Experimental Design - Online Tutor, Practice Problems & Exam Prep

An experiment is a scientific procedure designed to test a hypothesis, involving independent and dependent variables. The independent variable is controlled by the researcher, such as the amount of water given to plants, while the dependent variable, like plant growth, is measured. To prevent false positives and negatives, scientists use control groups: negative controls show no expected response, while positive controls show a known response. This ensures accurate results and reliable conclusions in scientific investigations.

Variables

In this video, we're going to talk about experimental design by focusing on the variables within an experiment. First, we need to define what an experiment actually is. An experiment can be defined as a scientific investigation or procedure that's designed to test the validity of a hypothesis or a theory. Within a well-designed experiment, there will be variables. As its name implies, the root of the word variable is "vary," which means it is going to involve changes that vary over time. Variables are changeable elements of the experiment and will change throughout the course of the experiment.

Scientists typically investigate the relationship between two main types of variables. Below in this table, we're going to introduce those two main types of variables. The first type of variable that scientists typically investigate is the independent variable, and the second variable is the dependent variable. The independent variable is a variable which means it is going to change throughout the course of the experiment. However, the independent variable is controlled or modified by the researcher.

Here, we have a few potential examples of what could be an independent variable in an experiment. One potential example is the age group of the people involved in the experiment. The researcher can control or modify the age group they focus their experiment on. Perhaps the researcher chooses to focus their experiment on the elderly group of people, or perhaps they focus on the middle age group, or perhaps they focus on the younger group. Because the age can be controlled or modified by the researcher, it could potentially be an independent variable in an experiment.

Another potential example is the amount of time someone is exposed to something. Perhaps the scientist chooses to expose the group to a chemical for a long period of time or a short period of time, but the time is something that can be controlled or modified by the researcher and so it could potentially be an independent variable in an experiment. Another example of an independent variable is the amount of a specific reagent that is used, which can be controlled or modified by the researcher, making it an independent variable.

The dependent variable, on the other hand, cannot be controlled or modified by the researcher. Instead of trying to control or modify this variable, researchers will measure or investigate it. The dependent variable can be defined as the variable that is measured or investigated by the researcher since they cannot control or modify it directly. Some potential examples of dependent variables for an experiment include the growth of a plant, which a scientist cannot directly control, making it a potential dependent variable. Another example is drug effectiveness or the effectiveness of a drug. Since a researcher cannot directly control or modify its effectiveness, they will measure drug effectiveness, making it a potential dependent variable.

We are going to analyze the independent and dependent variables in an experiment testing the effect of water on plant growth. You'll notice that in a typical graph, the independent variable is usually on the x-axis (the horizontal axis of the graph), and the independent variable again is going to be the variable that is controlled or modified by the researcher. In this experiment, you'll notice we've got two plants, both receiving the same amount of sunlight but different amounts of water. One plant is receiving little H₂O or a small amount of water, and the other is receiving high H₂O or a lot of water. Because the scientist is controlling and modifying the amount of water, we can say that the water amount is the independent variable.

In this example, the dependent variable will be plant growth, which is usually on the y-axis of the graph (the vertical axis of a graph). In the experiment, the plant that received little water did not grow nearly as much as the plant that received a lot of water. You'll notice here in the experiment, when there is little water, there is not a lot of growth, so we would expect a data point somewhere around here. But when there is a lot of water, there is a lot of plant growth, so we would expect a data point somewhere around here. We would expect a trend to be something like this. This graph shows the relationship between the independent variable, the amount of water—controlled or modified by the researcher—and its impact on the dependent variable, plant growth, which is going to be measured by the researcher.

This concludes our brief lesson on the variables within an experiment, and we'll be able to get some practice applying these concepts as we move forward. I'll see you all in our next video.

Jonathan wants to know which style/model of paper airplane is going to win the contest by traveling the furthest. He designs 5 different models of paper airplanes and drops each of them from the same height of 20 meters. He records the distance that each plane travels before it hits the ground. What are the independent and dependent variables of Jonathan's experiment?

            Independent Variable: ______________________________________ 

            Dependent Variable: _______________________________________

Alright. So here we have a practice problem that says, Jonathan wants to know which style or model of paper airplane is going to win the contest by traveling the farthest. He designs 5 different models of paper airplanes and drops each of them from the same height of 20 meters. He then records the distance that each plane travels before it hits the ground. What are the independent and the dependent variables of Jonathan's experiment? And so you can see that we can put the independent variable here in this blank and the dependent variable here in this blank. And so what we need to recall from our last lesson video is that the independent variable is going to be the variable that is controlled or modified by the researcher, and the researcher here is going to be Jonathan himself. And of course, the dependent variable, on the other hand, is going to be the variable that is measured by the researcher. And so when we go back and revisit this practice problem up above, what you'll realize is what Jonathan has complete control over is going to be the styles, the different models of paper airplanes, and so there are exactly 5 of them, and he can control exactly which models he decides to use. And so because he has complete control over this variable and he, can change it as he pleases, we can say that the independent variable is going to be the different models of airplanes. And so that is it for the independent variable, the different models of paper airplanes. Now when it comes to the dependent variable on the other hand, what Jonathan will be measuring is the distance that each plane travels before it hits the ground, and so that would be the dependent variable. So what we'll say is the distance traveled, by each airplane. That is what he's going to measure. He can't really physically control how far each of these is going to travel. So instead what he does is he measures that and he investigates that. And so really, these here are the answers to this practice problem and that concludes this practice so I'll see you guys in our next video.

In an experiment to test the effect of temperature bacterial reproduction rate, temperature would be the:

• A. Standardized variable.
• B. Dependent variable.
• C. Control variable.
• D. Independent variable.

The temperature at which an alligator's egg is incubated will determine the sex of the offspring. The dependent and the independent variables in this experiment are ________.

• A. Sex of the baby alligator and temperature respectively.
• B. Temperature and sex of the baby alligator respectively.
• C. Size of the incubator and size of the baby alligator respectively.
• D. Number of offspring and temperature in the incubator respectively.

False Positives/Negatives

In this video, we're going to talk about false positives and false negatives. And so what you guys should know is that a very well-designed experiment is going to contain control groups. Now, in our next lesson video, we'll talk a lot more about these control groups, but for now, what you should know is that the control groups that scientists use in a well-designed experiment are specifically supposed to prevent false positives and false negatives. So, what are these false positives and false negatives? Well, a false positive is defined as an outcome that falsely indicates the presence of a result. For example, if someone were to take a pregnancy test and that pregnancy test were to say that they are pregnant when in reality they actually are not pregnant, that would be a false positive. Now, on the other hand, a false negative would be the opposite. This is going to be an outcome that falsely indicates the absence of a result. For example, if the pregnancy test were to say that you're not pregnant when in reality you actually are pregnant, that would be a false negative. And so really, false positives and false negatives are bad, and scientists do not want to have false negatives and false positives in their experiments. And so in order to avoid or prevent false positives and negatives, scientists use control groups. And once again, we'll talk more about these control groups in our next lesson video. So, I'll see you guys there.

Negative & Positive Controls

In this video, we're going to talk about how scientists avoid or prevent false positives and false negatives in their experiments by using negative and positive controls. There are two main types of controls that are used in experiments: the negative control and the positive control. Ideally, these control groups will only differ from the experimental group in the one factor that's being tested.

Let's distinguish between the two types of control: the negative control and the positive control. In the first column, we have the control type, which again are going to be the negative control and the positive control. The negative control, as its name implies, is the control group where no response is expected; it's expected to react negatively to the test, which is why it's called the negative control. For example, this would be like using a placebo, such as a sugar pill that isn't supposed to do anything at all. The purpose of using a negative control is to prevent false positives, which we defined in our last lesson video.

The positive control, by definition, as its name implies, is going to be the control group where a response is expected. It's supposed to react positively to the test, and that's why it's called the positive control. This would be, for example, using something like a brand name pill that has been proven to work successfully in the past. The purpose of using a positive control is to prevent false negatives.

Let's look at this image down below, which is an example of an experiment where they're testing a brand new experimental pill to see its drug effectiveness on a toe injury. When testing this experimental pill, it's important to include a negative control group and a positive control group. The negative control group would be something where we have expectations that it will not react; there will be no response. Using a sugar pill as a placebo, which is not supposed to do anything, is an example. So, if you take a sugar pill like this one right here, it's not supposed to help heal your toe injury. This expectation that the sugar pill will react negatively and will have very little drug response defines it as the negative control.

