focus groups, case studies, and secondary literature exploration.
• Business Application: A fintech startup investigating why older consumers in rural Nepal hesitate to adopt mobile wallets.

2. Descriptive Research Design

• When Applicable: Used when the problem is clearly structured and the objective is to accurately map the characteristics, frequencies, or demographic profiles of a market or workforce.
• Techniques: Cross-sectional survey questionnaires, structured observational studies, and correlational studies.
• Business Application: A beverage company assessing customer demographic profiles, consumption frequencies, and brand awareness percentages across major cities.

3. Causal (Explanatory / Experimental) Research Design

• When Applicable: Used when management needs to establish unambiguous cause-and-effect relationships ($X \to Y$).
• Techniques: Laboratory and field experiments, A/B website testing, controlled market price variations.
• Business Application: Testing whether changing an e-commerce checkout button color from blue to green directly causes an increase in conversion rates while holding all other elements constant.

4. Mixed-Methods Design (The Business Gold Standard):

• Triangulates quantitative surveys with qualitative executive interviews to achieve both statistical generalizability and deep behavioral context.

Can Secondary Data Substitute for Primary Methods? (Evaluation & Illustrations)

Critical Evaluation:

“Secondary data can partially substitute for primary methods under specific research scopes (such as macroeconomic trends, industry benchmarking, and historical analysis), but it CANNOT serve as a complete substitute for primary methods in studies requiring proprietary, context-specific, or novel behavioral insights.”

+--------------------------------------------------------------------------+
|                  Primary vs. Secondary Data Substitution                 |
+-----------------------------------+--------------------------------------+
| When Secondary CAN Substitute     | When Primary is STRICTLY Required    |
| - Macroeconomic trend analysis    | - Proprietary customer perceptions   |
| - Industry competitor benchmarks  | - Testing a newly invented prototype |
| - Historical financial modeling   | - Internal corporate culture audits  |
+-----------------------------------+--------------------------------------+

---

1. Circumstances Where Secondary Data Successfully Substitutes Primary Methods:

• Cost and Feasibility: When collecting primary data across millions of citizens is financially prohibitive for an individual firm.
• Illustration: An entrepreneur planning to enter the commercial poultry market in Nepal can rely entirely on secondary census data from the Central Bureau of Statistics (CBS) and Nepal Rastra Bank quarterly reports to determine household meat consumption, feed import tariffs, and interest rates. Conducting a primary census would be wasteful and redundant.

---

2. Circumstances Where Secondary Data FAILS and Primary Methods are Essential:

1. Lack of Specificity / Fit: Secondary data was collected for a different original purpose and often lacks the exact variables or operational definitions needed.
2. Outdated Information: Rapidly changing consumer tastes render past published reports obsolete.
3. Absence of Proprietary Insights: Secondary sources cannot tell a firm how its specific target customers perceive a new packaging design or confidential product prototype.
4. Illustration: If a commercial bank experiences sudden employee dissatisfaction following an internal restructuring, published industry reports on general banking turnover cannot substitute for primary employee exit interviews and internal climate surveys.

Probability vs. Non-Probability Sampling Methods

Sampling is the scientific process of selecting a sufficient number of representative elements from a target population so that studying the sample allows researchers to generalize findings to the entire population.

                           Sampling Techniques
                                    |
         +--------------------------+--------------------------+
         |                                                     |
Probability Sampling                                  Non-Probability Sampling
(Known, non-zero chance of selection;                 (Subjective, unknown probability;
 Allows statistical generalization)                    Exploratory, fast, cost-effective)

---

1. Probability Sampling Methods

Every population element has a known, non-zero probability of being selected, eliminating researcher selection bias:

• Simple Random Sampling: Every element has an equal probability of selection, drawn via random number generators or lottery methods.
• Systematic Sampling: Selecting every $k^{\text{th}}$ element from a population list after a random starting point ($k = N/n$).
• Stratified Random Sampling: Dividing a heterogeneous population into mutually exclusive, homogeneous subgroups (strata) based on a characteristic (e.g., gender, income) and drawing random samples proportionally from each stratum.
• Cluster Sampling: Dividing a geographically dispersed population into heterogeneous natural groups (clusters, e.g., school districts), randomly selecting clusters, and surveying all elements within chosen clusters.

---

2. Non-Probability Sampling Methods

Elements are selected based on subjective researcher judgment or convenience; cannot statistically generalize to the population:

• Convenience Sampling: Selecting elements that are most readily accessible to the researcher (e.g., surveying shoppers walking past a mall entrance).
• Purposive / Judgmental Sampling: Selecting specific knowledgeable individuals based on expert criteria (e.g., interviewing exclusively senior CFOs about corporate taxation).
• Quota Sampling: Ensuring specified sub-group quotas are filled (e.g., 50 males and 50 females) using convenience selection within quotas.
• Snowball (Referral) Sampling: Initial respondents refer other individuals possessing the rare characteristic under study (ideal for hidden populations).