- Now, before we start
learning about atoms and how atoms come
together to form molecules, and the like, we first have to look at how we approach the discovery of any endeavor, not just biology, but anything because what you're gonna see
throughout the semester is, there's a lot of things
we know, very well. There's a lot of things
we know, fairly well. And there's a whole lot of
things we kind of know about. And the problem is our
method of investigation, sometimes, we just cannot
learn all of the answers, we either don't have
the technology for it. We haven't studied it enough. And so this lecture is all about how we go about learning new information, and it doesn't just apply,
like I said, to biology in fact, a good number of your questions are gonna have nothing to do with biology. They're gonna talk about
problem solving your car, forking out $500 for a stupid alternator, like I am right now. Whether you're a mechanic, and you get paid really
well to problem solve, whether you're a computer
designer or program designer, and you're figuring out
issues and all that, the same methodology applies. So this is gonna be important because we're not only gonna learn why
and how we gain information, but you're gonna see why we make so many mistakes today too. So you can see that, yes, we're not trying to
trick everybody by saying, "Oh, shoot, we got this wrong." It's just the methodology
is not always perfect and the way we executed
is not always perfect, because it's very difficult to do. So that's the goal of this lecture, is to see what the scientific
method is, how it applies, and very specifically, what is our level of acceptance
of the various concepts that we're going to learn
from hypothesis through law. All right. So let's start
with the first thing, the scientific method, is not as simple as we usually portray it, we're gonna put it forward simply today, but, and in a way of four steps, and you go through those four steps, it's not as simple as that. But we're going to study it
in this simplified version. Because the real thing is, as you go through these steps, you're always hitting hurdles. You've gotta discover new
technology, some new protocol, and you get stuck in various parts and have to go round
and round around until you can, plow through to the next level. And that's why it takes
so many freaking years to your graduate work and all that stuff. The scientific method is essentially a way of
gathering information. It's a framework of
repeatable experimentation. That's how mechanics are able to deduce, this is how things should work, this is out of place, this is the issue, at
least good mechanics can. And as far as scientists go, they're trying to observe what's going on and understand the natural
laws that are in our universe. But this can apply to
any type of endeavor, not just science, not just cars, not just computers, but any endeavor. We're trying to understand
how everything works, that's really what it's all about. So, this shows more of a cyclical nature of the scientific method
because it never ends, ever. Until we know everything, that's not gonna happen anytime soon. So the reason why it's cyclical is because we build upon our prior knowledge. We're constantly making greater
and greater discoveries, based upon what we've learned in the past. And that's one of the important aspects, as you can see here to
make known what we discover so that others can work off
of our shoulders, so to speak, and we can make greater
and greater advancements. If we lose that knowledge. And you have a society that's
completely isolated from that then they have to start all over again or start from scratch. I mean, that knowledge
has to be passed on, otherwise, it's lost. So there are four main steps. I know there's several things in here, but there's four main steps
that you're gonna be tested on. They're representative of again, how your multiple choice
questions are gonna be. And they're as follows, observation, hypothesis, experiment, and conclusion. So those are the four basic
steps of the scientific method. So let's start with
number one, observation. A lot of times people
just kind of miss this one as an actual step of the scientific method but let me give you
some background history as far as why that's so important. Until scientists had
actually been able to, create microscopes and
see the very, very small, we couldn't come up with cell theory, we didn't understand that living
things were made of cells, we couldn't see that small, the resolution of our eyes
is just not that good. So until we had the
technology to start observing, the very, very small, we didn't understand some
of the fundamentals of life. Let's go broader. Let's go out into the universe, scientists are constantly
trying to figure out how our universe works and whatnot, they've got massive telescopes and satellites and other things. They've discovered, they've
known this for quite a while, but it's fairly well
accepted among scientists is that, about 96% of our universe is unobservable by us, this constitutes two main substances which we call dark matter and dark energy. That means, when you look
out in the night sky, you see all those stars and you know that there's
other things and nebulas and other things out there
that's 4% of our universe. So there's quite a bit
that we have yet to learn because 96% of it is unobservable. How do we know that that even exists? We're only now beginning to see the influence that this
matter has on the 4% of observable matter but we cannot directly observe this yet. So there's a substantial
