If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

The scientific method

The scientific method is a logical approach to understanding the world. It starts with an observation, followed by a question. A testable explanation or hypothesis is then created. An experiment is designed to test the hypothesis, and based on the results, the hypothesis is refined. This process is repeated, leading to a deeper understanding of the observed phenomenon. This method ensures our knowledge is built on a solid foundation. Created by Sal Khan.

Want to join the conversation?

  • orange juice squid orange style avatar for user Ric Hamel
    Are there true explanations for phenomena that do not conform to the scientific method? Or rather, is the scientific method the only valid way to explain things?
    (66 votes)
    Default Khan Academy avatar avatar for user
    • mr pants purple style avatar for user Tanzina
      The scientific method is more widely accepted because it works almost all of the time but under certain circumstances, general theories and explanations, if they can be redone later using the scientific method, will be acceptable.
      (48 votes)
  • blobby green style avatar for user Mir Vg
    Would it be a good experiment for the example to take a cup of water from the pond and also from the ocean, then cool them both under the same conditions (until they were at the same tempt), then putting them at 0 Celsius to see what happen? (maybe a control cup with destilled water too)

    I think the solutes would also influence the result but it would be more realistic to answer the original question, wouldn't it?

    Please let me know the several caveats of this idea (it's just a question :), thank youu!!
    (30 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Jorge Fuentes
    How is biology similar to chemistry and mathematics? Can you explain this to me....
    (16 votes)
    Default Khan Academy avatar avatar for user
    • winston baby style avatar for user Ivana - Science trainee
      Imagine a pyramid.

      Mathematics sits on the base of everything and holds other Sciences since it explains the Universe itself.

      Physics is above and uses Mathematics to explain physical phenomena.

      Chemistry sits on top of Physics and explains chemical reactions and bond within atoms, molecules.

      Biology sits on Chemistry and explains life with all the foundation of Mathematics, Chemistry, and Physics. Biology is on top of the pyramid.


      Everything is interconnected and cannot exist without one another.
      (48 votes)
  • female robot grace style avatar for user Rachna158🦄
    Was the scientific method developed over time? Or was it created by a particular scientist?
    (20 votes)
    Default Khan Academy avatar avatar for user
    • piceratops ultimate style avatar for user Colin Andrews
      Good question, Rachna.

      Francis Bacon (yes, his name is Bacon) is given most credit for developing the Scientific Method. However, he did not do it alone. He was greatly influenced by the works of Nicholas Copernicus and Galileo Galilei.

      I hope this Helps!

      -GeekWiz
      (28 votes)
  • blobby green style avatar for user minahil zulfiqar
    what if the hypothesis was true but there were defects in experimental methods ?
    (14 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Tanto
    Are you saying the scientific method for Biology is just to do with common sense?
    (12 votes)
    Default Khan Academy avatar avatar for user
  • duskpin tree style avatar for user A.J.
    what is the diffrence between a dependent and independent variable in science? im still so confused about this
    (7 votes)
    Default Khan Academy avatar avatar for user
    • piceratops ultimate style avatar for user Hecretary Bird
      In an experiment, you would adjust the independent variable and measure the effect that the adjustment has on the dependent variable. If your experiment wanted to determine how the color of a room somebody was in affected how well they work, the independent variable would be the wall color, and the dependent variable would be the subject's productivity. You adjust the wall color and see how productivity is affected.
      (16 votes)
  • blobby green style avatar for user Angie Wiebe
    Is it possible to add more elements to the Table of Elements?
    (5 votes)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user Iron Programming
      Yep! One of the most recent discoveries was Oganesson (Og) which was discovered in 2006. One of the exciting aspects of space exploration or underwater exploration (or really exploration at all) is the idea of finding new amazing elements.

      We discovered radioactivity, but what if there are other elements out there with amazing unheard of abilities?

      The possibilities are pretty much endless, which is a pretty exciting part of Chemistry if you are interested in doing that type of thing.

      Who knows, there might be thousands more elements to come... or maybe only a couple more.

