Programming for Data Analysts Coursework 4 Summative Assessment Brief 2024-25 | BPP

PDA General Assessment Guidance

• Your summative assessment for this module is made up of this coursework submission, which accounts for 100% of the marks.
• Please note that late submissions will not be marked.
• You are required to submit all elements of your assessment via Turnitin online access. Only submissions made via the specified mode will be accepted, and hard copies or any other digital form of submissions (like via email or pen drive, etc.) will not be accepted.
• For coursework, the submission word limit is 2,500 words. You must comply with the word count guidelines. You may submit LESS than 2,500 words but not more. Word Count guidelines can be found on your programme home page and the coursework submission page.
• Do not put your name or contact details anywhere on your submission. You should only put your student registration number (SRN), which will ensure your submission is recognised in the marking process.
• A total of 100 marks is available for this module assessment, and you are required to achieve a minimum 50% to pass this module.
• You are required to use only the Harvard Referencing System in your submission. Any content that is already published by other author(s) and is not referenced will be considered a case of plagiarism.
  You can find further information on Harvard Referencing in the online library on the Hub.
• BPP University has a strict policy regarding the authenticity of assessments. In proven instances of plagiarism or collusion, severe punishment will be imposed on offenders. You are advised to read the rules and regulations regarding plagiarism and collusion in the GARs and UPPs, which are available on the HUB in the Help and Support section under Documents and Forms.
• Use of AI in assessments is only allowed to review a draft, correct language errors or if specified in the summative assessment brief. If you have used AI for any of these purposes, you should indicate this on the Assignment Cover sheet. For more information regarding acceptable and unacceptable use of AI, please enrol on the Generative AI Foundations course on the HUB.
• You should include a completed copy of the Assignment Cover sheet. Any submission without this completed Assignment Cover sheet may be considered invalid and not marked.

SUMMATIVE Assessment Brief

Assessment of learning outcomes

This assessment is designed to gauge your understanding, skills and application of common data analysis techniques used in business and other organisations today. As such you need to demonstrate your attainment in these areas according to the THREE Module Learning Outcomes (LOs):

• LO 1: Critically evaluate the principles of programming and apply them in a business context.
• LO 2: Critically evaluate the use of code libraries in programming for a business context.
• LO 3: Construct a programming solution to solve a defined business problem.

Assessment Tasks

This assessment is made up of TWO Parts

• Part 1 - A business report.
• Part 2 - A coding exercise in data analysis using a Python notebook.

You will have worked on both these Parts for your Formative Assessment. Now update both Parts for your Summative Assessment as set out below. You should act on any feedback you received on your Formative submission, together with your further learning and development across the module.

Scenario

SUCCESS Financial Services (SFS) is a local company that provides small business loans. Last year, loan applications increased by over 200%, largely because of a concerted online campaign to establish a strong digital presence. Almost all loan applications and business leads are generated from search engines and digital advertisements, reflecting the decision to increase advertising spend on SEO (Search Engine Optimisation) channels such as Google, Facebook, LinkedIn and similar platforms.

Despite a strong digital marketing approach, the current loan application process remains manual. It requires the online completion of information, including gender, marital status, number of dependents, education, income, etc. Loan decisions are manually categorised as either “approved” or “rejected.” To date, several of these factors have been considered in the approval decision. All applications are reviewed and approved by the loan team which, given the recent increase in volumes, has resulted in skills shortages, longer loan approval times and increased potential operational and control risk. The current operating model constrains further growth.

You are employed by SFS as a Data Team Manager, with programming and data analytics knowledge, as well as a deep appreciation for the need to balance business growth with a robust operational control environment. You will be leading this project with a team of programmers, with various
 
levels of experience, and have been tasked with providing a scalable solution that addresses key resourcing and control risks.

Specifically, the Board has instructed you to develop and maintain several partial automation processes that will help the existing loans team, freeing up their time for greater one-on-one customer contact. You need to provide a data-driven solution while working with a variety of key stakeholders, each with varying objectives such as marketing, internal audit and compliance. You need to assure the board that you have consulted widely in drawing up requirements and have considered how any code produced will be further developed and maintained. You also need to consider that any code developed will need to be accessible to (or used by) other staff such as sales, customer service and support personnel.

An in-house database administrator (DBA) was able to compile a PDF of past applications, which the loans team are hoping to map to previous loan approval outcomes.