amount of observation yet to come about once
we start figuring out how to make those observations of what our universe is really made of. It'd be like living in the ocean, and knowing what the
individual salt particles are, but not knowing what water is. I mean, that's how ridiculous it is. Observation is the very first step. You need to be able to observe something if your going to learn about it. So without the necessary
tools or instrumentation, I mean, we have radio waves
going through us all the time, we have all sorts of electromagnetic
energy passing through. And without the proper tools, we're completely unaware of it. You just can't detect radio
waves unless you have something that can pick that signal
up and interpret it. All right. So once we make an observation, the natural tendency is to ask, what is it that we're observing? But that's not an actual step
of the scientific method. That's just a, hey, what is that? What's that dark matter? What is that cell or whatnot? The second stage of the scientific method is to formulate what we call a hypothesis. A hypothesis is your educated guess and we stress educated, but it's a guess, nonetheless, of what it is you're observing. So you see something and you're like, "Hey, I think it's this." The more knowledge you have about a particular phenomena, fixing the car, computer programming, the more knowledge you have, the better your hypothesis. For example, let's look at a car. You wake up as I did six
o'clock this morning, you go out to your car, try to start it and it doesn't start, as always. There's a number of reasons
on why a car won't start. Your observation is the car won't start. Your hypothesis is, what is wrong? What ultimately is making it
so that the car will not start. So let's come up with
some viable hypothesis. Lemme throw a really bad one out there. The tires are flat. Would that be a good hypothesis? Why not? - 'Cause it has nothing to do-- - 'Cause it has nothing to do with whether the car will start. What if I'm from the 18th century or 17th century whatnot? I don't have any prior knowledge on what what matters in the car. But today, most of us have a
good understanding at least, of some of the basic mechanics of a car. But that's where, a good mechanic can get
right to the problem, a bad mechanic or a
dishonest one muddles around and doesn't quite get
to the actual reason. So let's look at some viable hypotheses, on why the car won't start. - The battery. - Okay, battery's dead. What else? - Your alternator died. - My alternator is not working. - Starter. - Actually it fried my car, instead of not being able to. Starter, so you're
dealing with things like the solenoid and other instruments. What is another easy reason?
- Gas. - Gas, okay, out of gas. Being cold, we usually know that that tends to have an effect on batteries, especially when they sit there
for long periods of time. So, let's just choose this one. There's many we can choose from. And there's actually little warning signs when you try to start up the car that the more you know, you can even do a diagnostic in the moment that you're trying to
figure out what's going on. It's not necessarily an experiment. But again, the more knowledge you have, the more you can be like, "Well, the car doesn't even turn over, "so it's very likely the battery." But if the car is turning over, then you're dealing with maybe
some other issue or whatnot. So, the more you know about it, the more you can get to
the answer right away. So let's deal with an experiment. Let's say that the battery is dead. What are some good approaches
to testing your hypothesis and here's the other
aspect about a hypothesis. It must be testable. There are many hypotheses out there that people have regarding
various observable phenomena, and a good majority of
them can't even be tested. So they're not good hypotheses because, a good hypotheses has even
the capacity of being tested, theoretically or otherwise. Sometime experimentally, where we actually perform something but there's a lot of
theoretical physicists out there that are able to say, theoretically this is possible and so that just leads us in
the right direction or whatnot. So experiment wise, what can you do? - Replace the battery. - What's that? - Replace the battery, - Okay, you can replace it. - Try and jumpstart it. - You can try to jumpstart it. Those are two, anything else you can do? - You can test other things and if they're not bad then, process of elimination. - Okay, so if you have the instrumentation you can, weed out some of the other factors to try to get it all the way down to one. I don't know how I write that
up there, but won't do that. - The way it sounds. - Okay, so as you start it, the sounds that it makes. Okay, so obviously, most of us don't have the
knowledge that a mechanic has, and we usually go with
the simplest solution. What do we usually do
when the battery's dead? What's the most obvious thing? Before even try to jumpstart it, because we don't wanna pay
the money to replace a battery if we're not sure that that's what it is. We just wanna say, "Hey, maybe it's just so cold outside." That, "It's just ran out of energy, "and we just need to give it a jumpstart "and it'll work." As long as the alternator works. Let's go through the experiment process. You have to make sure this is
where things start getting, where you're gonna understand why scientist make so many mistakes. Why we have drugs that get
out there that are recalled. Why we keep revising
