 Ladies and gentlemen, since Afghanistan is technically not part of the Middle East, I hope you will forgive me if we try to start more or less on time. I realize this is generally inappropriate, but I'm going to try anyway. It's hard to introduce Gary Langer in one sense. I don't think I have seen anyone do as dedicated or as comprehensive a job in structuring polling, particularly in complex areas where there are conflicts, like Iraq and like Afghanistan. So I'm going to let him introduce his methodology and his expertise and his results. But I think one of the key points in dealing with Afghanistan is to understand both the trends and the complexity. I think as some of you know, the situation has deteriorated far more quickly than we have yet made public. The national intelligence estimate basically has been sitting in the wings now for some six months. It was leaked in part in the New York Times, but quite frankly I think the reason that it has not had the publicity that would normally be the case is as we move towards the campaign season, being frank about what is happening in Afghanistan was not politically correct. For a long time NATO and ISAF have not produced clear results. Strangely enough in January for the first time they suddenly released a series of updates that they have not released for nearly eight months. We saw in January the Department of Defense issue a partly time or part update on Afghanistan and yet when you look at it, the report is dated in October and it clearly was something that people sat on throughout the political and the campaign season. And so let me introduce the poll by showing to some extent the pace of events. You can't see the details of this slide but the basic conclusions are obvious. The yellow and orange areas were the areas of Taliban, Hig, Akhane influence in 2005. The red areas you see in the map on the right are what happened in Afghanistan in 2007. In general, since 2005, the area of Taliban and insurgent influence, according to UN, NATO, ISAF, and US estimates, has increased by more than 50% a year each year through 2008. The rise in violence is hard to map. The incidents that are counted here are essentially incidents that attack NATO and ISAF forces. The very high range of attacks that occur on Afghan tribal leaders, people in the field are not counted in these statistics and they grossly understate what is actually happening. This map was released by NATO ISAF in January. A similar version of it is available in Department of Defense data. You can see the areas of influence, you can see the areas of attack. But let me note, these are not areas of political influence or where the Taliban or other groups have a major or key role in the countryside. And any analysis that is based purely on combat incidents is a horribly misleading picture of how much the threat has grown. There are other maps that provide a clearer picture of how bad the situation has been in Kabul, different counts of violence, different counts of casualties. And these are in the document that we've handed out along with Gary's results that he'll be briefing on showing just how bad the situation has gotten. This is a map drawn by another group, once called Senwas. It attempts to estimate the areas of political influence, the areas where the Taliban has been establishing a major presence as of 2008. This map is probably somewhat exaggerated. But in mapping that I have seen that has not been released, I would have to say that it is far more realistic than the kinds of maps which only show the areas of violence. And all of this is really a preface to just how important it is to understand what has happened in terms of Afghan perceptions, Afghan attitudes, towards NATO, towards the U.S., towards the Taliban, towards the government, and the war. We tend to see these kinds of wars in terms of kinetic events, tactical combat. The guerrilla wars are wars of attrition fought for political influence and political control over territory. They are essentially wars of perception. And if I may make a transition to Gary, the late Colonel Harry Summers, who was one of the leading analysts of the war in Vietnam, once told me of an exchange he had with the North Vietnamese officer long after the war, Colonel Summers turned to the Vietnamese officer and said that basically the U.S. had won virtually every tactical battle it had ever fought in Vietnam, which was true. And the Vietnamese officer replied to Colonel Summers, yes, but it was irrelevant. These are issues which I think we need to come to grips with very carefully. Neil, if I can add you to put Gary's presentation on, let me introduce Gary Weiner. Thank you, Tony. And thank you all for coming. I appreciate it. At ABC News, we've conducted now five national public opinion polls in Iraq since 2004, and this our fourth in Afghanistan. We do this for a simple reason. We think it's essential for us as a news organization to independently measure, know, and understand and report the progress of the U.S. and allied intervention in those countries. I'd note that the governments here in the United States and other governments and other entities as well conduct their own extensive polling opinion research in these same countries. They want and need to know how they're doing. And I would suggest simply that so do we. You described the project to you, this most recent survey I've been going to be telling you about today is our fourth in Afghanistan and it's serious since 2005. The field dates, as you can see, this survey research was completed actually one day short of only a month ago. These are very fresh, very recent data, and I think can bring us quite up to the moment in our understanding of public opinion there. And the fact that it is our fourth in a series enables us particularly to look at trend over time and invaluable aspect of this work. We repeat many of the same questions from survey to survey. Our first three were essentially fielded around October 05, October 06, October 07. This one happened a little later of finishing here in January. 1500 face-to-face interviews, full national coverage. Very quickly on the methodology, I can take questions on this or if you need details, contact me later. 194 random sampling points scattered about the country, stratified urban rural, proportional to the population size of district and then simple random sampling to settlement level. We conducted oversamples in those provinces to enable ourselves to do some additional, more discreet analysis in those areas of interest, household selection by random route, respondent selection by a random process called Kish grid. 176 interviewers, half male, half female. Male respondents were interviewed only by male interviewers, female respondents, but female interviewers with supervision and back checks. 33-minute interview, very high cooperation rate, far above the levels we see in our U.S. surveys. And I do want to point out the field work carried out for us by the Afghan Center for Socioeconomic and Opinion Research in Kabul, Aksar, a subsidiary of D3 systems. They've been doing our polling for us in Iraq as well as in Afghanistan and they're highly talented and really courageous individuals who we have a lot to thank for producing these data. I'll show you a few pictures. Tony hates photos. They're not data, you know, but I think an important point here is to remember that when we do these surveys, two things to keep in mind. One is the data that we report are actually the expressions of individuals. It's not just numbers. It's the people behind them. Another is that I think perhaps too many policy makers will jet into Kabul, visit Bagram, get out of town and think they've learned something about what Afghans think. Our interviewers went to every corner of the country, all 34 provinces conducting a truly random sample of individuals, as you can see here. And I think it's just important to keep in mind the very personal and granular nature of these interviews. And they give you a sense, I think, as well of what life in some of these areas is like. This last one actually is Tony's driveway. I don't know how that kind of looked. We also had reports from the field. Our field teams that traveled, supervisor and interviewers, do give us reports of their, summary reports of their experiences or notions in the field. These are anecdotal. In this case, I think they did very well support and illustrate some of our empirical findings. I'm not going to read these to you, but you might want to look through some of them. I think they're fascinating in their own right and help us again understand the data. Just today, I don't think Tony arranged this, but you never know, 19 killed and 54 injured in perhaps the most brazen attacks we've seen in Kabul proper. About three hours after that, this is an email I received this morning from one of our correspondents who was in Kabul at the time, recording to us that our bureau in Kabul received a call from the Taliban shortly after the attack, claiming that there were more than 20 suicide attackers in the city. This is not the kind of information we did or would ever report. It's the equivalent of a bomb scare. But I'm showing it to you to give you a sense of the tactics that are being employed in the attempts to instill fear in the population of this country. OK, so let's go on to some of our data. One of the most basic measures we use in this or almost any country, do you think things in this country are going in the right or in the wrong direction? You can note the trend over time. We've gotten from still heady days after the overthrow of the widely disliked Taliban in 2005, 77% said Afghanistan was going in the right direction, 55, 54. We're down now to 40%. Tremendous drop, a 37-point decline, nearly by half in the number of Afghans who believe the country is headed in the right direction. We started in 2005 again in those heady post-Taliban days with a 83% favorability rating in the United States. The United States was, in many ways, Afghanistan, you could say, was the poster boy for American popularity in the Muslim nation with that 83% approval. Ambassador Wood, looking at our data and commenting on them just this week, said that that, well, 83% is unsustainable. Perhaps so, but that doesn't really explain why it goes down to 47. Tremendous drop in favorability ratings, and indeed, for the first time now in our own data, we have a bare majority of Afghans holding an unfavorable opinion. Performance ratings, similarly. Top line, this is a rating for Hamid Karzai as president of Afghanistan. 83% favorable to 52, the Afghan central government from 80 to 49, the United States, in terms of its performance in Afghanistan, 68% favorable when we first asked that question, 32% today. The sense that U.S. or NATO and ISAF forces have a strong presence in Afghanistan, in your local area, we ask, do these groups and duties have a strong local presence? From 57% to 50, now 34%, a sense of weakening presence of these forces. And then do you think these forces are effective in providing security in your area? Obviously, their fundamental mission has gone from the 67% positive rating to 42, in our most recent. And what is support in your area among the people in your area for various groups, and in this case, for U.S. and NATO ISAF forces, from 67% to 37%, reported local support for the presence of these forces. I'm sorry, not for the presence, but for the effectiveness. The Taliban were and remain very broadly unpopular in this country. Nonetheless, was this a good thing or a bad thing for Afghanistan that the United States led coalition ousted the Taliban, it's got from 87% favorable and held steady, the first to hold, now down to 69%, a still pretty significant nearly 20-point drop in the most fundamental view that the country is better off without the Taliban. And on a personal level, barely about half of Afghans expect a better life for themselves in the next year. That's down from 67% when we first asked the question. And just under half expect a better life for their children. These dashed expectations, I think, as much as perhaps many of the other measures we have in the survey help explain the course of public opinion in this country. But reason one, security, and I think we need to take a look closely at these results. Ratings of local security, gone from 72% positive to 55% in our data, the substantial significant decline. There's an enormous amount of variability by region and province of this country. And it's important to keep in mind, you'll find it throughout the data I'm reporting to you, and it's, can be as Tony will tell you as well, kind of misleading to think of Afghanistan simply an aggregate given the different needs and the different issues we see across the country by province and by region. Positive ratings of local security, for example, in some conflict areas such as Helmand and Kandahar, where we oversample, compare those to Clintus and Balkan and North. Experiences of violence in the past year, suicide attacks. Most of these have not increased. If we're not showing increases here, there's a few possible reasons for that. One is this is national data, and when we disaggregated to the provincial level, we do see increases in some areas, these conflict areas, and less violence in some other areas, particularly in the North. Nonetheless, 26% of Afghans, that's one in four, report suicide attacks in their own local areas, Sniders, Kidnappans, Bahamas, these are very high levels of personal experience of strife and violence that do fuel the attitude. 38% in net of nearly 4 in 10 Afghans say there have been civilians killed or wounded in their area. They blame equally, roughly equal numbers say it's coalition forces or Taliban that were responsible with the ANA and ANP, a little bit less. You might have seen that bombings or airstrikes, number 16% of Afghans total report coalition airstrikes in their area. But again, look at the variability by region now in the Southwest and East, the two highest conflict areas of the country, compared to the North and North. More broadly though, we ask Afghans if coalition bombings and airstrikes are an acceptable means of attacking insurgents, because they're effective in that regard, or unacceptable because they put too many civilians at risk. As you can see, three-quarters of Afghans consider these airstrikes unacceptable. And when we ask who they blame when civilians are harmed in these airstrikes, the U.S. and NATO for bad targeting or the insurgents themselves for seeking themselves in a civilian population, they're more apt to blame the U.S. and NATO forces, public relations, coup if you will, perhaps, for the insurgents. That leads us then to this next view, and the change over time I think is one of Syria's concerns. Who do you think broadly is mainly to blame for the violence that's occurring in the country? The number who mainly blame the U.S., Afghan governments or forces has increased from 26 to 36 percent. The number who mainly blame the Taliban has decreased to the point now, this year, as compared to 2007. More blame the U.S. and NATO than the Taliban for the violence that's occurring in Afghanistan. In one response to that view and the sort of violence, 25 percent of Afghans now say that attacks against coalition forces are justified. That is double what it was last year. And as Tony will tell you, 2 percent would be sufficient for these sort of attacks to occur, but the fact that a quarter of the population finds them acceptable is another cause for concern. Note also the sort of conditions that fueled this, or they're at least associated with it. In areas, among people who report coalition bombings in their area, the number who say it's acceptable to attack coalition forces is not 25, but 44 percent. In the higher conflict provinces, mainly the southwest and the east, 38 percent say such attacks are acceptable. Compare that to more pacified areas. And the resurgence of the Taliban is the biggest danger facing Afghanistan, 58 percent of Afghans. Open-ended question volunteer that it's the Taliban. 