![]() ![]() So you send a survey out to your newsletter subscribers, right? Of course, interacting with your audience is important (I send out surveys to my Newsletter Subscribers every now and then, as well), but there’s an issue! When it gets to survey results, you should be aware that your newsletter subscribers do not represent your full audience. Before spending a lot of time and money, you want to know whether your audience would pay for it at all. Let’s say that you want to create a new product on your website. Here’s another example of selection bias. Note 2: If you want to read another classic selection bias story, check how Literary Digest made a similar mistake (also referred to as undercoverage bias) ~80 years ago! Data science related example of selection bias: ![]() Note 1: I do recommend blocking your Facebook feed for many reasons, but mostly so you don’t get narrow-minded: FB News Feed Eradicator! (In fact, it’s even narrower, because they see only the opinion of friends who are active and posting to social media – so a certain segment of their friends are overrepresented.) That’s classic selection bias: easy-to-access data, but only for a very specific, unrepresentative subset of the whole population. Very bad and sad practice, because what they see there does not show the public opinion – it’s only their friends’ opinion. Unfortunately for many of them, the source of their “research” is their social media feed. Most people have an immediate and very “well-informed” answer for that. Please answer this question: What’s people’s overall opinion about Donald Trump’s presidency? There are many underlying reasons, but by far the most typical I see is collecting and working only with data that is easy to access. Usually, this means accidentally working with a specific subset of your audience instead of the whole, rendering your sample unrepresentative of the whole population. Selection bias occurs when you are selecting your sample or your data wrong. I see these to affect the job of data scientists and analysts everyday. Why? Because these nine types of statistical bias are the most important ones. ![]() I have chosen to show you only 9 of these. There is a long list of statistical bias types. The most important statistical bias types Everything I will describe here is to help you prevent the same mistakes that some of the less smart “researcher” folks make from time to time. For ease of understanding, I’ll provide two examples of each statistical bias type: an everyday one and one related to data science!Īnd just to make this clear: biased statistics are bad statistics. But I promise that even if you are not looking for a data science career (yet), you will profit a lot by reading this. This is blog for aspiring data scientists, so in this article I’ll focus only on the most important statistical bias types. Even so, at least we can be a bit smarter than average, if we are aware of them. The most obvious evidence of this built-in stupidity is the different biases that our brain produces. We all are, because our brain has been made that way. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |