Get the Data Ready for Sharing

Based on the insights gathered during the scoping phase:

Data Masking $($De$-$identification$)$: Apply comprehensive de$-$identification techniques to the data you chose. Utilize concepts learned throughout the course, including equivalence classes and quasi$-$identifiers, to ensure that the data is suitably anonymized for sharing. Experiment with different masking strategies to optimize data privacy.

Risk of Re$-$Identification: Evaluate the likelihood of re$-$identification within your de$-$identified dataset. Assess how effective the anonymization techniques used are and document the residual risk associated with the dataset.

Show your work, it$$s OK to show that the level of de$-$identification you had chosen initially was really not enough once you ran a risk assessment.

Justification of the anonymization and Scenario Detailing:

Audience Consideration: Identify the intended audience of the dataset. Is it intended for a general internet audience, a specific partner, or a regulatory body? The target audience will dictate the required level of data anonymization.

Level of Anonymity: Decide on and justify the necessary level of anonymity for the dataset. Quantify this level using statistical or data science metrics and explain why this level is appropriate given the data's sensitivity and the audience.

Residual Risk Assessment: Discuss the residual risk of data re$-$identification. Evaluate whether the chosen anonymization methods sufficiently mitigate these risks and if further precautions are needed.

Data Utility vs$.$ Anonymity: Reflect on whether the anonymized dataset remains faithful to the original's utility. Ensure that the dataset retains sufficient detail to be useful for the intended purpose without compromising privacy. Comment on this balance in depth.

Technical Requirements: Consider whether the anonymization techniques used require additional resources such as hardware, software, or enhanced governance measures, comment on how the data will be shared, what your counterparty will need to do$,$ etc.

This assignment encourages you to apply critical thinking and tailor the de$-$identification process to a specifically designed scenario, emphasizing the balance between data utility and privacy.

Amended Data file to be shared:

Excel file your data. This is the file you$$d be sharing with your counterpart. Be sure to have proper labels and even formatting if necessary $($say your scenario is to share the data in a paper or a word report, so it$$s OK to have $$$$12,345\mathrm{.}67$ instead of $12345\mathrm{.}67.$

For ease of my review, let's cheat a bit and include the original data in a second tab called "Original data".

Make sure there are no extra tabs in your excel spreadsheets.

Do not reorder the rows, this makes it harder for me to compare the original with your dataset.