 Ladies and gentlemen, let me remind you that the details will be up on the web and we will provide both the trends in the Force and Gary's presentation. I would also ask that if you have a brilliant technical question on the subject of polling, you email Gary because the answers are almost as long as the questions and it is very difficult. In general if I could ask that each person ask one question with one part since I have noticed that people have gotten around the one question with nine part questions. Please look around you, we don't have time for interesting speeches but the steps outside are available for those of you who wish to give one. And finally if I may as you ask a question would you please identify yourself and your organization with those limited caveats who would like to start please in the front row. Wait for the microphone please if you don't mind. Mike Delaney, officer of the U.S. Trade Representative. It has been my experience in southern Afghanistan that there is a fairly strong positive correlation between tribal or sub-tribal affiliation and support for the Taliban. Have you tried to inject a tribal or sub-tribal dimension into your polling? We have to some extent although I haven't gotten a lot out of it and it may be simply a function of sample size. One thing of course is you've got, there's not a lot of variation in terms of ethnic identity. There's sub-tribal differences but there are not a lot of Hazaras in helmet for example right. We ask ethnicity and then we did ask a tribal element, Durrani etc. But our sample sizes aren't really adequate for provincial level analysis of that data much less, well not provincial, not even regional and nationally we don't see substantial differences. In fact for instance in Durrani versus other Pashtun for example. But it's a good question and by the way I really value this part of the presentation because I can take some of these suggestions, run our data, see if indeed we do see something and be happy to share that with you if it shows up. The gentleman in the second row over there. I'm Malik Nugent from the Treasury Department. You mentioned that 80% of the Afghan population is rural and one of the things that we're looking at at Treasury is expanding the reach of the formal financial sector out to those rural populations. And one question that we have though is to what extent do rural populations look to formal institutions based in Kabul, based in the cities for services, for public services, to what extent do rural populations look to the formal financial sector or would they look to the formal financial sector as opposed to traditional lending arrangements and then extending that not just to financial systems but to other public services as well. Yeah, I would say you should do a poll on that. These are specific topics to which you could get excellent data. All I would be able to give you are suggestive data but certainly the changes that we see have to occur in rural areas. There's not enough people in the urban areas to drive these differences. So the improvements in ratings of local conditions, the index of rebuilding that I discussed, the greater sense of some jobs and economic opportunity clearly do exist. The improvements have occurred in rural areas at least as much if not more than in urban areas. But in terms of specifics on the interest and reliance and microfinance or other sorts of loans, you'd have to go in and do that. But the guys from D3 are over there, you can get together. Let me just comment that there is a problem in defining rural areas. Some are near roads, some are relatively dense because they're in river valleys. Others are highly dispersed. One answer to your question that would come I think from talking to Canadian, British and other aid workers and the PRTs is that anything that's more than two hours walk, which is often the practical test, isn't accessible. But the question is who is two hours walk from what? And mapping that is not something where it's easy to generalize by going urban to rural. Let's see, the second row, the gentleman there. Chris, the Angel side also from the Office of US Trade Representative. I'll be going at the Pakistan in the summer. So I'm wondering, one of the key issues we've heard is problems with justice and the judiciary functioning effectively, particularly in the countryside. That's one reason it seems that these people turn to the Taliban because they can get something done even if it's some rough justice but they can get something done. Do you have any data on how people perceive the justice system working and the government's role in it? Sadly I don't. And the reason is we made a specific decision not to go there because it's a complex enough area of questioning that you can't do it in a question. You would need a battery to really pick apart the concerns and the solutions that are in place and the preferences for alternatives. And this was about a half hour questionnaire which is about as long as we can comfortably go without getting responded fatigue and others have done longer and I start worrying about data quality. The Asia Foundation has done survey research in Afghanistan that tends to be less focused than ours on political issues and on broader public attitudes and does drill down to some extent into social preferences and concerns. And that may be a place to look if you check the website. There's another way to look at this. There have been surveys done within the embassy and within the various aid teams. Basically you have at this point something approaching a court presence. That doesn't necessarily mean it's effective but approaching something like a judicial system that affects about 9% of the population. So