Statistical Ethics and the Outlook for Doctors

We have talked in class about the affect of journals and the manipulation of data. Awareness of this can change the view of the public and the scientific community at large. I was sitting in Biometry, statistics for biology, when he played the video bellow.

I was not surprised but it felt like something to be proud of. There are people in the world who know that there are problems and ways to fix it. It also helps that Dr. Goldacre is an entertaining presenter. I also learned about a different test that can tell us if data is missing from the results based on how much error is.

Above is a type of plot called a funnel plot, used to show how much information is being shown based on error and precision. Ideally, according to Dr.Goldacre's presentation transcript and a cross reference of what looks like an educational website and Science Direct's review of the topic, the shape of the pyramid can show bias and missing studies. Dr. Goldacre states that the smaller the sample size, the more error and variety in the data. In simple terms, the bottom should be wide and even, ideally. The larger populations should be grouped in one area to show similar results, also visualized as a point of the pyramid. If the pyramid is lopsided or multi modal(two pointed, two clusters), bias can be suspected. The American Statistics Association has a set of ethical guidelines for the use and presentation of data, available on their website. Statistics are everywhere, we use them to determine markets, play games, tell the weather, and even predict outbreaks of disease in our own field. To ignore the potential facts because of potential bias is foolish, teaching analysis and critical thinking to our doctors and the presenters is more proactive.