Over here, we have the positive control, and we have expectations that it will react positively. It will show a response. Using something like a brand name pill that has been successful in helping with toe injuries would be an example of a positive control because we have expectations that it should react positively and it should have some level of drug effectiveness.

Notice on the right-hand side, we have a graph where on the x-axis, we have the independent variables, which the scientists have control over, namely the exact type of pill they decide to use. On the y-axis, we have the dependent variable, which is what the scientists are going to measure, which would be the drug effectiveness and how good it is at healing the toe injuries. The sugar pill here is going to be our negative control because we have expectations that it should react negatively. It should not help with the toe injury at all, and it should show 0% drug effectiveness. If we were to actually use this sugar pill on a group and it showed that it had 100% drug effectiveness, then this would be an example of a false positive. Hence, including a sugar pill or a negative control group and having this group respond as expected is helping to prevent false positives. On the other hand, we have the brand name pill, which we said is going to be the positive control here because we have expectations that it should react positively and it should give some level of response. This proven effect in the past allows us to anticipate some level of drug effectiveness.

If this brand name pill were to somehow show 0% drug effectiveness, then that would be an example of a false negative. By including the brand name pill here and having it respond as expected, we're helping to prevent false negatives in our tests.

The experimental group here would be the brand-new experimental pill that we're testing for maybe the first time, and this experimental pill might respond at any level, from 0 to 100%. If it responded below the brand name pill, if it had less drug effectiveness, then perhaps the scientists would recommend not using the experimental pill as it's not as good as the brand-name pill. But if the experimental pill responded better, with better drug effectiveness, then perhaps the scientists would advise trying this experimental pill because it might help with your toe pain better than the brand-name pill. This demonstrates how positive and negative controls can be very helpful and useful for a scientist.

This concludes our introduction to the difference between negative and positive controls, and we'll be able to get some practice moving forward in our course. See you guys in our next video.

Experimental Design Example 1

Alright. So here we have an example problem that says, a scientific researcher designs an experiment to test the effectiveness of a new sleeping pill compared to the brand name product. Match each of the following controls or outcomes in the experiment, which we have over here, to their appropriate description which we have over here. And so notice that the first description here says a sugar pill that should have no effect on the patient. And so we have an expectation that this sugar pill should have no effect or it should respond negatively to the test. And so because we have an expectation that it should respond negatively this would be our negative control. And so in this blank we can put number 1, so that it matches with number 1 and then we can cross off 1 from our list.

Now moving on, the second description says a patient does not fall asleep after taking the brand name pill. Now remember that the brand name pill has been proven to be successful in the past and so we have expectations that it should react positively. And if it does not react positively, remember the brand name pill is supposed to be a sleeping pill, so it's supposed to put people to sleep. And if the patient does not fall asleep that means that we're getting a negative result even when we're supposed to have a positive result, and so this would be an example of a false negative and so that would match with option d so we could put 2 here and then cross off 2 from this list.

Number 3 says a brand name pill that has proven to work on patients. Once again we have an expectation that this brand name pill should react positively, it should work and because we expect that it should react positively this would make it the positive control and so we can put number 3 here and cross 3 off our list.

Last but not least, the false positive is going to be number 4 which says a patient falls asleep after taking the ineffective sugar pill. The sugar pill here would be like the placebo, which would be the pill that is not supposed to have any effect since it's like a fake pill. And so this is going to match with 4 being a false positive. The patient falls asleep, that's the positive result, when they're not supposed to fall asleep. And so, here are the answers to this example problem, and that concludes this example, so I'll see you in our next video.

A scientist wants to study the effects of nitrogen on wheat plants. They set up an experiment with 4 groups of plants: group A gets 20 pounds of nitrogen per acre, group B gets 40 pounds per acre, group C gets 60 pounds per acre, and group D gets 0 pounds per acre. Which of the following is the control group? Is it a positive or negative control group?

• A. Group A.
• B. Group B.
• C. Group C.
• D. Group D.