Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

COMM1190 Assessment 2: Team Report

Final submission: Week 8 Friday, 3rd  November 2023 at 5:00 pm (AEDT).

This assignment is graded upon 30 marks (i.e., 30% of the course total marks)

The assignment should be submitted in the form of a written report and to be undertaken in groups of 5 students. The team members must belong to the   same tutorial class.

Maximum five pages (~2500 words), excluding tables, figures, references, and Appendix (please refer to the detailed requirements provided in the section “Word Limit”)

Submission via Turnitin on the Moodle course page.

Objective

In this team assessment you are required to demonstrate your critical thinking and ability to conduct predictive analytics using R software and generate insights in the form of actionable recommendations to solve a business problem.

You are required to work with your team members (your COMM1190 group) . Your team is approached by SwiftFood to conduct a predictive analysis using R to predict the speed of food deliveries based on key variables identified from the dataset.

In this assessment, you will continue to use the same dataset as in Assessment 1 (the team  leader’s  personalized  data).  Based  on  the  discussion  of  key  findings  of  your predictive analysis, you are required to derive actionable insights on how to improve the speed of food delivery.

The learning content needed to complete this assignment is covered in the course up until the end of Week 7.

Requirements & assessment criteria

1.  Business problem identification (10%)

Accurately identify the problem, formulate insightful research questions, articulate the purpose of the analysis and the method undertaken. Think of why your analysis is important and how you plan to undertake it.

2.  Data selection and preparation (10%)

Refer to the findings of the descriptive analysis in Assessment 1 to discuss the relevant variables you would consider in your predictive analysis and any data transformations you might need to perform.  For each variable available  in the dataset, think of whether it should be transformed to better serve the needs of the predictive analysis. e.g. turning a continuous variable into a categorical variable,


time variable into specific time slots, the variable Order_Time_Placed could be transformed into Order_Hour (e.g. 23:25:00 into 23), etc.

3.  Predictive Data Analysis (30%)

Conduct a predictive analysis to forecast how relevant variables impact the speed of food delivery. Generate at least 2 models and explain why these models are suitable for your analysis. The 2 models can be generated using either two different modelling techniques (e.g. linear regression, logistic regression, classification tree, etc.) or the same technique by trying out different independent variables to predict the outcome variable.

Compare the generated models and decide what would be the best performing model to select along with justifications.

4.  Business Recommendations (20%)

Interpret the findings of your analysis to identify the key factors influencing the outcome variable (speed of delivery). Explain how these key factors influence the speed of food delivery.

Generate insights in the form of actionable recommendations on how to improve the speed of food delivery.  Provide justifications of your arguments  based on examples and/or academic references.

Please note that the insights should be derived entirely from the data and the model selected.

5.  Ethical Considerations (20%)

SwiftFood decides to use the outcomes of your predictive analysis to predict the delivery duration of each  new order  received from  customers.  However,  after application of your model, they noticed that some orders were not delivered within the predicted time frame. To address this issue, SwiftFood asked your team to consider collecting the real-time GPS locations of each delivery person.

Discuss the ethical implications you would consider during the data collection, analysis and communication (10%). Apply relevant ethical theories to justify and support your arguments. Provide suggestions on how to prevent and/or mitigate these ethical issues (10%).

Supplementary  reading:  Consult  Danish  Design  Centre’s  Digital  Ethics Compass       (https://ddc.dk/tools/toolkit-the-digital-ethics-compass/)       to understand the nuances of data ethics in the context of digital products

6.  Communication and Organization of your written report (10%)

•    Demonstrate proficiency in writing in English.

•    Develop a logical structure to organize the sections of your report.

•    Develop an executive summary using jargon-free language.

•    Use figures and/or tables to convey qualitative and quantitative information effectively and accurately.

•    Use  academic  referencing  in  Harvard  style.  Refer  to  UNSW  guideline:

https://www.student.unsw.edu.au/harvard-referencing

•   Attach the codes of your R programming (not a screenshot) in the Appendix of your report.