43 percent, these numbers are about the same as they were last year. 43 percent of Afghans say that the Taliban have grown stronger over the course of the last year. Almost double the number, certainly substantially more than the number who say they've grown weaker. And again, look at the difference by province here. In Helmand, interestingly in Harat as well, in Kandahar, compared to Kandahar. The Taliban activities reported here, night-letter bombings, again, when you disaggregate this, and I'll show you some of that data, you get that enormous variance across province and across the region of the country. Again, high levels of experience. You see the differences we're talking about across the country. Only 7 percent of Afghans have a favorable opinion of the Taliban overall. It's not a popular movement, quite the opposite. Nonetheless, again, there is substantial difference by province, both the Southwest. Beyond their own sentiment, 2 in 10 roughly Afghans report local support for the Taliban in their areas. Some levels of that kind of support. It's about the same as last year. But with increases in the Southwest, and specifically, and most impressively in Kandahar, of course, the base of the Taliban. A quarter of Afghans support the notion that the Taliban have become more moderate. That means that the substantial majority reject it. But one in four buy into that premise. Another, I think, potential concern for policymakers, as you'll see, for a few reasons. Again, look at the variability in this case. We're giving this some specific promises. 64 percent believe that the government should negotiate with the Taliban, although the vast majority of that group say only if the Taliban first lay down their arms and discontinue fighting. But the support for negotiations at some level is quite high. And I think one reason for that is there's not a lot of confidence that the Afghan government, with support of its western allies, will outright win. Only 33 percent. One in three Afghans believe that Afghanistan, with the support of its allies, will ultimately defeat the Taliban. More, as many, I'm sorry, expect a negotiated settlement. If you expect the Taliban to win outright in two and ten, perhaps the worst-case scenario depends how your view is. I think the fighting is just going to lag on. Not a lot of confidence there. Okay, so we've talked about security, the rise of the Taliban, and then let's look at local conditions. This is simply a demographic profile, if you will, of the Afghan population. I think it's been helpful for us to keep in mind the level of poverty, the deprivation in terms of basic utilities, possession. I think it's essential to understand the baseline experiences of this population as we then look ahead and see other lines again. Ask to rate local conditions in their areas. Some are rated pretty positively. This is the availability of the food, the presence of schools, the quality. But they drop off in some areas also quite dramatically. Only half say they have adequate medical care availability infrastructure. And it goes down. Economy of very important central concern is rated quite low. Economy and jobs, economic opportunity. And electricity has been the longest-standing and most severe complaint. Now look at the trend in these over time and some of these key ratings. A number of Afghans would say rate their living conditions positively while a majority is down 20 points. Security we saw earlier. Jobs and economy has not improved in terms of the experience of average. Another related to security is simply the question we ask about the freedom of movement. Do you have the ability to go where you wish safely? It's declined by 10 points in the last year. And while a majority say they do have the ability to go where they wish safely, four and ten Afghans don't. Other conditions, local conditions, are better rated or have not taken a negative course. An increase in our most recent data in the ratings of the availability of fresh water supplies. Medical care has been essentially flat although it's still not very high. And there's been an improvement in infrastructure particularly. We've gone from 24% to 42% positive. And electricity again has been flat. The problem I think of the challenges here is you don't really want to take one of these and say look ratings of the local infrastructure improved fairly dramatically. Therefore things are getting better in Afghanistan. I think it's essential for us to look at the totality of these data before we reach that kind of conclusion. There are other good measures in terms of development or release indications of progress. Schools rebuilt and reopened, mosques roads. In particular there's been an increase in our most recent data in reported road building of clinics opening and police stations. But these don't seem to mitigate the problems that exist. One important example I think is the great concern we see over rising prices and the affordability of food. 63% of Afghans in this survey tell us they're unable to afford adequate supplies of food and that has increased over the last year. It's hard to celebrate the new road when you can't buy enough to eat local corruption. Really the problem by 85% of Afghans is a big problem by 63% and that number in particular has increased dramatically over the last year. Indeed 50% of Afghans tell us that in their experience local corruption has increased in the last year. Again with some differences by area and I thought this was particularly interesting. I direct your attention to the big problem ratings particularly in Harat and Kabul. Is it too urban area? 82% call it a big problem. It declines in the rest of Afghanistan pretty dramatically. I think if you visited Kabul you might get a somewhat distorted picture of the level of concern about corruption. It peaks in Kabul, it peaks in Harat. It's very significant but not quite so intense in the rest of the country other than the primary concern. Measuring the effects. I'm going to go through this quickly. It's a little technical but we did a regression analysis trying to predict these dependent variables with the countries going the right direction, the performance of the central and provincial governments, performance of the United States. We experimented, we've only had these data for about two weeks so there's more work on the modeling to do. But we used this assembly of attitudes to see which held constant against the others best predicts these dependent variables. Ratings of local conditions, security, economic conditions, food, fuel and the rest. My fundamental point here is that the trouble, the problems in Afghanistan are decidedly not about one thing. It's not as if we could instill security and the problems would go away. Or building a road or a bridge or a school somehow magically solves the difficulties in Afghanistan. Each of those, local conditions, the sense that the CATH Alabama has grown stronger, local security, all independently predict whether you think the country is going in the right direction. So for example I must say it is living in Kabul negatively because the people in Kabul is through the event we saw today are a little bit freaked out. The positive ratings of the central government have really a wide range, I think the bottom line point here, of independent predictors. From regional, the ratings of conditions, overall security, development, index of violence, jobs in the economy. There's a complexity of issues that inform these negatives. Ratings of the prudential governments similarly, with particularly development, I think worth noting, expectations, but security as well. Positive ratings of the United States and Afghanistan predicted