Under Secretary of Defense David S.C. Chu Briefing on Armed Forces 2002 Sexual Harassment Survey
(also participating Timothy Elig and Rachel Lipari, Defense Manpower Data Center)
MR. CHU: Good afternoon, ladies and gentlemen. My apologies for being tardy. I'm, as you know, David Chu. I serve as the undersecretary of Defense for Personnel and Readiness in the Department of Defense. And I'm here this afternoon really to introduce the main speakers, Rachel Lipari and Tim Elig. Rachel is a co-author of what we are releasing today, which is the "Armed Forces 2002 Sexual Harassment Survey." And Tim helped supervise this and several other important survey efforts.
These survey efforts are crucial to the Department of Defense in understanding what our people think about the important issues in front of us, how they see our programs and policies. In some sense it's a way for them to tell us what's really going on out there that official statistics -- official reporting channels can't do. And that's particularly important in the area that we're focused on this afternoon, which is sexual harassment, inappropriate sex-related behaviors, including sexual assault, which was the subject of this morning's hearing.
Let me say just a word of background, if I might, about the survey, why we do it, what its purpose is, and where the department goes from here, especially back to the hearing subject this morning.
This is a congressionally mandated the survey. The law, as most recently revised, requires us to conduct this survey every four years. We did this survey in calendar year 2002, and we'll do the next one, therefore, four years from that point in time. It is one of the few -- in fact, it may be the only broad-based survey in this country that asks these kinds of questions in this way. And that's important, because one of the issues out there is comparisons. Well, how does the Department of Defense look relative to other parts of our society? And the short answer is, you really can't tell with the instruments we have at hand. That's one of the issues we'll work with our sister federal agencies over time to confront.
It's -- I would argue -- and granted I'm a little biased -- it's a distinguished instrument. It's benefited from over a decade's worth or work. As to how you ask these questions, these are sensitive subjects, people don't like to talk about these subjects. Women will in some circumstances, I'm told, not even agree it wasn't necessarily assaults. You don't start out, as some surveys do, by asking, well, were you assaulted. You start out by trying to bring the subject up in a more indirect way. And Rachel and Tim can speak to that in terms of how this survey was conducted.
We aimed at a sample of about 60,000 individuals in the military. We got better than a one-third response rate. So we have close to 20,000 responses, which gives us for most issues that the survey confronts a pretty good statistical base for estimating what the population as a whole looks like. And so, what's reported in this volume -- and this is really a preliminary printing of these materials. As part of this volume are population estimates based upon the samples that were taken, and then reweighted to get population estimates based upon the importance of each segment in the larger -- in the larger whole.
One of the powerful elements in what the department has done here is similar questions were asked in 1995. So we can compare trends that we're talking about. It's one of the benchmarks we can establish: how do we look today versus how we looked seven years earlier. A professional job. We are very grateful to our people for having responded, because it does take a little time to make these surveys out. This is our opportunity to give back to them, give back to you, some picture of how is the department performing in this important area.
I do want to say a word about this morning's hearing and the focus. The focus of this department is on the future. There was a lot of discussion this morning about the recent past. We regret every single offense that was committed, and we will do our best to corral the perpetrators and ensure that justice is done and the victims are properly cared for. But the focus of the department's effort, the task force the secretary directed be constituted -- and it's already doing its work in the Central Command area of operations even as I speak -- is on the future, is on how do we do a better job in caring for the victims of sexual assault, and how do we do a better job at preventing sexual assault from occurring in the first place. Those are the immediate and the longer-term objectives we have out there. But this kind of survey gives us an insight into what is happening, some insight as to why it's occurring, and specifically how well our training programs are working was one of the important issues that Rachel and Tim will cover.
I apologize. I will have to duck out in about 20, 25 minutes. But they are the experts, they know the answers, they wrote this report, and I think they'll give you a very good picture of what's going on.
So, Rachel, are you going to do the presentation? I'll give you the podium, so to speak.
MS. LIPARI: (To staff.) Next slide, please.
To measure sexual assault and sexual harassment in the military, we use a 19-item list of behavioral experiences that men and women indicate they experienced. This was developed by leading experts in the field of sexual harassment at the University of Illinois, Urbana-Champaign.
The 19 items can be broken down into four basic categories: sexual harassment, sexist behavior, sexual assault, and an "other" category.
Sexual assault, as we measure it, is actual or attempted sexual relations with you. We asked the women and men to indicate that someone attempted to have sex with them. We avoid the use of the word "rape."
