Skip to main content
SWStudyWalks Search Sign in

Getting Ready: The Exam and the Nature of Science

The narration between entries paraphrases; it never reads an entry aloud before the entry plays. Every entry in this lecture plays exactly as it does on its own page — same file, same speed, nothing laid over it.

What you can do after this

  1. Describe the exam's length, its two halves and its weights, and answer an EXCEPT question by testing each choice
  2. Tell a scientific, falsifiable question from one outside science, and qualitative data from quantitative
  3. Run the scientific method from hypothesis and prediction through a controlled experiment with matched groups
  4. Explain why results support but never prove a hypothesis, and tell inductive from deductive reasoning
  5. Distinguish hypothesis, theory, law and model, and describe peer review and basic and applied science

What it covers

18 entries in 4 topics. Each number links to that entry on its own.

  1. 1. The Exam Itself

  2. 2. What Counts as Science

  3. 3. The Method and the Experiment

  4. 4. Theories, Models and the Scientific Community

Check yourself

How should you answer an "all of the following EXCEPT" question?

Test each choice as a true-or-false statement, then pick the one false choice.

Does falsifiable mean false?

No; it means some possible result could show the claim to be wrong.

In a fertilizer test, which is the dependent variable?

Plant height, the quantity measured in response.

Can an experiment prove a hypothesis?

No; it can disprove one, but agreeing results only support it.

Does a theory graduate into a law once it is well tested?

No; a law describes one pattern, while a theory explains many.

Lecture transcript

The narration as recorded, in order. Where an entry plays, the transcript links to it rather than reprinting it — the entry plays from its own file, unaltered, and says the same thing on its own page.

This course prepares you for the CLEP Natural Sciences exam, and it begins with two things: the exam itself, and the habits that make knowledge scientific. The governing idea of this lecture is that a scientific claim earns trust by facing tests it could fail, so the best it can ever be is supported, never proved. The lecture moves in four steps. First the exam: its length, its two halves, its weights and its question forms. Then what counts as science, what makes a claim falsifiable, and the two kinds of data. Then the method: hypothesis, prediction, variables, controls and a fair experiment, arriving at why a result supports a hypothesis and never proves it. And last the words science uses for its explanations, theory, law and model, and how peer review and replication decide what is trusted. By the end you will be able to design a fair test and say exactly how far its result reaches.

We begin with the exam: how long it is, how its questions divide, and the forms they take.

First, the exam in a single view.

NSC 0001 The CLEP Natural Sciences Exam at a Glance

The exam has about 120 questions in 90 minutes, some of them unscored pretest questions. That pace is about 45 seconds a question. It tests basic principles for students who are not science majors, not memorized detail. Half is biological science and half physical science, so neither half can be skipped.

Now the four areas of the biological half.

NSC 0002 The Biological Science Half

Biological science is 50% of the exam. Origin, evolution and classification of life take 10%; cells, the gene and bioenergetics, 10%; organisms with heredity, 20%; populations and ecology, 10%. The 20% area holds roughly 24 questions. Heredity sits with organisms in the largest area, not with cells.

Then the six areas of the physical half.

NSC 0003 The Physical Science Half

Physical science is the other 50%. Atoms, nuclei and particles take 7%; elements, compounds and reactions, 10%; heat, states of matter, mechanics and relativity, 12%; electricity, magnetism, waves, light and sound, 4%; the universe, 7%; and the Earth, 10%. Earth science outweighs the universe.

Here are the skills the exam weighs.

NSC 0004 Skill Bands and Question Forms

The exam weighs three skills: about 40% knowing facts, concepts and principles; about 20% interpreting graphs, tables, equations or passages; and about 40% applying principles, more often in words than with numbers. Each question has five choices. Some join several sciences, some cover the history and philosophy of science, its methods, technology, and problems such as pollution and resource depletion, and some are laboratory questions. For an all-of-the-following-EXCEPT question, test each choice as true or false, then pick the one false choice. In the water example, only "mostly carbon" was false. Picking the first true statement you read is the classic error.

Now what makes a question scientific, what makes a claim testable, and what counts as data.

First, a working definition of science.

NSC 0005 What Science Is

Science is knowledge about the natural world, built from observation, experiment and evidence. The natural sciences divide into life sciences, such as biology, and physical sciences, such as physics, chemistry and astronomy. Whether a fertilizer makes beans grow taller is a science question; purely moral, aesthetic or spiritual questions lie outside, because no measurement can test them.

A scientific claim must be able to fail.

NSC 0006 Testable and Falsifiable

A scientific hypothesis must be testable and falsifiable: some possible result could show it to be wrong. Claims built to avoid any such test are pseudoscience, however scientific they sound. Astrology is the standard example; tests find no evidence that birth signs predict anything. Falsifiable does not mean false; it means a test could disprove it.

Observations become two kinds of data.

NSC 0007 Qualitative and Quantitative Data

Qualitative data describe in words, such as a color or a texture; quantitative data are numbers from measuring or counting. "The leaf is pale" is qualitative; "the leaf is 6.5 centimeters long" is quantitative. Scientists prefer numbers where possible, because they can be compared and analyzed. "The leaf is pretty" is opinion, not data.

