1. ## Craigslist Word Problem

I have no clue if I am posting this in the right place, forgive me!

If I post an ad on craigslist for 24 hours in New York City and I get an average of one response per minute from that ad, and the population of the metro area of NYC is about 18 million, is there a way to extrapolate about what percent of the population is showing an interest in my ad?

TIA!
~Marcie

2. Sure! But you need a whole lot more parameters than those you have suggested.
• How many COULD see the list or even own a computer?
• How many are in town?
• How many are too busy to care that day?
• I suspect you can think of many, many more.

3. Thanks - yes - I understand that, but given that the people who did respond to my ad were in town, and cared, and had a computer, can't I expand the sample based on this number? TIA!

4. You did not quite absorb my previous response.

If you are to extrapolate the responses to the whole of the population, you must gross up your estimate by some factor that recognizes some folks cannot participate. If you intend to extrapolate only to the population that could have participated, that will take different factors. It depends on where you are looking and how much homework you are willing to do.

Sample Population and Results:
10 were out of town and didn't see your content.
8 were busy and didn't look.
4 looked at your content, one only because he misspelled another URL.

What can we tell of the 34 when we've knowledge of only the 6 hits?

1) You must recognize the difference between hits and people. They are not the same.

2) You can make whatever assumptions you want. Get some population data and start to work. There are often government data on living conditions, ownership of various important devices (televisions, cell phones, etc.), travel habits, and on and on and on...

5. Originally Posted by Marcie0305
I have no clue if I am posting this in the right place, forgive me!

If I post an ad on craigslist for 24 hours in New York City and I get an average of one response per minute from that ad, and the population of the metro area of NYC is about 18 million, is there a way to extrapolate about what percent of the population is showing an interest in my ad?

TIA!
~Marcie
Originally Posted by TKHunny
Sure! But you need a whole lot more parameters than those you have suggested.
• How many COULD see the list or even own a computer?
• How many are in town?
• How many are too busy to care that day?
• I suspect you can think of many, many more.
Originally Posted by Marcie0305
Thanks - yes - I understand that, but given that the people who did respond to my ad were in town, and cared, and had a computer, can't I expand the sample based on this number? TIA!
Are you asking for the number of responses (which is the number of minutes in 24 hours) divided by the population 18 million? This is 24 * 60 / 18000000 = .008%.

6. Originally Posted by JakeD
Are you asking for the number of responses (which is the number of minutes in 24 hours) divided by the population 18 million? This is 24 * 60 / 18000000 = .008%.
Yes, basically, now let's say that no one would ever respond to my ad more than once, and different people would be looking at my ad each day - is it fair to say that in a year 2.92% of the population would respond to my ad?

TIA!

7. Originally Posted by JakeD
Are you asking for the number of responses (which is the number of minutes in 24 hours) divided by the population 18 million? This is 24 * 60 / 18000000 = .008%.
Originally Posted by Marcie0305
Yes, basically, now let's say that no one would ever respond to my ad more than once, and different people would be looking at my ad each day - is it fair to say that in a year 2.92% of the population would respond to my ad?

TIA!
That seems a huge leap to me. But I don't know anything about advertising or survey research, so I can't offer any advice on that.