Module 6: Final Milestone

BUA 6315: Final Project Guidelines and Grading Guide

Overview

The final project for this course is the creation of a data analysis report.

Data is not very useful if not converted into insightful information that is clearly articulated in written or verbal language. The final project for this

course is analyzing a big dataset and writing a data analysis report based on observations and analysis of a big dataset to convey the information

in written form.

This project is designed to provide you with hands-on strategic analysis experience. You assume the role of data analyst whose primary task is to

characterize an opportunity given the dataset and implement a data driven solution that could generate insights or value that may affect the firms or

organizations that use this dataset. Your written report is intended for a nontechnical audience who may not be familiar with the details of the

statistical and computational methods.

The project is divided into 2 milestones, which will be submitted at various points throughout the course to scaffold learning and ensure quality final

submissions. These milestones will be submitted in Modules 2 and 4.

In this assignment you will demonstrate your mastery of the following course outcomes:

● Explain data exploration, data preparation, and data analysis techniques in descriptive, predictive, and prescriptive modelling.

● Apply data exploration, data preparation, and data analysis techniques in descriptive, predictive, and prescriptive modelling to support

business-decision making.

● Interpret technical information to both technical and non-technical audiences in writing and graphically.

● Apply data analysis tools such as Excel and Analytic Solver to a given data set to estimate descriptive, predictive, and prescriptive models.

Prompt In your first milestone, you will first select a dataset from the following three data sets (which can be accessed in Blackboard), which you will use for

all milestones and your final project. You can find the description of the variables and data dictionary in the Appendix A of your textbook.

● College Admission Data

NOTE ABOUT SECTIONS 3-5

In Sections 3-5, the Regression Analysis subsections should be revisions based on feedback received for Milestone 2: Regression Analysis.

In Sections 3 and 4, for detailed instructions on how to complete the data mining subsections, please refer to the Final Project handout for your

specific dataset (available in Blackboard):

1. College Admission Data

2. Tech Sales Rep Data

3. Longitudinal Survey Data

● Tech Sales Rep Data

● Longitudinal Survey Data

Your final data analysis report submission must include the following sections, including all subsections and critical elements listed:

1. Executive Summary (one page in length or less)

This section should be a succinctly written management summary or overview of data analysis you have done in such a way that readers can

rapidly become acquainted with a large body of material without having to read it all. NOTE: Even though this section is at the beginning, it is usually

written last.

To complete this section on your report, you must include the following critical elements:

A. Describe in detail the opportunity that may impact the firm or organization’s operations or strategy that use the data set you have chosen for

your analysis.

B. Explain why and how data analysis offers new opportunities for this firm or organization.

2. Overview of the Data (two to three pages)

The overall objective of this section is to clearly explain the details of the data set you choose. Specifically, you will inspect, clean, visualize, and

transform the data. NOTE: This section should integrate feedback received from Milestone 1: Data Exploration, Visualization, and Pre-processing.

To complete this section on your report, you must include the following critical elements:

A. Determine whether there are missing values and explain how you handle missing values.

B. Identify potential outliers and explain how you handle outliers.

C. Subset the data based on variables and compare summary statistics of subsets.

D. Create visualizations of the data.

3. Methodology (one to two pages)

In this section, you will describe in detail the methods used to analyze your data. You will specify the motivation for using each method and explain

what manner of relationship you expect to find.

To complete this section on your report, you must include the following critical elements:

A. Regression Analysis: For this subsection, you will discuss the models (multivariate regression, logistic regression) that you applied in

Milestone 2. For your final submission you should integrate feedback received and include the revised version in your report. Specifically, be

sure to address the following:

i. Using the best models you chose for both multivariate and logistic regressions, specify the motivation for using both multivariate

regression and logistic regression to explain the relationship between the target variable and the predictor variables.

ii. Using the best models you have chosen for both multivariate and logistic regressions, explain what manner of relationship you expect

to find between the target and predictor variables.

B. Data Mining: You will use one supervised and one unsupervised data mining methods. Refer to the specific Final Project handout for your

dataset (available in Blackboard) to learn how to apply data mining methods in a step-by-step guide for completing this subsection.

Specifically, be sure to address the following:

i. Explain the motivation for using the supervised and unsupervised methods you have chosen.

ii. Explain what you want to predict using data mining methods that you have chosen.

4. Analysis and Results (three to four pages)

The overall objective of this section is to summarize your data analysis and include a detailed explanation of the performed analysis along with

tables and figures of the analysis results. You will present a summary of your results and evaluate the performance of each model used to analyze

the data. Be sure to interpret the performance evaluation measures.

To complete this section on your report, you must include the following critical elements:

A. Regression Analysis: For this subsection, you will include a revised version of Milestone 2. Specifically, be sure to address the following:

i. For multivariate regression, explain why the model you have chosen is the best.

ii. For multivariate regression, explain if any model assumptions are violated or not.

iii. For multivariate regression, include the table and interpret the table results (significance tests, R-squared, F-test).

iv. For logistic regression, explain why the model you have chosen is the best.

v. For logistic regression, include the accuracy rate of the best model.

B. Data Mining: Refer to the specific Final Project handout for your dataset (available in Blackboard) for a step-by-step guide for completing

this subsection. Specifically, be sure to address the following:

i. For supervised data mining, include the table that reports the accuracy, specificity, sensitivity, and precision rates for the test data set

and interpret your results. Comment on the model you have chosen using the performance charts.

ii. For unsupervised data mining, include the table that reports the averages of numerical variables that you use to cluster the data.

5. Discussion and Application (one to two pages)

The goal of this section is to discuss possible application of the results for the firm or the organization that uses this dataset. You will recommend to

the firm or the organization possible ways to implement the expected findings and explain how the results may impact the business or decision

process.

To complete this section on your report, you must summarize your results of models you have used and recommend a course of action to the organization or the firm that uses the data set you have chosen for your analysis.

i. Regression Analysis: Include a revised version of your suggestions and recommendations of your regression analysis in

Milestone 2.

ii. Data Mining: Include your suggestions and recommendations for the organization using this data to make decisions.

Milestones

Milestone 1: Data Exploration, Visualization, and Pre-Processing

In Module 2, you will choose your dataset, and then inspect, clean, visualize, and transform the data, documenting the key, and relevant steps or

plans. This Milestone is graded with the Milestone 1 Rubric.

Milestone 2: Regression Analysis

In Module 4, you will fit and estimate a regression model to predict the response variable. You will document the key and relevant steps or plans,

and include them in your report and appendices. This Milestone is graded with the Milestone 2 Rubric.

Final Submission: Data Analysis Report

In Module 7, you will submit a data analysis report based on observations and analysis of your big dataset to convey the information in written

form. It should be a complete, polished artifact containing all of the critical elements of the final product. It should reflect the incorporation of

feedback gained throughout the course. This milestone will be graded using the Final Project Rubric.

Deliverable Milestones

Milestone Deliverables Module Due Grading

1 Data Exploration, Visualization, and Pre-Processing 2 Graded separately; Milestone 1 Rubric

2 Regression Analysis 4 Graded separately; Milestone 2 Rubric

Final Project: Data Analysis Report 7 Graded separately; Final Project Rubric

Rubric

Requirements of Submission: Your final project report must be submitted as a Microsoft Word document, 9-12 pages length, double spacing,

12-point Times New Roman font, 1-inch margins, and no appendices.

Instructor Feedback: Students can find their feedback in the Grade Center.

Critical

Elements

Satisfactory (100%) Proficient (85%) Needs Improvement (55%) Not Evident (0%) Value

Executive

Summary

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of an executive

summary that describes in

detail the opportunity that

may impact the firm or

organization’s operations or

Executive Summary describes

in detail the opportunity that

may impact the firm or

organization’s operations or

strategy that use the data set

you have chosen for your

analysis; and explains why

and how data analysis offers

Executive Summary may be

lacking in detail or clarity, or

may be missing some of the

following: describes in detail

the opportunity that may

impact the firm or

organization’s operations or

strategy that use the data set

Did not submit an executive

summary.

15

strategy that use the data

set you have chosen for

your analysis; and explains

why and how data analysis

offers new opportunities for

this firm or organization.

new opportunities for this firm

or organization.

you have chosen for your

analysis; and explains why

and how data analysis offers

new opportunities for this firm

or organization.

Overview of

the Data

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of an overview of

the data that integrates

feedback previously

received and includes the

following: determines

whether there are missing

values and explains how to

handle them; identifies

potential outliers, explaining

how to handle them;

subsets the data based on

variables and compares

summary statistics of

subsets; and includes a

visualization of the data.

Overview of the data

integrates feedback previously

received and includes the

following: determines whether

there are missing values and

explains how to handle them;

identifies potential outliers,

explaining how to handle

them; subsets the data based

on variables and compares

summary statistics of subsets;

and includes a visualization of

the data.

Overview of the data may be

lacking in detail or clarity, or

may be missing some of the

following: integrates feedback

previously received;

determines whether there are

missing values and explains

how to handle them; identifies

potential outliers, explaining

how to handle them; subsets

the data based on variables

and compares summary

statistics of subsets; and

includes a visualization of the

data.

Did not submit an overview of

the data.

