Network News

X My Profile
View More Activity

About Those Iran Polls

Public opinion surveys are central to the Iranian opposition's argument that the elections there were rigged for incumbent President Mahmoud Ahmadinejad: they cite unspecified polls showing the main challenger, Mir Hossein Mousavi with a "strong lead in the final days of the campaign," according to the New York Times.

Now, a competing poll conducted by two American groups is being used as part of the pushback. In an op-ed in today's Washington Post, Ken Ballen and Patrick Doherty write-up the results of their telephone poll carried out in mid-May, showing Ahmadinejad ahead "by a more than 2 to 1 margin - greater than his actual apparent margin of victory in Friday's election."

The validity of the unreleased Iranian surveys cannot be assessed in detail, but a closer look at the one sponsored by Terror Free Tomorrow and the New America Foundation reveals ample reason to be skeptical of the conclusions drawn from it.

Methodologically, this survey passes muster as it's relatively straightforward to pull a good sample of the Iranian population, using the country's publicly available population counts and listed telephone exchanges. But the poll was conducted from May 11 to 20, well before the spike in support for Mousavi his supporters claim.

(See here for a summary of available Iran polls that finds some evidence for Mousavi momentum late in the campaign.)

More to the point, however, the poll that appears in today's op-ed shows a 2 to 1 lead in the thinnest sense: 34 percent of those polled said they'd vote for Ahmadinejad, 14 percent for Mousavi. That leaves 52 percent unaccounted for. In all, 27 percent expressed no opinion in the election, and another 15 percent refused to answer the question at all. Six Eight percent said they'd vote for none of the listed candidates; the rest for minor candidates.

One should be enormously wary of the current value of a poll taken so far before such a heated contest, particularly one where more than half of voters did not express an opinion.

By Jon Cohen  |  June 15, 2009; 9:08 AM ET
Categories:  Polls  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Virginia Governor: The Undecided Masses
Next: New Analysis Points to Fraud in Iran

Comments

"One should be enormously wary of the current value of a poll taken so far before such a heated contest, particularly one where more than half of voters did not express an opinion."

Nicely summarized. The main question about the election is how the counting was done so quickly. I understand Iran employs a rather unsophisticated paper balloting system that should by rights have taken several days to tabulate.

Posted by: Samson151 | June 15, 2009 10:53 AM | Report abuse

Over 24 million hand written ballots. One, thousand one, One thousand
two... 48 million seconds, at least... 360 seconds per hour ...a
million hours at less than ten seconds per ballot.. 6,0000 polling
stations? Ten thousand vote counters? Any witnesses? Monitors? The
results were announced two hours after the polls closed? Were ballots
counted as the ballot boxes filled and were emptied? Where are the
ballots? That who hasn't reported on any of this makes who wonder on
'realistic' whose revolutionary aspirations are. Productions and
Operations Management.

Edit Comment


volley2.ind 92: ?>*:\ ...//2009:06:12:02:08:75*W:70*F+/-?
#411 of 428: William Hale (hinging0) Mon 15 Jun 2009 (06:17 AM)

Note math error: 24 million times 10 seconds per ballot is 240 million
seconds only 2/3 rds the estimated 360 million seconds. Still 10,000
ballot counters is an estimate based on last knowing there were 3,000
US post office branch offices. So the idea that it would take 100 hours
to count the ballots might still be reasonable. No mention was made of
statistical sampling techniques to estimate the total during the
process of counting. Who did point out the difference in the style of
supporters and how the presence of 100,000 highly active internet
posters could make whom appear to be a much larger group than who
actually is. "Best out of five"; xref: "Love is the score in a game
where neither player has the advantage, "Point, set, match, game
tournament,"

Posted by: randomsample | June 15, 2009 11:08 AM | Report abuse

Let's see: "...telephone poll carried out in mid-May".

