QBUS6600

Data Analytics for Business Capstone

Semester 1, 2025

Assignment 1 (Individual Assignment)

1. Key Instructions

Required submission:

• Written report (in pdf, due date: Monday, March 31 by the end of the day).

• Confidentiality Deep Poll online form (deadline for submission: Monday, March 17). Detailed submission instructions for the report will be posted on Canvas in Week 5.

Weight: 30% of your final grade.

Length: Your written report should have a maximum of 12 pages (single spaced, 11pt). Cover page, references, and appendix (if any) will not count towards the page limit. Please keep in mind that making good use of your audience’s time is an essential business skill: every sentence, table or figure should serve a purpose.

2. Problem Description

Please read through the Project Outline document for your selected industry project, which you can find on the ‘Learn about our industry projects’ page in the Week 1 module on Canvas. Focus on the Problem Description section of the Project Outline, especially the first and third bullet points (EDA and Strategy), which are the most relevant bullet points for Assignment 1. You are not required to build a predictive model  for Assignment 1.  Both your analysis and  your recommendations should be in line with the requirements/suggestions provided in the Project Outline.

As a business analyst, you will conduct Exploratory  Data Analysis  (EDA) of the data corresponding to your industry project. You should aim to find or reveal all possible properties, characteristics, patterns, and  statistics hidden in the data, supporting your findings  with insightful plots and relevant statistical output.

Use the results from your EDA to outline a preliminary strategy or provide  preliminary recommendations to the UNICEF team corresponding to your selected industry project. You will have a chance to refine these recommendations in Assignment 2.

3. Written Report

The purpose of the report is to describe, explain, and justify your findings to the UNICEF team corresponding to your selected industry report. You may assume that team members have training in business analytics, however, they are not experts in statistics or machine learning.

The team’s time is important: please think about the most structured way to present your findings. Be concise and objective.

Suggested outline for the main parts of the report (further details below):

1)  Problem formulation.

2)  Data processing.

3)  Exploratory Data Analysis (EDA).

4)  Conclusions and preliminary recommendations.

You should consider breaking down the longer parts into smaller sections.

4. Marking Scheme

5. Rubric (basic requirements)

Business context and problem formulation. Your report gives a detailed description of the problem that is being investigated, providing the context and background for the analysis.

Data processing. You describe the data processing steps clearly and in sufficient detail, justifying and explaining your choices and decisions. You handle missing values and other data issues appropriately. You describe and explain your data transformations and/or your feature engineering process (if any). Your choices and decisions are justified by data analysis, domain knowledge, logic, and trial and error (if necessary).

Exploratory data analysis (EDA). Your report provides a comprehensive description of your EDA  process,  presenting selected results. Your analysis is sufficiently rich, and  your visualizations are insightful. You study key variables and relationships among them using appropriate plots and descriptive statistics. You note any features of the data that may be relevant for model building in Assignment 2. You note the presence of outliers and any other anomalies that can affect the analysis. You explain the relevance of the EDA results to the underlying business problem and your subsequent recommendations. You clearly describe and justify  the  methods in your analysis. The choice of methods is logically related to the substantive problem, underlying theoretical knowledge, and data analysis. You interpret the statistical outputs that you provide. You report crucial assumptions  and whether they  are potentially violated.

Conclusions and recommendations. The reasoning from the analysis and results to your conclusions and recommendations is logical and convincing. Your conclusions and recommendations are written in plain language appropriate for non-technical audience.

Writing. Your writing should be concise, clear, precise, and free of grammatical and spelling errors. You use appropriate technical terminology. Your paragraphs and sentences follow a clear logic and are well connected. If you use an abbreviation or label, you define it first.

Report layout. Your report is well organised and professionally presented, as if it had been prepared for a client later in your career. There are clear divisions between  sections  and paragraphs.

Tables. Your tables are appropriately formatted  and have a clear layout. The tables have informative row and column labels. The tables are relatively easy to understand on their own. The tables do not contain information which is irrelevant to the discussion in your report. The tables are placed near the relevant discussion in your report. There is no text around your tables, and your tables are not images.

Figures (plots). Your figures are easy to understand and have informative titles, captions, labels, and legends. The figures are well formatted and laid out. The figures are placed near the relevant discussion in your report. Your figures have appropriate definition and quality. There is no text around your figures, and your figures are not screenshots.

Numbers. All numerical results are reported to suitable precision (typically no more than three decimal places, in some cases fewer).

Referencing. You follow the University of Sydney referencing rules and guidelines.

Python code. The text of your report should be entirely free of Python code.

Note: you are required to use Python for all the steps of your data analysis. While there is no Python code submission for Assignment 1, you should keep your code well-organized, so that you can easily extend/modify/reuse this code for the purposes of Assignment 2 (which will have a Python code submission requirement).

6. Deductions

Marks may also be deduced from each item in the marking scheme in the following cases:

•   The report is disorganised and/or has a poor layout.

•   There is an excess of abbreviations or labels that the reader may be unfamiliar with.

•   The report has an excessive number of grammatical or spelling mistakes.

•   The tables are difficult to read, for example, due to poor layout or labelling.

•   The figures are difficult to read, for example, due to poor layout or labelling.

•  Numbers are not appropriately rounded.

7. Late submission of the report

Late submissions are subject to a deduction of 5% of the maximum mark for each calendar day after the due date. After ten calendar days late, a mark of zero will be awarded.