NB: A detailed rubric is included on the last pages of this document.


Guidance on Data Analysis

1.  Critically and collaboratively reflect on each team member’s feedback from their individual assignment and use them to develop your team project where applicable.

2.  Refer to the results of the descriptive analytics in assessment 1 to identify the key variables that may impact the speed of food delivery. Descriptive Analytics refer to statistics and visualization techniques. For example, a box plot and a bar chart are different techniques.

3.  Conduct predictive analytics to estimate the key factors that influence speed of food delivery. Predictive Analytics covered in this course include linear regression, logistic regression, and decision tree modelling techniques. It is required that you use the modelling techniques discussed in lectures and workshops (i.e., do not use modelling techniques beyond the scope of this course).

4.  Consider trying out several models / modelling techniques (at least 2) and explain which model can be considered the “best”. To select a model to be the “best” out of your candidate models, you can assess it based on the model’s goodness offit and/or its performance in predicting the outcome variable. You should use methods and criteria learned from this course to test the goodness offit and its performance (i.e., do not use methods and criteria beyond the scope of this course). You may also compare several models using the same modelling technique by trying out different independent variables to predict the outcome variable.

5.  Depending on the modelling technique chosen, you may need to perform data transformation. e.g. some modelling techniques require transforming a numerical dependent variable into a binary or categorical variable.

6.  Develop coherent logic from your business issue identification to your variables and modelling techniques selection, your data transformations, insights derived

from the findings of your analysis in the form of actionable recommendations.

7.  Explicitly state any key assumptions that impact your data analysis.

Project Management

During your work on this project, please make sure that you follow theUSNW Guide to Group Work(must read) to participate in the team project.

Develop and record a project management plan by specifying key milestones and each team member’s responsibilities.

Nominate a team project leader to facilitate collaboration and submit the report on behalf of the group.

Note that if any issue emerges from the collaboration and requires the teaching team’s support, a team should report the issues to the teaching team as early as possible by involving all team members.

Submission Instructions

The team leader or a designated team member needs to submit the written report via the Turnitin submission link on Moodle. Please Note that only ONE report from a group is required to be submitted.

You are required to submit your report in a Word format, accompanied by a cover sheet (provided on Moodle). Please note that you need to nominate a team lead on the cover sheet by specifying their name and zID.

Please include all relevant R codes in the Appendix. The codes should take the raw data file provided as the input and must be able to reproduce all the data analysis reported.

Late Submission

1.  Late submission will  incur a  penalty  of 5%  per  day  or  part thereof  (including weekends) from the due date and time. An assessment will only be accepted after five days (120 hours) of the original deadline if special consideration has been approved. An assignment is considered late if the requested format, such as hard copy or electronic copy, has not been submitted on time or where the ‘wrong’ assignment has been submitted.

2.  Please  note  that   no  extensions  will   be  granted  except  for   serious   illness, misadventure,  or  bereavement,  which  must   be  supported  with  documentary evidence. Requests for extensions must be made to the Course Convenor by email and be accompanied by the appropriate documentation 24 hours before the due date of the assignment. Students must apply for Special Consideration if this is not possible.

3.  Applications for Special Consideration must be submitted via myUNSW to be valid. Information on when and how to apply for Special Consideration can be found here: https://www.student.unsw.edu.au/special-consideration.   If  your  email  asks  to confirm receipt of an application, please be aware that we will only reply if we have not received your application.

4.  The  Course  Convenor  is  the  only  person  who  can  approve  a  request  for  an extension. If you request an extension, the Course Convenor will email you the decision. Note: A request for an extension does not guarantee that you will be granted one.

Word Limit

Your report will be evaluated based on the depth and the quality of the analysis (you are required to perform a rigorous analysis) and the explanations of the findings & insights. Hence, we suggest a maximum number of five pages. A penalty will not be applied if your report stays below five and a half pages.

Over and above the leeway, a penalty of 5% of the available marks for the assessment will be deducted for each extra half page.

Please note that 1% of the available marks for the assessment will be deducted for this assessment if you do not include a completed and signed cover page as part of the report.

