Business Research Methods - Unit 5 Notes

## 1. Data Analysis

### Definition

- Data Analysis: The process of systematically applying statistical and logical

techniques to describe, summarize, and evaluate data (Shamoo & Resnik, 2003).

- Purpose:

- Draw inductive inferences from data.

- Distinguish the signal (phenomenon of interest) from noise (statistical fluctuations).

### Types of Data Analysis

- Quantitative Analysis: Uses statistical procedures to analyze numerical data.

- Qualitative Analysis: Involves iterative processes where data collection and analysis
occur simultaneously, identifying patterns in observations (Savenye & Robinson, 2004).

- Approaches: Field study, ethnography, content analysis, oral history, biography,

unobtrusive research.

- Data Forms: Field notes, documents, audiotapes, videotapes.

### Importance of Data Integrity

- Accurate analysis ensures valid research findings.

- Improper statistical analyses can:

- Distort scientific findings.

- Mislead readers (Shepard, 2002).

- Negatively impact public perception of research.

- Non-statistical data analysis requires equal attention to integrity.

### Considerations in Data Analysis

- Skills: Researchers need analytical skills to interpret data.

- Concurrent Methods: Select data collection and analysis methods together.

- Bias: Ensure unbiased inferences.

- Subgroup Analysis: Avoid inappropriate subgroup breakdowns.

- Norms: Follow discipline-specific standards.

- Statistical Significance: Determine significance accurately.

- Outcome Measurements: Use clearly defined, objective measures.

- Honesty: Provide accurate analysis.

- Presentation: Consider environmental/contextual issues, recording methods, and text

partitioning (for qualitative data).

- Staff Training: Ensure reliability and validity through trained analysts.

- Extent of Analysis: Balance depth and breadth of analysis.

---

## 2. Data Processing

### Definition

- Data Processing: Preparing collected data for analysis through editing, coding,
classification, and tabulation to ensure efficiency and accuracy.

### Steps in Data Processing

1. Editing:

- Reviewing and adjusting survey data to control quality.

- Purpose: Ensure consistency, adequacy, and correctness; detect errors and outliers.

- Methods: Manual, computer-assisted, or hybrid.

- Types of Editing:

- Validity and Completeness: Ensure no missing or empty fields.

- Range: Verify data falls within specified boundaries.

- Duplicate Data Entry: Prevent repetition of data entries.

- Logical Consistency: Check connections between data fields.

- Outliers: Identify and verify extreme or unusual values.

2. Coding:

- Definition: Assigning short words or phrases to describe the meaning of sentences,

phrases, or paragraphs.

- Purpose: Quantify qualitative data, simplify analysis, and extract meaning from
responses.

- Process:

- Derive preliminary codes from observed data (e.g., interviews, questionnaires).

- Refine preliminary codes into final codes for evaluation.

- Assign numerical values or percentages to codes for analysis.

- Types of Codes:

- Preliminary Codes: Initial, broad codes to condense data.

- Final Codes: Refined, meaningful themes or topics for analysis.

- Note: Coding summarizes data meaningfully without losing critical information.

3. Classification: Organizing data into categories (e.g., qualitative, quantitative,

temporal, spatial).

4. Tabulation: Arranging data in tables for easy interpretation.

---

## 3. Tabular Representation of Data

### Definition

- Tabular Representation: Organizing data in rows and columns for clarity and ease of
interpretation.

- Advantages: Attractive, easy to read, handles large datasets effectively.

### Components of Data Tables

1. Table Number: Unique identifier for referencing.

2. Title: Describes the data, time period, location, and classification type.

3. Headnotes: Additional information about units (in brackets).

4. Stubs: Row titles indicating data content.

5. Caption: Column titles indicating data content.

6. Body/Field: Main content, with each item called a “cell.”

7. Footnotes: Supplementary information (rarely used).

8. Source: Citation for secondary data.

### Construction Guidelines

- Title aligns with study objectives.

- Place comparable rows/columns close together.

- Stubs can be on the right for lengthy rows.

- Use singular headings (e.g., “good” not “goods”).

- Include footnotes only if necessary.

- Maintain uniform column sizes.

- Avoid abbreviations in headings.

- Clearly specify units above columns.

### Classification Types

1. Qualitative: Based on non-quantifiable attributes (e.g., rural-urban, boys-girls).

2. Quantitative: Based on numerical attributes (e.g., income levels).

3. Temporal: Based on time periods (e.g., yearly sales).

4. Spatial: Based on location (e.g., regional sales).

### Frequency and Frequency Tables

- Frequency: Number of times a data value occurs (e.g., score of 80 appears 4 times,
frequency = 4).

- Frequency Table: Lists data values in ascending order with their frequencies.
- Class Intervals: For spread-out data, group values into intervals (5–10 rows
recommended).

- Start intervals at multiples of the group size (e.g., 5, 10, 15 for size 5).

- Frequency of a group is the number of values in that interval.

Example: Marks of 20 students: 6, 7, 5, 7, 7, 8, 7, 6, 9, 7, 4, 10, 6, 8, 8, 9, 5, 6, 4, 8.

| Marks | Frequency |

|-------|-----------|

|4 |2 |

|5 |2 |

|6 |4 |

|7 |6 |

|8 |4 |

|9 |2 |

| 10 | 1 |

---

## 4. Graphical Representation of Data

### Definition

- Graphical Representation: Visual presentation of data using graphs to show trends,

comparisons, or distributions.

- Uses: Time series, frequency distributions, mode/median identification.

### Construction of Graphs

- Axes: X-axis (horizontal, categories), Y-axis (vertical, values).

- Origin: Intersection of X and Y axes.

- Point Location: Defined by distances from X and Y axes (e.g., point P at (3, 5)).
### General Rules

1. Title: Clearly indicate the subject.

2. Unit of Measurement: Specify below the title.

3. Scale: Choose a scale to represent data accurately.

4. Index: Explain colors, lines, or designs used.

5. Data Sources: Cite sources at the bottom.

6. Simplicity: Ensure a layman can understand.

7. Neatness: Use appropriate sizes, lettering, and colors.

### Merits of Graphs

- Easy to understand at a glance.

- Attractive compared to tables.

- Saves time in interpretation.

- Facilitates comparisons across time or regions.

- Requires no prior statistical knowledge.

- Useful for forecasting, interpolation, and extrapolation.

### Limitations of Graphs

- Lacks complete accuracy.

- Shows only selected characteristics.

- Cannot support statements alone.

- Not a substitute for tables.

- May be difficult for laymen to interpret.

- Actual values may be unclear due to trends.

### Types of Graphs

1. Bar Charts:

- Rectangular bars compare categories (vertical or horizontal).

 - X-axis: Categories; Y-axis: Values (e.g., percentages).

- Types:

- Grouped Bar Graph: Shows subgroups (e.g., double bar for two subgroups).

- Stacked Bar Chart: Subgroups stacked on the same bar.

- Segmented Bar Chart: Stacked bars showing 100% of a value.

- Use: Categorical data (e.g., movie genres).

- Note: Leave gaps between bars to distinguish from histograms.

2. Histograms:

- Used for continuous data (e.g., height).

- No gaps between bars, unlike bar charts.

3. Pie Charts:

- Circular graph showing contributions to a whole.

- Each wedge represents a category’s proportion.

- Use: Nominal or ordinal data (e.g., county of birth, satisfaction levels).

- Guidelines:

- Best for 6 or fewer categories.

- Include percentages next to slices.

- Avoid 3D effects or exploded charts to prevent distortion.

- Use for part-to-whole comparisons, not group-to-group.

- Consider “other” category for small slices.

---

## 5. Hypothesis

### Definition

- Hypothesis: A specific, testable prediction about what will happen in a study.

- Types:

- Formal Hypothesis: Used in confirmatory studies.

- Exploratory Studies: May lack hypotheses, aiming to develop them for future
research.

- Note: A study may have one or multiple hypotheses.

### Qualities of a Good Hypothesis

- Close to observable phenomena.

- Simple and clear.

- Testable using available techniques.

- Relevant to the research problem.

- Specific and focused.

- Fruitful for new discoveries.

- Consistent with existing knowledge.

---

## 6. Report Writing

### Definition

- Reports: Documents communicating research findings, analysis, and

recommendations to a specific audience for decision-making.

- Purpose: Transmit information clearly and concisely to experts, executives, or

laypeople.

### Considerations

- Audience Needs: Determine formality, length, data types, explanations, and visual
sophistication.

- Clarity: Be thorough, concise, and specific (avoid vague statements like “results were
promising”).
### Body of a Research Report

1. Introduction:

- States the problem, its significance, and technical goals.

- Includes background, purpose, history/theory, literature review, or methods.

2. Review of Literature:

- Discusses prior work or theory relevant to the study (if not in the introduction).

3. Research Methodology:

- Research Design: Qualitative, quantitative, or mixed; exploratory, descriptive,

analytical, or experimental.

- Sampling Design: Target population, sample size, sampling method, justification.

- Data Collection: Primary (surveys, interviews) and secondary (books, journals)

sources.

- Research Instruments: Questionnaires, interview guides, scaling techniques.

- Pilot Testing: If conducted.

- Data Analysis Techniques: Statistical tools (e.g., regression, ANOVA).

- Limitations: Biases, constraints, or data limitations.

4. Data Analysis & Interpretation:

- Presents data (tables, graphs) and interprets findings.

- Considers the audience’s context and needs.

- Analyzes trends, uncertainties, limitations, and assumptions.

5. Conclusions & Recommendations:

- Conclusions: Generalizations deduced from results.

- Recommendations: Actions or further work needed.

- Example:
 - Results: Shoes got wet, leather cracked, soles separated.

- Conclusions: Shoes are not waterproof or durable despite high price.