Sexist behavior is verbal and nonverbal behaviors that are insulting and are based on gender, such as "People of your gender are not suited to the type of work you do."
Sexual harassment is composed of three subfactors: crude and offensive behavior, unwanted sexual attention, and sexual coercion.
Crude and offensive behavior is the verbal and nonverbal behaviors of a sexual nature that were offensive or embarrassing to the respondent. These include repeatedly telling jokes or stories of a sexual nature.
Unwanted sexual attention are attempts to establish a sexual relationship. For example, we asked if someone made unwanted attempts to stroke, fondle or kiss the respondent.
And sexual coercion is your classic quid pro quo.
Now as we calculate sexual harassment, we asked the members to indicate that they experienced one of the 12 sexual harassment behaviors and to indicate that they considered that behavior to be sexual harassment.
(To staff.) Next slide, please.
These are the components of sexual harassment. You'll note that, as Dr. Chu mentioned, we had to recalculate the 1995 rates. We improved the measure between 1995 and 2002. We reduced the survey burden by reducing the number of items from 25 behavioral items to 19. So all the numbers we present today have been recalculated to be as comparable to 2002 as possible.
And as you can see, the rates have gone down for women in each of the three categories that make up sexual harassment and for men in both crude and offensive behavior and in unwanted sexual attention. Sexual coercion already -- there was a low number of men responding that they experienced this. In the lower right-hand corner of the slide in front of you we provide the margin of error, so you can estimate how well we think these numbers -- how strong we think these numbers are.
Next slide, please.
Here are the sexual harassment incident rates for each service by year. As you can see, overall for women in the services, we've seen a 22 percent decline since 1995. The largest decline has been experienced by Marine Corps women, which has dropped 30 percentage points since 1995. And when we look at this by pay grade, there have been declines for women in all of the pay grades.
Next slide, please.
As I said, sexual harassment, we consider this part of 19 behaviors, broken down into four components. One of the components is sexist behavior, which is not considered part of our sexual harassment measure but is the leader behaviors. If you think of this as a continuum of behaviors, starting with sexist behavior as the least egregious, sexual harassment as egregious, and sexual assault as the most egregious, they represent our continuum.
As you can see, sexist behavior has declined for women in each of the services as well. And as I said before, this is treating someone differently because of their gender. Later on, Timothy Elig will show you a measure of gender discrimination, which is a similar concept. We consider sexist behavior to be attitudinal behaviors, non- verbal, non-verbal behaviors, whereas gender discrimination or sex discrimination are actions that could affect a person's career ratings, assignments, evaluation.
On this slide we give you the sexual assault numbers by service. As you can see there, there has been a decline since 1995, from 6 percent to 3 percent. The Army has experienced the largest decline, down six percentage points since 1995. And when we look at this by pay grade, there have been significant declines for women in the pay grades. For example, for junior enlisted we see the largest decline, down from 10 percent in 1995 to 5 percent in 2002.
When we measure sexual assault and sexual harassment, these unprofessional gender-related behaviors, we first ask the respondents to tell us all of their experiences, whether they've experienced any behavior during the past 12 months. And then we ask them to focus in on a particular incident that has occurred for them in these past 12 months and to provide details about this experience. This experience can include multiple behaviors and can go over multiple days, months, locations.
Next slide, please.
On this slide I give you some of the details about this one situation that the service members have experienced. Although most of women said that at least some or all of the behaviors are occurring on base, at work and during duty hours, the numbers have (not ?) declined since 1995 in the percentage saying this is the case.
Q (Off mike.)
MS. LIPARI: And when you look at the 2002 numbers and the 1995 numbers, and when you add up the red and the yellow bars, the majority of women are saying that these behaviors are occurring on base during duty hours and at work. But if you compare the percentage from 1995 to 2002, things have gotten better. Fewer incidences are occurring in the areas that the military has direct control over.
Next slide please.
To provide you a more detailed examination of where -- what kinds of behaviors are occurring in which places, we broke this out by the different types of behaviors that women experience. Let me preface this by saying that 57 percent of women have said they experienced behaviors that would fall into multiple categories -- the sexist behavior, the sexual harassment behaviors, or sexual assault.
On this slide here, I provide the sexist behaviors for those who only experienced sexist behavior, for those who only experience sexual harassment behaviors, and I provide the numbers of everyone who experienced sexual assault. Sexual assault does not occur -- typically does not occur without one of the other leader behaviors occurring as well. In our survey, only two people responded that they experienced only sexual assault. So for this slide I was not able to break that out for sexual assault only.
