9.1 - Variables & Controls

Identifying the variables in an experiment

• A common skills question asks you to name the variables in an experiment, or to give a variable that should be kept the same.
• Every experiment has three kinds of variable, and the rules for telling them apart never change - so these are reliable marks to score.
• The mark goes to the exact variable, named precisely - not a vague description.

The three kinds of variable to name

1. Variable changed - The one thing the experimenter deliberately alters
2. Variable measured - The result recorded to see the effect of the change
3. Variable kept constant - A factor held the same, named specifically (not just "keep everything else the same")

A typical question

A student investigated how the concentration of hydrochloric acid affects the rate of diffusion into cubes of agar. Cubes of the same size were placed in acid at five different concentrations, and the time for each cube to change colour was measured.

(i) Name the variable that was changed.

(ii) Name the variable that was measured.

(iii) Suggest one variable that should be kept constant.

The answers

The mark goes to the exact variable, named precisely:

1. The variable changed is the concentration of hydrochloric acid.
2. The variable measured is the time for the cube to change colour.
3. A controlled variable, named specifically, such as the size of each cube, the volume of acid, or the temperature.

The three kinds of variable

• Every experiment has one variable you change, one you measure, and several you keep the same.
• Read the aim as "how does X affect Y": the X is the variable you change and the Y is the variable you measure, every time.
  • Example: "How does acid concentration affect the time to change colour" hands you both.

Spotting each variable

Controls and a fair test

The word "control" covers two different things in an experiment - keep them apart:

1. Controlled variables - The things you keep the same
2. Control experiment - An extra sample set up with the tested factor left out

Controlled variables and validity

• Controlled variables are the things you keep the same - for example the cube size, the volume of acid, the temperature.
• Keeping them the same makes the investigation a fair test, which is what gives a valid result.
• SQA's wording: the other variables are controlled "to ensure any effect is likely to be due to the variable being changed".
• So if only the acid concentration changes and everything else stays put, any difference in the time to change colour must be down to the concentration and nothing else.
• Name a controlled variable in full - "the temperature" and "the volume of acid" score; "the conditions" does not, because the marker cannot tell which condition you mean.

Validity is not reliability

• This is a mark-critical point:
  • When asked why the other variables are kept the same, the answer is about validity - it keeps the test fair.
  • Answering "to make it reliable" is wrong here, and the marking instructions do not accept it.
• Reliability is a different idea - it means getting consistent values in repeats, earned by repeating readings and taking a mean.

The control experiment

• A control experiment is an extra tube or sample set up with the factor you are testing left out, run alongside the real ones for comparison.
• Example: testing whether sucrose draws water into dialysis tubing, the control is a tube of water with no sucrose in it.
• If water moves into the sucrose tube but not the water-only tube, you can be sure the sucrose caused it.

Common exam questions

Question 1

A trial compared a genetically modified cotton crop with the original cotton, measuring the yield of each. Identify the variable that was altered.

The type of cotton grown, modified or original. That is the one variable deliberately changed; the yield is the variable measured.

---

Question 2

Suggest one variable that would have to be controlled to ensure the results are valid.

Any named factor kept the same for both crops, such as the amount of water, the amount of fertiliser, the soil type, or the area of each field. Keeping these the same makes it a fair test, so any difference in yield can be put down to the type of cotton.

---

Question 3

The fields with the original cotton were used as a control. Give one reason for using a control.

So there is something to compare the modified crop against; the control shows that any change in yield was caused by the modified cotton and not by the growing conditions.

---

Question 4

Describe how the reliability of these results could be improved.

Repeat the trial over more fields, or over more growing seasons, and take a mean, so that consistent values across the repeats show the result is reliable.

---

Question 5

Why is "to make it reliable" wrong when asked why the other variables were kept the same?

Because keeping the other variables the same is what makes the test fair, so it is about validity, not reliability; reliability comes from repeating the trial and averaging.