Figure 5: Elementary and Secondary Education: Half of On-Budget Education Dollars

Fiscal Year 1997

Other ($5.1 Billion)
7.0%

Research ($15.9 Billion)
21.8%

Elementary and

Secondary ($36.6 Billion)
50.1%

Postsecondary ($15.4 Billion)
21.1%

Our next chart gets to the issue of the targeting of the programs and really the heart of my testimony today. Yet this is only an illustration of some of the multiple programs that exist. Most of the Federal money is targeted to specific groups, but many education programs administered by different agencies can target the same group or attempt to provide the same service. We have comprehensive figures, this kind of information on all of the programs. What we have had to do each time someone wanted to know is go out and do a study. Teacher training programs, how many programs are there? How many agencies are responsible?

Figure 6: Three Target Groups Served by Multiple Programs and Agencies

So just to illustrate, we have put three targeted groups on this chart, and as I say, it is simply illustrative it is not comprehensive of the multiple programs, and you have described some of these briefly.

For example, in the center circle for at-risk and delinquent youth, as you have indicated, we have found 127 programs in 15 different departments and agencies.

Similarly for at-risk and delinquent youth, these are programs that might provide any one of a wide range of services for this group-substance abuse prevention, violence prevention, counseling, a variety of services. But the Department of Health and Human Services provides 59 of those, and Justice 22. So those, along with a few programs from Labor and Education, account for about 77 percent of these programs. However, you do have another 11 departments that are providing something targeted at that group.

For the ones where we were able to obtain some data on the amount of the program funds targeted for this group, we were able to count about $4 billion in fiscal year 1997 for 110 of these programs. The Labor Department spends the most money on the programs, that is, $2.2 billion.

Let me turn to teacher training on the left. We know this is a crucial need in this country, to improve the skills of the teachers. We looked at this for fiscal year 1993, and we found 86 programs across 9 agencies, and there the dollars were smaller, but still the 42 main programs used funding of $280 million. Often what we have to caveat all this. We think we have found the programs. We perhaps did not find them all. Again, this is how much money we are able to count. There is probably more and we cannot find it all right off.

On early childhood, as you mentioned, also we found over 90 programs there in 11 agencies in fiscal year 1992 and 1993. As with many other types of programs or target groups, one disadvantaged child might be eligible for as many as 13 different programs, and that eligibility could change from 1 week to the next because of the very complex eligibility requirements. So if they move from one place to another, they become eligible for another program or not eligible. But at the same time, most of the programs reported that they were not able to serve the entire population that might benefit from their program. Again, the dollars here, more than $3.5 billion.

Our concern is that this multiplicity of programs scattered over many different departments produces a very substantial potential for inefficient service and for ineffective service delivery. We believe these should be closely coordinated, at least because the uncoordinated program efforts we believe can waste scarce funds, it confuses and frustrates the people who are trying to use the programs, and it really limits the overall effectiveness of the Federal effort. So I think the idea of coordination and consolidation has merit. I will say some more later in my statement about some opportunities that now exist to look in a cross-cutting way to consolidate or make some changes. At a minimum, the coordination needs to be better than it is now among these programs.

Senator FRIST. Thank you, Dr. Joyner.

So that we may vote, the committee will stand in recess for approximately 15 minutes. Thank you for an excellent presentation, a great platform for what is to come later this morning.

We stand in recess.

[Recess.]

Senator FRIST. Dr. Joyner, thank you. We will be able to go straight through now, and you may continue with your testimony. Thank you.

Dr. JOYNER. Thank you.

The second general topic that I would like to address is the additional information for planning, implementation, and evaluation that is needed by agencies and the Congress about these multiple programs, these on the chart and the many others. The large numbers of programs and agencies involved make this information even more crucial.

Agencies need reliable, timely program performance and cost information and the analytic capacity to use this information when they get it. We also think it is important to think about information needs in terms of the multiple levels of need for information or levels at which information might be obtained. And the first of these would be to find out what program approaches really work, what models for delivering a particular service to a particular target group really work better than others and under what circumstances do they work.

At the next level-the program level-though, which is the level at which the Congress makes funding decisions, one needs to know whether a program such as Safe and Drug-Free Schools and Communities Act, whether those programs are working nationwide, whether we are getting what we hoped to get from them.

And then at the third level-and this is the one where I appreciate the interest your committee is showing to it-we have moved up from any one specific funded program to overall the level of Federal effort and an overall mission. Let's say we are trying to prevent substance abuse in children, and we have a whole set of programs, each of them funded separately. What are we getting out of this set of programs to address this important issue or this important target group? We believe that there is not as much information as there should be at any one of those levels about most the Federal spending on education.

We do have some information about the pre-kindergarten through high school programs about what seems to be not working as well as we had hoped and some that seem to be showing promise. You will be hearing more later about compensatory education programs, for example, Title I. The concern there that, as it has been implemented and funded in the past, it has not shown the kind of result people would like to have in closing the gap between disadvantaged children and their more advantaged peers.

In our studies at GAO, we have often tried to find out what seems to be working or what information is needed, and we have looked at several different areas. We have looked at promising practices or approaches in the areas of school safety, preventing violence in the schools, substance abuse prevention, and school-towork transition are some examples, and we also in 1995 issued a report looking more broadly at successful schoole what hawa atud

es shown, and that includes a bibliography of some key studies looking at what seems to make for successful versus unsuccessful school as a whole.

Some of the themes that have come out of this are the need for a clear and comprehensive approach, a clear sense of mission about what the program is intended to accomplish, strong program leadership, and often a linkage between the program and the community, the community and the family, especially, as a key part of that larger community.

The Education Department, as I said, has used the research on Title I in part to identify where there are schools in high-poverty areas that are performing well, what seems to characterize those, and, while they understand that this is limited in its representativeness, there do seem to be some indications of some characteristics of those schools in their initial study that were more high performing. They also looked at ten promising alternatives for Title I practices in another study identifying, for example, the Success for All Program and Comer School Development Programs.

Your next panel could talk a bit more about some details of what seems to be working in some Title I schools.

There are some significant information gaps I wanted to focus on, though, at two levels. One is just the information about the programs themselves, where are they, what is happening in them, who is being served, and, at the other level, knowing what impact the programs are having, what success or lack of success. This was another theme from our report on successful and unsuccessful schools and workplaces: few evaluations of successful strategies exist and the lack of methodological rigor in much of the research that has been done.

We can speak of this more specifically on Head Start, for example. That is a program that is serving a very important target group that Congress has felt is important to the extent of substantially increasing its funding. The funding has doubled since 1990, up to $4 billion in fiscal year 1997. And during this period Head Start has received money to change its programs in some ways to try to improve the quality of the programs and expand. But we have found in a systematic review of the literature looking for impact evaluations, that the body of research available on Head Start really is inadequate for use in drawing conclusions about the impact of the current Head Start Program.

We believe that we really do not know enough about what is working now with the modern-day Head Start. The early research that was done over 20 years ago was not done on the Head Start Program that exists today. And we recommended that the Head Start bureau use some of its funding. With a $4 billion program, we thought it was worth spending some of their research funding to design some impact studies. They do not really think it is necessary, but have agreed to take a look at the possibility. That was the last we heard from them.

Now, it would seem obvious that information is good. Knowing where the programs are and what is happening is good. Knowing what impact it is having is good. So the question arises of why don't we know more. So a part of what I would like to point out

reasons why there are the information gaps that there are. A very fundamental one is competing priorities. This is a matter of balancing, for example, the desire to limit the burden-the regulatory burden on the local schools, school districts, and on the States, and reduce paperwork burden and promote flexibility in program implementation with the priority to obtain information that can be aggregated over multiple locations to let you know what the effect of the program is.

I could say more about that later, if you would like, with respect to particularly Safe and Drug-Free Schools. We have issued a report on that program just recently and we subtitled it "Balancing Accountability and Flexibility."

Another is the cost of data collection. This is not inexpensive. And neither is doing really good impact evaluations, which is more difficult and more costly than just obtaining data on program proc

esses.

Also when you look at all the education programs, education is not the primary role of some of them, or at least there are multiple roles. Head Start has more goals and a broader mission than just the cognitive development of children.

A key issue as well, we think, is the multiplicity of programs that in itself, makes it harder to know what the overall result is from the set of them. Another major issue has been that Federal programs have not had an emphasis in the past on results and accountability. That is where we think things have changed a bit in the last few years.

You may have seen the articles in the Post today about the report on the Department's strategic plans required under the Government Performance and Results Act. That is a major change over the last few years, putting in place really an overall statutory framework of which the Government Performance and Results Act is the centerpiece, along with the Paperwork Reduction and Clinger-Cohen Act as it relates to information technology, and also the Chief Financial Officers Act which has been expanded.

So in combination, there is now a structure to require better information, more reliable information and focus on results. And we think that does have some potential, very clearly, to provide information that the agencies need and that the Congress needs, because the Results Act, for example, requires the agencies to define their mission, to lay out their strategic goals, and how they are planning to achieve them. And then, as you know, beginning with the budget that will be submitted for fiscal 1999, agencies will move from the broad goals to specific performance plans and indicators for measuring their performance and linking it to their budget requests.

And particularly important in the world of education funding with the multiple programs is that they are required to coordinate across other departments with programs serving similar missions. They are supposed to show evidence that they have done that. Then, when a Government-wide plan gets produced, it will potentially to be able to address the cross-cutting nature of some of these missions. It has the potential, we believe, to allow the agencies to better understand for themselves the cross-cutting nature and move them toward better coordination But it also has the potential