just in a very crude sense, is there anybody there doing anything? Those are mapped and the answer is it affects maybe 10% of the population on a functional level. Let's see. There was a young lady in the third row there. Anika Arabidzanska on the part international. What was the percentage of women respondents in your sample and also how were you able to reach women? Okay, so the sample is about 50-50 men and women. We send out separate teams of male and female interviewers to conduct interviews of male respondents and of female respondents. But the ultimate ratio is 50. The women interviewer teams I should say are escorted of course. But only women interview women and only men interview men, I believe. And indeed we have separate sampling points for men and women in order to prevent rivalry. But yeah, it does reflect. The survey across the board reflects as best we can the population values in the country. One of the challenges is the population values in the country are a little sketchy, right? There hasn't been a census done in Afghanistan since 1979, I believe. But we do have a CSO, Central Statistics Office estimates of the population and the distribution of the population. And those include the roughly 50-50 division of men and women which we then reflect in our sampling. The gentleman in the third row in the center. Incidentally, I have a blind spot on the left, so please raise your hand because I can't see you otherwise. Marvin Wyman of the Middle East Institute. Probably one of the most surprising findings that you have is the popularity of Karzai, which is obvious. I think you reflected that in your remarks, particularly because of our concern about the legitimacy of the central government. How important that is in terms of support for our counterinsurgency. Trying to understand that further, is it possible that many of the respondents saw that as do you prefer Karzai to the alternative? So that it was in a sense there a choice there Karzai or someone else. And also, is there some explanatory value in the fact that it is well noted that in the move up to the election, Karzai deliberately in good fashion, political fashion, went and spent a lot of money in the development area. Clearly, people are, especially with electricity, are experiencing a better outcome because of that. And it has a strong effect. And I think I mentioned that, or tried to mention that, that the effect of campaign promises can have a positive impact on... Delivery. Delivery as well, on expectations and on assessments as well. We asked about Karzai both directly and elliptically, if you will, in a variety of ways. And the data are pretty consistent. We ask about his personal favorability. We ask, as we have consistently now for five years, about ratings of the performance of the government he leads. And we ask about, not about Karzai, simply are you satisfied or dissatisfied with the outcome of the election? A question that we didn't want to tie to Karzai, again, to see if we get any difference. And we get consistent readings across. I would again, there is a halo effect of winning, of emerging victorious from a period of conflict, even if an election conflict, if you will, that we've seen in previous cultures including our own strange culture. So that certainly is not unheard of by any means. Separately, this is a society that seems to, I don't want to go too far beyond our data, that I think can be said to value a strong and assertive leadership. And the emergence of a strong and assertive leader after a period of uncertainty, again, can indeed inform positive ratings of that leader. The question is, do they stay positive? President Obama has gone from 69% approval to around 50%. It depends on performance. And that's why it's necessary to continue to take these measurements and assess in an ongoing way. The gentleman on the background. Phil Mastretta, Office of the Special IG for Iraq. Are you still polling in Iraq? And how much comparative analysis do you do between Iraq and Afghanistan? I noticed one or two slides on this point. But it's become quite a cottage industry here inside the bell way. Yeah, that's true. My last poll in Iraq was about a year ago. And I'm always looking for another opportunity in the right time. We've in the past polled around the anniversary of the invasion, but very frankly, the elections were in the way this time. I didn't want to poll in the midst of the national elections there, which I would have needed to do to get results in time for the anniversary of the invasion. So we let that go. We do have continued interest in conducting surveys in Iraq, but it's obviously of less pressing, if I may say, news value, which I have to keep in mind given my employer. But I think it's important. The value we can provide in doing these independent surveys as news organizations is by opening a window that many of the decision makers in our government, frankly, already have opened and are looking out. There is a lot of data available in Iraq and Afghanistan, but not a lot publicly available. And I think by doing this, we can get a better public understanding by disseminating these data of the decisions that are being made and the information that's informing them. And that's frankly why we do it, why I think it's an important project, and I'd like to continue it. I've not done a lot of comparison, some general stuff, but we've seen similar trends to some extent in Iraq and Afghanistan. In Iraq, as you know, some reasonably high expectations. And no one was, other than Sunnis, no one was glad to see. It was sad to see Saddam gone. But then that spiral of despair in that country is it fell into sectarian violence after the Samarra bombing and then a recovery, a dramatic recovery in our last poll, which I suspect is continued because violence is lower and there's a greater sense that the country is going to stay together. But there are very fundamental differences. The attitude driver in Iraq in many ways was the sectarian attitudes, the Kurds and the Sunnis and the Shiites. And the question is whether they could find some accommodation. That's much less the case. There's much more uniformity across groups in our data in Afghanistan and it's much more about the level of violence, the development, and the other measures I've discussed. I should note that both the State Department in dealing with Iraq and the State Department in dealing with Afghanistan is zero-basing its models of how to measure progress in both countries with new polling elements as well as fixed indicators, which is kind of interesting after nine years. It is truly zero-basing. It is also doing so without any coordination with the rest of the United States government or between the Iraqi and Afghan efforts to see if there are areas of common effort. So let me note that your question is a very good one, but I would be much happier if the United States government showed even the slightest interest in coordinating within itself. The lady in the back there. Read our latest white paper. I have unfortunately read the latest papers from State in defense as of this week. They have evidently not read your white paper. They have, but they don't agree with something. All right. I'd love for you to send it to me. The lady in the back there. Gail Moragato from SCRS, Coordinator for Reconstruction and Stabilization within the Department of State. My question is you spent a lot on public perception on military forces and the U.S. as a whole, but what about, did you ask about what the perception was of civilians working in the field and of the U.S. as well as international community that's there and is ramping up and if that's noted by Afghans at large or if that's simply not... Yeah. We had a favorability rating, I believe, of, I don't know if we did NGOs, but we did of the United Nations, I believe, the performance of the UN. Let me see if I can find it for you. Here it is, sure. And it was a little underwhelming. We asked ratings of the presence and the performance of foreign aid organizations. Half of Afghans... This would make more sense, too, if we dissected it regionally, but half of Afghans said there was a strong presence of foreign aid organizations in their area. About half said there was not. And we didn't do a lot of other metrics on it. You know, our report is up at ABCNews.com. Our questionnaire, again, even in a presentation of this detail, I can't hit it all because it's a half-hour interview of 60 or 70 item questionnaire, but the full questionnaire and the marginal results are up at our website. In this document, you're welcome to take a look at. And I'm happy to run any cross-tabulations anyone's interested in. So you can look through our questionnaire and say, what did that look like in the Southeast or in Kunduz? And we can run that data. I'm happy to do it. Again, I'd be very cautious about this, where efforts were made last year to map where the civilian aid workers were and where they reached out to. When you talk about polling nationally, you have to understand that the actual presence of civilian aid workers in the field, which is very different from seeing a project somewhere, is very limited in terms of most of Afghanistan. The other is that when you ask questions like civilian versus military or U.S. versus international components of ISAF, often people think U.S. as ISAF unless they happen to know it's German or Swedish or what have you. Nationality differences and perceptions are extremely uncertain. So when you start doing French, German, and U.S. presence in Afghanistan, you're more likely to get misleading than positive results. We've tried some of that and it's essentially the same. In other words, if you ask about the performance of the United States or you ask about the performance of NATO ISAF, we have those separate measures and they're almost identical. I think they're conflated. So gentlemen, back. Jim Doros from the Office of Military Affairs at AID. I enjoy your polls. Two things I'd like to ask you, not in numerous parts. One, you've got some data on development and things are going up. And I'd ask, so what? Dr. Korzman has talked about this all the time and his stuff about just having metrics of output things. The second thing is I'd ask, when you look at favorable response to the Taliban, did you get a sense of why? Why? Well, two things. So first, in terms of development improving, as we show in our regression analysis, it is a significant predictor of the positive public attitudes in terms of the sense that countries go in the right way, confidence in the government in the U.S. and NATO as well. I think it's an essential element, in fact, of public confidence and attitudes more broadly, where development is seen as occurring. I think this was the promise, or the perceived or taken promise of the invasion. The notion that A, the yoga, the Taliban would be removed and B, that much needed development in a very impoverished country would follow and that slow pace of development, in addition to the ongoing violence, has been one of the great sources of frustration among the population. So to see those improve does in a very important way, I think, inform our broader judgment and understanding of how people are reacting to the situation. There's nothing like local development that is going to hit you personally, that's going to affect your personal life. If you've got electricity when you didn't erode, so you can get