1. When an operator says "all circuits are busy" - this counts as a vote.

2. If someone picks up the phone while you are dialing and the call does not complete - this counts as a vote.

3. DialIdol does not take any text message votes into account.

• It is not reasonable to expect DialIdol to be 100% accurate when two Idols are separated by only a few percentage points. The primary reason DialIdol declares a margin of error is to "levelset" that the results are actually too close for DialIdol to have any real confidence that the "measure of the busy signal" is even right. If the measure of the busy signal isn't even right - then there is no way this measure can reflect who gets voted off.
• All polls/surveys have margins of error and DialIdol is no different. A margin of error loosely reflects the confidence of the accuracy of the data. At DialIdol it means that if the percentage points separating two or more Idols is within the margin of error then the ranking order might not be accurate for the contestants in that margin.
• Just because DialIdol declares a margin of error does not mean that DialIdol guarantees' the accuracy of the results. In fact DialIdol will never guarantee the accuracy of the results and if you'd like to know the results - watch the results show.
• Prior to week 10 of season 5 the MOE was just decided on. Starting with week 10 of season 5 MOE follows this calculation which is the actual calculation for MOE = 2.58 * sqr(((p)*(1-p))/n) where p = each Idols DialIdol score (percent) and n = the actual score for each Idol. 2.58 is used for polls with a 99% confidence which is what DI is using.

• No, no way, no how and because DialIdol is not scientific it's results are for entertainment purposes only!
• DialIdol's predictions are based on measures of the busy signal, not actual votes. There are known problems with measuring the busy signal.
• If you want to know who is going to get voted off the only way to know for sure is by watching the results show.
• Scientific polls go so much further beyond what DialIdol is capable of it's not even funny. Scientific polls would measure text message votes. Scientific polls would consider some/all of the following different things: geography, age, sex, weight, population, race, financial position, etc. A scientific poll would only include specific results from a small pool of people that represented the larger pool of people voting - DialIdol includes everything.

DialIdol is graded based on these questions.

1. For each contestant that was voted off - did DialIdol predict them to be SAFE?
2. Did DialIdol absolutely predict that someone would be voted off that didn't get voted off?
   • Absolutely predicting someone will be voted off means their name and bar graph shows up in red.
   • This does not happen frequently.
3. Did DialIdol absolutely predict someone would not be in the bottom three that was absolutely in the bottom three?

Any time an answer to either question is YES - DialIdol got it wrong, anytime the answer is NO - DialIdol got it right. These questions don't usually apply to every contestant per week - instead they usually only apply to a few contestants.

DialIdol does not grade itself by asking For each contestant who was safe, did DialIdol predict them to be safe? because this would result in a disproportionately high grade.

An example... During the week with 24 contestants left on American Idol Season Six, four contestants were voted off (Nicole, Amy, Rudy and Paul). DialIdol did NOT predict that any of them would be SAFE - so all four of those predictions were correct. That same week DialIdol made a definitive prediction that Alaina Alexander would be voted off however she was NOT voted off so this prediction was incorrect. In total this week DialIdol got 4/5 right - or 80%.

I've considered including many other factors in the grading system however they would either lead to unfairly high scores that anyone with a dart board could achieve, were unfairly critical of DialIdol or barely ever produced enough data to be graded on.

This is DialIdols grading system - you may judge DialIdols accuracy however you want. While doing so - please consider that these predictions are based on statistical data. Under these circumstances it is generally customary to allow some margin of error. In fact there are actual mathematical calculations that mathematicians have come up with to calculate margin of error and DialIdol uses those to calculate margin of error.