But when you focus on the women experiencing sexual assault, you'll see that fewer reported these -- that all of these behaviors are occurring on installation -- that's the 40 percent versus the 54 and the 60 -- at work and during duty hours.
Next slide, please.
Tim, would you like to take over? And then -- (off mike).
MR. ELIG: Following the Aberdeen incident, when the Army did the study of what was going on at that time, they found that there were -- other than sexual harassment, there seemed to be other problems that were more prevalent. They termed it at that time "gender discrimination." We've decided in this survey following up the mid-'90s period to explore this concept further, and we developed a separate measure based on a measure that was used in equal opportunity research to measure discrimination. This includes 12 items in three different areas: evaluation or items such as receiving a rating lower than they deserved because of their gender; having unjustified negative comments on their evaluation form -- again, that they said because of their gender.
In assignments we have areas such as the current assignment has not made good use of their skills because they were treated that way because of their gender. Or they did not get an assignment that they were qualified for and that was open to women -- they didn't get it because of their gender.
And in the career area we asked questions such as they did not learn until too late of opportunities that would have helped their career or they did not have a mentor to help them through career planning. And again, in each case they identified this is happening because of their gender.
If you recall the earlier findings on sexist behavior. Sexism is attitudinal, it's more prevalent. Fifty percent of the women said they had experienced sexist behavior, the attitude, comments about not being qualified to do their job. However, only 18 percent of women said that at least one of these types of discriminatory behaviors happened. We did not ask them if it was discrimination, so we don't have that same kind of breakout that we could do on sexual harassment. All they are telling us is that these types of behaviors happened.
The rates are much lower for men saying that these kinds of behaviors happened to them because of their gender.
In the area of training, 79 percent of the men said they had been trained in the past year; 77 percent of women said they had been trained. Training got good marks. At least 75 percent of women and men agree that their service training conveyed these types of things: a good understanding of what sexual harassment was, an understanding of what behaviors were offensive and could not be tolerated, that sexual harassment makes it difficult for service members to perform their duties.
In the area of being safe to complain, 76 percent of women, 83 percent of men did agree that training covered this. However, as you will see in the report, there is still a problem with those who said they were harassed are less willing to agree that they actually know how to complain, where to report.
Leadership is important. Even perhaps more important than training is that people understand and accept and know that their leaders are taking this seriously. As you can see in this slide, there was an improvement at all areas, from immediate supervisors up to senior leadership, the belief that they are taking this seriously. In all areas, women's evaluations increased by at least 10 percentage points.
The slides we have gone through here back up information that were in the undersecretary's written comments. There is additional information in the report.
There are questions? Thank you.
Q I'd like to ask Dr. Chu a question, if possible.
MR. CHU: Sure.
Don't go away -- (off mike).
MR. ELIG: We'll stay.
MR. CHU: Yes?
Q In the review that Secretary Rumsfeld ordered -- I understand it's not complete, but it is under way?
MR. CHU: Just started, right.
Q Has it become apparent yet the degree to which some in the military have been callous to sexual assault victims, the degree to which prosecutions have not been pursued against attackers, or victims not getting adequate care or being retaliated against?
MR. CHU: I think that's a multi-part question. On the more straightforward element, which is do we prosecute when we know if that's been committed, the answer is yes.
Now, the degree depends on the particular circumstances. On the more -- on the other elements of your question, which I believe were is there underreporting and how much, and how are victims cared for, that's one of the issues we want to look at, is are we doing a good enough job at providing the victim the support that she or occasionally he needs in those circumstances; how well are those programs going forward? Our policy is that you get that kind of care. We have a series of programs to deliver that care. The issue the task force has to look at is are those policies strong enough, are the programs good enough? To the extent there are shortcomings, what do we need to do to improve the performance of the department in this regard?
Q What I was trying to get at, basically, is from what you've seen so far, can you decide whether this is a bigger problem than you thought it was, whether it's a littler problem than you thought it was?
MR. CHU: It would be too early to say. They are just starting.
Q The congressional mandate for this report being done every four years, when did that start?
MR. CHU: Every four years -- that is with the 2003 authorization act.
(To Ms. Lipari.) Is that correct, Rachel?
It actually had been mandated to be administered annually, if I believe -- I remember correctly part of that. The department in -- but we did conduct it in '95, and then for whatever reason -- that's before my time, I should acknowledge -- the department didn't do it in the late 1990s.
