Final Exam Notes: Complete and Expanded PDF

Final Exam Notes: Complete and Expanded ======================================= Field Work & Errors in Research ------------------------------- \*\*When to Use and Criteria for Good Onsite Data Collection:\*\*\ - Onsite data collection is necessary when the target population cannot be reached thro...

Final Exam Notes: Complete and Expanded ======================================= Field Work & Errors in Research ------------------------------- \*\*When to Use and Criteria for Good Onsite Data Collection:\*\*\ - Onsite data collection is necessary when the target population cannot be reached through other means (e.g., phone or online surveys).\ - Examples include observing consumer behavior in a retail store or conducting interviews at an event.\ - Criteria:\ - Diverse locations for broader representation.\ - Systematic sampling to reduce biases.\ \ \*\*Difference Between Sampling and Non-Sampling Errors:\*\*\ 1. \*\*Sampling Errors\*\*:\ - Result from how the sample is selected.\ - Measurable and controllable with probability sampling.\ 2. \*\*Non-Sampling Errors\*\*:\ - Arise from human or process errors during data collection.\ - Examples: respondent misunderstandings, interviewer mistakes.\ - Cannot be precisely measured but can be reduced.\ \ \*\*Types of Non-Sampling Errors:\*\*\ 1. \*\*Respondent Errors:\*\*\ - Intentional (e.g., social desirability bias, nonresponses).\ - Unintentional (e.g., misunderstanding, fatigue, distractions).\ 2. \*\*Fieldworker Errors:\*\*\ - Intentional (e.g., falsifying data).\ - Unintentional (e.g., unclear instructions, fatigue).\ \ \*\*Data Collection Quality Control:\*\*\ - \*\*Intentional Fieldworker Errors\*\*:\ - Supervision and validation of responses.\ - \*\*Unintentional Fieldworker Errors\*\*:\ - Role-playing during training and enforced breaks.\ - \*\*Intentional Respondent Errors\*\*:\ - Assuring anonymity, offering incentives, and using validation checks.\ - \*\*Unintentional Respondent Errors\*\*:\ - Clear instructions, reversed scales, and prompters.\ \ \*\*Types of Nonresponse Errors:\*\*\ - Refusals: Participants decline to take part.\ - Break-offs: Participants stop midway.\ - Item Omissions: Skipping specific questions. Database Structure, Coding, and Data Validation ----------------------------------------------- \*\*Principles of Coding Response Categories:\*\*\ - Assign unique numeric codes to each category for consistency.\ - Codes are comma-separated, with no spaces.\ \ \*\*Database Structure in XLDA:\*\*\ - \*\*Columns\*\*: Represent survey variables.\ - \*\*Rows\*\*: Represent individual respondent data.\ - \*\*Data Sheet\*\*: Stores raw data for analysis.\ - \*\*Define Variables Sheet\*\*: Contains variable descriptions and coding.\ \ \*\*Steps to Validate Data:\*\*\ 1. Check for missing entries or duplicates.\ 2. Ensure proper formatting in variable names and labels.\ 3. Confirm consistency in Value Codes (e.g., no extra spaces).\ 4. Verify labels for proper capitalization.\ 5. Save the file before submission. Descriptive Data Analysis ------------------------- \*\*Five Main Types of Data Analysis:\*\*\ 1. \*\*Descriptive Analysis\*\*: Summarizes data using averages and percentages.\ 2. \*\*Inferential Analysis\*\*: Draws conclusions about the population.\ 3. \*\*Difference Analysis\*\*: Identifies significant differences between groups.\ 4. \*\*Associative Analysis\*\*: Examines relationships between variables.\ 5. \*\*Predictive Analysis\*\*: Forecasts future trends based on current data.\ \ \*\*Why We Summarize Data:\*\*\ - To make large datasets understandable.\ - To identify trends and patterns efficiently.\ \ \*\*Measures of Central Tendency:\*\*\ 1. \*\*Mean\*\*: The average value.\ 2. \*\*Median\*\*: The middle value in a dataset.\ 3. \*\*Mode\*\*: The most frequently occurring value.\ \ \*\*Measures of Variability:\*\*\ - Range: The difference between the highest and lowest values.\ - Standard Deviation: Measures data dispersion around the mean.\ \ \*\*Grid Questions Analysis:\*\*\ 1. Use averages to identify trends.\ 2. Percentages to identify dominant categories.\ 3. Cross-tabulations for segment-specific insights.\ \ \*\*Analyzing Open-Ended and Multiple Response Questions:\*\*\ - Categorize responses into themes.\ - Calculate percentages for recurring themes. Calculating Sample Size ----------------------- \*\*Definitions of Sampling Method and Sample Size:\*\*\ - Sampling Method: The process used to select participants.\ - Sample Size: The number of participants.\ \ \*\*Key Concepts:\*\*\ - \*\*Representativeness\*\*: Ensures the sample reflects the population.\ - \*\*Confidence Interval Method\*\*: Defines the range of acceptable error (e.g., ±5%).\ - \*\*Population Variability\*\*: Measures diversity within the sample.\ - \*\*Confidence Level\*\*: Typically 95% (assumes 95% accuracy).\ \ \*\*Steps to Calculate Sample Size in XLDA:\*\*\ 1. Input population size, confidence level, and variability.\ 2. Interpret the output to ensure acceptable error margins. Associations Between Variables ------------------------------ \*\*Purpose of Analyzing Associations Between Variables:\*\*\ - To identify relationships between different data points (e.g., sales and marketing spend).\ - Helps in predicting outcomes and understanding trends.\ \ \*\*Four Types of Relationships Between Variables:\*\*\ 1. \*\*Non-Monotonic\*\*: General association without a consistent direction (e.g., brand and type of product purchased).\ 2. \*\*Monotonic\*\*: One variable consistently increases or decreases as the other changes (e.g., age and income).\ 3. \*\*Linear\*\*: A straight-line relationship (e.g., price and quantity sold).\ 4. \*\*Curvilinear\*\*: A nonlinear relationship (e.g., sales and promotional spend showing diminishing returns).\ \ \*\*Three Characteristics in Crosstab and Correlation Analyses:\*\*\ 1. \*\*Presence\*\*: Is there a relationship? (e.g., statistical significance).\ 2. \*\*Direction\*\*: Is the relationship positive or negative?\ 3. \*\*Strength\*\*: How strong is the association?\ \ \*\*Statistical Significance and Interpretation:\*\*\ - Significance (p ≤ 0.05) indicates the likelihood the result is not due to chance.\ - Higher significance means stronger confidence in findings.\ \ \*\*Correlation Analysis:\*\*\ - \*\*Purpose\*\*: Measures the strength and direction of a linear relationship between variables.\ - \*\*Correlation Coefficient (r)\*\*: Values range from -1 to +1:\ - Positive (+1): Perfect positive relationship.\ - Negative (-1): Perfect negative relationship.\ - Zero (0): No relationship.\ - \*\*Steps in XLDA\*\*:\ 1. Select variables.\ 2. Interpret \"r\" value and significance (p ≤ 0.05). Crosstab Analysis ----------------- \*\*Purpose of Crosstab Analysis:\*\*\ - Identifies relationships between categorical variables (e.g., age and brand preference).\ - Displays data in a table format with row and column percentages.\ \ \*\*Components of a Crosstab:\*\*\ 1. \*\*Column Variable\*\*: Independent variable (e.g., gender).\ 2. \*\*Row Variable\*\*: Dependent variable (e.g., product rating).\ \ \*\*Steps in XLDA:\*\*\ 1. Select variables.\ 2. Analyze row and column percentages.\ 3. Evaluate significance using Chi-Square tests.\ \ \*\*Evaluation:\*\*\ 1. \*\*Presence\*\*: Use Chi-Square (p ≤ 0.05).\ 2. \*\*Direction\*\*: Identify general trends (e.g., higher percentages for specific groups).\ 3. \*\*Strength\*\*: Look for patterns across rows/columns. Inference, Confidence Interval, and Hypothesis Testing ------------------------------------------------------ \*\*Concept of Inference in Marketing:\*\*\ - Uses sample data to draw conclusions about a population.\ \ \*\*Parameter Estimation vs. Hypothesis Testing:\*\*\ 1. \*\*Parameter Estimation\*\*: Estimates population parameters (e.g., average sales).\ 2. \*\*Hypothesis Testing\*\*: Tests assumptions about populations (e.g., campaign effectiveness).\ \ \*\*Confidence Interval (CI):\*\*\ - Range of values likely to contain the population parameter.\ - Example: \"We are 95% confident that average sales are between \$10K and \$12K.\"\ \ \*\*Steps in Hypothesis Testing:\*\*\ 1. Formulate Null Hypothesis (H₀) and Alternative Hypothesis (H₁).\ 2. Collect data and calculate test statistic.\ 3. Reject or fail to reject H₀ based on p-value. Segmentation: Testing of Differences ------------------------------------ \*\*Why Testing for Differences is Important:\*\*\ - Identifies variations in behavior, preferences, or demographics.\ - Provides actionable insights for marketing strategies.\ \ \*\*Tests of Significance:\*\*\ 1. \*\*Non-Metric Data\*\*: Compare percentages using Chi-Square.\ 2. \*\*Metric Data\*\*: Compare averages using t-tests or ANOVA.\ \ \*\*Steps in XLDA for Analysis:\*\*\ 1. Select target (dependent) and grouping (independent) variables.\ 2. Run appropriate tests (e.g., t-test, ANOVA).\ 3. Interpret results for significance and actionable differences. ANOVA and Conjoint Analysis --------------------------- \*\*ANOVA (Analysis of Variance):\*\*\ - Compares means across multiple groups (e.g., satisfaction ratings by age).\ - \*\*Grouping Variable\*\*: Non-metric (e.g., age groups).\ - \*\*Target Variable\*\*: Metric (e.g., satisfaction).\ \ \*\*Conjoint Analysis:\*\*\ - Evaluates trade-offs among product features (e.g., price, color).\ - Identifies feature preferences and forecasts product success.\ \ \*\*Steps:\*\*\ 1. Rank feature combinations.\ 2. Analyze segments for preferred attributes. The Marketing Research Report ----------------------------- \*\*Importance of the Marketing Research Report:\*\*\ - Communicates findings, insights, and recommendations.\ - Acts as the final deliverable to stakeholders.\ \ \*\*Sections of the Report:\*\*\ 1. \*\*Front Matter\*\*: Title page, table of contents, executive summary.\ 2. \*\*Body\*\*: Objectives, methodology, findings, and recommendations.\ 3. \*\*End Matter\*\*: References and appendices.\ \ \*\*Guidelines:\*\*\ - Use visuals for clarity.\ - Group analyses by research objectives.\ - Write in a clear and concise manner.