Now the method itself, from a testable explanation to a fair experiment, and the limit on what any result can show.

Every test starts from an explanation.

NSC 0008 Hypothesis and Prediction

A hypothesis is a suggested explanation that can be tested. A prediction states what should happen if the hypothesis is right, usually as if-then. The lamp is dark because the bulb burned out; if a new bulb goes in, the lamp lights. The hypothesis explains; the prediction is the testable result it implies.

Here are the steps of the method.

NSC 0009 The Steps of the Scientific Method

The scientific method is a cycle, not a fixed recipe. An observation raises a question; the scientist reviews what is known and proposes a hypothesis; the hypothesis yields a prediction; an experiment or careful observation tests it; and the results are analyzed. A contradicted hypothesis is rejected or revised and the cycle runs again, while a supported one is shared for review and further testing. In the tomato example, sunny plants averaged 3.2 kilograms of fruit and shaded plants 1.1 kilograms, which supports the light hypothesis without proving it and raises the next question: how much light is enough?

An experiment needs variables and a control.

NSC 0010 Variables and Controls

The independent variable is the one the experimenter changes; the dependent variable is the one measured in response. A control group gets no treatment and is the baseline for comparison. Testing a fertilizer, the amount of fertilizer is independent and plant height is dependent. The control must match the treated group in every way except the treatment.

Here is how to design and judge one.

NSC 0011 Designing and Judging a Controlled Experiment

A fair experiment takes five steps. State the hypothesis and prediction; name the independent and dependent variables; set up treatment and control groups that differ only in the independent variable, giving the control a harmless substitute when the treatment adds something; use several identical subjects per group; and compare. In the bean example, ten seedlings got plant food and ten got the same volume of plain water. The fed plants grew 9.0 centimeters and the controls 6.0, which supports the hypothesis. Judge the design before the result: without a matched control, no difference can show its cause.

This is the limit on every experiment.

NSC 0012 Supported, Never Proved

An experiment can disprove a hypothesis, but it can never prove one. Results that agree with a prediction support it, and a later, better test might still overturn it. A thousand white swans support "all swans are white"; one black swan disproves it. Science does not deal in proofs the way mathematics does.

Science reasons in two directions.

NSC 0013 Inductive and Deductive Reasoning

Inductive reasoning moves from many specific observations to a general conclusion: every metal tested conducts electricity, so metals conduct. Deductive reasoning moves from a general principle to a specific prediction: copper is a metal, so copper conducts. A deduction holds only if its principle holds; from a false premise it gives a false conclusion.

Last, the words science uses for its explanations, and how the community decides which results to trust.

Three words that are often confused.

NSC 0014 Hypothesis, Theory and Law

A hypothesis is a tentative, testable explanation. A theory is a broad, well-tested explanation of many related observations. A law is a concise description, often one equation, of how nature behaves. The law of conservation of energy describes; the theory of evolution explains. Theories do not graduate into laws.

The everyday word means something else.

NSC 0015 A Theory Is Not a Guess

In everyday speech a theory is a guess. In science a theory is a testable explanation backed by evidence and confirmed by independent researchers, and it changes only when good evidence contradicts it. Evolution and relativity are theories in this strong sense. "It's only a theory" misreads the word: theory is the best-tested kind of explanation.

Some things are shown through models.

NSC 0016 Models

A model represents something too large, too small or too complex to show directly. Models help explain observations and make predictions, but each is accurate only within limits. A globe models Earth's shape and continents, but not its size or mass. A model is not the thing itself: the atom is not a tiny solar system.

Results face other experts before they count.

NSC 0017 Peer Review and Replication

Scientists report results in journals that use peer review: experts in the same field check the work before it is published. Other scientists then try to replicate the results, and a reported effect gains standing when other labs repeat the study and get the same result. Unreviewed results are not a reliable basis for further work.

Science seeks knowledge and solves problems.

NSC 0018 Basic Science, Applied Science and Society

Basic science seeks knowledge for its own sake; applied science, or technology, solves practical problems. Some discoveries come by chance to an observant mind: Alexander Fleming noticed a mold killing bacteria in a dish left open, and the mold yielded penicillin. Basic research is not useless; learning how DNA copies itself led to DNA tests.

This first part set out the exam and the habits of science. We described the exam: about 120 five-choice questions in 90 minutes, half biological and half physical, and the true-or-false test for an EXCEPT question. We told science from questions no measurement can test, a falsifiable claim from pseudoscience, and quantitative data from qualitative. We ran the method from hypothesis and prediction through a controlled experiment, with a matched control and a harmless substitute. We saw why a result supports a hypothesis but never proves it, and how induction and deduction run in opposite directions. And we separated hypothesis, theory, law and model, with peer review, replication and basic research deciding what is trusted. Part two turns to measurement: SI units, conversions, scientific notation, accuracy, significant figures and graphs, then the road to a Sun-centered universe and one investigation start to finish.