Total Posts:4|Showing Posts:1-4
Jump to topic:

If two studies have different results...

themohawkninja
Posts: 816
Add as Friend
Challenge to a Debate
Send a Message
10/17/2013 9:20:52 AM
Posted: 3 years ago
If two studies contradict each other (one with a negative correlation, and one with a positive correlation), I imagine that the most logical course (aside from finding more studies) would be to just throw out the piece of evidence, since the data is inconclusive.

However, what should someone do if one study has a correlation (either negative or positive), and the other study shows no correlation? Do you cite that there is a net correlation, or do you make the argument that these were two different sets of people, which explains the two different outcomes, and therefore the data is inconclusive?
"Morals are simply a limit to man's potential."~Myself

Political correctness is like saying you can't have a steak, because a baby can't eat one ~Unknown
TheAntidoter
Posts: 4,323
Add as Friend
Challenge to a Debate
Send a Message
10/17/2013 10:35:56 AM
Posted: 3 years ago
I would say usually show it as inconclusive.

More tests all around.
Affinity: Fire
Class: Human
Abilities: ????

Nac.

WOAH, COLORED FONT!
chui
Posts: 511
Add as Friend
Challenge to a Debate
Send a Message
10/17/2013 11:42:06 AM
Posted: 3 years ago
I would think the first step would be to compare the studies in terms of quality. For example how much data was collected, how good was the collecting method, how statistically significant is the result etc. If both are equal in quality then further studies are required.
Bullish
Posts: 3,527
Add as Friend
Challenge to a Debate
Send a Message
10/17/2013 4:54:01 PM
Posted: 3 years ago
At 10/17/2013 9:20:52 AM, themohawkninja wrote:
If two studies contradict each other (one with a negative correlation, and one with a positive correlation), I imagine that the most logical course (aside from finding more studies) would be to just throw out the piece of evidence, since the data is inconclusive.

However, what should someone do if one study has a correlation (either negative or positive), and the other study shows no correlation? Do you cite that there is a net correlation, or do you make the argument that these were two different sets of people, which explains the two different outcomes, and therefore the data is inconclusive?

Contradicting studies occur all the time. If this happens, then the first thing to do should be to compare them. Sometimes the studies could have a single different variable that makes a huge difference. Some times it's simply a wording, definition, or semantics problem. Sometimes it could be that one study messed up. Sometimes it could be both studies used diverging methods. There are many possibilities.

The solution is usually to do more studies and record procedure and data more diligently.
0x5f3759df