to the market in a few hours rather than a few days, all these make very important differences in living conditions and therefore are reflected in attitudes. And I'm sorry, your second... Does that lessen support for the Taliban? I mean, we can do all these things. There's lessen support for the Taliban and that's what we're trying to do. Right. Why is the Taliban a fraction of it? Because they're not providing jobs, programs, putting girls in school over the road. Right, support for the Taliban A is not at all broad around the country. It peaks in conflict zones and there may be some cultural or religious or political reasons for that but I think, I mean, in Kandahar it's the birthplace of the Taliban. It's where they're from and it's where their support is essentially strongest. I don't know that there'd be a surprise a cultural and allegiance that's related to a variety of preferences but also that these are the areas where not only is development challenged but where violence is highest and where violence is highest people are going to look to some solution to that violence and if they don't see it coming from one source there may be an increased inclination to seek it from another. It's drawn from a different source but I think we need to be extremely careful here. First, the Taliban does provide key jobs for young men and until really late in 2009 it was paying more than the local place or the army was. So it was a major employer. Prompt justice is, for many Afghans the key to local stability. The Taliban was covering about five times the area that the Afghan government was in terms of this because we focused on formal justice systems rather than traditional justice systems and the other problem is when you talk about Taliban presence you have to remember that and again the maps here are not generally distributed and in an awful lot of areas the Taliban is the only presence. There is no competing Afghan government presence of any kind whatsoever. So when we talk about Taliban versus NATO ISAF or Jiroa we have to keep in mind exactly what the areas are and these and I think one of the great problems we have had in general is the failure just in the crudest terms to map where we are active or the Afghan government is active because quite aside from the qualitative issue if you're not there at all strangely enough you're not taken very serious. Again as I think the table I showed you our data suggests that attitudinally it's very clear that where you're seen as having a strong presence or any presence you're much more likely to be rated positively for your performance. You'd be in real trouble if the opposite were the case. Let's see I think we let me take the question over here on the left sorry I couldn't see it. Craig Adair just an interested member of the public I have a follow up question on ethnicity and if that's if I understood your response earlier that your sample sizes weren't large enough but the numbers that you did get didn't pick up significant differences across ethnicity. If that's the case how do you or do you have any explanations to offer for the difference that you the clear difference that you saw in the south and the east. Yeah there are regional differences I haven't teased out if those are ethnic or regional differences and we could run the data to establish whether you're right the Pashtun in the south as opposed to Hazara and Tajiks in the north that may well be associated. I don't know if it's regional or ethnic and it's an analysis I haven't done but it's one worth doing. Initially in Iraq for example there were suggestions that the differences we saw there attitudinally were regionally based rather than ethnically or ethno sectarian based and we could dispel that pretty quickly when we ran that data. It was clearly ethno sectarian. There are some I shouldn't say we haven't seen any differences I haven't seen any that have risen a level for me to spend a lot of time on them. Rural Pashtuns are more conservative more negative about the government to some extent particularly rural to the extent I've looked at them rural Pashtun men but I haven't done depth and in depth analysis of those. We've got by the way a good robust data set now we can combine these surveys and look at some of the basic metrics over time and even though we've had change in the outcome we can therefore increase the sample and look at some of the individual groups and differences among groups can stand out better. Ladies and gentlemen I think we have kind of run over the normal hour but let me take one last question I think the gentleman in the second row there. Vijay Nilekani, NEI. The scope of your survey include the perception of neighboring countries? Like what do the Afghans think of the Iranians or the Tajiks or Uzbeks or Chinese or Indians? Yeah we did include some favorability ratings of other countries and consistently in our data the country that is seen most favorably by Afghans this should not be a surprise I hope is India and I think that's because India is seen as a counterweight to the Pakistanis who are viewed with a great deal more suspicion there's that less in the sense that Pakistan is harboring the Taliban but there is still not a lot of warm feeling about Pakistan as a check our data 81% of Afghans see Pakistan unfavorably I guess you could look on the bright side and note that it was 91% the last time we asked. I hope are the Iranians and Chinese? Yeah we did ask about China we got an even split on Iran half see around favorably about half unfavorably and again the one that really jumps out of me is India's 71% favorable. Ladies and gentlemen thank you very much for coming and may I ask you to thank Gary in a sort of traditional manner.