Fabricated Data

I learned about this concept today in my Environmental Econometrics course. Fabricated data refers to data that is manipulated to suit a specific audience. For instance, imagine I’m conducting a drug trial for approval. Fabricated data analysis involves altering the data to create the appearance that the drug is effective while disregarding any trials where the drug failed.

As a data scientist, recognizing fabricated data is crucial because it can be misleading. Fabricated data is akin to dressing up information in fabric to make it more appealing and convince the intended audience. In the example I mentioned, it’s about persuading regulatory bodies like the FDA that the drug is effective.

However, relying on fabricated data leads to poor sampling, and any models built based on such data will be inherently deceptive. This is why I’m passionate about data science. I aim to delve into the intricacies of statistics, assess data accurately, and empower individuals to make well-informed decisions.

I’m aware that the media often sensationalizes stories, and people manipulate statistics to deceive audiences with false information. Understanding this motivates me to explore how I can utilize data to enact positive changes, much like running a business.

DRAFT colloquial version:

I learned about this word today.

Fabricated data is data that is fixed for an audience.

Lets say I have a drug that I’m trying to test for approval. A fabricated data analysis would consist of manipulating the data so that it will look as if the drug is working, and cleaning any trials where the drug failed.

As a data scientist, this is misleading and you can use the word fabricated data for this.

I learned that fabricating data is like dressing up your ‘data’ or information in fabric so that it can look better and appease the audience you are trying to convince.

In the example I brought up it was to convince like the FDC that the ‘drug’ I am testing is ‘working.’

This is poor sampling, and the linear model made for the drug will be deceiving.

This is why I like data science. I would like to get to the root of the statistics, make evaluations of data, and help people make better informed decisions for themselves.

I know that the news tends to have headliners that can be misleading, and people can be using % to convince the audience at a glance false information.

To get to the root of this, you don’t have to study data science, but this particularly engages me in knowing how can I use these big numbers to convince people to make changes. It’s like a business.