Source: Wundt Research Group; PD-1923 http://en.wikipedia.org/wiki/File:Wundt-research-group.jpg Milgram experiment: public domain; http://en.wikipedia.org/wiki/File:Milgram_Experiment_v2.png
Hello, class. So in today's lesson, we're going to be looking at two key aspects of scientific research. And that's predicting and controlling.
So let's start with predicting. First of all, the purpose of scientific research is to understand the causes of things within the world. We want to know what's happening in the world around us. So that understanding should mean that we should know about something relatively well enough to be able to predict what's going to happen or to say whether or not something will happen given a certain situation or time.
So in other words, the goal of science is prediction. We want to be able to understand it so well that we can say whether or not it will occur. And usually this means predicting so that we're able to help people or change things. So for example, we want to know when an animal might attack or if somebody will get sick so we can prevent it in the future.
So research always starts with a prediction. And that prediction of what might happen as a result of that research is what we call a hypothesis or an educated guess about what will or will not occur within a given situation and with given variables.
However, it's important to note that there are multiple influences on things within scientific research. Nothing is actually as simple as it might seem in the world itself. So anything that can change or be measured or can affect research is what we call a variable. And these variables, if in scientific research, require us to control them or to keep the variables the same so that certain ones can be examined in more detail, because we don't want to examine everything. We want to focus on specific things so we can understand them better.
For example, within a scientific experiment, often we'll have different groups of people that are placed in different groups which we call control or an experimental group. A control group is a group of people that receive all of the conditions of an experiment except the variable that's being tested. So the thing that we're zeroing in on is not given to a control group. In other words, they don't get the experimental factor that's being studied. So the experimental group, on the other hand, would be the people that receive all of the conditions of the experiment plus the experimental condition or the variable that's being studied.
So let's say I was testing out a new drug. So I would take two different groups of people. Well, I'd take one group of people, split them into two groups, and put them both into the exact same type of room. And then in the control group, I would give the people a glass of water. And I would give them something relatively benign, something like what we call a sugar pill, a pill that wouldn't do anything on its own. And then the experimental group, I would give the people a glass of water. And I'd give them the pill that's being studied.
So you notice I control all of the other conditions in the experiment to make sure nothing else is causing the effects that I measure. so I'm trying to make sure the rooms are the same. I'm trying to make sure they're both being given water. And the only thing that I change is the pill that I want to study.
And this control is done as best you can. So obviously, not everything can be controlled. But scientific research requires us to try to do the best we can so we can make sure that we understand the effects of that one variable.
Now, sometimes within a scientific experiment or other scientific research, it's not necessarily the variables that need to be controlled. But sometimes, it's the researcher themself that also needs to be controlled. Researcher bias means when a scientist doing the research affects the research and the results, either intentionally or unintentionally. So the researcher themselves, the scientist has that kind of effect that changes the results.
And this can come from many sources. For example, there's design bias, when the researcher designs it in a way that doesn't necessarily measure what it's supposed to be. There can also be selection bias, which is to say the researcher chooses one particular group for the experiment or might focus in on one type of group that might affect the results of the research.
There's also what we call measurement or reporting bias, where the researcher is looking for specific changes that they expect and so they report those things more than they don't, or they emphasize the change more because they're expecting to see it. And this can be overt, like they can actually be trying to do this. Or it might be accidental just because that's what they're looking for. And finally it can be the presence of the researcher themselves that can affect what the subject does. So if a scientist is in the room, that might be different and might affect the way a person might normally act.
So this is why certain scientific methods are reported to other scientists and examined and repeated before they can become a theory or become more widely accepted by others. And this is why it's important that scientific research is what we call repeatable. It's something that needs to be done over and over before we accept the results as being true.
The ability to forecast outcomes; establish a hypothesis (an educated guess).
Accounting for variables so that only the selected effect is tested
Conscious or unconscious effect the researcher has on the experiment; through design; selection of subjects, or measurement or reporting emphasis, or even the presence of the researcher.