1. Using a telephone poll in a country that jails dissenters: If you lived in Iran, to what extent would you trust an anonymous person who called you on the phone to ask who you plan for vote for, for president?
2. Mid-May: That was eons ago. Mousavi came on strong during the 4 weeks preceding the election.

Ballen's and Dougherty's poll was defective.

Michael Herrinton
Oakland, California

Posted by: michaelherrinton | June 15, 2009 11:12 AM | Report abuse

Polling can work in a democracy. But as Michael Herrinton says, in a police state a phone call or a knock at the door is cause to be suspicious, not to offer an honest opinion. What would the results been if they would have polled in neighboring Iraq, say 9 years ago? 100% support for Saddam? Does that mean anything? Does this poll?

Yes, this poll does mean something. It means that the actual electoral vote was better than the greatly flawed poll showed it should have been. If given all the reasons not to do so, 1/3 say they are for the opposition, then why would the opposition only get 1/3 in the election?

Maybe the Mullahs need to get some training in Chicago or Florida on how to properly and realistically rig and election.

Posted by: cyberfool | June 15, 2009 11:34 AM | Report abuse

ABC has done a better job taking apart these numbers to reach a very different conclusion.

See "The Nunmbers" at http://blogs.abcnews.com/thenumbers/2009/06/irans-election-the-odds-of-fraud.html

Posted by: wizinit1 | June 15, 2009 12:15 PM | Report abuse

"Using a telephone poll in a country that jails dissenters: If you lived in Iran, to what extent would you trust an anonymous person who called you on the phone to ask who you plan for vote for, for president?"

It is clear to me from reading twitters that people in Tehran are talking to each other on the phone freely even now.

Besides, in the absence of a poll you approve of, you feel free to believe whatever you like? The only remotely reliable data anyone has mentioned so far says the result conforms to their poll.

This says to me that the government-claimed result is not impossible.

Posted by: geneven | June 15, 2009 12:35 PM | Report abuse

Irrespective of what you think of President Ahmadinezhad he has demonstrated a very sophisticated ability to communicate numbers and statistics, frankly better than most of our politicians.

That said, if in fact he won the votes, he would have jumped at the opportunity to make a detailed and sophisticated presentation of this accomplishment. Instead, they prevented international monitors from this election, they prevented the opposition monitors from attending most stations, had his own man counting the votes, and as advanced as Iran is in technology there is no way in hell they would have had the results so quickly.

This appears more like a well planned coup by Ahmadinezhad. After all why would Ayatollah Khamenei, the supreme leader, want to have a president with whom he clashed throughout Mousavi’s premiership? Now there are millions of young Iranians who are shivering at the thought of having been caught on the cameras of intelligence ministry shouting “Ahmadi bye bye”!

As much as we are told that the Leader Ayatollah Khamenei is the most powerful person in Iran, I bet he also lost some of his power to President Ahmadinezhad in this coup. It will be much harder for the Leader to fire a fundamentalist president who “won” 62% of the vote of “the oppressed”!

The impact of this election on the foundation of Islamic Republic is profound. If the votes were real Ahmadinezhad has succeeded in Thailandization of Iran, if the votes were fraud, he has planted the seed of an entirely new generation of opposition leaders who would have no loyalty to the principles of the Islamic Republic and at the next opportunity will go for the broke, and that may not be through participating in elections.

Posted by: MichaelAkbar | June 15, 2009 12:47 PM | Report abuse

Jon Cohen wrote:
More to the point, however, the poll that appears in today's op-ed shows a 2 to 1 lead in the thinnest sense: 34 percent of those polled said they'd vote for Ahmadinejad, 14 percent for Mousavi.

Yes, but by the same token, a candidate polling less than 20% with a month to go before the election is somewhat unlikely to pull off a victory.

While the fact that there were irregularities in the count cannot be ruled out as a mathematical impossibility, the results on the surface actually seem quite plausible.

Posted by: sharatch | June 15, 2009 12:49 PM | Report abuse

The tools may be there to pull sample in theory, but can it truly be representative. Surveying is much more than just sample. That is only the starting point. I have lived and worked in the the Middle East and many other countries. Accurate surveys are hard to accomplish. I posted the following in response to the original article:
A 3.1 confidence level with a thousand surveyed? Hmmm. That's better than a great many American surveys. A telephone survey from a neighboring country? By native Farsi, Azeri, Uzbec speakers? No discernable Turkish, Syrian or Iraqi accent? Who supervised the surveyors? Who monitored the quality of the calls? Did they speak the language(s)? Was the sample truly nationwide representing all ethnic groups, rural/urban, age groups? Were people who had cell phones contacted (prevalent where land lines are more difficult to acquire in some areas)? Especially prevalent among 18-24 year olds. In a country that is a quasi police state do you really think a call from a stranger (perhaps with an accent) is going to elicit an anti Ahmadinajad response from many respondents? Talking about democracy or freedom generally is not so dangerous. Asking whether the current leader should be gotten rid of is quite another matter. I for one would be very careful of drawing sweeping conclusions from your survey. Ahmadinajad may have had more support than many thought, but it is definitely open to question as to whether he had over sixty percent. Especially when these numbers were given out by the Ministry of the Interior before the votes could have been counted. A ministry that locked out many of its own employees during the counting process as reported by many western reporters who interviewed MOI employees.

Posted by: tarryh | June 15, 2009 12:57 PM | Report abuse

If Ahmadinezhad had won he would have eagerly offered an audit of the results to bury his opponents for ever.

The chatter on your polls is a nice intellectual exercise that does not take into accoutn the realities in the field. The bitter Iranian voters have a less polite term for such exercise.

Posted by: MichaelAkbar | June 15, 2009 1:13 PM | Report abuse

I attended college with several Iranian students. They were nice people who only wanted to get rid of the Shah. Unfortunately, they and their children have now been screwed by their current political and religious leaders. I hope they can rid themselves of Ahmadinejad and the crooked mullahs before they become slaves of the state.

Posted by: original_maddog | June 15, 2009 1:28 PM | Report abuse

What editor left that out of the article?!

That editor should be fired immediately.

Posted by: bug45 | June 15, 2009 2:11 PM | Report abuse

This man should be FIRED for attempting to DISTORT the facts to the readers.
He has no shame and now? He has lessened the readership of the Washington Post.
STOP LYING TO US

Posted by: Grissom1001 | June 15, 2009 2:22 PM | Report abuse

Posted by: MichaelAkbar
If Ahmadinezhad had won he would have eagerly offered an audit of the results to bury his opponents for ever.

The chatter on your polls is a nice intellectual exercise that does not take into accoutn the realities in the field. The bitter Iranian voters have a less polite term for such exercise.


RESPOND
Actually of the guys in turban (some ayatolah), is ordering investigation. But, the question you should ask is; WHO WILL DO THAT INVESTIGATION / AUDIT? Is it UN or same people that could have rigged this election.

I think we all know who will do that!!!!

Posted by: BOBSTERII | June 15, 2009 3:57 PM | Report abuse

Any country that has religious prefix in the main name, should tell us everything about that country's real ambitions. In this case its "Islamic" Republic of Iran. If they can't represent other religions like the rest of the world countries, we should make a mess to them. There are a lot of ways to play dirty games and it looks like Iranian religious turban guys and Ahmedinejad, want that exactly.

Posted by: BOBSTERII | June 15, 2009 4:09 PM | Report abuse

"One should be enormously wary of the current value of a poll taken so far before such a heated contest, particularly one where more than half of voters did not express an opinion."

Good point, nicely sidestepped by Sampson.

"Nicely summarized. The main question about the election is how the counting was done so quickly. I understand Iran employs a rather unsophisticated paper balloting system that should by rights have taken several days to tabulate.

Posted by: Samson151 | June 15, 2009 10:53 AM"

The main question is how the counting was done so quickly.