Q So why between '95 and 2002 nothing?
MR. CHU: Well, I came to office and we confronted this because this is an important set of issues for the department. My instructions to poor Ms. Fites and company were to organize a stronger program of surveys across the board, not just in this area, and the deputy secretary was kind enough to provide the funding to do that. And one of the most important, of course, is this survey. The Congress, I think properly has put a greater period between surveys because these behaviors don't change rapidly over time. I'd like to believe that in months you could change these situations, but it does take time. And so I think four years is about right.
Now, the mandate extends both to the active and Reserve forces, and we'll do a Reserve forces sexual harassment survey this year.
Q With the same, going out to 60-some-thousand, do you think?
MR. CHU: I presume.
Is that right, Rachel, a similar --
MR. ELIG: About 65,000.
Q How will Reservists be -- how will they know that they've been selected? Is this by mail? Done by Web?
MR. ELIG: In similar methodology, they will receive letters. We use the address files from the DEERS, the Defense Eligibility and Enrollment Reporting System. We use change of address files. So all the notifications are by postal mail. They are sent a paper survey, but the letters give them a website where they can go and take the survey by Web.
One reason we do the survey over a period of four or five months is so that the mail can catch up to them, and in fact so that the mail can be forwarded to them if they are deployed, or wherever they are over the world, and they would still have time to take the survey during the fielding period.
Q Are you doing both Guard and Reserves?
MR. ELIG Yes. And it's a sample that is -- it's a stratified random sample designed to cover everyone who has at least six months of service at the time they take the survey, up to 0-6, that is excluding flag officers, generals and admirals.
MR. CHU: Sir?
Q Now, you guys have just compiled the statistics. Have you come to any conjectures or conclusions why, for instance, the Marines were higher than -- so much higher than the other services, or the Air Force was lower?
MR. CHU: Let me start and answer, then let the experts fill in, if I might.
Some issues we can come to conclusions on. And Tim touched on, I think, one of the most important, which is our people see the training they received and the training their commanders received as largely effective. And that's a good news story, because we've been investing heavily for some years now, as you heard in this morning's testimony, in training as one of the ways to change behaviors. Maybe we should do more -- that's back to your question, sir. Maybe we should do it differently -- that's all in front of us for the future.
In terms of why one service looks different from another, I'm not sure we have the wherewithal yet, but I'll turn to Rachel and Tim for their view. I am impressed that in terms of the improvement you see in the survey in all types of sexually inappropriate behavior, that we see this happening in every service. It means the across-the-board effort the department has made is paying off. It's not because one service did well and is pulling everybody along in its halo.
Tim? Rachel? Do you want to say something about -- (off mike)?
MR. ELIG: Part of the facts of life is that each of the services are unique in their composition. The Marine Corps is 70 percent first-term enlisted. So they are overwhelmingly 18, 19, 22-year old, and they've had much less time to be under the training of the service compared to the other armed services, who tend to be older, more experienced populations. That's all I can at this point say is -- demographic differences between the services is at least part of the story.
MR. CHU: But, as an ex-researcher, I would always add "further research is necessary."
MR. ELIG: We do make the data set available. It is used by any number of universities. We do fund some research by our colleagues at the University of Illinois at Urbana-Champagne, who for us look at antecedents and consequences of the behaviors. The instrument is designed to measure those things. We don't just go and ask did this happen to you, but we ask kind of what kind of experiences they've had prior and what their health outcomes might have been. We don't directly ask them to identify the health outcomes. That can be biasing. But by asking all the questions of the same people, you can study the interrelationships of the antecedents and potential consequences.
MR. CHU: Yes, ma'am?
Q Dr. Chu, you said your focus now is on the future. What do you do with this?
MR. CHU: Well, we use this for a variety of purposes. One purpose is to understand where we are relative to where we've been. A second purpose is to -- and therefore, where we need to go. In other words, where are the areas where we need a special emphasis? And this is broader, now, than sexual assault. It goes to all the behaviors encompassed here. These are all behaviors that we reject. I think that's the point I really need to start with. The reason we're doing a survey like this is these are behaviors that we condemn. And so our ambition is to root them out, to change attitudes, to change how people treat each other and to ensure that everyone is treated with the respect that he or she deserves.
What this helps us do -- and I think Tim's focusing on gender discrimination is very helpful in that regard -- it shows an area where we clearly have a long way to go still, in terms of what needs to be done. It might be the least serious, as Rachel points out, in the spectrum here, but it doesn't mean we shouldn't address it.
