Hi. This tutorial covers selection bias. So recall that just bias in general is the tendency for collected data to differ from what is expected in a systematic way. Biased data can often favor a specific group of those studied. Bias jeopardizes the accuracy of the data collected. Studies can be carried out and evaluated better if the different types of bias can be recognized.
So let's define selection bias. Selection bias is a type of bias that occurs when some groups in the population are left out of the process of choosing the sample. Selection bias yields an unrepresentative sample. So if some groups are left out of the sampling process, they won't be reflected in the sample, so the features of the sample may differ significantly than the features of the population.
So let's take a look at an example of a study where selection bias may come into play. So suppose that a large university is interested in learning about credit card debt among their students. A random sample of students is selected from the undergraduate dormitory residents list and contacted.
So selection bias has occurred here. Undergraduate students who do not live in the dorms, as well as graduate students are unable to be selected for the survey. So because they're taking their sample only from these undergraduate dormitory lists, the students that don't live in the dorms, as well as the graduate students are not able to be part of the study.
Now, that's important, because these students may have very different financial situations than those who can afford to live in the dorms. So perhaps graduate students, because their tuition is generally more expensive and they already have, maybe, debt still from their undergraduate degree, they might be more likely to have higher amount of credit card debt than generally younger undergraduate students. So because these portions of the population were unable to be selected, selection bias had occurred.
So a better way of selecting the sample would be to randomly dial phone numbers of enrolled students. This technique would allow all of the students to be part of the sample, giving you a much more representative sample. So this idea of randomly dialing phone numbers is a common method of selecting samples-- generally called random digit dialing. And usually that's a good way of avoiding selection bias. So that is your tutorial on selection bias. Thanks for watching.
Hi, this tutorial covers deliberate bias. So let's just quickly review what biases first. So recall that bias is the tendency for collected data to differ from what is expected in a systematic way. Biased data can often favor a specific group of those studied. Bias jeopardizes the accuracy of the data collected. Studies can be carried out and evaluated better if different types of bias can be recognized.
So let's just start with a definition of deliberate bias. So deliberate bias is a type of bias that occurs when a study, in its design or reporting, is deliberately done in a way to advance an interest, be it a financial interest an ideological interest or some sort of personal interest.
Contrasting that with unintentional bias, unintentional bias or types of bias that occur that are not motivated or intentional. Most of the named type of biases are unintentional. So you may have learned about response bias, nonresponse bias, selection bias. A lot of those are unintentional biases, still biases. But these would be more unintentional.
So let's take a look at an example of where you sometimes see deliberate bias. So an EIS, which is an environmental impact statement, must be prepared before construction can begin on many public works projects, such as roads. By law, and eia as is required to consider several alternative ways in which construction could potentially occur.
Typically, writing an EIS involves an analysis of previous existing data. Unfortunately, the data an EIS presents is often chosen to support a decision that has been already made, not to fairly investigate all possibilities.
So since they're using pre-existing data, they're going to just simply use the data that's going to benefit them the most, so the one that's going to cost the least for them instead of investigating possibilities that might have less than an environmental impact. So this is one way where deliberate bias occurs because of, generally, financial interests of these construction companies. So that is the tutorial on deliberate bias Thanks for watching.