Sara
Glad to help.
So the way I see it is as follows. You have made
calculations to determine the concentration of a
substance (x value) based on a y value. You have plotted
these points on a graph (scatter graph most appropriate
here as you are dealing with bivariate data points of the
form [x,y]) and you have now obtained a line of best fit
for the data. This is the basis of regression analysis. I
am presuming that there is some kind of relationship
between x and y that is evident from the chart e.g. high
values of y associated with high concentrations of x.
Even if there is no discernible relationship it is of
little consequence.
The next step is to take a value of y and predict the x
value from the line of best fit? I think that is what you
mean. (I presume you are not going to experimentally
determine the x value from the y). So, given that value y
is known, you can read off the predicted x value from the
line of best fit. Best to just use the equation of the
line of best fit and do the simple algebra to obtain the
x value, given y. So if y = 2x + 10 and y is 30, then we
know x = 10.
Now to get your data point onto the line of best fit. In
Excel spreadsheet you should have placed your y value and
predicted x value into cells. e.g. cell A1 = y A2 =
predicted x.
Now in your chart, right click the mouse and select
<Source Data>. You will most likely just have
one 'Series' on the chart known as 'Series 1'. At any
rate you need to select <Add Series> and in the adjacent
boxes type in the cell references for your x and y values.
So y may be =sheet1!$A$1 (use name of sheet if you have
one instead of 'sheet1')
and x may be =sheet1!$A$2 (based on the example I gave
above)
This will add your new data point onto the chart as a
separate series. Now back on the chart you can double
click that data point and assign it a different colour
etc. so that it stands out from your original data.
I hope this helps and is what you were requesting. Again,
please reply if your are still struggling and I will
endeavour to assist.
Thanks
Alex
-----Original Message-----
Thanks a lot. That did help. Now, I have a follow up question: (this one you
may not be able to do, but its worth a shot, right?) Can you also plot a
point on the line of best fit based on a y value and get an x value for it? I
dont know if that's clear or not, but I have a lab, and I have a set of data,
and I created the line of best fit. Now, I need to take an unknown substance
and find the concentration (the x value) based on the y value I found and
make it show up on the graph. Does that make any more sense? I mean, I could
do it by hand, but that would not look too good, and if I have the program, I
might as well use it, right? Thanks for the help so far-
~Sara
.
