# OT no such thing as a statistical dead heat

• posted on April 2, 2014, 9:08 am
OT Unless the percentages match, no such thing as a statistical dead heat
http://iase-web.org/documents/papers/isi53/292.pdf
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 2, 2014, 10:21 am
but non-statistical live colds are ok I'm sure
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 2, 2014, 12:07 pm

Absolutely
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 2, 2014, 2:59 pm
wrote:

The more I think about this, the more things become clear. When the say the difference is within the margin of error, is there a 95% chance the guy who is ahead, is ahead? No. But I just realized that the chance never goes below 50+%, because if it were lower than 50, the odds would be that he is behind. And if his poll numbers exceed the other guy, the odds are he's ahead, not behind.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 3, 2014, 1:48 pm
On Wednesday, April 2, 2014 10:59:39 AM UTC-4, micky wrote:

Consider a poll on candidates Smith versus Jones. We poll one person who pi cks Smith. Our poll indicates Smith with 100% of the vote, but we wouldn't put much confidence in the result of that poll.
So we poll another person who picks Jones. Now, with a sample size of 2, we have 50% Smith and 50% Jones but how reliable is a poll of two people when there are thousands of people who will be voting?
We go out ringing doorbells again and end up with a hundred responses, 52 f or Smith and 48 for Jones, a 4% difference between the two. Based on the sa mple size and other parameters*, we might find that the results of this pol l give us a 99% confidence that the results of the sample are within 5% of the results had we polled the entire electorate.
You could think of it as meaning that Smith should end up with somewhere be tween 57% (52% + 5%) and 47% (52% - 5%)of the vote and Jones should end up with somewhere between 53% (48% + 5%) and 43% (48% - 5%)of the vote. Since we're 99% confident that either candidate could end up with over 50% and wi n, it's a "statistical dead heat."
* Here's a good reference on all that: http://en.wikipedia.org/wiki/Sample_ size_determination
Paul
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 3, 2014, 1:26 pm
On Wednesday, April 2, 2014 5:08:26 AM UTC-4, micky wrote:

I do a lot of statistical analysis at work. Crunching the numbers is the ea sy part; explaining the results in terms that non-statisticians will unders tand is difficult. I agree that "statistical dead heat" is a misleading phr ase but it gets the point across to people who don't understand standard de viations, confidence intervals and sampling techniques.
Paul
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 3, 2014, 1:43 pm
On 4/2/2014 9:59 AM, micky wrote: ...

Let's see if I can muddy it up, then... :)
The probability of the obverse being the true situation is a Type II error...
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 3, 2014, 7:54 pm
wrote:

Okay, we agree on that.

No it doesn't. It's misleading, like you said.

If you want to get the point across without using fancy words, say "Smith is probably ahead now, but because we didn't poll every person, Jones might actually be ahead."
Using the example from your other post, you could say " There's an 80%** chance that Smith is ahead now and a 20% chance that he's behind.
How's that?
**Whatever the number really is. And maybe I should have included a 1% chance they were tied. I don't know what the chances are for that.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on April 4, 2014, 12:29 am
On 4/3/14 8:43 AM, dpb wrote:

And you did it without using any ten dollar words.