The two files provided by the DBA are:

• A file in PDF format called ‘SUCCESS_Loans_Database_Table.pdf’
• An Excel file, called ‘SUCCESS Loans Data.xlsx’

(Note: These same two files are used in the Formative and the Summative).

The first file has been extracted from business loan records from the previous year, and it includes a loan approval status field (Yes/No) for each application, allowing the business to map inputs to outcomes for a possible supervised machine learning exercise.

The Excel file is maintained by the Sales team, and it is currently being saved in a shared folder. This increases the chance of duplication and missing values.

You will need to reflect on the learnings throughout this module and consider the learning outcomes, particularly LO 3: Construct a programming solution to solve a defined business problem as you create your answer.

Report - Business Proposal (LO1, LO2)

Using the scenario given, develop a report that proposes and justifies the continued development and maintenance of the code that has been written by the programming team.

Task 1 (LO1 and LO2) Business Report:

• Identify the problem issues in the given scenario.
• Discuss requirements, challenges, and data quality of the data sets.
• Explain what steps should be taken to implement the proposed solution.
• Your Approach should include cleaning and preparation, design, test, and maintenance.
• State the potential benefits of your approach and the implications of not addressing the problem using programming principles (e.g. OOP, performance optimisation, version control, code maintenance, etc)
•  Provide clear recommendations based on your insights to enable automation of the loan approval.

Programming Solution (LO3)

Task 2 (LO3): Programming Solution (this is not included in the word count)

• Load both data files and combine them.
• Explain the steps taken to load the data, how you treated the PDF data, how you cleaned the data, the libraries called and the overall procedure.
• Show outputs from the EDA (exploratory data analysis).

Report Structure and References

Follow the guidelines given in Section 3, Report Structure and Referencing
 
Business Report Guidance Notes:

The format of the report should include:

TASK 1 – Business Report (LO1 and LO2): (2500 words)

First, set out the purpose of the report and its structure. It must then cover the current business environment of companies like SFS, the problems that the programming solution has addressed, and what impact and benefits your proposed development and maintenance of the code might have on the business. You should also mention the implications of not proceeding and the kind of human resources needed. There is no need to discuss project finances, but your introduction must comment on the following:

• What user requirements have been considered (what tools will be used)?
• the source and nature of the supplied dataset and its suitability for data analytics;
• The challenges the team faces in further developing and maintaining the code, the benefits of using reusable code and any regulatory and ethical implications.

Analyse the approach you took to implement your solution. This should be based on a suitably named data analysis lifecycle or framework (e.g. PPDAC, CRISP-DM, SDLC or similar frameworks). Discuss, especially, the key steps taken in loading and preparing the data for analysis. Use the outputs (calculated values and any charts and visuals) from your EDA in Part 2 to explain the key findings. Key findings can relate to any business insights that might be gained from the analysis of the historical loan records, as well as any general lessons about the approach being taken for this project (may include comparisons/contrasts to similar real-life examples). The SFS board wants to know if your approach works, is sustainable and worth building on.

Your Approach discussions should include:

• The code libraries used (what these are and why they are needed);
• The reasons for the choice of language and platform;
• The design (e.g., using pseudo code) to aid understanding and allow the design to be implemented in other languages if needed (code agnostic);
• Testing the code (e.g., by using Excel) to confirm that the programme works; and
• Maintenance of any code to allow controlled and efficient development (e.g., by version control using Git and good commentary)

Recommendations for future work: Briefly explain how your solution could be further developed to build a predictive model using the historical data provided. (i.e. A model that can be trained to predict the outcome of any future loan application.)
This should include a short explanation of the techniques, libraries, tools, and objective functions used to evaluate the precision of your recommended predictive model.

Note: There is no need to implement your recommendation ideas in code or build a predictive model
 
Your recommendation must also discuss:

• How existing data collection and data quality may be improved;
• if other libraries might be used to increase model accuracy;
• additional training or support the staff will need to use these new tools.
• safeguards needed to ensure compliance with regulatory and ethical standards.

Conclusions: A brief conclusion summarising the main points in the report.

In writing your report, use the insight and knowledge provided in this module, but also leverage sound independent academic research to support your report. As you develop your work, you should self-evaluate your developing draft against the criteria set out in the Marking Guide below.

Programming Solution Guidance Notes:

TASK 2: Construct a Programming Solution (LO3)

You must deliver a pre-run Jupyter notebook, also known as an Interactive Python Notebook (a .ipynb file), using Google Colab, with the code used, with comments, to explain the scripts, the libraries used, and the logic. All such commentary should be written using the built-in markup language (Markdown text) or embedded using the standard Python syntax (preceded by a #).

The notebook that you create should highlight some of the key findings that you found from the given data and their insights, which you can provide to the business.

The Python Notebook should include the following:

1: Loan Data Solution Development

Python Code: Copy and paste the entire contents of your .ipynb file – screenshots of the notebook with the executed code and markdown text.

Test Results: Copy and paste any output from Excel showing if it has confirmed the correct working of your code.

Pseudocode: to communicate the design logically.

The Python Code should be presented in a new.ipynb Jupyter notebook within Google Colab. Programmatically extract the two datasets from these two files, which contain information about past loan records. The integers shown in each column of the loan datasets are to be interpreted as follows, and you may read them as strings or numbers in your code as you deem appropriate.:

• Gender: 1-Male, 2-Female
• Married: 0-Single, 1-Married
• Dependents: 0, 1, 2, 3+
• Graduate: 0-No, 1-Yes
• Self_Employed: 0-No, 1-Yes
• Credit_History: 0-No, 1-Yes
• Property_Area: 1-Urban, 2-Semiurban, 3-Rural

You should use Python to load the information from these datasets into memory. You should also add comments to your notebook, explaining the steps taken to load the data, how you treated the PDF data, the libraries called and the overall procedure, bearing in mind this will be used for training colleagues in future.

2: Descriptive analysis

First, check the datasets and make sure the data that comes from these two files is valid. Ensure your loan data is correctly indexed on the LoanID column.

Then, clean the loan data. Explain the steps taken to ensure data preparation for analysis, such as the correction of duplicates, missing values, outliers, etc.

Next, carry out Descriptive analysis on the current loan data. Your notebook file should contain the code and output of some basic Exploratory Data Analysis (EDA), which must include the following:

• The total amount loaned by SFS.
• The average amount loaned.
• The average loan term.
• The total number of applicants broken down into Approved and Rejected, the number of males and females in each case, and shown in an appropriate chart.
• The maximum and minimum amounts loaned and shown in an appropriate chart.
• The number of self-employed who had their loan approved, expressed as a percentage of all who had their loan approved, and shown in an appropriate chart.

Report Structure and Referencing

In addition, ten marks are awarded for the overall professionalism of your report and the adoption of academic standards.

Guidelines:

• Your report should follow the section naming structure and order set out in the Brief. You should also add your subheadings as you see fit to demonstrate your ability to develop structure and content.
• Your report should include an auto-generated contents page including section headings and sub-headings. The contents page should also include a page-referenced list of all tables, charts and figures provided in your report. Remember to number all pages in your report, for example ‘Page 8 of 12’.
• Ensure you develop your discussion in a logical progression: Findings, inferences, conclusions, recommendations.
• Do not make general assertions without supporting evidence.
• Zero spelling errors and grammatical mistakes
• Include a blend of industry research, case studies and academic references.

You should set out your Business Report in a Word document, according to the following heading structure. The sections in bold count towards your word count.

• University Cover Page
• Table of Contents
• Introduction
• Approach
• Recommendations
• Conclusions
• Programming Solution
• Appendices (Optional)

You should add subheadings under this overall structure as you feel fit to demonstrate your ability to develop the section themes and to provide meaningful substructure. But you must use this overall structure to provide a consistent framework against which your marker will allocate marks. You will be deducted marks if you do not follow this structure. Also, note that there is NO requirement for producing an Executive Summary.

Total word count: 2,500. The Cover Page, Table of Contents, References, Appendices (Optional), Tables, Charts and Figures do not count towards word count.

The code output, test validation, pseudocode and content of the Python Notebook are not included in the word count.
 
IMPORTANT: Adding your pre-run code to your report before uploading to TurnItIn

Locate the report file and embed your Pre-run Python notebook into your Word document. If you are unable to embed your Python notebook for any reason, you must provide a shared link to the file. This is easily done within Google Colab by selecting the ‘Share button’ in the top right-hand corner of the screen:

Mapping Learning Outcomes to Assessment Tasks

The table below sets out the mapping between the three Learning Outcomes and the key tasks in your Summative Assessment, which are designed to test your achievement against these Learning Outcomes.

Marking Guide

The assignment is marked out of 100 and counts towards 100% of your module mark. The following table shows the tasks, marks and marking rubric. You should iteratively self-assess your performance against the Marking Guide as you develop your draft submission to evaluate your performance against your target grade.