When the Fascists all over the World, including the American Right Wing, wish to steal an election, they drag out the count until the opposition anger fizzles out.

Monte Haun mchaun@hotmail.com

Posted by: mchaun | June 15, 2009 6:13 PM | Report abuse

Everyone knows that conducting poll in Iran is not possible for many reasons e.g. people’s fear of the regime. Iran is not free like the U.S in which you can express your ideas. Look at today's rally in Tehran. Look at the results. For instance Ahmadinejad got up to 15 times more vote than his rivals in their own cities????? The night of the election, they declared the rival the winner and then they changed their word. They cooked up the numbers. How can you believe Ahmadinejad? Let me refresh your mind about the honesty in his claims: there is no gay in Iran, Iran is a free country, Holocaust didn’t exist, Islamic regym has the solution for the problems of the world and ....
Go to twitter and see what they are doing to people these days. A confident winner does not kill people. Please do not close your eyes to the violation of human right that is happening right now in Iran.

Posted by: simontavasoli | June 15, 2009 7:41 PM | Report abuse

I like how some people are defending Ballen and Doherty's indefensible op-ed. Who cares whether Iranians are free to talk on the phone or worry about the secret police. Ballen and Doherty lied about the numbers, plain and simple, by not giving them. When you say that one candidate beat another in a poll by a 2 to 1 margin, people assume that you're saying 60% for one, 30% for the other and some undecideds. When you reach the limits of a useful poll -- like when half don't respond but 70% go to the polls, meaning that you had no prediction data on 1/5 of the population -- you need to convey that to readers.

Unless your evidence doesn't support your point. Then you don't tell them.

Posted by: simpsonsteve75 | June 15, 2009 8:51 PM | Report abuse

Much is made of a poll taken a month before the election but this ignores the fact that most of the world has campaigns of 4-6 weeks, e.g., UK and Canada and not 2 years like the US. I also agree that it was dishonest to say Ahmadinezhad had a 2 to 1 advantage without making it clear that they are talking 15% & 30% and not 30% & 60%.

Posted by: joesewell | June 15, 2009 10:39 PM | Report abuse

Sorry, I have to be blunt.
First care for your own s***t then mingle in the affairs of other countries.

Posted by: mixedbreed | June 16, 2009 1:05 AM | Report abuse

Jesus. Wow. I didn't believe the original article could possibly be as bad as it is. I agree that someone needs to be fired here.

Posted by: Drew_Miller_Hates_IDs_That_Dont_Allow_Spaces | June 16, 2009 10:58 AM | Report abuse

Iran is not a socialist country, they abhor capitalism, so a democratic process doesn't stand a chance there, what they need is for a neutral country like N. Korea, because of its ties to Islam, to show them how to set up an election that will be fare for the people but also benefit the state.

Posted by: edtroyhampton | June 16, 2009 4:06 PM | Report abuse

Posted by: edtroyhampton | June 16, 2009 4:22 PM | Report abuse

"While the fact that there were irregularities in the count cannot be ruled out as a mathematical impossibility, the results on the surface actually seem quite plausible. Posted by: sharatch"

No, on the surface the "results" of the poll show everything but. That 34% who definitely would vote for Ahmadinejad is comprised of the sum of Ahmadinejad voters and non Ahmadinejad voters who figured it was better to lie than to die. Any reasonable pollster getting ore rejections than answers would call off the poll and try to figure what he was doing wrong.

Phone polling is only accurate given to circumstances not available in Iran.

First, you need a way to insure that your sample actually looks like your population. from the sixties to the early eighties landline phones in the U. S. could actually do that, but even tn the U. S. that isn't true, thanks to cell phones. In Iran the problem would be throw away cell phones, buy them with minimum minutes and don't reload. That makes it harder for whatever replacement for the Shah's secret police the Mullahs rely on to get a handle on you. Since phones aren't distributed in proportion to the population, your sample is degraded and your results are meaningless.

