Which statistical test should I use?

Hi. I just have a quick question about which statistical test I should use.

Say my friend and I go counting the number of bikes that pass by in a busy road each day (mon-fri) for a month (so 20 data entry) between the same hours in the morning, say 7am-8am

We both stand on the same street but approximately 200 metres apart. This 200 metres a part, has no no roads or path that will led to different paths. It is just two roads. However, my friend and I so happen to have continuously different counts (because there are so many bikes and we may miss some along the way).

So I am wondering, what kind of statistical test can I use to test whether the counts between my friend and I agree with each other?

I was thinking of using a two-sample paired t-test as we both have the same number of data entries and treating each day as like a different 'subject' and testing the before and after scenarios of the count.

Or should I be using a two-sample independent t-test?

Thank-you

Re: Which statistical test should I use?

If you both collected data at the same times on different days then it appears that a paired t-test is more appropriate. Make sure you compare number of bikes per hour for each day. Also check whose record are consistently lower than the other person...he is the one who causes bikers to disappear!

Re: Which statistical test should I use?

i was thinking, would a Wilcoxon test be better?