G2 Science K223, K224 and K225 experimental reasoning becomes sharper when learners can say exactly what a control condition is doing. “Keep it the same for a fair test” is a useful beginning, but advanced examination answers need the next step: which alternative explanation does that control remove?
This ninety-ninth Learner’s Guide develops control-condition logic. It complements Vol 0065 Counterfactual Causality, Vol 0083 Diagnostic-Test Evidence and Vol 0087 Converging Evidence.
The control-logic question
For every control, complete this sentence: “If this factor differed, it could also explain the outcome; therefore we hold it constant or provide a comparison condition.” This connects procedure to causal inference.
1. No-treatment control
Design: treatment group receives factor; control does not.
What it does: helps separate treatment effect from background change when other conditions match. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
2. Vehicle control
Design: both groups receive carrier; only treatment has active factor.
What it does: rules out carrier as simple explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
3. Blank control
Design: reagents/apparatus without sample.
What it does: checks whether procedure itself creates signal. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
4. Positive control
Design: known-positive condition should produce signal.
What it does: checks whether test can detect target under procedure. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
5. Negative control
Design: known-negative condition should stay negative.
What it does: checks background signal and specificity. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
6. Same temperature
Design: temperature is held equal.
What it does: reduces temperature as alternative explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
7. Same light
Design: light exposure held equal.
What it does: reduces light as confounder. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
8. Same duration
Design: groups measured for equal time.
What it does: reduces exposure-time difference. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
9. Same volume
Design: sample/reagent volume held equal.
What it does: reduces amount differences where relevant. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
10. Same concentration except target
Design: background concentration matched.
What it does: isolates target factor more cleanly. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
11. Same organism type
Design: groups use same species/type.
What it does: reduces biological-type variation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
12. Same age/stage
Design: developmental stage matched.
What it does: reduces age as alternative cause. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
13. Same starting size
Design: initial size matched.
What it does: improves comparison of growth/change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
14. Same apparatus
Design: same equipment used.
What it does: reduces equipment differences but shared calibration bias can remain. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
15. Same instrument model
Design: measurement method matched.
What it does: improves comparability but does not guarantee accuracy. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
16. Calibrated instrument
Design: reference check is performed.
What it does: addresses measurement accuracy rather than biological control. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
17. Random allocation
Design: subjects assigned without systematic bias.
What it does: reduces pre-existing group differences. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
18. Matched pairs
Design: similar subjects paired across conditions.
What it does: controls selected individual differences. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
19. Repeated trials
Design: same condition repeated.
What it does: assesses repeatability; repetition is not itself a control condition. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
20. Independent replication
Design: study repeated independently.
What it does: tests reproducibility; not a substitute for within-experiment control. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
21. Baseline before treatment
Design: same subject measured before change.
What it does: provides within-subject reference but time effects may remain. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
22. Concurrent control
Design: untreated group measured over same period.
What it does: helps distinguish time/background effects. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
23. Historical control
Design: comparison uses earlier data.
What it does: weaker against changing conditions than concurrent control. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
24. Control with no independent variable
Design: baseline condition lacks manipulated factor.
What it does: supports causal comparison if all other relevant factors match. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
25. Control at standard level
Design: factor held at ordinary/reference value.
What it does: allows comparison with changed level. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
26. Multiple treatment levels
Design: several levels plus control.
What it does: reveals dose-response or threshold pattern. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
27. Sham procedure
Design: control experiences procedure without active intervention.
What it does: separates procedural effects. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
28. Placebo-like school-science analogy
Design: control mimics treatment experience without active factor.
What it does: conceptually separates expectation/procedure where relevant, though not all school experiments involve this. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
29. Dark control
Design: photosynthesis setup kept dark.
What it does: helps distinguish light-dependent gas/starch change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
30. No-plant control
Design: apparatus without plant.
What it does: checks whether apparatus/environment creates signal. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
31. Boiled enzyme control
Design: inactive enzyme condition.
What it does: helps show active enzyme is needed for observed reaction. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
32. No-enzyme control
Design: substrate without enzyme.
What it does: checks non-enzymatic background change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
33. No-substrate control
Design: enzyme without substrate.
What it does: checks whether product signal requires substrate. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
34. Room-temperature reference
Design: treatment temperatures compared with standard condition.
