PCE - Well Designed Experiments Lesson

Math_Lesson_TopBanner.png Well Designed Experiments

1.  Control the effects of lurking variables by comparing several treatments (include a control group if possible).

2.  Use randomization to assign subjects/units to treatments.

3.  Replicate (repeat) the experiment on many subjects/units to reduce "chance" variation in the results.

CONTROL - RANDOMIZATION - REPETITION

Another step MAY be required if the response to the treatment is expected to be different for different groups. For example, men and women may react differently to medications or teens and senior citizens might respond differently to different types of music.   In situations where mixing all the participants together could potentially cause HUGE variation in results it is best to group subjects with similarities into a BLOCK and keep them separate throughout the experiment. An effect or outcome is called STATISTICALLY SIGNIFICANT if it is too great to be caused simply by chance.  Even a well-designed experiment can contain hidden bias, so it is extremely important to handle the subjects/units in exactly the same way.  REMEMBER: one way to avoid hidden bias is to conduct a DOUBLE-BLIND experiment.  In a double-blind experiment, neither the subjects nor the people who have contact with them know which treatment has been received.

Math_APStatisticsBottomBanner.png