by region, area, and again, ratings of local jobs and economy as much as ratings of local security. I think this is a fascinating other way to look at the data. We can basically break the country here into four groups. Those who say their security and their development are good. Those who say their security and their local development are both bad. Those who say their security is good, but development is bad. Those who say their security is bad, but their development is good. It breaks the country into roughly equal sized groups, interestingly enough. Seven and ten say something's not going poorly. And these do inform judgments, but first let's look at it by region. Security is good in some of the areas we've been talking about. Compare it to the both bad here or the security bad numbers you're seeing here. The division by region of the country, again, is very important to keep in mind when we analyze the problems in Afghanistan and when policymakers start thinking about directions. The way these inform attitudes, if you think both are good, you're much more likely to think the country is going off in the right direction. Twice as likely as you think both are bad. Security is good. You're closer to being positive and security is bad overall. You're closer to being negative. However, when we look at ratings of the central government, we see the difference between all good and all bad. But security and development are about equally powerful. And on the provincial government as well, there's a great deal of similarity between whether the security good and development bad or security bad and development good affect your attitude. In terms of rating of the U.S., however, we're more judged on the issue of security. It's, by the way, a coherence in the data set that I think is encouraging since that's our task in the country that seems to be borne out in our region. Further challenges beyond what we've discussed, security, the Taliban, local conditions. Only 32% or only one in three Afghans say it's acceptable, particularly if nothing else is available economically, to cultivate opium poppy. But look how that grows in the top producing provinces and then in Helmand, which produces two thirds or more, I think, of the country's opium. Very few, even in Helmand, say it's acceptable in all cases, but rather, say growing opium poppy, it's acceptable only if there's no other way to earn a living. That suggests that there are, there's a possibility of acceptance among the population of alternatives if any were available that were economically viable. Ratings of neighboring countries, another problem, another concern I put in India and Iran, mainly as a counterpoint to views of Pakistan, viewed unfavorably by 91% of Afghans. Today's events can't help. The Director of Intelligence in Kabul accused the perpetrators of this morning's attack of calling minders in Pakistan just before they took action. 86% is near unanimity rate that Pakistan's role in Afghanistan negatively. Two thirds think Pakistan is allowing its element to operate. An enormous amount of, as if Afghanistan's own problems weren't enough, there's an enormous amount of tension in this relationship. So what are some ways forward? One question is, does the election of Barack Obama's President of the United States make things better? In and of itself, in the eyes of Afghans, the answer is no. Two in ten expect this to somehow, in and of itself, be an improvement of essentially as many or nearly as many think it'll actually make things worse and the rest are just going to wait. An important point, I think, is that while the performance ratings, which I've shown you, of the U.S. and Afghanistan are pretty dismal and have declined dramatically, there is still fundamental support for the presence of these forces. 63% of Afghans support the presence of U.S. forces in Afghanistan. Now that is down from 78% in 2006. Also, the number who's strongly supported has declined from 30% to 12%. And I think that one motivator in the view that the presence of U.S. forces is something to support is a concern about the alternative, particularly with the fairly broad sense that Taliban is gaining strength in the country. So while the performance ratings are pretty bad, there is still a majority support for the presence of the United States if it could only do better, I think it's fair to say. However, when we ask if coalition forces, U.S. and NATO forces in Afghanistan should be increased, decreased, or kept at the current level, there's very little support to intend for increasing the deployment, as is now being, as you know, mooted. In fact, it's gone. And indeed, 44% of plurality of Afghans would rather see the level of these forces decreased. We look at this by experience, and I think this is informative, among Afghans who find coalition airstrikes unacceptable, they're far more likely to favor a decrease in the U.S. troop deployment. Among those who say the United States and NATO forces are mainly to blame when civilians are harmed, even more. I think these troops should be decreased in deployment. The paradox in this is that where these entities are seen as having a strong presence, they are better rated. That's perhaps because these are less likely to be conflict areas. But where the central government is seen as having a strong presence, its performance ratings are very markedly higher than where it's seen as weak, the same for the provincial government, and the same indeed for U.S. forces. It has noted the Taliban are not seen as a viable or remotely popular alternative. Who would you prefer to see rule Afghanistan? The current government of the Taliban, very few, almost no one, prefers the Taliban. Although preference for the current government has declined somewhat, and other someone else has grown in some choice. Another fact to keep in mind when anticipating ways forward or thinking about progress in Afghanistan is the simple fact that while their ratings too have declined and declined fairly substantially, local confidence in the country is much higher in national and indigenous organizations than it is in the United States coalition forces. Confident they can provide security, the local police, the Afghan government, the provincial government or all still have substantial majority ratings and strong reports of local support for these entities in their areas compared to coalition forces, local militias and the Taliban. The extent perhaps to which these entities were seen more in the forefront of efforts to restore Afghanistan's security and economy, public acceptance might be improved. And I think finally, I find this an interesting result and I think perhaps a helpful one to end on. There's been a lot of talk and suggestions by some analysts that there really isn't an Afghanistan, that the people, the population of this country don't have any true national identity and think of themselves only on an ethnic or tribal level. We ask that question in this survey, this is a new question, we ask people their tribal or ethnic identity and then we ask them, well, do you think of yourself more as a blank of Pashtun or Tajik or Hazar or as an Afghan? And I was interested to see that 72% think of themselves as Afghans first. Thank you very much. I think Tony and I both would be happy to take any questions. Let me, before we start, just make two quick points. First, I think that this can somehow be perhaps more discouraging than it should be. The fact is that we are not talking about strong opponents as much as weak, disorganized and ineffective efforts on the part of ourselves, our allies and the United Nations. When you look at the breakouts of resources and we've made this one of the papers given to you, it's interesting to note that seven years into this war, the best we can do according to the Department of Defense is provide roughly a third of the U.S. and allied military advisors that are needed by the Afghan