The payoff: $650. The odds: 1 in 500,000.

(by Andrew Gelman)

I just received the following (unsolicited) email:

Dear colleague;

Please take two minutes to tell us how you work with public relations professionals and how your practice journalism in today’s social media environment and we’ll enter you to win a new iPad 2™. The lucky winner will be selected on Wednesday, March 23.

We’re reaching out to more than 500,000 working journalists around the world to learn more about their relationships with public relations professionals. At ***.com, we are here to connect you, the members of our working journalist database with the press releases and resources that you need to write your stories.

Please go to the brief 15 – question survey now . . .

This survey is your forum to tell us what’s working for you. Tell us what needs to be changed. We want to hear from you. To say “thank you” for participating, we are going to place your email address into a drawing for the new iPad 2™.

Ummm . . . 500,000 journalists and one i-Pad 2 (I refuse to type TM here), which, according to ebay, seems to be selling for about $650. That would be an expected value of $650/500000 = 0.13 cents. Not 13 cents. 13/100 of a cent. If the survey takes 15 minutes, that comes to . . . a half cent an hour! I don’t even know if Mechanical Turk people will work for a half cent an hour.

But, hey, maybe the response rate is only 1 in a 1000, in which case it’s worth $1.30, which comes to a respectable $5/hour (assuming you want an I-Pad 2, that is). On the other hand, what could be the possible use of a survey with a 0.1% response rate???

Either the prize drawing is a ripoff or the survey is a joke. Or both.

If I didn’t know better, I’d think that these people aren’t interested in our opinions at all . . .

P.S. This is relevant to the Statistics Forum because it is an example of how the statistical idea of survey sampling, used so effectively by George Gallup and so many others, can also be used as a pseudo-scientific cover story for just about anything.

It’s also a fine example for your teaching, to demonstrate the idea of an expected value.

Advertisements

2 Responses to “The payoff: $650. The odds: 1 in 500,000.”


  1. 1 DavidC April 23, 2011 at 3:53 pm

    Not that it matters, but that’s all if the thing is even worth $650 to you. If you haven’t bought one already, it’s probably not.

  2. 2 Stas Kolenikov April 24, 2011 at 3:24 pm

    It’s about the same rate per hour that you’d be paid for writing a 1000-word encyclopedia entry for a honorarium of $300, as I have seen recently in a call for submissions to some sort of “mathematics-for-high-school-students” encyclopedia. “Mathematicians of Oceania” were given the same 500 words as “Mathematicians of Eastern Europe” in the History of Math section. I can write 1000 words on a topic of my professional qualification in one day (I thought of volunteering for “Rock climbing” which I am moderately good at), but you don’t want to publish it in an encyclopedia as is. It is at least a week’s work, so it boils down to about $8/hr assuming 40 hr week (a European one, not an American one).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




About

The Statistics Forum, brought to you by the American Statistical Association and CHANCE magazine, provides everyone the opportunity to participate in discussions about probability and statistics and their role in important and interesting topics.

The views expressed here are those of the individual authors and not necessarily those of the ASA, its officers, or its staff. The Statistics Forum is edited by Andrew Gelman.

A Magazine for People Interested in the Analysis of Data

%d bloggers like this: