Andria wants to conduct an experiment to determine whether a new medication is effective in reducing swelling. Explain to Andria the best sampling method, what statistics she should study, and what conclusions she can draw from the experiment.

Question

undeadknight26 · Answer

The best sampling method she could use would be experimentation.
She should study how long it takes for the swelling to reduce.
Some conclusions she can draw the the experiment is if the product actually works, who it works best on and some possible side effects of the medication

undeadknight26 · Answer

@amistre64 @sleepyjess @dan815 @e.mccormick @robtobey Is this right?

amistre64 · Answer

experimentation is not a sampling method ... at least none that im aware of.

undeadknight26 · Answer

x.x
I was having trouble finding the list of sampling methods ;-;

amistre64 · Answer

take a simple random sample of the population of swelled people ... then run an experiment using a control group etc ...

amistre64 · Answer

the best sampling method is one that allows you to gather a group of people that is similar in properties to the population as a whole.

a simple random sample allows us to gather randomly from the population such that each member of the population has an equal likelihood of being picked.

bias comes in when we try to take  subset of an otherwise biased subset.  like taking a random sample of callers to a radio station only gets a sample of radio station callers and not the desired population as a whole.

e.mccormick · Answer

Ami, the name for that is multistage sampling. Here, these are the official names they probably need to deal with: http://stattrek.com/survey-research/sampling-methods.aspx

amistre64 · Answer

official names .... fight the power!  :)

e.mccormick · Answer

Umm... 64 = 8^2, so you rely on the power for your name.

undeadknight26 · Answer

So im guessing use Voluntary Sampling?

amistre64 · Answer

i use the brute method:  just count up 8, 8s

Voluntary sampling is akin to the radio station caller.  you only get a sample of those who are willing to volunteer.

now which sample you choose, im not going to really know if its good or not.  i was never adept at that part of this stuff.

amistre64 · Answer

there are statistical measures that help even out the bias in a selection process.  i just never got that far into it.

e.mccormick · Answer

Well, swelling and inflamtion happen for a wide range of things.  Like Ami said, you need people both with it and without.  You may do stratified sampling with the strata being if they have regular inflamation or not... or systematic random sampling.  Yah, that one is probably what you need for who gets the drug or placebo.

undeadknight26 · Answer

So is everything else in the answer alright?

amistre64 · Answer

what statistics she should study:  what are some things that we would be looking for?  sideeffects, people satisfied, effectiveness of therapy, cost of therapy ... etc

amistre64 · Answer

what conclusions she can draw from the experiment.  well, she would be able to conclude who well the statistics meaure up  and if the therapy is useful overall.

undeadknight26 · Answer

Alright thank you my good sir!