forces. So we have no consistent funding profiles for any aspect of Afghan force development that aid funding is erratic and reactive. And that when you look at the problems here, you are talking essentially a war about the scale of Iraq with about 20% or less of the combat forces in much more difficult terrain. These are factors that it is easy to forget because we often look at the threat side rather than the aid side. But I really would look at some of the figures in the data and we've given you the web references as well as all of the data that Gary has given. He's been kind enough to put his presentation on the web. The second point I would raise is this. If you look around, you can see how many people are here. I would ask you to give us one question rather than several. A question does tend to end in a question mark. If you have a speech to give, we have a very delightful set of steps outside the building. And it would help everyone to hear your name and organization when you ask the question. Ladies and gentlemen, with that, let's start the lady in the front row. I just have one question. Who do you see using these data and how do you see them using these data? In our news reporting. The reason we do this poll is so we as a news organization can effectively understand and report on public attitudes in the course of the progress of the US of policy and intervention in Afghanistan. So we report it. We've got on our ABCNews.com websites some fairly extensive reporting on the results and in venues like this as well, we can pass the work. Other entities that conduct research in that places like Afghanistan, I think so, do so much for the same purpose to evaluate and analyze their effectiveness, what's working and what isn't, and how the local public is reacting to their efforts. But the difference is much of that we never see publicly. It's classified. It's used for internal intelligence. I think the reason that we at ABC go to this effort is partly to get a look into that playbook, but also to come to our own independent evaluation and understanding of the public attitude. Well, to support what Gary says, if you go out and talk to US forces, for example, in eastern Afghanistan, you will find that they have very similar polling down to even more detail with very similar results. If you look at efforts today to move toward a approach to clear, build, and hold strategy, a lot of it will be focused on the areas where polling results support analysis of the threat and violence to provide that kind of focus simply because there simply will not be enough forces to be allocated on any other basis. Tom Parker from the Defense Department. May I ask how confident you are in the data? There's always, I would think, when you're polling in the third world, people being polled might ask themselves, I get into trouble if I said something bad. And then secondly, how do you control for your own pollsters going into the field and the accuracy of the data that they report back to you? Thank you for a great presentation, both of you. Thank you. Appreciate that. A couple of things on that. One is we have a high level of confidence in our data. The methodology we use is the best possible, I believe, in the country. There are limitations and challenges in conducting public opinion research in a country like Afghanistan, particularly where the census data are not particularly current or good. But accepting the limitations, we think it's the best we can do. Also, because we've followed the same methodology continuously now over these four polls, I think the trends are very significant. When you do the same approach each time, then the trends can be quite revealing. As far as the questionnaire, our full questionnaire, by the way, is up and available for inspection at abcnews.com in English. But I could actually get to the Darian posture versions if you'd like to see them. We work very hard on the questionnaire and consultation with our colleagues in Kabul and elsewhere to make sure that it sounds and reads as if Afghans were interviewing Afghans, that we work closely on the back translation to make sure that the language is appropriate and works. The interviews themselves are carried out by trained Afghan interviewers working for Aksar in Kabul, which has done an extensive amount of polling in the country for a variety of clients other than ourselves. And we've had our own producers in Kabul attend training sessions and even a company on some interviews. Those interviews are not retained in the dataset, but simply to check and confirm that the field work is being carried out to our specifications. So we do, frankly, everything we can to ensure that it's the best quality data possible. And then finally I'd say there is, I think I indicated in some of this presentation, when you get into the cross tabulations and the regression analysis, there's a great deal of internal coherence in the data. You'd be worried if people who felt insecure over here at the same time gave you a positive rating on another security or safety measure over there, or if the cross tabulations were nonsensical and instead we see a great deal of coherence in the dataset. And that gives us a lot of confidence as well. I hope that helps. I'll wait for the mic, please. David Izby. Years of talking with Afghans, one of the things which I have found in my own experience is Afghans are reluctant to criticize Afghan institutions or actions before anyone's. Afghans are foreigners. I noted the relatively high level of confidence in the police, yet at the same time high levels of corruption. One question being, the need there to follow up, who is doing this corruption? Who's doing it? And I think one of the ways where your methodology may be, the police have to get a pass there, and there would be much greater acceptance expressed to a question than in a more anecdotal approach. Well, a couple of things about that. One is if Afghans are less likely to criticize local institutions, we nonetheless have seen the trends that I've showed you in terms of ratings of central and provincial governments and others having declined by 20 and 30 points over the last few years. So we do have significant change there. Corruption is an interesting issue. It is widely reported, seen as growing, and a significant problem. At the same time, in the regressions I showed you, it does not independently predict ratings of the national provincial governments or of general attitudes about the course of direction in the country. I don't know that corruption, I don't have old data from Afghanistan. I don't know that the existence or experience of corruption in Afghanistan is a particularly new thing. And if it is indeed a part of life in a long going sense, it may not therefore be unusual or different enough to independently drive attitudes, ratings, even of the police, as well as these other entities. All I can tell you is that I was interested to see that it didn't independently predict ratings indeed of the provincial government. I didn't run the regression on ratings of the police to see if that particular entity is independently associated with corruption, and it's a good one to do. As a Chicagoan, I have to ask, what's wrong with corruption? The gentleman over in the back there. Sir, in the living conditions part seems to be fairly good, if not benign. What factor were the provincial reconstruction teams in that? Did you ask at all about them? If I understand correctly, they were operating there since the intervention. They were operating in Iraq too, where they were very successful, but started only there in 2005. Did they play a factor in your question? We couldn't ask specifically about the reconstruction teams. We did ask about the level of reconstruction, and as you can see in some cases it's positive. The reports of the schools rebuilt or reopened, the police stations, etc. There are some positive results there. I think the particular, to me at least, the one that stood out was the report of improved infrastructure, roads and bridges, particularly up in fact indeed in the past year, and better ratings for those in terms of an increase or an improvement than in any other area. Also true for water, but supply of clean water. But I think our broader conclusion seems to be that these aren't enough, that they have perhaps not matter-kept pace with expectations, and that they've been counterbalanced, if you will, by the negative views in other areas in security, but also in, for example, the fundamental jobs and economic opportunity. I think you also have to be very careful about watching your funding streams. PRTs get, in Afghanistan, a relatively limited amount of the aid money, and in the Department of Defense report issued in January, out of the PRT manning we had, our part of the PRTs, there are over a thousand military and less than 40 U.S. civilians. So when we talk about PRTs, these are essentially military operations, funding through third and similar aid. They are not really, in the classic sense, civilian aid programs. Let's say the gentleman there in the second row. Are you covering from the Voice of America Afghanistan Service? Do you ask the Afghans about more U.S. troops in their country? Were they in favor of more U.S. troops in Afghanistan? And what I've learned from my work with the majority of the Afghans is that they don't see it as a very good solution if the strategy is not changed. And I would also like to hear from Mr. Korsman. You are testifying before the House Armed Service Committee tomorrow and the same issue would be talked about. What is your insight about it? How many troops can do this job done in Afghanistan and should they be sent? Just to recap mine quickly, I did put this up. Only in our data, 18% of Afghans support increasing the deployment of U.S. forces to Afghanistan. 44%, rather, I think it should be decreased. And I think the data that informed this is the notion that they're not being particularly successful when, for example, the performance ratings of the United States in Afghanistan have gone from 68% positive in 2005 to 32% now. It is perhaps hard for Afghans to support an increase in deployment of forces that they see as largely ineffective in providing security. I think there's another way to look at it, to answer your question. In those areas where there have been sufficient ground troops to actually provide something on the order of what we call clear-hold build, as Gary hinted in his polls, you get a different set of attitudes and polls because people see the ground forces are providing protection. There is a development component and a governance component. Wherever people see violence in kinetic terms, in other words, people come in on sudden raids where there aren't sufficient troops to hold or where is often the case in air operations where air operations have to be used because you don't have sufficient ground troops and you have to go in on a tactical and emergency basis to support NATO forces that would otherwise be defeated or where you do not have sufficient human intelligence to actually target the airstrike in ways where you can be sure of the enemy, you get very different reactions. The same is true when you have time to plan systematically airstrikes and you can be absolutely certain of your target base as distinguished from providing close air support under tight pressures. Wherever you have very limited combat forces and in most of Afghanistan we have very limited combat forces in the field, particularly in the provinces in the south where, as Gary pointed out, you have the highest levels of unpopularity relative to the threat you get different reactions and this is borne out in the kind of polls run by NATO by governments which are not publicly released and where it's obviously impossible to describe the results. So I think that when you look at the impact of a clear hold and build strategy where there is a hold and build strategy, not just a clear, what you see in Afghanistan is almost exactly what you saw earlier in Iraq and people tend to forget that we were becoming very much more unpopular in Iraq under very similar conditions until we changed our strategy and tactics there. So this is something where I think perceptions are to some extent a result of what happens. There's an axiom in government I think we often forget. There is no such thing as a good intention. There is only a successful action. That affects public opinion and war fighting as it does public opinion and every other aspect of government. Yeah, I just want to add, I think Tony is exactly right and this cuts to the paradox that I reported earlier when I was showing you our data which is that where U.S. forces are seen as strong they are far more popular. So public opinion, I don't think anyone suggests should independently in and of itself drive policy. Only that I think it deserves and needs a place at the table in terms of understanding the implications of this policy and I think it would be wise not to expect an increase in deployment of U.S. or NATO forces there to be greeted with bouquets of flowers by the Afghan population given the data now. These ratings and these attitudes are performance based. So the gentleman in the second row there. Scott Worden from the U.S. Institute of Peace. Keeping in mind the earlier question about corruption and approval of the police I was particularly interested in the one graphic about expectations that local militias would improve security was only 18%, local support for militias was 17%, given the discussion in the press and other places recently about potentially using local militias or something along those lines to compensate for lack of police or security presence. Did you get more data than that one bar and can you comment on how you see that attitude influencing our security plans? Yeah, we do have more data on it. I think rather than do it now if you maybe send me an email after we can pull it out at the regional and even provincial level and I imagine what you'll see is again some substantial differences. Let me just note that there are other polls which show that if you look at police and then look at do you have functioning courts and detention and functioning government services you can't separate them, at least in terms of regular police functions. So one of your problems is that any survey that only deals with police and you can't deal with everything at once depends to misstate the problem because there are often police present where there is no criminal justice system and no government presence in the classic sense of governance. Let's see, the young lady in the third row. Thank you for your presentation. My name is Anya Kaslova. I come from International Foundation for Electoral Systems. My question is, was there any reference to upcoming elections in your polling and what is the mood among the general population toward the presidential elections? Whether people registered or think these elections, it's all in the reference to elections and if people think they will be effective. Thank you. The bad news is I can answer them all with one word, which is no. We didn't measure election attitudes. We felt that it's far enough from the elections and now I think set for August and also the scheduling of these elections was being debated just as we were preparing the questionnaire preparing to go in the field. It was, I think, just announced that date that it didn't seem an effective for our purposes a line of questioning. We don't, election, pre-election polling, I can tell you from my own experiences just this past year, tend to get people to hyperventilate a little bit and to do effective polling on presidential preference, for example, you really have to do some detailed modeling of who actually is going to participate with some relative degree of certainty to get a good estimate. And I didn't want to not do that modeling and throw out some estimate of Karzai's, for example, reelection prospects that would get people very excited but probably not be a truly accurate representation of true voter intention. And to do that kind of modeling takes another battery of questions that I didn't want to put in this questionnaire. This is a 33-minute questionnaire. You can drag people into hour- and two-hour-long surveys, but frankly, I don't think it's a good idea. You tend to fatigue them. So the gentleman in the second row here. Mike Miyazawa. It's been more than seven years since the Taliban was removed from power but they continue fighting. And for them to continue fighting, they need weapons and bombs. My question is, where and how do they obtain those weapons and do you or does anybody have proof that Iran is one of the major sources of those weapons? I think first, nobody has charged that Iran is a major source of those weapons. You can trace certain weapons coming out of Iran. Second, this is an armed society and it is an armed society on both sides of the borders. It's all very well to talk about controlling the inflow of small arms but when everybody has one, it is not that much of a challenge. I think the other thing is that we have a fairly good picture that between narcotics trafficking by groups like the Taliban, fundraising from outside going into groups like the Haqqani network or Haqmatia groups or particularly the two centers of Taliban activity and the federally administered tribal areas in Afghanistan. There's a lot of money being moved in not by governments and this gives them significant ability to buy arms, particularly in areas where there is very weak control if any by the Pakistani government. They try to be randomly distributing the gentleman in the back in the center. That's my Jeff. Chuck Berry from the National Defense University. I was going to ask a question about corruption but since I realize that Tony is from Chicago as I am, I guess I'll save that question until CSIS has a future program that may be inviting Rod Blagojevich here before he goes to prison. So my question is going to be about the chart you showed for Afghan support for the central government versus local provincial governments that may be headed by leaders of tribes or clans. We hear a lot about in the recent Congressional Research Service report and USIP reports that came out in the last couple of months that were attempting too hard to push democracy into Afghanistan and sweeping aside maybe local customs and what might be regarded as legitimate governments. What was your sense in questioning people about how the mix should be between a central government that may be more distant and less hands-off than more present in terms of the needs for desires for infrastructure and a good economy and so forth? Well, as you can see, I think some of this is borne out by the regression analysis. Again, I do want to work on that and perfect it. But one point I think is that similar measures are used or significant in predicting each of those of popularity both the central and the provincial government and a much greater mix. Remember the long list for the central government? People basically are throwing everything at it and there's a lot of independent factors that predict attitudes about it. The views of the provincial government were rated by a somewhat smaller range of factors having to do perhaps a little less with security although it was in the mix and more with development. I don't know if that helps answer your question but I think there are different responsibilities that are being assigned to some extent. That seems to be the result or a suggestion of that regression. One other thing I'd mentioned by the way is that while just as a corruption didn't show up in our model as a significant predictor of these attitudes, neither did ethnicity. We didn't see that being a posthum or Tajik or Hazara or any other group independently predicted attitudes about the country's direction or its institutions. Another interesting result. I think too we need to be very careful about the word democracy because there was a significant group at AID which strongly recommended that you have local and provincial elections rather than relying on one central election. None of those took place. So we are really talking about central government people who often have no constituent base operating in a very diverse and scattered country which was a problem we also saw built up in Iraq. So I think we need to be very careful about terms. Let's see. Let me try that gentleman in the front row. Hi there. My name is John Stinner. I'm with the Asia Foundation. Let's go back to the poppy cultivation. Did you guys do any polling on poppy eradication programs and how Afghans feel about their implementation or how well they're doing things of that nature? We did ask an eradication question and what we found is very little support. We found support for eradication of the crop particularly outside of its cultivation areas but we found very little support specifically for spraying of herbicides. That's a case that seems to have not been made in public opinion. There's a possibility of it because there was a lot of indecision on it. It's one of the questions on which we had a high number of Afghans saying they didn't know. But among those who had an opinion I think there was more opposition than support for spraying and certainly the support level for spraying if I recall correctly was 13 or 14 percent. So look, most two-thirds of Afghans overall across the country don't find it acceptable to grow the crop. Therefore it does follow logically and we have in our data that a majority do support eradication but the process and means of that eradication are still open. The case hasn't been made for herbicide spraying. And that's in the data report. I get your copy. You can find it online. Ladies and gentlemen, we've already gone over an hour if I extend this by ten minutes and I take a quick poll of the audience is that acceptable? Yes, no, maybe. Why don't we go on then? I'm going to jump on in sort of the middle of the back row. Greg Sanders, CSS. There may be a sample size issue but could we look at the conflicted provinces and figure out the impact of opinion on US troops of security versus arrow of strikes? In a similar manner but you did the security and development slide. Obviously that only really makes sense because the conflicted provinces of sample size might be low to low there. Well, we do. I'm not sure I follow you. Maybe we can talk about it more after but we do find as I think I should certainly at the provincial level much higher incidents of area strikes in the areas where you would expect in particular places like Helmand and Kandahar in the southwest, I think to some extent of the east as well. And those are areas where US forces are particularly unpopular in the conflict areas again. But again, for anyone who's got a question if we run out of time you're welcome to send me an email gary.langer at abc.com and I'll be happy to follow up with you personally. I would say that other polls would pretty well confirm these kinds of results of German second row there. Hi, I'm Martin Ellsman from the Embassy of Afghanistan. I was curious, you mentioned a lot about the geographic variance in some of the questions. Could you also talk about the kind of internal variance and the answers to the questions? Because it wasn't just do you think security is either A, good, B, bad. It was somewhat good. It was either really good, somewhat good, somewhat bad, really bad or no opinion. So how do you think that affects the way we perceive the outcome of the poll? Well, I'm not sure. I'm giving you, I gave it nets here, the net positive or the net negative. And we can do closer analysis of the range at the provincial level. Again, I don't have those numbers in front of me but I'm happy to dig them out for you. But I think what you'll find is what you would expect. And one thing that's nice, I guess, about these data is that they're very coherent with the independent data that Tony presented at the start of the presentation, where you see areas of greater conflict. Indeed, we find security ratings at their most negative and at their most intensely negative. The gentleman in the third row there. Wait for the mic, please. Now you was an formerly World Bank representative in Afghanistan. Did your survey detect any differences among the responders of the women's answers and men's answers? Only in one area did we see significant differences between men and women worth talking about. And it's in one area which I didn't report on at all today, which is in views of women's rights. Views of women's schools, of women working, of women holding government office. There are significant differences between men and women in views on those questions. There are also significant differences in those questions at the provincial level, also at the ethnic level between Pashtun and Tajik, although a lot of that is regional as well. Much less of it in the southwest, very conservative in Kandahar. But even in Kandahar, striking differences between men and women in their attitudes about women's rights is almost to, you could have a war of the sexes on top of all the other battles we have in that country right now. Not to make light of it, but that's the area in which you saw the greatest differences. There were a few differences elsewhere, but not enough for us to call them out. That is in the analysis, that discussion of views on women's rights. Mohammad Kasim, retired for World Bank. Pakistan. What is the perception of Pakistan? Because the grand line does not exist. And you therefore have 30 million Pashtuns on the Pakistan side, and 60 million also Baluch for the south. Do these people regard that as Pakistan or is Pakistan those damn Punjabis? That's a good question. We didn't tease out a definition of Pakistan, but one thing I thought was interesting is that when you have a 91% unfavorable rating across the country, that's about as close to unanimity as you get in survey research. So it seems as though our respondents had certainly their own sense of what we meant by Pakistan and were in general agreement on it. Clearly in a country that's what 40% Pashtun, Pashtun, Tajik, Hazara, there is near unanimity across groups to get that kind of ultimate result of a 91% negative rating. Can you tell me about the third row of it? Can you do it again? Just like to return to airstrikes, it's a truism that's repeated a lot that this is a campaign that can't be won by military means alone. And yet obviously, our power is a major source of U.S. and NATO superiority. So can it be won by military means at all? If every time there's an airstrike, a figure showed it produces massive unpopularity. Yeah, I can't comment on military tactics. Tony certainly can. All I can tell you is that attitudinally we see very powerful negative effects of those strikes. I think we need to be very, very careful. There are airstrikes which are based on the combination of human and technical intelligence which have a great deal of planning time for restraint. Those can sometimes be manipulated by the people who are attacked. You get charges that there are civilian casualties, but they have a completely different impact from having to go in when you are basically dealing with ground forces which are isolated which suddenly find themselves in an ambush and basically have to use air power to bail themselves out because there aren't sufficient ground troops present. You have a broader problem with airstrikes wherever you do not have a continuing security presence where there's no aid presence. So often people are looking at the airstrike per se rather than there's no governance, there's no aid presence, there's no security, or what you have are horribly insufficient ground forces often that don't have their own national helicopters or air support which in some cases do not have adequate armor which are suddenly thrust in the position of having to survive. So we are talking about, again, you have a maximum of 30,000 U.S. troops in there. Today if you put in the contract slice you have to compare that with around 45,000 people against over 200,000 in Iraq and we basically were only dealing with central and western Iraq and the problem is not just a matter of airstrikes. The young lady in the second room. Hi, Meethu Mahill from the Stimson Center. I was wondering if I could touch on religion. Did your survey include at all some sort of questions that would allow people to identify themselves as having moderate, secular, or highly religious views and could that be more predictive than, let's say, corruption or ethnicity on how they felt about their government? That's an interesting question. We didn't in this survey because we did in a previous one and we didn't find that it had a strong independent effect and I think that that's because we found a fairly high level of religious devotion across the board. I could check back at the data and again send me a note and I will. We ask, asking something simple like mosque attendance doesn't work because you get essentially 50% of the population always attends mosques and then when you look at data you find that that's all the men. But no, we didn't, this time go deeper into religious devotion we did in a previous survey and it didn't seem to work much for us but I could take another look at it. Let me make this the last question from the gentleman there in the third row back. Thanks, Rick Hill with RTI. One of your early slides had to do with the increase in sensible support or approval of the Taliban. We tend to think of the Taliban as a push-ton phenomena and yet there were, you could see these pockets over in the east around Harat and then there was one that looked like it was south of Mazar al-Sharat which is primarily a Zara area where there was an increase in approval ratings or something like that for the Taliban. Any explanation for that? Well I think some of this could be incident specific or related as Tony might put it to kinetics but in a place like Wardak province for example there's a lot of transport, at least there has been a transport route on the way to Kabul and I think a lot of violence there and what I've heard anecdotally is that there's been some resentment about the directed toward US and allied forces in terms of the violence that's been occurring along the highways in Wardak and things like that can increase or tip support in another direction and that may be an example in a place like Harat I think there's been some activity recently with the guys at D3 and in fact the field report talked about police officers there some defecting to Taliban and others being arrested. So the best I can say is that these can be incident specific or related to transient events but they certainly do seem to be real and I think if you dig in you can find supporting reasons or rationales for it. I think you also need to be extremely careful unless I'm mistaken the last time we released any of the data which showed the growth of presence as distinguished from levels of violence was in 2006 as the situation deteriorated people stopped releasing any of the maps which showed the increases of influence outside the main combat area so one of your problems is that there are indicators which go into more detail a significant push towards the north and the west but they are of support areas and supply areas not of high levels of violence and aside from one United Nations map which I think probably got declassified by mistake it would be extraordinarily difficult to find this out because effectively we have stopped providing transparency on this war since the situation began to go sour Ladies and gentlemen let me ask you to thank Gary and the usual matter