What it does: shows change relative to reference, not universal optimum by itself. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
35. pH reference
Design: one standard pH condition.
What it does: supports comparison across pH levels. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
36. Distilled-water control
Design: sample receives water instead of tested solution.
What it does: helps isolate solute/treatment effect. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
37. Solvent control
Design: same solvent without dissolved treatment.
What it does: rules out solvent as simple cause. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
38. Unheated control
Design: heated sample compared with unheated sample.
What it does: isolates heating when other conditions match. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
39. No-catalyst control
Design: reaction without catalyst.
What it does: supports catalyst effect on rate when other factors controlled. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
40. No-current control
Design: electrical setup without current.
What it does: helps distinguish electrical heating/effect from background. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
41. Open-circuit control
Design: component present but circuit incomplete.
What it does: tests whether observed effect requires current flow. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
42. No-force baseline
Design: system measured without applied force.
What it does: provides reference for force response. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
43. Zero-load reference
Design: extension measured at no added load.
What it does: anchors change in length. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
44. No-magnet control
Design: object tested without magnet/field.
What it does: checks background movement. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
45. Shielded/background reading
Design: detector records baseline without source.
What it does: allows subtraction/interpretation of background. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
46. Ambient-temperature baseline
Design: system measured before heating/cooling.
What it does: reference for temperature change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
47. Closed-system comparison
Design: mass measured with matter retained.
What it does: tests conservation under controlled boundary. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
48. Open-system comparison
Design: matter can leave/enter.
What it does: shows why mass reading can differ from closed case. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
49. Same initial mass
Design: groups start equal.
What it does: supports fair comparison of mass change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
50. Same surface area
Design: reaction samples matched in exposed area.
What it does: reduces surface-area effect. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
51. Same particle size
Design: solid pieces matched.
What it does: reduces rate difference from surface area. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
52. Same stirring
Design: mixing held equal.
What it does: reduces mixing as alternative explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
53. Same pressure
Design: gas-related conditions matched.
What it does: reduces pressure effect where relevant. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
54. Same humidity
Design: transpiration groups matched.
What it does: reduces humidity as alternative cause. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
55. Same airflow
Design: air movement held equal.
What it does: reduces airflow effect. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
56. Same leaf area
Design: plant material matched.
What it does: reduces area-driven transpiration/photosynthesis difference. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
57. Same seed number
Design: groups contain equal counts.
What it does: makes germination proportions/counts comparable. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
58. Same seed viability source
Design: seeds drawn from comparable batch.
What it does: reduces viability differences. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
59. Same soil
Design: growth groups use same soil conditions.
What it does: reduces nutrient/water-holding differences. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
60. Same water
Design: watering held equal.
What it does: reduces water as alternative cause. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
61. Same nutrient level
Design: nutrient supply matched.
What it does: reduces nutrient differences. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
62. Control variable accidentally changes
Design: a supposed constant differs between groups.
What it does: causal attribution weakens because another explanation remains. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
63. Control variable measured poorly
Design: factor may differ despite intended control.
What it does: state limitation rather than assuming equality. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
64. Control variable unnecessary
Design: a variable is held constant though it cannot affect outcome.
What it does: harmless but may add procedural complexity. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
65. Control variable impossible to hold exactly
Design: natural variation remains.
What it does: measure or randomise where appropriate rather than claiming perfect control. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
66. One control rules out one class
Design: a matched condition addresses a specific alternative.
What it does: do not claim the experiment controls everything. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
67. Control does not prove mechanism
Design: treatment-control difference exists.
What it does: difference supports effect but mechanism may need additional evidence. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
68. Control does not prove universality
Design: effect appears in one setup.
What it does: scope remains population/range/condition limited. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
69. Control does not fix measurement bias
Design: both groups measured by same biased instrument.
What it does: comparison may remain fair while absolute values are inaccurate. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
70. Control does not fix contamination
Design: both groups may be contaminated.
What it does: shared contamination can distort both. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
71. Control does not fix wrong variable definition
Design: measurement targets wrong quantity.
What it does: perfect group matching cannot rescue invalid endpoint. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
72. Control does not fix tiny sample
Design: groups are matched but too small.
What it does: random variation can still dominate. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
73. Control does not fix short duration
Design: groups are matched but response needs longer.
What it does: null result may remain weak. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
74. Control does not fix ceiling
Design: both groups near maximum.
