QUESTION BANK
DEPARTMENT : COMPUTER SCIENCE AND ENGINEERING
SUBJECT CODE : OMG 355
SUBJECT NAME : MULTIVARIATE DATA ANALYSIS
SEMESTER : 7th
COURSE COORDINATOR MODULE COORDINATOR PROGRAM COORDINATOR
(Signature with Name and date) (Signature with Name and date) (Signature with Name and date)
IQAC COORDINATOR
(Signature with Name and date)
UNIT I
PART A
Blooms Level
Q. No. Questions CO (BTL1/BTL2)
1. Define univariate analysis. CO1 BTL1
2. What is bivariate analysis? CO1 BTL1
3. List any two examples of multivariate techniques. CO1 BTL1
4. Explain the importance of univariate analysis. CO1 BTL2
5. Differentiate between bivariate and multivariate analysis. CO1 BTL2
6. What is the purpose of multivariate analysis? CO1 BTL1
7. Identify two guidelines for conducting multivariate analysis. CO1 BTL1
8. Explain the term 'classification of multivariate techniques. CO1 BTL2
9. State the difference between dependent and independent variables in CO1 BTL2
multivariate analysis.
10. What is the primary objective of multivariate interpretation? CO1 BTL2
PART B
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
11. Describe the steps involved in conducting a univariate analysis with a real-life CO1 BTL3
example.
12. Apply bivariate analysis to a dataset of your choice and explain the CO1 BTL3
relationship between the variables.
13. Analyze the role of multivariate techniques in market research with CO1 BTL3
appropriate examples.
14. Compare and contrast the different types of multivariate techniques used in CO1 BTL3
data analysis.
15. Evaluate the effectiveness of multivariate analysis in solving complex CO1 BTL4
business problems.
16. Discuss the challenges faced during multivariate analysis and suggest ways to CO1 BTL4
overcome them.
17. Create a multivariate model to predict consumer behavior and justify the CO1 BTL6
choice of variables.
18. Design a research study using multivariate analysis to address a public health CO1 BTL6
issue.
19. Interpret the results of a multivariate analysis performed on a given dataset CO1 BTL5
and provide recommendations based on the findings
20. Explain with examples how multivariate techniques can be used in decision- CO1 BTL5
making processes.
PART C
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
21. Apply factor analysis to a set of data and interpret the factors derived. CO1 BTL3
Develop a cluster analysis to segment a customer base and explain the CO1 BTL6
22. segmentation results.
Critically analyze the impact of multivariate analysis on predictive analytics in CO1
23. financial sectors. BTL5
Discuss the ethical considerations in the interpretation of multivariate analysis CO1
24. results. BTL4
UNIT II
PART A
Blooms Level
Q. No. Questions CO (BTL1/BTL2)
1. Define a research model in the context of multivariate analysis. CO2 BTL1
2. List any two common types of variables used in multivariate analysis. CO2 BTL1
3. Explain the importance of conceptualizing a research model before data CO2
BTL2
collection.
4. What is missing data in the context of multivariate analysis? CO2 BTL1
5. Identify two approaches for dealing with missing data in datasets. CO2 BTL1
6. Differentiate between dependent and independent variables in a CO2 BTL2
research model.
7. Describe what is meant by the assumptions of multivariate analysis. CO2 BTL2
8. Give an example of a multivariate analysis technique. CO2 BTL1
9. State two reasons why testing assumptions is critical in multivariate CO2 BTL2
analysis.
10. What is the role of data collection in the research process? CO2 BTL2
PART B
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
11. Illustrate the steps involved in conceptualizing a research model with CO2 BTL3
variables.
12. Develop a hypothetical research model and explain how you would CO2
identify and classify variables. BTL4
13. Evaluate the strengths and weaknesses of different methods for CO2
handling missing data in multivariate analysis. BTL5
14. Explain the procedure for testing multivariate analysis assumptions and CO2
its impact on the results. BTL3
15. Design a strategy to collect data for a multivariate analysis study on CO2 BTL6
consumer behavior.
16. Analyze the impact of missing data on the validity and reliability of CO2 BTL4
multivariate analysis results.
17. Discuss how to ensure that a dataset is suitable for multivariate CO2 BTL5
analysis.
18. Construct a detailed plan for handling missing data in a large dataset, CO2 BTL6
using multiple imputation methods.
19. Apply a multivariate analysis technique to a given dataset and interpret CO2 BTL3
the results.
20. Critically analyze the importance of testing for multicollinearity in CO2 BTL5
multivariate analysis.
PART C
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
Formulate a research question and outline a complete multivariate CO2
21. analysis plan to address it.
BTL6
Discuss the ethical considerations in data collection and handling CO2
22. missing data in multivariate research. BTL4
Compare and contrast the use of factor analysis and cluster analysis in CO2
23. multivariate research.
BTL4
Create a research design that incorporates multivariate analysis for CO2
24. evaluating the effectiveness of a new teaching method. BTL6
UNIT III
PART A
Blooms Level
Q. No. Questions CO (BTL1/BTL2)
1. Define Multiple Linear Regression. CO3 BTL1
2. List the assumptions of Multiple Linear Regression Analysis. CO3 BTL1
3. What is meant by the estimated regression function? CO3 BTL1
4. Explain the term 'Residual' in regression analysis. CO3 BTL2
5. Differentiate between simple and multiple linear regression. CO3 BTL2
6. Define Factor Analysis. CO3 BTL1
7. What is the purpose of factor analysis? CO3 BTL2
8. List two methods used in factor analysis. CO3 BTL1
9. Explain the term 'Eigenvalue' in the context of factor analysis. CO3 BTL2
10. What is meant by the validation of a regression model? CO3 BTL2
PART B
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
11. Describe the steps involved in conducting a Multiple Linear Regression CO3 BTL3
Analysis.
12. Develop a regression model using a given dataset and interpret the results. CO3 BTL3
13. Evaluate the significance of regression coefficients using hypothesis testing. CO3 BTL4
14. Assess the goodness-of-fit of a regression model using R-squared and CO3 BTL5
Adjusted R-squared values.
15. Formulate a strategy for validating a Multiple Linear Regression model. CO3 BTL6
16. Compare and contrast Principal Component Analysis and Factor Analysis. CO3 BTL4
17. Interpret the output of a factor analysis including factor loadings and CO3 BTL5
communalities.
18. Design an experiment to collect data suitable for factor analysis. CO3 BTL6
19. Analyze a case study to identify factors affecting a particular outcome using CO3 BTL4
regression analysis.
20. Evaluate the impact of multicollinearity on regression models and suggest CO3 BTL5
remedies.
PART C
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
21. Critique the use of stepwise regression in model selection. CO3 BTL4
Construct a factor analysis model and interpret the scree plot and factor CO3 BTL6
22. rotation results.
23. Propose methods for improving the accuracy of a regression model. CO3 BTL6
Examine a real-world application of factor analysis and discuss its CO3
24. implications. BTL4
UNIT IV
PART A
Blooms Level
Q. No. Questions CO (BTL1/BTL2)
1. Define Confirmatory Factor Analysis (CFA). CO4 BTL1
2. List the key components of Structural Equation Modelling (SEM). CO4 BTL1
3. Describe the role of latent variables in SEM. CO4 BTL2
4. Differentiate between a mediation model and a moderation model. CO4 BTL2
5. Explain the purpose of longitudinal studies in research. CO4 BTL2
6. Identify two advantages of using CFA over Exploratory Factor Analysis CO4 BTL1
(EFA).
7. What are the assumptions underlying Structural Equation Modelling? CO4 BTL1
8. Describe the significance of path diagrams in SEM. CO4 BTL2
9. State the importance of fit indices in Confirmatory Factor Analysis. CO4 BTL1
10. Mention two common software tools used for conducting SEM. CO4 BTL1
PART B
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
11. Explain the process of conducting Confirmatory Factor Analysis and CO4 BTL3
discuss how it differs from Exploratory Factor Analysis.
12. Discuss the steps involved in Structural Equation Modelling and CO4 BTL3
provide an example of its application in social sciences research.
13. Analyze a hypothetical research scenario and demonstrate how a CO4 BTL4
mediation model can be used to explain the relationship between
variables.
