expt

Debate and discussion of any biological questions not pertaining to a particular topic.

Moderators: honeev, Leonid, amiradm, BioTeam

Post Reply
MIA6
Garter
Garter
Posts: 43
Joined: Thu Sep 14, 2006 1:15 am

expt

Post by MIA6 » Fri Sep 15, 2006 3:32 am

Why when we do the experiment, we need to have large sample sizes?

Thanks.

User avatar
mith
Inland Taipan
Inland Taipan
Posts: 5345
Joined: Thu Jan 20, 2005 8:14 pm
Location: Nashville, TN
Contact:

Post by mith » Fri Sep 15, 2006 4:13 am

You don't if you did it correctly the first time, once should be enough.
Living one day at a time;
Enjoying one moment at a time;
Accepting hardships as the pathway to peace;
~Niebuhr

baikuza
Coral
Coral
Posts: 429
Joined: Tue Aug 30, 2005 10:09 am
Location: Yogyakarta, indonesia, south-east asia
Contact:

Post by baikuza » Fri Sep 15, 2006 7:42 am

hm.. i do not know any about you.

maybe you are thinking about this when you were come to the lab. of your faculty/school
and it is your first experience.
more chemicals for more practican
by the way.. what poison said is right.
we do not need to spend more bugs(money) for addition of extra chemicals if we do it correctly.

pardon me if "i step my foot to far on your teritory".

User avatar
canalon
Inland Taipan
Inland Taipan
Posts: 3909
Joined: Thu Feb 03, 2005 2:46 pm
Location: Canada

Post by canalon » Fri Sep 15, 2006 7:04 pm

Well in some case, even if done properly, an experiment need to be repeated enough time to give a statistically significant result.
You should read about statistical power and how to detrmine the smallest possible sample size to be able to decide if the difference you expect between your control and the experiment is significant.
Patrick

Science has proof without any certainty. Creationists have certainty without
any proof. (Ashley Montague)

truevisionmicroscopes06
Garter
Garter
Posts: 5
Joined: Mon Sep 25, 2006 2:30 pm

Post by truevisionmicroscopes06 » Mon Sep 25, 2006 3:12 pm

You're right canalon, experiments must be done twice or more to come up with the correct results. I think large or small samples doesn't really matter.

The hobbyist!

User avatar
MrMistery
Inland Taipan
Inland Taipan
Posts: 6832
Joined: Thu Mar 03, 2005 10:18 pm
Location: Romania(small and unimportant country)
Contact:

Post by MrMistery » Mon Sep 25, 2006 5:34 pm

of course samples matter. If you flip a coin twice and it lands on heads both times, you won't find it suspicious. Flip it 2000 times and if it always lands on heads, you should ask yourself some questions about that coin. Get my point?
"As a biologist, I firmly believe that when you're dead, you're dead. Except for what you live behind in history. That's the only afterlife" - J. Craig Venter

User avatar
canalon
Inland Taipan
Inland Taipan
Posts: 3909
Joined: Thu Feb 03, 2005 2:46 pm
Location: Canada

Post by canalon » Mon Sep 25, 2006 11:13 pm

truevisionmicroscopes06 wrote:You're right canalon, experiments must be done twice or more to come up with the correct results. I think large or small samples doesn't really matter.

The hobbyist!


Huhh, I was saying that the size of the sample is important. The more the better, but economical reason (money and time) suggest that we can try to minimize the numbers of repeats. That is why I was suggesting to calculate the power of a test. Some differences are easy to spot and can be solved with a very few case (effects of 500mg/kg cyanide vs placebo) while others need much bigger samples (effects of low radiation on the incidence of cancer).
Patrick

Science has proof without any certainty. Creationists have certainty without
any proof. (Ashley Montague)

User avatar
mith
Inland Taipan
Inland Taipan
Posts: 5345
Joined: Thu Jan 20, 2005 8:14 pm
Location: Nashville, TN
Contact:

Post by mith » Mon Sep 25, 2006 11:28 pm

Of course I was being facetious when I answered in my first post, since the question is a homework question that MIA6 should have thought about him/herself.

Obviously it would be nice to do one correct experiement and be done with but many times you don't know whether it's correct or not even when you do your best to control all the variables there's still chance error, so you do more...
Living one day at a time;
Enjoying one moment at a time;
Accepting hardships as the pathway to peace;
~Niebuhr

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests