Answer:
Experiments A and D are completely randomized. And experiments C and E are also randomized enough to pass as a randomized design.
But the design for experiment B is nowhere near a completely randomized design.
Step-by-step explanation:
In random sampling, each member of the population would have an equal chance of being sampled. In a perfect random sampling scenario, they would use computer generated random numbers to select members of the population to be sampled.
a) Experiment A randomly selects all the fishes with epidermal abrasions for the experiment and control experiment. Here, all the fishes with epidermal abrasions have an equal chance of being sampled; hence, this is a randomized design.
b) Experiment B gives plants on the outer edge of the garden old fertilizers and gives plants in the middle of the garden new fertilizers and intends to check for growth based on the two fertilizers. This is not a randomized design, as all the plants do not have equal probability/chance of being tested for either fertilizers. A randomized design would have randomly picked plants all over (outer edge, middle, everywhere) with no giving which plant which fertilizer based on where it is growing.
c) Experiment C intends to check the effects of a new weight loss remedy. But they group the participants in the study into genders and then randomly select which weight loss plan to administer to them.
The design to administer the weight loss plan is randomized alright, but grouping the participants into gender groups removes the air of complete randomization. But this design should still pass for a randomized design as every member of the population has an equal chance of getting any one of the 3 weight loss programs.
d) Experiment D uses a computer to randomly assign people into an experiment group and a control group, to test whether a new supplement helps individuals increase egg production. This is very very randomized!
e) Experiment E tests a new antivirus by randomly picking the sample population (10 each) from two groups of computers and then randomly testing 5 of the 10 selected per group while leaving the remaining 5 per group as Control. This is fairly randomized too.