14. Evaluate the effectiveness of moderation models in psychological CO4 BTL5
research, providing examples to support your argument.
15. Create a research proposal using longitudinal study design to CO4 BTL6
investigate a public health issue, detailing the rationale and
methodology.
16. Compare and contrast mediation and moderation models, and discuss CO4 BTL4
their application in behavioral studies with examples.
17. Illustrate the steps involved in validating a measurement model using CO4 BTL5
CFA, and critique potential challenges in the process.
18. Assess the role of latent variables in SEM and discuss how they CO4 BTL5
enhance the understanding of complex relationships in data.
19. Design a SEM model to study the impact of social support on mental CO4 BTL6
health outcomes, including both mediation and moderation effects.
20. Discuss the importance of longitudinal studies in tracking changes over CO4 BTL4g
time and identify potential limitations of this approach.
PART C
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
Explain the concept of model fit in CFA and SEM, and discuss how CO4
21. different fit indices are interpreted in practice. BTL3
Propose a research study utilizing SEM to examine the effects of CO4
22. educational interventions on student performance, and describe the BTL6
hypothesized model.
Evaluate the methodological considerations when implementing a CO4
23. mediation model in psychological research, including sample size and BTL5
measurement reliability.
Analyze the challenges and strategies for dealing with missing data in CO4
24. longitudinal studies and their impact on research outcomes. BTL4
UNIT V
PART A
Blooms Level
Q. No. Questions CO (BTL1/BTL2)
1. Define Multiple Discriminant Analysis. BTL1
CO5
2. What is the primary objective of Logistic Regression? CO5 BTL1
3. List two applications of Cluster Analysis in marketing. CO5 BTL2
4. Explain the difference between Multidimensional Scaling and Conjoint CO5 BTL2
Analysis.
5. What is a centroid in Cluster Analysis? CO5 BTL1
6. Describe the term 'discriminant function' in Multiple Discriminant CO5 BTL2
Analysis.
7. Identify the key assumption of Logistic Regression. CO5 BTL1
8. How does Conjoint Analysis help in product development? CO5 BTL2
9. Mention two types of data suitable for Multidimensional Scaling. CO5 BTL1
10. State the purpose of the eigenvalues in Multiple Discriminant Analysis. CO5 BTL2
PART B
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
11. Apply Multiple Discriminant Analysis to a scenario where a company CO5 BTL3
wants to classify customers into different loyalty groups. Explain the
steps involved.
12. Analyze a dataset using Logistic Regression to predict customer churn. CO5 BTL4
Discuss the methodology and interpret the results.
13. Compare and contrast Cluster Analysis and Multiple Discriminant CO5 BTL4
Analysis with examples.
14. Design a marketing strategy based on the results of a Cluster Analysis. CO5 BTL5
How would you implement this strategy?
15. Evaluate the effectiveness of Multidimensional Scaling in creating CO5
BTL5
perceptual maps for competitive analysis.
16. Develop a research framework using Conjoint Analysis to determine CO5
consumer preferences for a new smartphone feature. BTL6
17. Interpret the outcomes of a Logistic Regression analysis in terms of CO5
BTL4
odds ratios and their implications for decision-making.
18. Assess the limitations and strengths of Multiple Discriminant Analysis CO5 BTL5
in comparison to Logistic Regression.
19. Create a clustering model to segment a customer base for a retail store. CO5 BTL6
Discuss how to determine the optimal number of clusters.
20. Explain how Multidimensional Scaling can be used to understand CO5 BTL4
consumer perceptions of competing brands in a marketplace.
PART C
Blooms Level
Q. No. Questions CO (BTL3 to BTL6)
Critique a real-world application of Conjoint Analysis in a product CO5 BTL5
21. launch. What improvements would you suggest?
Propose a framework for using Logistic Regression to improve a CO5 BTL6
22. company’s credit risk assessment process.
Demonstrate how Cluster Analysis can be used to identify distinct CO5 BTL3
23. segments within a large dataset. Provide a hypothetical example.
Analyze the results of a Conjoint Analysis study. How would you use these CO5 BTL4
24. results to recommend product changes?