the things, is because you really can't set up
the most perfect experiment but you can get close, the better you set up your experiment, the better your ultimate
knowledge comes into play. The worse you do your experiment, the worse things tend to be. So there are a number of ways in which you can do this experiment. For example, make sure that
the cables are set up properly or you're gonna get a nice little jolt. How long do you let it go for? Do you let it go for a minute? No, you gotta let it go for longer, experience shows that it
takes maybe even 10 minutes. Now there are other things
that can apply a larger voltage if you're not doing it car to car. There's many ways to do this. But there are right and wrong ways. If you do it too little, you haven't performed
the experiment properly to test your hypothesis
that the battery is dead. So we're gonna spend some time
a little bit later on today, looking at how to set
up an experiment well, so that you can reliably say, "Yeah, I did a good enough job "that these results are good." So we set up our experiment, we jumpstart the car, let's say we do it right as you're supposed to jumpstart the car and then at the end, we turn the ignition, the car starts up. This is a part that people
tend to get confused on, especially when it comes
to testing aspects. So I'm gonna make it
really simple for you. The starting up of the car, the eureka moment where you're like, "Yay, I can go to my class and," well, maybe not a eureka moment. But yeah, I get to go to
my class, and whatnot. All of that's part of the experiment. The conclusion, only
has one statement to it. There's actually two possible statements but it's one or the other. You either support your hypothesis, or you reject your hypothesis, that's the conclusion. It's not the crunching of your numbers, it's not even the analysis of your data, it's not any aspects. That's all part of the experiment. So usually, when in doubt, most of the time you
spend, be it in your PhD or other things like that
is in this part right here. And then you write, a 200 page, long paper and one sentence at the end, which is your conclusion
that you were right, or you were wrong. Usually, you have to be right to move on. So, it's not the analysis of your data. It's not, hey, the car started up. That's not the conclusion. It's, was I right? Was it the battery? Yes. So, we support our hypothesis. Let's say you go through this process, and the car still doesn't start up. And you you tested everything
you could about the battery, then what would you do? Would you support your hypothesis? No, you reject it, because you're like, "Well, I tested it, it's not the battery." Now what do you have to do? You go back, and you
take up a new hypothesis. So if the battery is not dead, what's another likely problem? - Alternator? - Alternator, yeah. Yeah, that could be another aspect of it. I usually choose the gas but my kids are not old
enough to drive yet. But when they do, I know they're gonna leave
it out of gas all the time. The main point of this is, when you're thoroughly convinced and you reject your hypothesis, you go back and do a new hypothesis. You don't go back to here and start messing around with
the data until you support it. That's when you start
getting into trouble. That's when you get certain dishonesty in pharmaceutical companies, where they just kind
of skip certain steps, skip certain data and
push something through. I'm not saying all
pharmacies are dishonest, just saying that dishonesty is out there. So in some situations, people
don't do what they should do, which is go back to the drawing board, because this going back to here, could mean millions of dollars. That's understandable why sometimes, people don't wanna do that. To make things a little easier for others, as we gain this data, we publish it because we need to work on
the shoulders of others, like I said. But that doesn't mean that, everything that comes
out in these journals is absolutely correct. There have been really bad PR things where people have come out with things that have been published
in really vetted journals, and they've been shown to be
a farce or faults or whatnot. And the reason why it's so bad is because it puts a bad
light on the other scientists that are honestly being
honest and trying to do well. So yes, there are situations where scientists will fabricate and lie. But my experience shows that
most do the best that they can. Yes, you've got always some bad apples in every type of endeavor you do. This is more of my soapbox, but you're not gonna be tested on this. But, this is really an important aspect. Maybe one of these days I will test on it. But one of the biggest
pet peeves that I have is that the media, also the media, but the media and other
sources of information will take what scientists publish and say, and use that for their own agenda. They'll say, "Hey, this
is what the science says "so therefore, we must do this." And the problem is they're implying that, because of certain results, that we have to do these things, and the reality is science is impartial. Science is just about gaining knowledge, it doesn't tell us what we
should do with that knowledge, nor how we should use it. So we have to make ethical and
moral judgments all the time, about what and how to do
with the new knowledge and new information we're getting. But it pisses me off to no end when, I'm not just saying politicians, but anybody uses the data to say, "They're meant to do this." It's not they should be separate issues. It should be, "Hey, this