10

Methodology:

Regression

Analysis

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of a methodology

section about regression

analysis that integrates

feedback previously

received and uses the best

models for both multivariate

and logistic regressions; and

specifies the motivation for

using both multivariate

regression and logistic

regression to explain the

relationship between the

Methodology section about

regression analysis integrates

feedback previously received

and uses the best models for

both multivariate and logistic

regressions; and specifies the

motivation for using both

multivariate regression and

logistic regression to explain

the relationship between the

target variable and the

predictor variables.

Methodology section about

regression analysis may be

lacking in detail, clarity, or

accuracy, or may be missing

some of the following:

integrates feedback previously

received and uses the best

models for both multivariate

and logistic regressions; and

specifies the motivation for

using both multivariate

regression and logistic

regression to explain the

relationship between the

Did not include regression

analysis in the methodology

section.

10

target variable and the

predictor variables.

target variable and the

predictor variables.

Methodology:

Data Mining

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of a methodology

section about data mining

explains the motivation for

choosing one supervised

and one unsupervised data

mining method, as well as

what will be predicted using

each method

Methodology section about

data mining explains the

motivation for choosing one

supervised and one

unsupervised data mining

method, as well as what will

be predicted using each

method.

Methodology section about

data mining may be lacking in

detail, clarity, or accuracy, or

may be missing some of the

following: explains the

motivation for choosing one

supervised and one

unsupervised data mining

method, as well as what will

be predicted using each

method.

Did not include data mining in

the methodology section.

10

Analysis and

Results:

Regression

Analysis

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of an Analysis and

Results section about

regression analysis

integrates feedback

previously received and

includes the following:

explains why the

multivariate regression

model selected is best;

explains if any model

assumptions are violated;

includes tables and

interprets table results;

explains why the logistic

regression selected is best;

and includes the accuracy

rate of the best model.

Analysis and Results section

about regression analysis

integrates feedback previously

received and includes the

following: explains why the

multivariate regression model

selected is best; explains if

any model assumptions are

violated; includes tables and

interprets table results;

explains why the logistic

regression selected is best;

and includes the accuracy rate

of the best model.

Analysis and Results section

about regression analysis may

be lacking in detail, clarity, or

accuracy, or may be missing

some of the following:

integrates feedback previously

received and includes the

following: explains why the

multivariate regression model

selected is best; explains if

any model assumptions are

violated; includes tables and

interprets table results;

explains why the logistic

regression selected is best;

and includes the accuracy rate

of the best model.

Did not include regression

analysis in the Analysis and

Results section.

10

Analysis and

Results:

Data Mining

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of an Analysis and

Results section about data

Analysis and Results section

about data mining includes the

tables that report the accuracy

for supervised and

unsupervised data mining,

Analysis and Results section

about data mining may be

lacking in detail, clarity, or

accuracy, or may be missing

some of the following: includes

Did not include data mining in

the Analysis and Results

section.

10

mining includes the tables

that report the accuracy for

supervised and

unsupervised data mining,

including the specificity,

sensitivity, and precision

rates for the test data and

interprets the results. Also

comments on the model

chosen using the

performance charts.

including the specificity,

sensitivity, and precision rates

for the test data and interprets

the results. Also comments on

the model chosen using the

performance charts.

the tables that report the

accuracy for supervised and

unsupervised data mining,

including the specificity,

sensitivity, and precision rates

for the test data and interprets

the results. Also comments on

the model chosen using the

performance charts.

Discussion

and

Application:

Regression

Analysis

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of a Discussion and

Application section about

regression analysis that

integrates feedback

previously received and

includes a comparison of

results, as well as

suggestions and

recommendations.

Discussion and Application

section about regression

analysis integrates feedback

previously received and

includes a comparison of

results, as well as suggestions

and recommendations.

Discussion and Application

section about regression

analysis integrates feedback

previously received and

includes a comparison of

results, as well as suggestions

and recommendations.

Did not include regression

analysis in the Discussion and

Application section.

10

Discussion

and

Application:

Data Mining

Demonstrates a

sophisticated knowledge of

data analysis through the

creation of a Discussion and

Application section about

data mining that includes a

comparison of the results

between two subsets of

data, as well as suggestions

and recommendations.

Discussion and Application

section about data mining

includes a comparison of the

results between two subsets

of data, as well as suggestions

and recommendations.

Discussion and Application

section about data mining

includes a comparison of the

results between two subsets

of data, as well as suggestions

and recommendations.

Did not include data mining in

the Discussion and Application

Section.

15

Articulation of

Response

Submission is free of errors

related to citations,

grammar, spelling, syntax,

and organization and is

Submission has no major

errors related to citations,

grammar, spelling, syntax, or

organization.

Submission has major errors

related to citations, grammar,

spelling, syntax, or

organization that negatively

Submission has critical errors

related to citations, grammar,

spelling, syntax, or

organization that prevent

understanding of ideas.

10

presented in a professional

and easy to read format.

impact readability and

articulation of main ideas.

Earned Total

Comments

100%