Reference no: EM132639816
MIS771 Descriptive Analytics and Visualisation - Deakin University
Learning Outcome 1: Apply quantitative reasoning skills to solve complex problems.
Learning Outcome 2: Plan, monitor, and evaluate own learning as a data analyst.
Learning Outcome 3: Deduce clear and unambiguous solutions in a form that they useful for decision making and research purposes and for communication to the wider public.
The Case Study
ANALYTICs7, a leading data analysis consulting company, has extensive experience in analysing data for both local and global, small to medium companies. By solving their business problems, ANALYTICs 7 helps these businesses to plan ahead and thrive.
Your Role in ANALYTICS7
Dr Hugo Barra, the lead data scientist at ANALYTICs7 has engaged you to lead the modelling component for the TPM and AP projects and construct a report of your key findings and recommendations in response to the questions posed in the meeting minutes of the last team meeting on the next page.
Datasets (accessible via T22020MIS771_A2Data.xlsx file)
There are two datasets available for this assignment: TPM_Employee_Attrition and Monthly_EnergyCon_MW
Employee Survey data (TPM_Employee_Attrition )- TassPaperMill (TPM), a subsidiary of Pinnon Paper Industries (PPI), is an Australian company with a long history of manufacturing paper rolls. To address numerous concerns raised in their recent employee survey TPM is currently reviewing how they calculate salary increments for their employees. TPM has hired ANALYTICs7 to extract a random sample of 1470 employee records from their HR database. Their ultimate goal is to adopt a more holistic rewarding system factoring the key relations between remuneration indicators and demographic characteristics, employment history and various other potential contributors to boost performance. In addition, human resource manager at TPM reported in her recent presentation to the company executive management team that the staff turnover rate at TPM is higher compared to their competitors. Thus, TMP wants to identify key contributing factors before they lose more talented, motivated and focused employees who contribute to the organisation's overall success.
Energy consumption data (Monthly_EnergyCon_MW) - Australian Power (AP) is one of the largest generators of electricity in Australia, servicing for more than three million households in Victoria. AP operates an electric transmission system that covers much of Victoria and serves over 30% of the electricity demand in Victoria. This dataset consists of monthly power consumption data in megawatts (MW) comes from AP's data warehouse during 2010-2019. AP wishes to review their current resources allocation strategy to plan and prioritise the provision of resources based on rapidly growing energy demand in Victoria.
A complete listing of variables is provided in the T22020MIS771_A2Data.xlsx file.
Note: All data, reports, people and scenarios in this assignment are either fictitious or have been modified from their original state. Any similarity to actual events is purely coincidental. It has been produced for the sole purpose of assessing performance of summative assessment task 2.
Purpose: Specifying and Allocating Data Analytics Tasks
1. Variable(s) description
2. Modelling PercentSalaryHike
3. Modelling the likelihood of an employee leaving the company
4. Forecasting monthly energy consumption in Megawatts
5. Producing a technical report
Who:
Modeller What:
1. Providing an overall summary of the following two variables:
Percentage increase in salary (PercentSalaryHike)
Attrition
2. Identify potential variables that may influence PercentSalaryHike:
Identify a list of possible variables that influence percentage increase in salary. Which three independent variables have the more impactful linear relationship with PercentSalaryHike? What form of relationship(s) exist between the independent variable(s) and PercentSalaryHike? Are there any potential multi-collinearity problems? If so, which variables are they?
Build a regression model to estimate percentage increase in salary.
Perform residual analysis. Based on your residual plots, does there appear to be any problems with the regression model?
3. Hugo has performed some preliminary analysis and discovered that the performance rating is a significant predictor of the Percentage increase in salary. Prior research shows that the strength of the relationship between performance rating and percentage increase in salary may vary according to satisfaction with the job. Generally speaking increased job satisfaction creates a more productive
workforce as they are more motivated to improve their job performance.
Therefore, Hugo believes that the relationship between performance rating and percentage increase in salary should be stronger for employees who are satisfied with their jobs.
Model the interaction between the variables to test Hugo's assumption.
Comment on whether there is sufficient evidence to conclude that the interaction term in the model is statistically significant.
4. A model to predict the likelihood of an employee leaving the TPM
Hugo has already performed an analysis with Attrition and Age, Environment Satisfaction, Overtime and Years in current role as the independent variables. Continue to refine his work and develop a model to ascertain the likelihood of an employee leaving the TPM.
Hugo is specifically interested in understanding how the following aspects drive employee attrition.
a) Medium satisfaction level with their working environment and job, and 5 years since their last promotion
b) Number of years in current roles and whether they work overtime
c) 45 years old married employee with a very-high level job classification and maintaining a good work-life balance.
In order to gain an edge in the current very competitive talent market, Hugo believes attaining a very good understanding in what drives employees to quit is well worth the time and investment. In addition, TPM should take prompt actions to mitigate increasingly high employee turnover costs which could be up to twice an employee's salary depending on their position.
Accordingly, your job is to visualise the predicted likelihood of employee attrition with the specific attributes described above.
5. Develop a time-series model to forecast AP's energy consumption for the next 12 months. How are summer predictions different from those for winter?
6. Provide a written report detailing ALL aspects of your analysis. The report should be as detailed as possible and should describe ALL key outputs of the analysis. The results of the analysis should drive the recommendations to the executives/decision makers at both TPM and AP.
7. The ability to submit work on time is a highly sought after skill at ANALYTICs7. As a part of your ongoing professional development, I would like you to report how you plan to deliver agreed outputs on or before the set date.
Attachment:- Descriptive Analytics and Visualisation.rar