{"id":1612,"date":"2024-07-11T08:18:55","date_gmt":"2024-07-11T08:18:55","guid":{"rendered":"https:\/\/britainwriters.com\/answers\/?p=1612"},"modified":"2024-07-11T08:18:58","modified_gmt":"2024-07-11T08:18:58","slug":"introduction-to-business-and-social-data-science-assignment","status":"publish","type":"post","link":"https:\/\/britainwriters.com\/answers\/introduction-to-business-and-social-data-science-assignment\/","title":{"rendered":"Introduction to Business and Social Data Science Assignment"},"content":{"rendered":"\n

Assignment Task<\/strong><\/h3>\n\n\n\n

Problem formulation<\/strong><\/p>\n\n\n\n

You will work on a selected and edited subset of job reviews that cover (anonymous) employee ratings and reviews of various jobs\/companies. You will have several numeric or qualitative ratings provided by employees, but also multiple text fields reporting more in detail their views and experiences. Your main task will be to assess job quality indicators in combination with the views expressed by the employees and evaluate various models that can help us better understand what predicts good job\/company evaluations.<\/p>\n\n\n\n

The data are accessible on Canvas under in multiple formats: either as one .rdata file containing two data objects, or two separate .csv files. You can either use the .rdata or the two .csv files. The two data frames in the .rdata file contain the same information as the two .csv files. Variables are described after this section.<\/p>\n\n\n\n

Tasks<\/strong><\/p>\n\n\n\n

You have considerable freedom in what questions you formulate and how you employ the data science skills you have learned during the course to offer a detailed and informative analysis of job reviews, however, there are some mandatory steps that need to be included in your assignment.<\/p>\n\n\n\n

1. You have to combine the information from the two different data frames. Pay attention to ID variables and joining, as you will need more than one variable to link the two data frames. Show data wrangling skills by creating new variables that are useful for you, provide various group-based statistics and at least 3 visualizations that are customized and interpreted. In this process, show that you can change between long and wide data formats when needed. Please include figures and tables in the main text of the assignment.<\/p>\n\n\n\n

2. You have to provide sentiment analysis of the pros, cons, headlines and\/or their combination and discuss the interpretation and insights generated by this analysis. Start with simple comparisons between pros and cons and then combine all textual information and their sentiment scores. Discuss your choices and provide at least three different substantive, data-driven calculations and reflections on what these sentiment scores mean and why do you think they can be useful. Please include figures and tables in the main text of the assignment.<\/p>\n\n\n\n

3. Pick one outcome that you can treat as continuous and fit, evaluate and interpret at least two linear regression models. Here, make sure that one of your predictors is related to the sentiment scores you calculated in step 2.<\/p>\n\n\n\n