      Happy learning! :-)

      EDIT: This answer was a little naive--most elements that have been discovered recently can only be created in a lab setting due to their instability. There is a practical limitation on the # of protons that can be in the nucleus and how many energy levels can exist for electrons. The limited range of the nuclear strong force is key.
      (15 votes)
  • aqualine ultimate style avatar for user ΣҳơʄყƖ
    Is the first step of the scientific method asking a question or making an observation? Some websites say that making an observation is the first step, while others say that asking a question is the first step.
    (6 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Marc-Dylan Lee-Williams
    If I was to heat up metal would it turn into a liquid ?....
    (6 votes)
    Default Khan Academy avatar avatar for user

Video transcript

Let's explore the scientific method. Which at first might seem a bit intimidating, but when we walk through it, you'll see that it's actually almost a common-sense way of looking at the world and making progress in our understanding of the world and feeling good about that progress of our understanding of the world. So, let's just use a tangible example here, and we'll walk through what we could consider the steps of the scientific method, and you'll see different steps articulated in different ways, but they all boil down to the same thing. You observe something about reality, and you say, well, let me try to come up with a reason for why that observation happens, and then you try to test that explanation. It's very important that you come up with explanations that you can test, and then you can see if they're true, and then based on whether they're true, you keep iterating. If it's not true, you come up with another explanation. If it is true, but it doesn't explain everything, well once again, you try to explain more of it. So, as a tangible example, let's say that you live in, in I don't know, northern Canada or something, and let's say that you live near the beach, but there's also a pond near your house, and you notice that the pond, it tends to freeze over sooner in the Winter than the ocean does. It does that faster and even does it at higher temperatures than when the ocean seems to freeze over. So, you could view that as your observation. So, the first step is you're making an observation. Observation. In our particular case is that the pond freezes over at higher temperatures than the ocean does, and it freezes over sooner in the Winter. Well, the next question that you might wanna, or the next step you could view as a scientific method. It doesn't have to be this regimented, but this is a structured way of thinking about it. Well, ask yourself a question. Ask a question. Why does, so in this particular question, or in this particular scenario, why does the pond tend to freeze over faster and at higher temperatures than the ocean does? Well, you then try to answer that question, and this is a key part of the scientific method, is what you do in this third step, is that you try to create an explanation, but what's key is that it is a testable explanation. So, you try to create a testable explanation. Testable explanation, and this is kind of the core, one of the core pillars of the scientific method, and this testable explanation is called your hypothesis. Your hypothesis. And so, in this particular case, a testable explanation could be that, well the ocean is made up of salt water, and this pond is fresh water, so your testable explanation could be salt water, salt water has lower freezing point. Has lower freezing, freezing point. Lower freezing point, so it takes colder temperatures to freeze it than fresh water. Than fresh water. So, this, right over here, this would be a good hypothesis. It doesn't matter whether the hypothesis is actually true or not. We haven't actually run the experiment, but it's a good one, because we can construct an experiment that tests this very well. Now, what would be an example of a bad hypothesis or of something that you couldn't even necessarily consider as part of the scientific method? Well, you could say that there is a fairy that blesses that blesses, let's say that performs magic, performs magic on the pond to freeze it faster. Freeze it faster. And, the reason why this isn't so good is that this is not so testable, because it's depending on this fairy, and you don't know how to convince the fairy to try to do it again. You haven't seen the fairy. You haven't observed the fairy. It's not based on any observation, and so this one right over here, this would not be a good hypothesis for the scientific method, so we would wanna rule that one out. So, let's go back to our testable explanation, our hypothesis. Salt water has a lower freezing point than fresh water. Well, the next step would be to make a prediction based on that, and this is the part where we're really designing an experiment. So, you could just view all of this as designing. Let me do this in a different color. Where we wanna design an experiment. Design an experiment. And in that experiments lets say, and let's see, the next two steps I will put as part of this experimental. Whoops. I messed up. Let me, I did my undo step. So, the next part that I will do is the experiment. Experiment. And there you go. So, the first thing is, we'll say I take, you know, there's all sorts of things that are going on outside. The ocean has waves. You know, maybe there are boats going by that might potentially break up the ice. So, I just wanna isolate that one variable that I care about, whether something is salt water or not, and I want a control for everything else. So, I want a control for whether there's waves or not or whether there's wind or any other possible