Physics EE error analysis help

[HKamal]

Hello everyone,

I am doing my ee in physics. I'll get straight into it without wasting anyone's time.

So basically in my results after conducting the experiment I have two values for the electron to mass ratio of an electron (e/m ratio). The independent variable was changed in the experiment and it was conducted in two parts therefore leading to two values, one for each part. 

The values are:

1.96*10¹¹ ±8.63*10¹⁰ 

2.01*10¹¹ ±1.09*10¹¹

Finding the values required using the gradient of the graph results. Therefore to get the errors the uncertainty on the graphs were calculated using the range (largest gradient - smallest gradient) and some propagation of errors was used as well. That was all good. Now, the uncertainty as you can see is very large about 50% for each value. There are reasons for that but that is also ok. 

First question:

To get the discrepancy, I first need an average of the two values. To get the uncertainty on the average I just took the average of the uncertainties (add both and divide by two). Is this correct?

Second question:

Now that I got the average value I calculated the discrepancy to be 13%. My teacher said that I should check to see if this lies within the %error on my results and if it does then my experiment is accurate. But the problem is I got the %error on the averaged results to be 50%. Does this mean that my results are accurate and not precise? How can it be accurate if the %error is so large, the discrepancy is bound to fall into the range of the error.

Hope someone can answer these confusions. I have my ee due in less than two weeks. Sorry for the long post.