How Science Works

How Science Works:

  • Theories
  • Data
  • Bias
  • Limits
  • Ethics


How Science Works 1:

Scientists are not always right. Increasing Scientific knowledge involves making mistakes on the way.

Scientists come up with a hypothesis (a theory) -THEN THEY TEST THEM.

1) You try to explain things. Anything.

2) Observe and think about something you don't understand.

3) Use your own knowledge and guess work to come up with a hypothesis that may explain it. (However it is not fact it's a theory or a belief.)

4) Try to convince others that the hypothesis is right and use evidence to back it up. The hypothesis has to fit evidence already avalible.

1 of 10


5) Use the hypothesis to use a prediction. If it predicts something and evidence from experiments backs this up it will convince people. However this still does not make it fact.

Other Scientists Will Test the Hypotheses too.

1) Other Scientists will use your hypothesis to make their own predictions. To do so they carry out their own experiments to find more evidence and may even reproduce earlier results to see if they are reliable. If all experiments lead back to the original hypothesis people will have a lot of faith in it. 

2) If someone finds an experiment that contradicts your hypotheses  and their results are reliable. Then people will lose faith in your hypothesis. So now scientists come up with a new hypothesis, either a modification of the original or a new one all together.

3) The point of trying to disprove a hypothesis is vital to the scientific process. Without "Healthy Scepticism" we would still believe very basic theories.

2 of 10


If Evidence Supports a Hypothesis, It's Accepted - for Now.

1) If nearly every scientist in the world believes in a hypothesis it is accepted and goes in a text book for students to learn. 

2) The currently accepted theories are ones that survived the "Trail by Experiments" They have been tested hundreds of times and survived while the disproved theories were ditched.

3) However they never become solid fact. It only takes one odd result and it can be disproved.

If scientists think something is true, they need to produce evidence to convince others - it's all part of testing a hypothesis. One hypothesis may survive these tests, while others won't - it's how things progress. Along the way some hypotheses will be disproved and shown not to be true.  

3 of 10


How Science works 2:                                                                                  

Evidence is the key to science - but not all evidence is equally good. The way it is gathered can have a big effect on how reliable it is.        

Lab Experiments Are Better Than Rumour or Small Samples:

1) Results from controlled experiments in laboratories for great. A lab is the easiest place to control variables so that they are all kept constant (Except for the one you are investigating.) This makes it easier to conduct a fair test. However there are things you cannot study in lab though like wind speed.

2) Rumour cannot be seen as fact - ever. For something to be proved it has to be tested in controlled conditions to be genuinely scientific.

3) Data based on small samples don't have much credibility as a sample should be the representative of the whole population and have similar characteristics a small sample just won't do that.

4 of 10


Evidence Is Only Reliable If Other People Can Repeat It.

Scientific Evidence needs to be reliable (or reproducible). If it isn't then it won't mean anything.

Reliable - Means that the data can be reproduced by others.

Evidence Also Needs to Be Valid. 

Valid - Means that the data is reliable AND answers the original question.

Controlling All Variables Is Really Hard.

To be accurate you need to control all the variables that might have be having an effect on the question. This is hard as for example, if your testing people, each person is individual and certain aspects may effect the result. Even if you make them all the same age, gender etc, somethings may still effect the results, and you cannot use lab experiments as it may not be ethical.

5 of 10


How Science Works 3:

Scientific results are often used to help people make a point. (For example politicians, environmental campaigners etc.) However some of these results are presented in a biased way.

You Don't Need to Lie to Make Things Biased.

1) For something to be misleading does not mean it dosn't have to be untrue. When you read a scientific statisic you believe it to be true.However there is more than one side to the truth.

If a scientist says 1 in 2 people are above the average weight, it sounds bad. This is a scientific analysis of the facts and is true.

2) The average is a kind of middle value of all the data. If some readings are higher than the average (normally half) the other half is lower than the average. So if a scientist says 1 in 2 people are below the average weight it sounds completely different but still very true.

6 of 10


You can easily turn a statistic into something really good or really bad even though it's not either. To make something bias you can...

1) ...Use only some of the data, rather than all of it. - "Many people lost weight using SlimFast!" "Many" could mean anything - e.g: 50 out 5000 (1%) but ignoring most of the data.

2) ...Phrase things in a - "Leading way" 90% fat free! Would you buy it if was 90% Cyanide free? The 10% could be very important.

3) ...Use a statistic that supports your point of view. - "The amount of energy wasted is increasing." "Energy wasted per person is decreasing." "The rate at which energy waste is increasing is slowing down." These describe the same things but two are positive and the other is negative.

Think About Why Things Might Be Biased.

1) People who want to make a point can sometimes present biased data to suit their own purposes sometimes with out knowing it. 

7 of 10


2) There are all sorts of reasons why people present data in a biased way. For example...

  • Governments might want to persuade voters and journalists so evidence might be ignored if it could create political problems, or emphasized it it helps them.
  • Companies might want to "big up" their product and make impressive safety claims.
  • Environmental Campaigners - might want to persuade people to behave differently.

3) Not all scientific evidence is bias but is why all scientific evidence has to be looked at carefully. Look at the reasons it might be biased:

  • Does the experimenter stand to gain or lose anything?
  • Might someone have ignored some data for political or commercial reasons?
  • Is someone using their reputation rather than evidence to help make their case?
8 of 10


How Science Works 4:

Science can give us amazing things but there are some questions that you just cannot answer, whether it be an ethical problem or an unsolved mystery.

Some Questions Are Unanswered By Science - So Far.

1) We don't understand everything we never will. We will find out more but that will just lead to more questions. As more hypotheses are suggested and more experiments done we will progress but never actually know everything.

2) Complex questions such as why is climate change happening? Has caused a lot of controversy between scientists with different hypotheses. However eventually we will have an answer after a lot of trial and error.

Other Questions Are Unanswerable By Science.

1) There's one other unanswerable question type, one that all the experiments in the world couldn't answer. The ethical ones. 

9 of 10


2) Take embryo screening It's possible but is it right?

3) There are two opinions :

  • It's a good thing, a child with cancer in a desperate need of a bone marrow transplant can't find a donor. So the parents have a child viaembryo screening to save the child. So now the parents have two healthy children.
  • Others say it's bad as a child id born into a world with only one purpose to save it's sibling. This child won't have any rights to their own body. So what do you do?

4) This question of whenther something is ethically or morally wrong does not have a "right" or "wrong" answer.

5) The best we can do is get a consensus from the public a generaljudgement that most people are happy to live by. science can only provideevidence to try and help the judgement change over time. But in the end it's up to your conscience.

Other factors to influence may include, economic, socialand environmental factors.

10 of 10


No comments have yet been made

Similar Science resources:

See all Science resources »