I am calculating the solubility of benzoic acid at different temperature from a set of experimental data.

However,the relationship of solubilty and benzoic acid i found was inversely proportional.

I try to change the data of final Sodium Hydroxide used a bit and I found that the wrong trend mostly due to titration error.

Could you convince me how to deal with the titration error?i.e. how to calculate the possible percentage error?..

This is the 1st time I encounter this kind of problem.I just wonder if the error can be determined by analytical method.Or I just need to explain some experimental errors and convince what change could make (as shown in the pdf file) to obtain the correct trend in the discussion part?

Heap of thanks.