which is responsible for an increase or decrease in utilization of the G.I. Bill. To the extent that the efforts of the VETS programs are resulting in enrollment of those who would not normally be influenced by VA activities, then any increase can be attributed to the VETS program. Since one VETS program premise is that there is a pool of veterans who are "turned off" by the VA but can be assisted through peer counseling, it is assumed for evaluative purposes that this is the case.

(2) In cities where the VA was operating for a year prior to the start of the VETS program, the base rates derived from the July-December 1970 separation group may have been influenced upward. As a result it is more difficult for these cities to reflect an increase than it is for cities where the earlier VA activity did not exist. Again, if the VA activities do not influence the same group on which the VETS programs are targeting, the influence on the statistical data should not be major.

It will be noted that there are both increases and decreases in utilization rates in both target and control cities. Obviously, it is unlikely that the efforts of a VETS center could cause a decrease in utilization. Other factors, especially the economy and employment situation in an area, of course impact on utilization rates. In Seattle, for example, the staff of the VETS center recognized that at least some veterans were enrolling in VA programs as a means of getting some money to live on when they were unable to find a job. As the employment picture brightens, as it has in Seattle, some veterans can be expected to find jobs rather than enter school. The extent to which such factors influence participation rates is unknown at this time. It is clear, however, that there are fluctuations in the utilization rates for which the VETS programs cannot take either credit or blame.

Definition of program goals in particular cities clearly had an influence. It is quite clear from the statistical analysis that, as a whole, the VETS target cities have not out-performed the control cities as measured by percent increase/decrease in G.I. Bill utilization. Nor, overall, has there been a dramatic increase in G.I. Bill utilization rates in the VETS target cities. However, in several VETS cities, there were substantial increases in G.I. Bill utilization, consistent with the outreach goals in the particular cities. Conversely, in several VETS cities, where outreach goals were clearly defined, there were no appreciable and consistent changes in G.I. Bill utilization rates, when examined in the light of the goals of the VETS program and the performance of the particular control cities.

In addition, there were clear regional trends in G.I. Bill utilization, which influenced the comparisons. Perhaps the most dramatic of these were the increases in utilization rates in Brooklyn and Newark. In Brooklyn, for example, there were substantial increases in utilization among almost every ethnic and educational subgroup. Yet, other than the VA's own outreach efforts, no known organization is of sufficient size to have affected the utilization rates in the manner described. The same is true in Newark. Although Newark is a VETS city, the increases in utilization rates must be examined in light of the size of the VETS program, its outreach goals, and the increase found in Brooklyn, a city geographically very close to Newark. When such an examination is made, it is clear that the scope of the Newark VETS program is not great enough to have effected the substantial changes seen among veterans in Newark. The presence of what appears to be forceful regional trends in VA utilization makes it more difficult to assess the impact of the VETS program.

In light of the above discussion, it is felt that the most productive way of presenting the statistical results would be on a city-by-city basis, with heavy emphasis on changes in G.I. Bill utilization seen in the context of the goals of the particular VETS program. The 11 VETS programs that are the subject of this evaluation do not form an homogeneous group. Each program is to some extent unique, and the range of differences among the

11 programs examined is very great. There are differences in the size of the program, the organizational affiliation, and the outreach goals. All of these differences impact on the utilization rates.

PERFORMANCE OF INDIVIDUAL VETS CITIES

The report on each VETS city consists of three parts. First, the changes and patterns of G.I. Bill utilization in the VETS target city and its control city, as shown by the statistical analysis, will be outlined.

Next, the size, organizational affiliation, and philosophy of the particular VETS program will be described. Mention will be made of the educational opportunities in the city, the relationships existing between the VETS program and other organizations in the city, and the emphasis within the program other than signing veterans up for educational benefits. As noted earlier, the information contained in these descriptive narratives was obtained through a questionnaire, and from an on-site visit early in 1973. Notes have been added to provide updated information furnished by the National VETS staff.

Finally, the performance of each city is evaluated, on the basis of the size and orientation of the program and the trends apparent in the control city.

The data for all target and all control cities for the 1970/71 period are summarized in Table 1. The corresponding data for 1971/72 appear in Table 2. The data for all target cities without Los Angeles and all control cities except Brooklyn are given in Tables 3 and 4. The overall results shown in these four tables are discussed in the section following the presentation of results for the individual cities.

A word is necessary about the textual treatment of the results of the statistical analysis. The tables present complete data on each target city and on each control city. In each table there are 48 increases or decreases in utilization rate. Since it would be impractical to describe all of these in detail, only highlights and selections from the full set of data are discussed.

The evaluators used two criteria for selecting the results to be treated in the text. First, since the VETS program is directed at educationally and economically disadvantaged veterans, results from the restricted area of the target city and its corresponding control city are usually given, and non-high school graduates are usually highlighted. In addition, since the VETS program is concentrated on minority veterans, results for Black and Spanish-surnamed veterans are usually included in the results and discussion sections.

The second criterion used for selecting results to be treated textually was the emphasis of the particular VETS program. Since some programs, for example, concentrated their efforts on Chicano veterans, the evaluation emphasizes Spanish-sumamed veterans. In short, the report focuses on results that make sense in terms of the program under examination.

The data, however, did have impact on the results presented. As mentioned previously, since small numbers produce unstable utilization rates, alternative data were presented in these cases. When, for example, the restricted area of the city did not have sufficient numbers to produce stable rates, the city limits area was used in the text.

The tables presented with each VETS city indicate the size of the veterans population and the full set of results. Highlights of the results are presented graphically for important subpopulations in each VETS and control city in Appendix A.

42-710 - 74 - 3

Table 2

Table 3