What it does: treatment difference may be hidden. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
75. Control does not fix floor
Design: both groups near minimum.
What it does: decrease may be impossible to observe. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
76. Control plus dose response
Design: control anchors zero/reference and treatments vary.
What it does: stronger evidence for graded relation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
77. Control plus time course
Design: control and treatment followed over time.
What it does: helps distinguish background temporal change. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
78. Control plus mechanism
Design: group difference plus mechanistic evidence.
What it does: supports stronger causal explanation than comparison alone. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
79. Control plus counterfactual
Design: removing factor removes effect.
What it does: strengthens necessity claim under conditions. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
80. Control plus rescue
Design: restoring factor restores effect.
What it does: can strengthen mechanism inference. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
81. Control plus independent measure
Design: same effect seen through different measurement.
What it does: reduces single-method explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
82. Control plus replication
Design: same controlled pattern repeats.
What it does: confidence in repeatability rises. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
83. Control group changes too
Design: both treatment and control shift.
What it does: look for common background cause. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
84. Treatment changes more
Design: both shift but treatment changes further.
What it does: estimate treatment-specific difference rather than ignoring baseline shift. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
85. Control changes opposite
Design: groups diverge.
What it does: check whether conditions truly matched and whether mechanism predicts divergence. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
86. No difference
Design: treatment and control remain similar.
What it does: weaken large-effect claim if manipulation and sensitivity are valid. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
87. Control unexpectedly positive
Design: negative control shows signal.
What it does: suspect contamination/non-specific test. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
88. Positive control unexpectedly negative
Design: known-positive fails.
What it does: test sensitivity/procedure is questionable. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
89. Control sample mix-up
Design: labels swapped.
What it does: data ownership must be repaired before interpretation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
90. Control timing mismatch
Design: control measured at different time.
What it does: time becomes alternative explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
91. Control location mismatch
Design: groups kept in different places.
What it does: environment becomes alternative explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
92. Control handling mismatch
Design: one group handled differently.
What it does: handling becomes confounder. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
93. Control observer mismatch
Design: different observers use different criteria.
What it does: observer effect may confound comparison. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
94. Control instrument mismatch
Design: different instruments used.
What it does: instrument differences become alternative explanation. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
95. Control selection bias
Design: groups differ systematically before treatment.
What it does: post-treatment difference may pre-exist. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
96. Control attrition
Design: different subjects drop out.
What it does: remaining groups may no longer be comparable. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
97. Control range mismatch
Design: control and treatment cover different input ranges.
What it does: comparison may not isolate factor. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
98. Control denominator mismatch
Design: rates use different reference populations.
What it does: apparent difference may be normalisation artefact. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
99. Control graph baseline
Design: control line provides reference trajectory.
What it does: compare treatment deviation from baseline, not just final value. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
100. Control table ownership
Design: control values must remain attached to correct column.
What it does: source misassignment destroys comparison. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
101. Control conclusion wording
Design: say effect under tested conditions.
What it does: do not write treatment always causes outcome. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
102. Control final rule
Design: name the alternative explanation the control is designed to remove.
What it does: then state what alternatives still remain. For practice, name one alternative explanation this control reduces and one alternative it does not address. This prevents the phrase “fair test” from becoming a substitute for reasoning.
When evaluating a result, ask whether the control actually remained comparable, whether the manipulation succeeded and whether measurement sensitivity was adequate. A nominal control is not useful if its conditions drift or its data are assigned to the wrong group.
Controls narrow explanations; they do not create certainty
A strong control can make one causal explanation more plausible by removing competitors. It does not automatically prove the mechanism, generalise to every population or repair measurement bias. The conclusion should remain matched to the alternatives genuinely ruled out.
Links
Use the Science Hub, Vol 0079 Negative Evidence, Vol 0095 Scale Translation, the Examination Craft hub and the PSLE Learner’s Guide.
Official-source discipline
For the current 2027 SEC G2 school-candidate framework, use the official SEAB G2 syllabus directory and linked K223–K225 Science syllabuses. Control-condition logic is an eduKateSengkang reasoning framework used to organise experimental evaluation, not an additional SEAB syllabus topic.
Final rule
Do not write “keep variables the same” and stop. Name the alternative explanation each control is designed to remove, check whether the control actually worked, and keep the final causal claim inside the evidence the design can support.
