Scientific
Inquiry
Anyone can have an opinion about the ocean. Scientists go further — they follow a method that turns a hunch into a result other people can check, repeat, and trust. By the end of this topic you'll know that method, how to use it, and why it's the rulebook for your own Marine Sciences Expo project.
Imagine you read online: "Sharks are 10× more aggressive on a full moon." It sounds dramatic. It might even be true. How would you actually test that — without just believing it because someone said so?
You'd need a method. You'd look at the data — every recorded shark interaction in SA over the last 20 years — and check the moon phase on each date. You'd compare full-moon weeks to other weeks. You'd check whether the difference is real, or just noise. And you'd let other scientists try the same thing.
That whole process — observation, question, hypothesis, experiment, analysis, conclusion, communication — is the scientific method. It's why we trust some claims about the ocean and dismiss others. And it's the exact rulebook you'll use for the Marine Sciences Exposition project you do every year of this course.
Science is a Method, not a Memory Bank
Every claim about the ocean has either been tested — or it hasn't.
Science isn't a list of facts. It's a rulebook for finding out. The rulebook says: don't trust your gut. Don't trust the loudest voice. Trust evidence — and trust it most when it's been checked, repeated, and survived attempts to disprove it. The internet is full of confident-sounding ocean claims; only some of them have been through the rulebook.
The South African coast is one of the most-studied stretches of ocean on Earth, but huge questions are still open. Why are sardines moving east? Why are some MPAs recovering faster than others? Why is the Agulhas warming so quickly? Marine scientists are answering these the way you're about to learn — one hypothesis at a time.
The Seven-Step Cycle
The scientific method is a loop, not a line. A conclusion almost always sparks a new question — and the cycle starts again. Tap each step to see what happens at that stage, with one running SA example threaded all the way through: "Why are sardines moving east?"
Local connection — your annual Expo project
Every year of this Marine Sciences course, you'll do the cycle yourself. The Marine Sciences Exposition is the showcase: pick a topic from the year's curriculum, formulate a hypothesis, design and run an experiment, analyse what you found, and present a poster. Sitting in Gqeberha on Algoa Bay is an unfair advantage — you have rocky shores, an estuary, two penguin islands and a working harbour all within a Saturday morning's reach. Pick something you can actually go and measure.
Three Variables. One Sentence. Two Tests.
A good experiment isolates one cause and one effect. Get sloppy with variables and your data tells you nothing. The trick is to keep three categories straight from the start.
Independent
The single factor you change between trials. Also called the manipulated variable. There should only be one of these per experiment.
Dependent
What changes in response. Also called the responding variable. This is what your data table records.
Controlled
Everything else that could change but you deliberately don't let it. Lots of these. They make the test fair.
Writing a hypothesis
A hypothesis is one short sentence linking your independent variable to your dependent variable, with a direction. It must be testable — there has to be a result that could prove it wrong.
As water temperature increases, the mass of mussels after four weeks will decrease.
A hypothesis is never "right" or "wrong" — it's supported or not supported. If your data rejects it, that's a useful result. You've eliminated one possible explanation and earned the right to ask a sharper question.
Validity vs Reliability
Two different tests every experiment must pass. They're not the same word, and Grade 10 examiners love asking which is which.
Did you actually test what you said you'd test?
A fair test. Only the independent variable is changing. If two things changed at once, you'll never know which one caused the result.
~ if you change the temperature and the food, the test isn't valid.
Would you get the same result if you ran it again?
Repeatability. If you ran the experiment three times and got similar results each time, it's reliable. If the result was a one-off, it might just be luck.
~ minimum three repeats. Five is better. More is more reliable.
Two more terms examiners love: outliers and inferences
An outlier is a data point that sits far away from the rest. Don't just delete it — explain it. Maybe one tank had a leak, or one mussel was sick. An inference is the conclusion you draw from your data — not the data itself, but what you reason from the data. "Mussel mass dropped" is data; "warmer water stresses these mussels" is an inference.
Hypothesis, Theory, Law — and Why Science Never Says "Proved"
In ordinary speech a "theory" is a hunch. In science it is almost the opposite — it is the strongest thing there is. Getting this distinction right is worth easy marks, and getting it wrong makes an otherwise good answer look naive.
Hypothesis
A tentative explanation of what might be causing something — written as one short, testable sentence, before you collect data.
~ "Sardine shoals peak further east in years when sea-surface temperature is higher."Theory
An explanation proposed on the basis of evidence obtained from research — one that has survived many independent tests by many scientists over years. Not a guess. A guess that won.
~ plate tectonics (Topic 11), evolution by natural selection, the germ theory of disease.Law
A description of what always happens, often as an equation — but a law does not explain why. That is the theory's job. A theory never "graduates" into a law; they are different kinds of statement.
