Small Group Tutorials

Here to help students catch up, keep up, and move ahead. Book a consultation here.

How to Perform in the new G2 SEC Examinations | Learner’s Guide Vol 0115 | Science: Baseline Drift — Check Whether the Reference Condition Is Moving Too

Three students working with a tutor in a small-group learning setting

G2 Science K223, K224 and K225 experimental reasoning can fail when the reference condition itself changes. A baseline is not automatically stable because it was measured first. Temperature, instruments, organisms, reagents and environments can drift over time.

This one-hundred-and-fifteenth Learner’s Guide develops baseline-drift control. It extends Vol 0099 Control-Condition Logic and Vol 0103 Model-Validity Boundaries.

The baseline-drift question

Ask what would have happened to the measured outcome over time if the treatment had never been applied. A concurrent control, repeated baseline or calibration check can reveal whether the reference process is moving.

1. temperature baseline

For temperature baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

2. light baseline

For light baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

3. humidity baseline

For humidity baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

4. airflow baseline

For airflow baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

5. background sound

For background sound, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

6. background light

For background light, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

7. instrument zero

For instrument zero, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

8. balance zero

For balance zero, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

9. thermometer calibration

For thermometer calibration, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

10. pH calibration

For pH calibration, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

11. voltage supply

For voltage supply, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

12. battery condition

For battery condition, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

13. lamp output

For lamp output, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

14. reagent strength

For reagent strength, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

15. enzyme preparation

For enzyme preparation, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

16. organism activity

For organism activity, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

17. plant water status

For plant water status, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

18. seed viability

For seed viability, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

19. room-temperature trend

For room-temperature trend, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

20. time-of-day biology

For time-of-day biology, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

21. practice or fatigue

For practice or fatigue, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

22. apparatus warming

For apparatus warming, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

23. evaporation baseline

For evaporation baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

24. natural cooling

For natural cooling, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

25. natural heating

For natural heating, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

26. control-group drift

For control-group drift, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

27. treatment pre-trend

For treatment pre-trend, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

28. different starting baselines

For different starting baselines, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

29. different baseline slopes

For different baseline slopes, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

30. historical baseline

For historical baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

31. concurrent baseline

For concurrent baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

32. interleaved control

For interleaved control, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

33. randomised order

For randomised order, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

34. counterbalanced order

For counterbalanced order, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

35. repeat baseline

For repeat baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

36. calibration before

For calibration before, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

37. calibration after

For calibration after, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

38. calibration during

For calibration during, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

39. blank over time

For blank over time, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

40. positive control over time

For positive control over time, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

41. negative control over time

For negative control over time, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

42. sensor warm-up

For sensor warm-up, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

43. sample settling

For sample settling, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

44. reaction already started

For reaction already started, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

45. delayed measurement

For delayed measurement, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

46. unequal exposure duration

For unequal exposure duration, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

47. sequential sampling

For sequential sampling, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

48. environmental trend

For environmental trend, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

49. batch effect

For batch effect, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

50. day effect

For day effect, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

51. operator effect

For operator effect, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

52. observer calibration

For observer calibration, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

53. camera exposure

For camera exposure, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

54. scale drift

For scale drift, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

55. baseline subtraction

For baseline subtraction, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

56. difference-in-change

For difference-in-change, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

57. ratio-to-baseline

For ratio-to-baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

58. percentage change

For percentage change, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

59. rolling baseline

For rolling baseline, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

60. baseline noise

For baseline noise, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

61. baseline trend

For baseline trend, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

62. baseline step change

For baseline step change, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

63. baseline outlier

For baseline outlier, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

64. drift mistaken for treatment

For drift mistaken for treatment, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

65. treatment mistaken for drift

For treatment mistaken for drift, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

66. shared drift

For shared drift, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

67. opposite drift

For opposite drift, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

68. drift plus interaction

For drift plus interaction, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

69. drift plus ceiling

For drift plus ceiling, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

70. drift plus floor

For drift plus floor, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

71. drift and rate

For drift and rate, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

72. drift and amount

For drift and amount, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

73. drift and model

For drift and model, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

74. drift and conclusion

For drift and conclusion, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

75. drift and improvement

For drift and improvement, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

76. baseline drift final rule

For baseline drift final rule, identify the reference quantity that is assumed to stay stable. Then ask what physical, biological, environmental, instrumental or procedural process could move that reference even without the treatment.

For practice, draw a baseline trajectory and a treatment trajectory across the same time axis. Decide whether the relevant treatment effect is the final difference, the change from each group’s start, or the difference between changes. State why that comparison controls the drift more effectively.

Do not automatically subtract the first reading. If the baseline continues moving, one initial correction can leave later bias. Match the reference to the time, apparatus and condition of the treatment measurement, and use concurrent controls or repeated calibration where the design allows.

Baseline drift is a time-dependent confounder

When treatments are applied sequentially, time can become correlated with condition. Randomising or counterbalancing order, interleaving controls and recalibrating can separate treatment from session drift. The correct repair depends on what is drifting.

Stable-looking data can still contain drift

A slow drift can be hidden by noisy measurements, rounding or a short observation window. Conversely, random fluctuation should not automatically be labelled drift. Look for a systematic time pattern, repeated reference measurements and a plausible process that could move the baseline.

Links

Use the Science Hub, Vol 0111 Rate Versus Amount, Vol 0083 Diagnostic-Test Evidence, 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. Baseline-drift control is an eduKateSengkang experimental-reasoning framework, not an additional SEAB syllabus topic.

Final rule

A baseline is not merely the first number. It is the reference process that would have happened without the treatment. Check whether that process is stable before attributing every later difference to the factor you changed.