hi welcome to the wonderful world of the scientific hypothesis my name is brad alger and today we're going to focus on the foundational concepts of hypothesis and prediction at the end of this short video you'll easily be able to tell them apart and you'll begin to see why it's so important to distinguish between them if you really want to understand scientific reasoning scientific papers and scientific reports where the distinctions are not always made clear and if you're a scientist or science student you'll come away with a much better appreciation for how these fundamental concepts can help you in your own scientific thinking experimental design and so forth okay hypotheses and predictions the basics we'll define them and compare them and then look at their relationships first off what is a scientific hypothesis and how is it different from a prediction well a hypothesis is a potential explanation for an observation or phenomenon it tries to say why things are as they are a prediction on the other hand is simply a statement of what you think will happen in the future maybe as a result of doing an experiment but it doesn't itself explain anything and this is fundamentally why no matter what you read or hear about sometime hypotheses and predictions are not the same at all these ideas are of course related we say that a hypothesis makes or implies predictions which is the same thing and for our purposes what that means is if the hypothesis is true a prediction that it makes must be true and this is the relationship of logical deduction if the prediction is false the hypothesis is false now here's where things get tricky because if the prediction is true the hypothesis might or might not be true and we'll look at several reasons why this is the case so for our purposes here for experimental science the most important relationship is this one where the hypothesis is false if the prediction it makes is false and in general these two ideas form the basis of the scientific reasoning that we're going to talk about these ideas have some important consequences the first is that a hypothesis must be testable that means that it must make at least one experimental prediction that could if tested and if the result came out a certain way show that the hypothesis is wrong and this is what we mean by saying that the hypothesis is falsifiable it also follows that a scientific hypothesis is about a problem that cannot be solved by making a single observation and it often involves unobservable factors that play a role in something called a mechanism that we'll get to in later videos let's consider an example of a hypothesis say there's a lake nearby and one day you notice that fish in the lake are dying and this raises the question of why on thinking about it you realize there's a state nearby where there's a lot of heavy industry and you formulate the hypothesis that the fish death is caused by acid rain resulting from industrial air pollution now notice what a really very complex idea this is it's a hypothesis that makes many predictions and you cannot test it directly by doing any single test for example it predicts that the material going up the smoke stacks into the sky forms acid rain in the clouds that the prevailing winds blow the clouds over the lake that enough rain falls from the clouds to change the ph in the lake and so on however if you break it down you realize that there are single predictions that you can test for example in this case the prediction would be that the lake water will be acidic and this you can test directly by getting a ph meter in a sample of the lake water and measuring the ph of the water notice that if the ph of the lake water were neutral the acid rain hypothesis would be false but also notice that even if the prediction is true the hypothesis might or might not be true maybe the fish tolerate the acidity quite well but the water lacks oxygen this example illustrates a number of principles that are quite important for example any hypothesis makes a large number of predictions and we test hypotheses indirectly by testing their predictions and if any of the predictions is false as we said the hypothesis itself is false we test a prediction directly however by making a measurement we can use the same kind of graph to illustrate a different point that i mentioned earlier here we have the case where you have a number of very different hypotheses all of which make the same prediction and you can see right away that confirming a prediction supports all the hypotheses that make the same prediction for example in principle it's possible that there are a number of different reasons that the lake water was acidic and therefore measuring lake water and finding it that it was acidic would not prove the acid rain hypothesis is true and in general you can't conclude that a positive prediction test confirms any given hypothesis now this table summarizes many of the points that we've made the hypothesis is explanatory whereas the prediction is non-explanatory the hypothesis is the primary concept it implies predictions and therefore predictions are secondary and they follow from hypotheses we can test both hypotheses and predictions but we test hypotheses only indirectly by testing their predictions whereas we test predictions directly by making measurements as a result of the testing we can only show that a hypothesis is false or that the results are consistent with it but we can't prove that a hypothesis is true a prediction on the other hand we can find is either true or false now in later videos we'll refer to the points made in this table and you'll see how useful it is to keep the distinction between hypothesis and prediction clear thanks for watching remember to give it a thumbs up if you like it and subscribe to hear more see you next time