Then you must find a way to statistically distribute the non responders. When you have more non responders than responders, THERE IS NO STATISTICALLY ACCEPTABLE WAY TO DO THAT! Margin of error? Plus or minus 200 %.

The idiot that wrote the article ended with3% margin of error? He must think that is some magical phrase like Amen, attached to all polls to tell the reader that the poll is over.

Posted by: ceflynline | June 16, 2009 9:37 PM | Report abuse

This article sounds like something written by people who've never been to a Potemkin village, never talked to secret police, and never been afraid to speak the truth to a pollster who might be a policeman seeking information to use against them later.

The fact is that if the Iranian Government were speaking the truth they'd have jumped at the chance to do a recount. They would have taken 3 days to count the ballots, not decided the winner before the election was counted, and not started arresting people who protested the irregularities. There is no need to do such things if one has the evidence and is not lying.

These people claim to be such geniuses and "realpolitik" savvy folks. Geniuses.

Hats off to the Iranians for being willing to risk their lives to stand up for what they believe in. We'll see if the Ayatollahs are genuine holy men or are the corrupt pretenders, autarchs, and tyrants, they are seen as by most people outside the country.

Chris

Posted by: chris_holte | June 16, 2009 9:52 PM | Report abuse

Ballen and Doherty's poll is evidence for what may be an "inconvenient truth." Amhadinejad may have won the election fair and square. The poll isn't conclusive - no polls are - but it was independent and objections that it was taken 3 weeks before the election don't hold water. Most people would have made up their minds by then.

Posted by: cloudtrapezer | June 16, 2009 9:54 PM | Report abuse

And by the way Iran is capitalist. Most Arabs were capitalist when Europeans were ruled by Monarchs and nobles who disdained "Money lending" and business.

The Persians were civilized before or at the same time as the Greeks, and have a civilization that rivaled Romes, Greeces, and has been a source of science and learning; both before the arrival of Islam and after. Before they converted to Islam they believed in Zoroastorianism and invented many of the ideas of Monotheism. They practiced religious tolerance when Europe was burning heretics at the Stake. And they contributed much of what we consider 'arab' to Arab culture.

We should respect Persia and its history. Persians are a lot like us.

Posted by: chris_holte | June 16, 2009 9:58 PM | Report abuse

In 2005, when Ahmadinejad beat the very powerful (not many folks are wear the religious robe, are filty rich, and have a power family) Rafsanjani, he got almost 62% of the vote.

If nothing else, Ahmadinejad's approval ratings have increased over the last 4 years, thanks to his accomplishments, and hte misrepresentations and threats by some in the formerly colonial countries.

Intuitively, I do not have a problem with the 62.5% number this time, against Mousavi, who is a close ally of Rafsanjani...

Posted by: beyoundbull | June 17, 2009 7:03 AM | Report abuse

I'm speaking from inside of Iran,
i approve result of suvey.
3 months before election i had many work trips to most of Iran's provinces, more that 60 percents of people were supporter of Ahmadinejad.
now most of iranian are sure, ahmadinejad won correctly because they voted him!, but mir hosein mosavi and his supporter don't want to accept this hard defeat.

Posted by: member5 | June 17, 2009 8:09 AM | Report abuse

The way this poll was conducted is highly flawed and it's data whatever it maybe can not be trusted. People in Iran would be extremely fearful to tell an anonymous caller who has called their home something too aggressive about a sitting president.

1. They don't know who they are talking to
2. They all know phones are often tapped and calls are listened to by the government
3. It is very easy to tell the call is from overseas ( if you have called to or from Iran once you know what I mean) which even makes the matter worst ( the last thing you want is to be be accused of taking to a foreign agent and telling them something bad about the government or the sitting president).
4.Phone polls are not common in Iran, there are no phone polls period, so the whole thing just sound very suspicious.

