Experimental Design · Strategy summary

Experimental Design: strategy summary

  1. Map each study first: changed (independent), measured (dependent), kept the same (controlled).
  2. A result is never a controlled variable: if the students couldn’t set it ahead of time, they measured it.
  3. The control matches the treated groups in every way except the treatment; check it column by column.
  4. For “why was this step done,” find the step and pick the reason its own words point to, not general lab advice.
  5. A useful new trial differs from an existing trial in exactly one variable; check it against every row.
  6. Averaging repeated trials shrinks random scatter, not a bias that pushes every reading the same way.
  7. If an instrument reads low, the true value is higher than recorded; if it reads high, the true value is lower.