Alicia598 Posted November 19, 2015 Report Share Posted November 19, 2015 I'm doing a lab, and my data is very messed up. Unfortunately, I don't think I'm allowed to share the actual graphs on here. So I'll just ask some questions: 1. If I'm looking at only two data points, can I say they have a direct relationship? Or would it be safer to say 'positive correlation'?2. As for a linear graph with a negative gradient - not an inversely proportional relationship, but is it still an inverse relationship?3. If the data points have significantly large error bars, to the point not being to plot a maximum line, is it fine to just refer to the minimum value? (There is a justified reason why the error bars are large). And yes, I've went to see my teacher about the lab, he was quite vague (understandably I guess), but told me not to redo my experiment... Thanks to anyone with answers.... Reply Link to post Share on other sites More sharing options...
CkyBlue Posted November 20, 2015 Report Share Posted November 20, 2015 1. Two data points is not enough to make a safe conclusion about a relationship. To be absolutely blunt, you can't even say the correlation is positive. The relationship could be periodic, making the graph behave in a wave; it could also be polynomial to a higher degree, so there could be turning points. If you're just strictly talking about the two data points, then yeah I guess you could say they have a positive correlation, but that doesn't give you any information about the relationship between the dependent and independent variables. 2. Yeah it would be an inverse relationship. I believe that just means that as one variable goes up, the other goes down. Proportionality just means the variables are bound by a factor of a constant; e.g. y=-2x is an inversely proportional relationship. But if you have something like y=e^(-x), that would be an inverse relationship. 3. Haven't really encountered this before, but I would at least try to graph the points electronically and make note of the maximum slope, just to stick to standard procedure as much as possible. I don't know the nature of your experiment so I can't really say anything about that, but if you have a justified reason then make sure you get your point through in the discussion section of your lab. Hope this helps 1 Reply Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.