Studiosity English Support

UNSW has partnered with Studiosity to provide online writing support. It is an online platform, freely available 24/7 for our students, providing focused feedback on structure, grammar, referencing, and choice of language but not on the course content. A link to the Studiosity is accessible on Moodle. Please note that you have up to 6 submissions per

term.     You      can     also     find      additional https://www.student.unsw.edu.au/referencing.

academic     writing     resources     at

AI Tools and Academic Integrity

The use of AI or ChatGPT to help you learnthe R codes is encouraged. However, the use of ChatGPT for writing the report will be subject to an AI-detective tool, which may lead to academic integrity investigations. In general, you must comply with academic integrity and avoid all forms of plagiarism. This includes buying essay/writing services from third parties, and engaging another person to complete your assessment, whether the latter is paid or not.

UNSW Guide to Group Work

“This page will inform you about the nature of group work, what you should expect , and the  expectations  teachers  have  of  you   in  group  learning  situations.”  Access   via https://www.student.unsw.edu.au/groupwork.   Furthermore,  please refer to the  Modus Operandi next page.

Modus Operandi

1. Group Formation

•      Students  are  expected  to  reach  out  to  the  other  class  participants  for  team formation,

•      Students are encouraged to choose the teammates who are aspiring to match their commitment in time and effort. The groups can be communicated to their tutors as soon as possible.

•      In every team, the maximum number of students is strictly limited to five. The tutor has the ultimate discretion to reshuffle a group if there are unexpected drop-outs or for other reasons. If your groups fall short of the size of 4-5, you can be granted an extended deadline from Monday 6 November.

2. Meetings

•      At the inception of the team formation by the end of week 4, shall initiate a first exchange of contacts by email and alternative means of communication (e.g., WhatsApp or mobile phone number). If any team member fails to respond to their teammates within 24 hours, the remaining members can contact their respective tutors.

•      All teams  are strongly recommended to conduct at least two meetings with an adequately set agenda at a mutually agreed time, the first at the beginning to discuss the task allocation and the expected milestones for deliverables, and the second at the end to fine-tune their work and bring them together.

3. Team Contract

•      The use of the Gantt Chart is instrumental in task scheduling.

•      The  team  members  should  sign  a  team  charter,  outlining  the  details  of  task allocation,  the   mode   and  frequency   of  communication,  and  the   rules   of collaboration. The team charter shall be added as an appendix to the finalized assignment report as evidence of effective team project management.

•      The  team  must  appoint a leader, whose zID data set file will  be used for the assessment 2. The role of the team leader is to coordinate that all members are participating and communicating with each other regularly.

•      The mark awarded will be assigned to all team members, but individual marks may be moderated if the peer assessment and subsequent investigation identified an uneven contribution and effort across the group members.

4. Communication

•      Every member of the team must attend a weekly 30-mins meeting with their team, to discuss and clarify expectations, as well as checking-in the progress of their respective deliverable.

•      A shared Drive can give access to all team members who can review and edit their common  report,  thus  fostering  real-time  collaboration  and  smooth  tracking  of members’ work in one place.

•      Team   bonding  is  a  crucial  element  of  effective   collaboration,  whereby   the teammates  agree  to  exchange   communication  details   (WhatsApp  or   phone numbers). They can opt for any preferred forms of social media or traditional means of communication.

5. Dispute and Resolution

•      Your tutor will intervene if there are unresolved contentious issues as a matter of last resort, whereby all communication channels have failed to achieve harmony and civility among team members. Yet, teams should involve respective tutors as early as possible whenever there are unequal contribution issues or disputes.

6. Peer Review

•      At  the end of each deliverable, each team member is expected to review and provide feedback on each other's work to enhance the quality of the overall report. The peer assessment review can be reported to the Course convenor in the event of obvious unequal contribution-namely non-participation by a team member or a member who is not meeting the deadline. Then, the team member can fill in a written statement of unequal contributions and emailcomm1190@unsw.edu.auon or before the assessment 2 report’s due date.