~ pressure rises 1 atm per 10 m of depth (Topic 6) describes; it does not explain.⚠️ But watch out: your CAPS Investigations column literally asks "Have we proved or disproved the hypothesis?", and a paper may well use that wording. Answer it anyway — just phrase your answer as supported / not supported. You get the mark for the reasoning, and you demonstrate you know the difference. Never write "this proves…" in your own conclusion.
Judging a source — because CAPS says "evidence from a variety of sources"
Half of scientific inquiry is deciding what to believe before you start. Four questions that do most of the work:
1. Who wrote it, and what do they gain? A fisheries lobby group and a university lab can report the same catch data and frame it oppositely. 2. Is it peer-reviewed? Published in a journal means other scientists tried to find holes in it first. 3. When? Ocean data ages fast — a 2009 sardine distribution map is a historical document, not a current one. 4. Does anyone independent agree? One study is a finding. Three separate teams agreeing is knowledge.
Cite what you use. An Expo judge who cannot see where your numbers came from will assume you made them up.
Showing Your Data
CAPS does not hedge about this one. It says it is "vital that students understand how to summarise data and the correct and appropriate use of Diagrams (drawing technique), Tables, Line graphs, Bar graphs, Histograms, Pie charts." And your booklet ends every single formatting list with the same five words: "Marks will be allocated for following these steps." This section is the most directly mark-bearing thing in the topic.
Which graph? Let the data decide
The commonest mistake is picking the graph you like rather than the one the data calls for. Tap each type — the question underneath each one is the test you apply.
Tables — the five rules
- Number every table, graph and diagram as Fig 1, Fig 2, Fig 3… in the order they appear.
- Give a descriptive title, e.g. "Fig 1: A table showing pressure and temperature of sea water in Table Bay, at different depths."
- Clearly label the row and column headings.
- Put units in the headings only, in brackets — never repeated beside every number.
- The body of the table holds numbers or comments only.
Graphs — the seven rules
- Choose the right graphing style for your data (the chooser above).
- Give the figure number.
- Give a descriptive title, e.g. "Fig 2: The change in heart rate of an athlete after intense exercise."
- Label the x (horizontal) and y (vertical) axes, with the unit in brackets.
- Use an appropriate scale — use most of the available page. A graph squeezed into a corner loses marks.
- Plot the points accurately.
- Draw the line, bars or sectors accurately to your plotted points.
Diagrams — the nine drawing rules
Identical to the Drafter's Code in Topic 14. Learn them once; they are examined in both topics.
- Use a dark, sharp pencil.
- Draw neat single lines, not sketchy repeated strokes.
- Do not overlap your starting point when closing an enclosed structure.
- Colour or shading only if instructed, and only to highlight a specific structure.
- Use a ruler for label lines — in pencil too.
- No arrows on label lines.
- Labels sit under each other, on one or both sides of the diagram.
- Labels in neat legible print.
- Give a suitable underlined heading, above or below the drawing.
The Report, the Poster and the Expo
CAPS is explicit that this whole block exists to get you to one place: "Learners must be trained in how to submit a credible research project to the annual National Sciences Exposition." Here is what that actually requires.
Four research skills CAPS names
Classify
Group related objects and ideas together. Before you can count or compare anything, you must decide what counts as the same kind of thing.
~ sorting a quadrat's contents into species before tallying them.Quantify
Use numbers and measurements — of time, length, width, volume and ratio. "Lots of mussels" is not data. "37 mussels per 0.25 m² quadrat" is.
~ tape along a transect · volume in a water sample · a ratio in a quadrat count.Communicate
Report results in writing and, wherever possible, visually in tables and graphs. If a reader cannot follow it, the work does not count.
~ the report and the poster — Section 5 is how.Interpret
Discuss the data and draw conclusions — explain the observation. This is the step that separates a data dump from science.
~ "mass dropped 12%" is data; "warming stresses these mussels" is interpretation.The research report — six parts, in this order
Straight from CAPS. The two highlighted in orange are the ones almost everyone leaves out — including, until this page was corrected, this very course.
Introduction ← most often missing
What is already known, from your reading. Sets up why the question is worth asking, and shows the judge you did background research.
Hypothesis and Aim ← the Aim gets forgotten
The hypothesis is your testable prediction. The aim states what you are setting out to do — and CAPS expects you to write it with one of these stems:
Method
Materials and procedure, in enough detail that a stranger could repeat it exactly. Include how you calibrated any equipment.
Results
The data itself — tables and graphs, formatted to the rules in Section 5. No interpretation here; just what you measured.
Discussion
What the results mean. Trends, outliers, and an honest account of flaws and inaccuracies in your procedure and where you could have done better.
Conclusion
Was the hypothesis supported or not supported? Then suggest at least two future studies the work points toward.
The poster is not the report
This trips people up every year. Your booklet defines it in one line: "The poster must be a visual graphic summary of the investigation." Not the report shrunk to fit — a visual summary. A judge walking past should grasp your question, your method and your finding in about twenty seconds, mostly from the diagrams, tables and graphs. CAPS lists this as a Display Skill in its own right: "Know how to present diagrams, tables & graphs in a poster for an exposition."