- Recommendations: Avoid puddles, check waterproofing warranties.

6. Bibliography: Lists all references cited.

7. Annexures: Supplementary materials (e.g., raw data, questionnaires).

### Types of Reports

1. Preliminary Reports:

- Problem/Needs Analysis: Examines complex or ill-defined issues.

- Project Plans: Outline project conceptualization and resource needs.

- Comparison Reports: Analyze relative advantages of options.

- Feasibility Reports: Assess viability of a specific option.

- Recommendation Reports: Advise specific actions.

- Proposals: Request support for projects (e.g., RFPs).

- White Papers: Provide background for decisions.

- Marketing Plans: Review and revise marketing strategies.

2. Progress Reports:

- Inform on project status and preliminary findings.

- Examples: Test reports, strategic plans, business plans, design reports.

3. Periodic Reports:

- Generated regularly or as needed.

- Examples:

- Annual Reports: Inform stakeholders about financial performance.

- Performance Reports: Evaluate individual/organizational performance.

- Activity Reports: Document employee accomplishments.

 - Trip Reports: Justify off-site activities.

- Presentation Reports: Use PowerPoint for reporting.

---

## 7. Publication Ethics

### Definition

- Publication Ethics: Rules ensuring integrity in publishing research, protecting

intellectual property, and preventing unauthorized reuse.

### Importance

- Promotes truth, knowledge, and error avoidance.

- Ensures trust, accountability, and fairness in collaborative work.

- Builds public support for research.

- Upholds moral and social values (e.g., social responsibility, compliance with laws).

### Perspectives

- Moral: Publishing false or fabricated data.

- Professional: Failing to credit contributors.

- Legal: Violating copyrights or intellectual property.

### Ethical Issues

1. Data Fabrication: Creating fake data or results.

2. Data Falsification: Manipulating data to misrepresent findings.

3. Plagiarism: Using others’ work without citation.

4. Self-Plagiarism: Reusing own work without citation.

5. Redundant/Salami Publication: Splitting data into minimal publishable units.

6. Dual Submissions: Submitting the same article to multiple journals.

7. Inappropriate Authorship:

- Gift Authorship: Including non-contributors as co-authors.

- Ghost Authorship: Excluding contributors from authorship.

8. Conflict of Interest: Undisclosed financial/personal interests affecting judgments.

9. Citation Manipulation: Excessive or irrelevant citations to boost metrics.

---

## 8. Summary

- Data Analysis: Involves quantitative and qualitative techniques to draw meaningful

inferences.

- Data Processing: Editing, coding, classification, and tabulation ensure data quality.

- Tabular Representation: Organizes data in rows and columns for clarity.

- Graphical Representation: Visualizes data using bar charts, histograms, and pie
charts.

- Hypothesis: Testable predictions with qualities like clarity and relevance.

- Report Writing: Structured communication of findings, tailored to audiences.

- Publication Ethics: Ensures integrity, accountability, and fairness in research

dissemination.

---

## 9. Daily Quiz Answers

1. Hypothesis cannot be stated in:

- Answer: General terms (hypotheses must be specific and testable).

2. Formulation of hypothesis may not be necessary in:

- Answer: Fact-finding (historical) studies (exploratory studies may not require formal
hypotheses).

3. Data organized in rows and columns:

 - Answer: Tabular.

4. Not a part of a table:

- Answer: Histogram (tables include headnotes, footnotes, stubs).

---

## 10. MCQ Answers

1. A good hypothesis should be:

- Answer: All of these (testable, precise, limited scope).

2. While writing a research report, a researcher must:

- Answer: All of the above (logical order, avoid numerical figures at sentence start,
compare results).

3. Method of presentation of data:

- Answer: All of the above (tabular, textual, graphical).

4. A particular data value which occurs number of times:

- Answer: Frequency.

5. Shows information about subgroups:

- Answer: Grouped bar graph.

6. Stacked bar chart showing 100% of discrete value:

- Answer: Segmented bar graph.

---

## 11. Expected Questions for University Exam

1. Explain the process of data analysis and its importance in research.

- Discuss quantitative/qualitative analysis, data integrity, and considerations like bias

and validity.

2. Describe the steps involved in data processing.

- Cover editing, coding, classification, and tabulation with examples.

3. What are the components and types of tabular representation of data?

- List components (table number, title, etc.) and classification types (qualitative,
quantitative, temporal, spatial).

4. Discuss the merits and limitations of graphical representation.

- Include bar charts, histograms, pie charts, and general rules for construction.

5. What are the qualities of a good hypothesis? Provide examples.

- List qualities and illustrate with a specific hypothesis.

6. Outline the structure of a research report and explain its key sections.

- Detail introduction, methodology, analysis, conclusions, and recommendations.

7. What are the types of reports and their purposes?

- Compare preliminary, progress, and periodic reports with examples.

8. Explain key publication ethics and their importance in research.

- Cover data fabrication, plagiarism, authorship issues, and conflict of interest.