Note: Your concluding statement does not “prove” your hypothesis, it supports it. The conclusion itself is often very short, do not worry if this is the case.

Conclusion:

Does the trend seen in the graphs support your hypothesis? What is the actual relationship - here you need to state the gradient with its uncertainty (is it a directly proportional relationship i.e., y=mx ?) and the intercept with its uncertainty (with a non-zero intercept it is a linear relationship so the intercept clarifies which sort of relationship you have discovered, which is why it is so important to report). 

Does this answer your research question fully, or partially, are there limitations to your conclusion? Is there more than one pattern in the data? 

How do your results compare to literature values (or other similar experiments reported on the internet / the model you have developed / a predicted gradient value using data from other sources)? (e.g., The measured time was 9.76±0.12ms^(−2), compared with the accepted value of approximately 9.81ms^(−2). As the accepted value lies within the experimental uncertainty range, the experimental result is consistent with the accepted value.)

The percentage uncertainty of the gradient is an indicator of the validity of your method, what is yours and what does it tell you? Can you use this to comment on the reliability of the method, does it indicate improvements could be made?

The conclusion should be short, just simple statements about the data found in your analysis section.