Hi, sorry but could you explain me why is it necessary to link purchase date and review date? The way I see it those are different data, a customer usually buys first and make a review later so, in fact almost always these dates shouldn´’t match.
Regarding the exercise particularly, I checked in excel spreadsheet and none of dicember 2017 purchase have a review so the first results would be ok.
Thanks
Hi Anna,
We will be unable to use the data about purchases if we use the Reviews file only.
@Lucia: Yes, the join isn’t a good fix in this case, this is why we use blending later on.
Great to see that both of you engaged with the course 🙂
Hi Ned,
Thanks for the answer.
I agree that we won’t be seeing data about purchases if we only use the Reviews table. I should’ve specified that I referred to the “Performing a Numbers Check – Attempt #2” exercise of the course.
The results that we obtain at the end are exactly the same as if we used just Review Date and Rating fields from only Review table. Which is understandable considering that we blended the data using Review and Purchase Date, but that’s how we get to the main question – why should we blend the data using those fields?
I share Lucia’s opinion that we shouldn’t be linking our tables with those fields and that for December 2017 zero (or null) is the correct result as none of the books bought during this period received any rating. The results obtained at the end of the exercise definitely don’t answer the question of how many ratings the books purchased in a specific month received.
I agree, it seems completely wrong. What we’re actually getting at the end of the exercise is the number of ratings per month according to Review Date (not purchase date) and the same result can be obtained with using only the Audiobook Review table fields.