explanation for why the pond freezes over faster. So, what I do, in a very controlled environment I take two cups. I take two cups. That's one cup and two cups, and I put water in those cups. I put water in those cups. Now, let's say I start with distilled water, but then this one stays, the first one right over here stays distilled, and distilled means that through evaporation I've taken out all of the impurities of that water, and in the second one I take that distilled water, and I throw a bunch of salt in it. So, this one is fresh, very fresh, and in fact, far fresher than you would find in a pond. It's distilled water. And then this is over here, this is salt water. So, you wouldn't see the salt, but just for our visuals, you depict it. Then we would make a prediction, and we could even view this as step 4, our prediction. We predict that the fresh water will freeze at a higher temperature than the salt water. So, our prediction, let's say the fresh freezes at zero degrees Celsius, but salt doesn't. Salt water doesn't. Salt water doesn't. So, what you then do is that you test your prediction. So, then you test it. And how would you test it? Well, you could have a very accurate freezer that is exactly at zero degrees Celsius, and you put both of these cups into it, and you wanna make sure that they're identical and everything where you control for everything else. You control for the surface area. You control for the material of the glass. You control for how much water there is. But, then you test it. Then you see what happened from your test. Leave it in overnight, and if you see that the fresh water has frozen over, so it's frozen over, but the salt water hasn't, well then that seems to validate your testable explanation. That salt water has a lower freezing point than fresh water, and if it didn't freeze, well it's like, okay, well maybe that, or if there isn't a difference, maybe either both of them didn't freeze or both of them did freeze, then you might say, well, okay, that wasn't a good explanation. I have to find another explanation for why the ocean seems to freeze at a lower temperature. Or, you might say, well that's part of the explanation, but that by itself doesn't explain it, or you might now wanna ask even further questions about, well, when does salt water freeze, and what else is it dependent on? Do the waves have an impact? Does the wind have an impact? So, then you can go into the process of iterating and refining. So, you then refine, refine, refine and iterate on the process. When I'm talking about iterate, you're doing it over again, but then, based on the things that you've learned. So, you might come up with a more refined testable explanation, or you might come up with more experiments that could get you a better understanding of the difference between fresh and salt water, or you might try to come up with experiments for why exactly, what is it about the salt that makes this water harder to freeze? So, that's essentially the essence of the scientific method, and I wanna emphasize this isn't some, you know, bizarre thing. This is logical reasoning. Make a testable explanation for something that you're observing in the world, and then you test it, and you see if your explanation seems to hold up based on the data from your test. And then whether or not it holds up, you then keep going, and you keep refining. And you keep learning more about the world, and the reason why this is better than just saying, oh well, look, okay, I see the pond has frozen over and the ocean hasn't, it must be the salt water, and you know, I just feel good about that, is that you can't feel good about that. There's a million different reasons, and you shouldn't just go on your gut, 'cause at some point, your gut might be right 90% of the time, but that 10% that it's wrong, you're going to be passing on knowledge or assumptions about the world that aren't true, and then other people are going to build on that, and then all of our knowledge is going to be built on kind of a shaky foundation, and so the scientific method ensures that our foundation is strong. And I'll leave you with the gentleman who's often considered to be the father, or one of the fathers of the scientific method. He lived in Cairo, and in what is now Egypt, nearly 1,000 or roughly 1,000 years ago. And he was a famous astronomer and phycisist and mathematician. And his quote is a pretty powerful one, 'cause I think it even stands today: "The duty of the man who investigates the writings of scientists, if learning the truth is his goal, ..." Let me start over, just so I can get the dramatic effect right. "The duty of the man who investigates the writings of scientists, if learning the truth is his goal, is to make himself an enemy of all that he reads, and ... attack it from every side. He should also suspect himself as he performs his critical examination of it, so that he may avoid falling into either prejudice or leniency." Hasan Ibn al-Haytham, and his Latinized name is Alhazen. So, he's saying be skeptical, and not just skeptical of what other people write and read, but even of yourself. And another aspect of the scientific method which is super important is, if someone says they made a hypothesis and they tested and they got a result, in order for that to be a good test and in order for that to be a good hypothesis, that experiment has to be reproducible. Someone can't say, oh it's only, you know, a certain time that only happens once every 100 years and not, that that's why it happened that day. It has to be reproducible, and reproducible is key, because then another skeptical scientist like yourself can say, let me see if I can reproduce it. Let me not just believe it, because that person looks like they're smart, and they said that it is true.