So that's the second major application of this.
Finally, it tells us something about what programs work and how much confidence our people have in those programs. That's very important to us, because, for example, if the training is not getting through, we clearly -- back to your question, sir -- have to do something about the training program. Well, the good news is, it is getting -- at least the majority, substantial majority, say it is getting through. "My commander gets it," you know.
Now that doesn't mean -- we still have people who are doubting, and we need to work on that segment and understand better -- that's the further research aspect -- why are they questioning, why are they less satisfied, why are they challenging us in terms of how effective we're being. But I think those are the kinds of applications you can make of these survey materials.
This is the first report. As Tim suggests, we'll do research. We'll use these data to understand better what we need to do as a department to improve further our performance in this area, in this set of areas.
Q How much did it cost to fund the report?
MR. CHU: I don't know, but not all that much.
Tim, do you have a cost estimate?
MR. ELIG: The postage alone, on a survey like this, is about $100,000, if you think of the number of times you have to contact people to remind them to take the survey and are contacting 60,000 people. The data collection, over and above the postage, would have been about another $250,000.
MR. CHU: I would emphasize, before we get too focused on exact dollar costs might be -- it's worth every penny to do this.
Now I was challenged this morning: Could we do it in a more high-end way, telephone surveys, instant analysis? Yes, you can. That's a much more expensive approach. We might do that for certain -- we do do that for certain particular subjects.
This is, as, I think, Tim's commented -- a survey that will yield dividends over a period of time here, as people exploit the richness of the questions asked. So that's one reason it takes a while to compile the answers. And I would say in fact we will speed up the tabulations next time around.
MR. ELIG: And it is very difficult to reach the military via telephone or -- any kind of instant survey is, you know -- we have a very mobile force, and if you do something very quick, then you can end up leaving out the very people you might most want to talk to, those who are being moved or deployed, or are away from home. So you have to be careful to properly represent the population, and not all of them have a chance to be heard.
MR. CHU: This will be the last question, because I do have to go.
Q Have you identified any aspects of training that might need to be improved upon, any specifics? I mean, 92 percent of the respondents said that behaviors that they learned from training is that behaviors that are offensive to others should not be tolerated. I mean, you have 127 reported cases out of Iraq and Kuwait. Is there any specific area that you've identified that would not have led to Iraq, for example?
MR. CHU: Not yet. I do think what you see is in the services -- and you heard this morning General Nyland's testimony on this -- a dedication to constant improvement. And so, the Marines are going to increase the periodicity in which people get training. I think one of the core lessons of this survey and the other indicators we have is it's the intervention with the junior force, especially junior enlisted force, that is most critical to getting these behaviors down. You look across the board in these behavioral areas, what you see is a pattern where it's worse in the junior ranks than it is in the more senior ranks. Some of that's the effect of age and general maturation, I grant that. But I do think some of it is -- you know, we do -- we do make a big investment in telling people, now, I don't care what you might have seen happen when you were a child, et cetera, this behavior is wrong, we don't do it, we won't tolerate it. And if you are a perpetrator, we will throw you out. In fact, we're might put you in jail first.
And that's a message you got to get through to people. It takes time to get through to people. And I think one of the issues that the task force is looking at is how we're being efficacious in getting that message through to people. Why do some people, as you point out, somehow not get the word and don't understand, or succumb to temptation, whatever the case might be? There are actually, when you get to these cases, all sorts of factors that lead to adverse outcomes.
One of the most important -- and the vice chiefs touched on it this morning -- is abuse of alcohol. Many of these incidents -- not so much in Iraq; we're not supposed to be drinking. But if you look at the other situations, these are often -- it's the Air Force Academy finding, as you know -- often occur in a situation where people have abused alcohol. And so, one of the things that our health community and our (inaudible) community will be doing is addressing the issue. We should -- we need to be doing that for other reasons, but I think there will be a payoff in this arena as well.
With that, let me thank you for your attention and your interest.
Q Thank you.
Copyright © 2003 by Federal News Service Inc., Ste. 220, 1919 M St. NW, Washington, D.C. 20036 USA. Federal News Service is a private firm not affiliated with the federal government. No portion of this transcript may be copied, sold or retransmitted without the written authority of Federal News Service Inc. Copyright is not claimed as to any part of the original work prepared by a U.S. government officer or employee as a part of that person's official duties. For information on subscribing to the FNS Internet Service, please visit www.fednews.com or call (202) 347-1400.