is what's happening." We as a society, what
should we do about it? And I think that's the biggest problem is, most people will reject the science because they reject the judgments of how we should use that science. And that's a big thing that
we're dealing with today. And I blame the media for that. But, sometimes we as scientists don't get our point across as well. We tend to want certain things, we're human, we want
this progress to be made and other things like that. So, just be aware that there are many out there that the reason why people, usually, reject certain
scientific theories is because of what people
are saying we should do, or what you should understand this to be, based upon that theory. And hopefully, as we go through here, you'll see that once you
understand the science of it, it's just about knowledge. It's about learning how things really are. Okay, now with that. This is on your quiz. So it's important that we
understand these aspects. As we go through the scientific method, there is a certain level of confidence that we have in our knowledge. Hypothesis is merely an educated guess, it has zero supporting data to it. As such, it is the weakest of any concept because anybody can make up a hypothesis, anybody can come up
with a reason, a guess. So, when we look at the progression
of the scientific method it has to start somewhere. We observe something, we've gotta at least try to reason it out and that's where hypothesis begins. If you go through one round
of the scientific method, it doesn't immediately become a theory. It has to go through, sometimes, thousands of vetted experiments before the general scientific
community accepts it as a valid hypothesis. At that point, then we call it a theory. The main difference between
theory and a hypothesis is, a theory is a more thoroughly tested and accepted hypothesis. That's really what a theory is. Which is also why when I
watch shows, it pisses me off when people use the word wrong. They come upon some scenario, they see, dead guy, they're like, "Hmm, well, here's my theory." No, it's your hypothesis is what he has, not a theory till we test that. Theory is essentially a
well accepted hypothesis, the scientific community or the community in that discipline, pretty much says, yeah, this is how it is. This doesn't mean, one, that
we know everything about it, but it is the best explanation that we have amongst all
possible explanations. And so when Reagan and others
say, "Oh, it's just a theory." That really hurts right here, you know, because it's not just a hypothesis. If you said "Oh, that's
just a hypothesis." Then yeah, have that but
it's not just the theory. And so there are several theories. For example, cell theory
does a good job of explaining what life is because as of yet today, we have not seen anything that exhibits the characteristics of life that is not made of cells. Remember, there is still debate
about viruses and whatnot. So again, that's where
theories come into play. Most scientists accept cell theory as the theory that all living
things are made of cells. But there are some that
still reject that and say, "Nope, viruses are alive. "They exhibit all the characteristics "that we consider life." And there's good healthy debate. We're not choking each other out and trying to pore each other whatnot. We're just trying to learn what life is. Now, Principles. Principles
are more concrete concepts, but they don't explain the broad subject. What do I mean by that? For example, we have
the theory of evolution. It describes many things, and it reasonably shows
how things progress. However, within that theory, there are some more concrete concepts, which we call principles. In fact, within the theory of evolution, there are five principles. Everybody knows about one
of them, natural selection, but there are other principles
that govern evolution, such as mutation, and gene
flow, and genetic drift. Don't worry about writing all those down. And sexual selection and whatnot. These are principles of evolution. So a principle is more
narrow in it's scope and what it defines. It's not as broad as a theory. You can almost equate
principles with facts. But we understand it better. So in that endeavor to
understand how a theory works, how the mechanics of evolution work for understanding the dynamics, there's many theories we're
gonna go through this semester. We usually incorporate
what we call principles, which are very well defined concepts, very well defined facts
within that theory. Now in the end, this is what we're after. We want to know how
everything absolutely works. And this is where law comes into play. There's many laws that we have in science, we have laws of what
we call thermodynamics, where we understand how chemicals and molecules
interact with one another. We have the law of gravity, where we understand how
gravity works so well, that we can predict. I throw this pen at a certain
distance at an angle whatnot, I know exactly where it's going to fall. So the more we understand the concept, the more we can predict what
the outcome is going to be. That's really what our endeavor is in the scientific method is to understand how everything works. even in biology, we have laws. They're called the laws of inheritance. We understand how biological
inheritance work so well, that we can make calculated
predictions about the outcome. That you have these two individuals, they have a 25% chance of having a child with cystic fibrosis. Those are the laws of inheritance. And we'll learn that later