What the Expo judges actually look for
That you did it yourself. Your booklet is blunt: "In some cases, parents think up (and complete) the project… and it is discovered at the exposition that the student does not have a clear idea about the subject matter, its aims or purpose. Judges at expos regularly find that projects have not been thought through carefully enough." Being able to defend your conclusions under questioning is the test — and CAPS names that skill directly.
That you were precise. The booklet's analogy: an experiment is like baking a cake. Leave out the teaspoon of baking powder and it will not rise. Precision and accuracy are not optional extras.
That you used the right words. CAPS: "Use the right terminology. All procedures and reporting are vitally dependent on the precise use of terms." Hypothesis, not guess. Supported, not proved. Independent variable, not "the thing I changed".
And they do not care whether you were right. "The judges are interested in your findings regardless of whether or not they support your original hypothesis." A well-run experiment that disproves your hypothesis is a success. Never fudge data to make it agree with you.
In the Lab — Your Own Investigation
Everything above is preparation for one thing: running your own investigation and writing it up. CAPS gives this block six hours and a full Investigations column — problem statement, hypothesis, experimental design, data collection, analysis, conclusion, extension. It is the only block in the year whose stated purpose is the practical.
🔬 The practical — a complete investigation, start to finish
- Practise on someone else's data first. A supplied Algoa Bay data set to tabulate, graph and interpret — so you meet the formatting rules before your own marks depend on them. 🟢 desk
- Pick the right chart, four times. Four data sets, four different correct answers. This is the skill Section 5 teaches, tested. 🟢 desk
- Write the aim and hypothesis for three scenarios, using the CAPS verb stems, and identify all three variables in each. 🟢 desk
- Run your own experiment — a full scaffold with three ready-to-go marine options you can do at home or on the shore, from designing it to collecting the data. 🟢 kitchen / shore
- Write the report — all six CAPS parts, with a checklist and the mark allocation. 🟢 desk
- Build the poster — a layout plan for the Expo board, plus the judges' questions to rehearse. 🟢 desk
This is the one practical in the course with no supplied-data fallback for the experiment itself — and deliberately so. Every other topic's practical can be done from provided results, because what CAPS assesses there is the analysis. Here, doing it yourself is the assessed outcome. The three options are all built from things you already have.
📄 Practical sheet: Open the Topic 5 practical →
Key Vocabulary
Tap each card to flip it. These eight terms are examinable.
A testable, tentative explanation — one short sentence linking an independent and a dependent variable, that an experiment can support or reject.
The act of noticing facts or taking measurements. The starting point of every scientific investigation — and what separates "I think" from "I noticed".
A statement of what you think will happen in the experiment. Flows directly from the hypothesis. If the prediction comes true, the hypothesis is supported.
Data points that sit far from the rest of the data. Don't just remove them — explain them. They sometimes reveal a flaw in the method, sometimes a real surprise.
A conclusion drawn from evidence by reasoning. Not the raw data — the meaning you read into it. "The water was 23 °C" is data; "this is warmer than usual" is an inference.
An instrument that measures how much light bends when entering a liquid — used to read the salinity of seawater in parts per thousand. Must be calibrated with distilled water before use.
A black-and-white disk lowered into water on a marked rope. The depth at which it disappears measures turbidity (water clarity) — useful for tracking pollution and plankton density.
A square frame placed at random spots in a study area to count organisms in a fixed area. Lots of quadrat samples → a statistical estimate for the whole rocky shore or sand beach.
An explanation proposed on the basis of evidence obtained from research, which has survived repeated independent testing. Not a guess — in science this is the strongest kind of statement there is. Plate tectonics is a theory.
A statement of what you are setting out to do, written with one of the CAPS verb stems: To determine… / To show that… / To investigate… / To find out… / To observe… / To measure… Distinct from the hypothesis, and required in the report.
The initial inquiry — the question or problem that starts the whole investigation, written down plainly before anything else. CAPS names it as the first item in this block's Investigations column.
A histogram shows values for related, continuous categories and its columns touch. A bar graph shows separate categories in the same units and its columns are separated. The gap between the bars is the difference.
A CAPS research skill: group related objects and ideas together. You cannot count or compare anything until you have decided what counts as the same kind of thing.
A CAPS research skill: use numbers and measurements of time, length, width, volume and ratio. "Lots of mussels" is an impression; "37 mussels per 0.25 m² quadrat" is data.
A measured line laid across a habitat — usually a tape from the low-water mark up the shore — along which you sample at set intervals. It maps how organisms change with position, which random quadrats alone cannot show.
An instrument that measures the temperature of the ocean at any depth. The expendable version (XBT) is dropped from a moving ship and radios back a temperature-vs-depth profile as it sinks. Largely replaced now by Argo floats.
Quick Check
Question 1
Well done!
Move on to Topic 6 when you're ready.