After all of that %66 did not say they would vote for Ahmadinejad. That folks says a lot

Posted by: pummy64 | June 17, 2009 4:23 PM | Report abuse

member5, cut out the B.S. you are NOT in Iran. Washington Post website is blocked in Iran. How is the weather in Israel today ? Why are you guys so interested to claim Ahmadinejad has won ? The head of Mossad Dagan, also today said it is better for Israel if Ahamdinejad wins. Why ? WTF...

Posted by: pummy64 | June 17, 2009 4:28 PM | Report abuse

beyondbull,
You don't really know what you are talking about. Where are you ? how much are you getting paid to hang around these blogs. How long is the cheat sheet they gave you ? That is some highly inaccurate yet specific information you have for an "average joe", here in the U.S. but again, you are not here are you ?

Posted by: pummy64 | June 17, 2009 4:33 PM | Report abuse

Jon, please take a look at this. I think you might find it interesting...

http://tehranbureau.com/2009/06/17/poll-indicating-legitimacy-of-ahmadinejads-victory-called-into-question/

Posted by: pummy64 | June 17, 2009 11:22 PM | Report abuse

As a non voter,by choice,I believe some of these comments are make believe rather than facts.one says it must takes so many days to read 39 M. votes!!. But he seems doesn't know that each box contains no more than 1000 votes. So reading the name of each should not take more than 10 second at most.So why not more than 2 hours rather so many days.Indeed this was reason of complain in the former elections that why waiting for the final announcements so long. As a non voter that I said before and stayed in home,I was not in direct contact,but I have first hand account of two villages around one hundred km. from Tehran.I tell you again;villages. One estimated that they have voted 350,400 for Ahmadinejad and 50(youngs that visiting Tehran very often!) for Mossavi. Final results later; 400 against 50. In second village near the first 600 against 100. The reason for this was explained that most pensioners got a substantial amount very recently and a new pipe line was laid for irrigation. I listen to guys from southern section of Tehran: HE is against Rafsanjani's family. please remember that in last election we made Rafsanjani' corruption a very big issue. looks that so called reformers have forgotten this for convenience;but not these people. Bottom line; Mossavi jumped in this trap, by not disassociate himself from him,rather say in debate that "do not talk about him since he is not present"! I think poor people did not buy it. And neither I ,non voter, now:LOL From Tehran

Posted by: Shabahang | June 18, 2009 7:25 AM | Report abuse

The mullahs may have long feared that change would eventually come in reaction to their abuse of the population. Many have moved the proceeds of their pilfering offshore, “just in case.” Some have built themselves Los Angeles and West Vancouver mansions, in anticipation that the gun might eventually not suppress the crowds in Tehran.

The potential for change is directly conditional on the persistence and endurance of the youth filling the streets of Iran. It will be unstoppable if the demonstrations move to the poorer rural regions of the country.

http://pacificgatepost.blogspot.com/2009/06/new-dawn-for-iran.html

This genie is out of the bottle. Change may be slow in coming, nevertheless, it will come.

Posted by: JamesRaider | June 18, 2009 9:40 PM | Report abuse

History shows how what is writ may be turned to whatever purpose is required.

Anyone heard of bias. To believe a phone poll, scientific or otherwise, in such a Society as Iran is valid is bordering on the ridiculous.

The science of probability is one thing.

The science of protecting ones skin is quite another.

Given:
1. the ferocity of the reaction against dissent in Iran exemplified by the deaths of demonstrators- fellow human beings,
2. the systematic locking up of opposition supporters and,
3. the intimidation and violent tactics of the Iranian Basij volunteer militia controlled by Ayatollah Khamenei’s 200,000 strong revolutionary guards.

frankly I would be saying I am for the man who is backed by the man with control of the black motorbike, helmet and club. Even if I truly believed otherwise.

Who knows in such a society who is really asking the question on the other end of the line?

2 to 1 - I am surprised there was a 1. Very brave fellows indeed.

Posted by: markjuliansmith | June 19, 2009 8:55 AM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company