on in lecture 15 or so. So, as we go through, you're going to see these
concepts over and over and over, cell theory, theory of evolution, Bayless electron shell theory, that's next lecture and whatnot. Then we've got natural
selection as a principle, we've got the Heisenberg
uncertainty principle which is part of the valence
electron shell theory. That's next lecture, whatnot. We've got theories of our laws of gravity, thermodynamics, entropy, I told you about entropy, that's a law, that's a pervading law
through the entire universe on how energy is transferred and the like. So that's really the goal
of the scientific method to understand everything,
and this is the progression. Now, for the last part of this, let's look at experiments. We're gonna talk about drug trials. We're gonna talk about, a number of different
things that come up today. Why did I get the H3N2 virus? Why did people who got the flu
shot get that virus as well? We'll show why we make mistakes
in creating vaccinations, and doing drug trials and
discovery and whatnot, because it is so difficult to work with as dynamic a system as is the human body. It's much simpler to work on a car, or to work on some other aspect to figure out what's going on, the mechanics of a car is much easier than the mechanics of a living being. When we do these experiments, there are many, many things that we have to control for, most of the time, we can't
control for them all. One of the biggest things that
we have to deal with first, is sample size. There are billions of
people on the planet. But when you're creating a drug, you can't test it and every single person. So you have to take a sample size, a representative group, and this is where we get
into problems because, if you're not representing
every ethnic group, or genetic diversity on the planet, if you're not taking
samples from people all over then your sample size only
represents a small portion of the general populace as a whole. This is why when drugs get vetted, and they go through, they're expecting that
there are certain parameters they didn't test for, with people with certain
genetic diversity, in India, or in Africa,
or in some other area where we might not have sampled
from and tested the drug in our human trials. So in that scenario, the more representative
sample you can get, the better your data's going to be. You cannot extrapolate and say, "Oh, I took someone from every
continent except for Asia." That's where most of the
people on the planet live. So, you need to have a
representative sample. When you get that representative sample that you're doing your experiment on, be it a drug trial,
whatever the case may be, there are the variables and this is where things
become so difficult because in any scientific endeavor, the only way to look at cause and effect, I give you a drug, I wanna
see what happens as a result. The only way to look at
repeatable cause and effect is to make sure that
there is only one variable that you're accounting for, everything else get
standardized, so to speak. So let's look at the different variables that have to be factored in. The independent variables
is what you manipulate. Lemme give you some examples. Let's say we create a vaccination and we want to determine
how much of the vaccine we want to give to allow the
children or whatever group, whether it's adults for the flu virus, or whatever the case may be. We wanna test, how much
of it we need to give for it to be effective or so effective. That's the independent variables. It's what you manipulate within
these independent variables, we have what's called the placebo effect, which is a big problem,
especially in humans, we have to do this, because due to the complex
nature of our physiology, the mere thought that
we're taking something that will cure us, causes
a chemical release, causes a physiological change that can actually get us better. So we think we get better,
it's the positive attitude, or whatever the case may be, we think we're getting some drug and we do and they shown that if they've got a group of individuals that take nothing whatsoever, and they give another
group, just some sugar pill, they'd say, "Hey, this is
gonna cure your migraine." There is a significant difference. They didn't give 'em anything but just a little bit of sugar,
but they didn't know that. And so when we look at drug trials, we have to see something above and beyond the placebo effect. Because if you're looking
at something like this, your drug is crap. It's no different than the placebo effect. So that's why we have to do that. Another thing that has to be done when we're doing placebo controls is what we call a double blind, where the person giving it, doesn't even know which drug it is, and why would we have to do that? Those are variables that
have to be accounted for. Because, if you're doing
something of this nature, and all of a sudden, you see something like this, then somebody probably
leaked what the placebo, which one's the placebo, which one's not, and people are like, "Well, I'm not getting the
placebo, type of thing." So it can really mess up your results by not accounting for
all of these variables. That's why when we do placebos, neither the person giving nor
the person receiving knows, which one's the right drug and which one's just an
inert substance like, sugar, salt or whatever
the case may be, all right. So a placebo or the actual vaccine, these are what we call
