General Guidelines for Analytics Written Assignments

Guidelines for written projects in my Analytics classes

This post outlines the general content and format guidelines for written assignments in my Analytics classes.

Deliverables

Submit your assignments on Blackboard. They are due at midnight on their due dates. Please submit:

A Markdown file that contains the text, wrangling, analytical, and reporting operations required to render your analysis
A formatted Word file that shows the audience-directed version of your report
The raw data set used in your analysis

General Guidelines

Submit a 800 – 1500 word report that includes the following:

Summary: A summary of the post. Les than 50 words.
Introduction: Establish the target audience, the informational need, and the relationship between your analysis and that need.
Data and Methods: A discussion of the data used in this analysis, as well as sample and measurement information.
Analysis: Statistical operations and interpretations
Recommendations or Conclusions: A natural-language information item based on analysis

Grading Rubric

My goal in evaluating your work is to steer you towards being someone who creates better analytical products in the real world. In this semester’s class assignments, we will focus on delivering better information products in terms of six qualities:

Does the assignment fulfill the professor’s/client’s informational request? Does the submission deliver the information and follow the guidelines or specifications set forth in the job request? Basically, did you give the client what they requested?
Plausibly Useful Product: Does it seem plausible and likely that the submission fulfills an informational need that helps a definable group perform a definable task or make a definable decision? In other words, is it believable that this is useful information?
General Methodological Quality: Does the discussion or analysis employ scientific reasoning? Does their reasoning show a cognizance of the theoretical and empirical elements of scientifically-derived information, and how these two elements interplay? Does the analyst demonstrate a cognizance of how sampling and measurement influence their inference-making?
Wrangling Quality. Did the analyst render a clean and complete set? Were wrangling operations well documented? Do they engage the issue of missing data mindfully? Were their wrangling operations sensible, transparent, and easily reproducible? Was the analysis ambitious in its use of data?
Analytical Quality. Did the analyst apply the proper operations, do so properly, and did they properly interpret the results? Does the analysis convey a nuanced contemplation of its subject matter and inferences? Are the qualitative inferences drawn from the data sensible, plausible, and data derived? Was the analysis ambitious in its use of analytical operations?
Communications. Does the analyst convey the problem in an engaging way? Do they articulate a clear target audience for this information, and is there a clear case made for the utility and rigor of the information being offered? Is the presentation aesthetically attractive and easy-to-process, yet still thoughtful and sophisticated in the information the content of its discussion?