the independent variables, what you manipulate. In this situation, you're
not changing the vaccine, you're just changing the amount. So those are the controls,
the manipulated variables. The second type of variables
are, what's the outcome? Do we call these the dependent variables? Because they depend upon
what you manipulate, or the independent variables? So the dependent variables
is, what's the outcome? What's the results of that? So they showed that, when they
gave children this placebo, it wasn't actually a virus, it was administered the same way whatnot. But there was no virus then
about 15% of them got sick, when they gave a low dose, a little less, and they're like, "Oh, this is the minimum "we need to give at this point "to be able to have the
virus be effective." So that's that was their conclusion but, there are so many other
things that go into it to make sure that you
get accurate results. Let's look at some examples. This is where the third
set of variables comes in. This is the most difficult
part of any experiment. We call them standardized variables. So what are standardized variables? Basically, all other factors
need to be held constant, so that only the independent variables and the dependent variables
become correlated. Lemme give you some examples. Let's say for some reason
that in this group, you just couldn't find 80 children. So they ended up just having like 10 and the others they had 86, 78 and 87. That would not be keeping
everything standard, because you would end up getting a different set of statistics
with just 10 children than you would with 80. So they try to keep these
numbers as close as possible. Ideally, they would be all the same. How do you administer it? Do you inject some
children with the needle and others give a droplet? That's gonna change your outcome. So you need to make
sure you administer it. I mean, you poke a child with a needle, you're gonna cause a
physiological reaction that might mess something up. I mean, we're very dynamic creatures. So if you give one group, a dropper with a vaccine, you better do it to all
and so on and so forth. If you give 'em five drops,
give five drops to all and all of that. Keeping everything standard, that's probably the most
difficult thing to do. Because you gotta also
have the representation. If you have 80 children, but
they're from one ethnic group and 80 from a different ethnic group. That's not good. You're not keeping things standard, you need to make sure
that all are represented in that group. Otherwise, you get too many
variables to account for. Because it could be that this ethnic group just has an
innate immunity to that virus, that you didn't account for. And you're like, "Oh,
this is the minimum," and then you try it out there,
and people are getting sick. So that's where problems come into play. All right. Let's do some more examples. Let's say you're doing some farming. Scientists wanna know whether or not they should add fertilizer to their crops. Plants don't have the same
physiology as you and I, they don't understand that
they're getting fertilizer versus not fertilizer,
but you still have to do a placebo like effect, you're like, "Why do we need to do a placebo effect?" Lemme show you. Let's say that you have
these crops in the dirt. And in this one where you're
not gonna add any fertilizer. This is your control. This is your placebo or whatnot. And over here, let's say you
add a pound of fertilizer, and here, two pounds of fertilizer. adding that extra pound or two might be the main reason, it might not be your fertilizer, it may just be the fact that
you've added more weight to the soil and other things. So one of the biggest issues
you have to deal with is, if you add a pound of fertilizer, you better add a pound of something inert, that's in the soil to compensate so that they're in the
same amount of soil, maybe the roots don't have as much space as these roots to grow, and these absorb your nutrients. Again, you're standardizing the variables between all of these. So even though you control
for the amount that you give, and those are your independent variables, how much fertilizer you give, you need to be making sure that, everything else remains constant, the volume of dirt that these
are in and so on and so forth. In the end, obviously, these are the dependent
variables, how was the yield? How many tomatoes did you get per bind, and as a result, you can
make some conclusions. The sample size wasn't great. They did three plants, it could have been much more. So it's not that great of an experiment, when you're only doing
such a small number, it doesn't represent the whole. Let's do one more,
lemme show you one more. Here's another experiment. All of these are potential
experiments on your quiz. Some of you might get one
where it describes this plant, they're very simple. They're very straightforward. And the only thing that you're
really gonna be tested on is, I'll describe this process and say, "Oh, when they gave no fertilizer, "low fertilizer or high fertilizer, "what type of variable was this?" That's the independent variable. Or when they looked at
the yield of the tomatoes, what type of variable was that? It's a dependent variable and the like. Let's do one more. Scientists have a question. They think that they have a hypothesis that people will get ulcers, primarily due to a
particular strain of bacteria that's in their gut. So they wanna test this hypothesis. in order to test this hypothesis, they're manipulated or
independent variables is they use a couple of
different antibiotics, because they want to see
which antibiotics might kill that strain of bacteria better. And then ultimately, what the outcome is, as far as ulcers go. The dependent variables is, do the ulcers, that
the people get decrease as a result of the treatment? So obviously, you're gonna
have your placebo group that gets an inert substance. It's not an antibiotic at all, but it's inert. And you find that 10% of them actually, have reduced ulcers just
by taking a placebo. And then you look at
antibiotic A and antibiotic B, and you see your respective results of how many people got
better from the ulcers. The hardest part here again is, the standardized variables. Here you've got a mix of people up there, of different races, different
ages, men and women, hopefully, you know we're very
different from one another. So it would not be a good idea to put all the women in one group, all the men in another group. And there's nothing left
for the third group. But if you have, men and
women in the third group and men represented that
way, it's not standardized, because there's gonna be
substantial differences in our overall physiology. So each group needs to be represented. Ideally, you would have
groups of triplets. The only triplets I know
are from my sisters, she does have triplets, I'll
get on more of that later. But identical triplets, because all of their genetics
are pretty much the same, and they're all the same
sex and so on and so forth. But we really can't do that. So we do the best that we can by reducing any variation
between ethnic groups, between age, between male
and female physiology and the like. That's really what standardized
variables is all about. Making sure that, if there are differences,
that they are representative in each of the groups, so that in the end, we manipulate a variable,
and we see some outcome. And we can draw a direct
correlation between those two. There's no other confounding factors. Think about the car scenario, again, if you were to have a hypothesis
that the battery was dead, but you're like, "I just don't have time "to just test the battery, "I've gotta test another hypothesis." You fill up the gas, and
you jumpstart the battery while you're filling up the gas. And then you start the car, you're like, "Yay, it was both." No, because you didn't account
for one variable at a time, you were looking at two different things, you wouldn't be able to
know which one it was. The same thing is true. And that's why you effectively look at one variable at a time, otherwise, you just don't
know which variable it was. All right. Now we've got
about five more minutes here, I'm gonna let you dissect
down this Calvin and Hobbes and come up with the
fundamental scientific method, the observation, the hypothesis, the experiment and the conclusion. Then we're gonna spend
a little bit of time, talking about what he did right
or wrong with his experiment from sample size to, the overline, design of it and all that kind of stuff. So what's his observation? It is hot. That's his observation. It's really hot outside. He has a hypothesis, what is it? (students voices chattering) That is hot enough on the
sidewalk to fry an egg. That's a pretty specific hypothesis, and from his observation that this very well may happen. Here's where he gets
into the experiment part. Let's dissect his experiment and we'll pick it apart in a little bit. How many eggs did he use? - One. - One? How long do you let it sit there? (students voices chattering) Probably a couple seconds, not nearly long enough or whatnot. He had to choose a spot on
the sidewalk to do it in. And he had to get a carton of eggs from somewhere and whatnot. And there's all these parameters, obviously, it could be done better. How many eggs would you have used? How to use all of them? In fact, I did when I was his age. Actually, twice that,
according to my (mumbles). I would use all of
them, large sample size, sample size, one not so good. The timing of it, the
choices of his location, what's the variable that he
may not have accounted for that you see in the first slide? - Under a tree. - Yeah, maybe got some shade in that particular spot and so therefore, it's not as hot. You didn't account for that. So we maybe needed to distribute the eggs, all along the sidewalk like I did. So, timing, let it go a
little bit longer boy, it takes a lot longer even in a hot pan for it to cook like that. How cooked did he want it to be? That was another factor
that he didn't address in his hypothesis or that he counted for, over easy or, a little
bit burnt or whatnot. Besides all of his problems
with his experiment that we can definitely improve upon, what was his ultimate conclusion? Do you support or reject this hypothesis? - Reject. - He rejected it. So what was his new hypothesis? It's hotter on the car dash. So instead of is it hot enough to fry an egg on the sidewalk, he changed it to say, maybe it's hot enough to
fry an egg on the car dash. Which is not a bad change
of his hypothesis because, maybe intuitively like we know, metal conducts heat lot
faster than concrete does. So not a bad change of his hypothesis. But if he does it the
same way he did here, he might still run into
issues that he dealt with. So that kind of gives you an idea, you will not have a Calvin
and Hobbes quiz question. But this is just to
get you thinking about, the process of problem solving that we all do on a daily basis. How do we approach an observation and our hypothesis about how
the world works around us?
