May 7, 1998
 
 

MEMORANDUM FOR:             RAYMOND J. UHALDE
                                                      Acting Assistant Secretary for
                                                          Employment and Training
 

                                                              / s /
FROM:                                         JOHN J. GETEK
                                                      Assistant Inspector General
                                                          for Audit

SUBJECT:                                 Profiling JTPA's AFDC Participants
                                                      Final Audit Report No. 06-98-002-03-340

The attached subject final report is submitted for your information. Because the report is informational only, it does not include recommendations and requires no response. If you request additional information or would like a personal briefing on the report's contents, please let me know.

Attachment
 

 

Acronyms and Abbreviations                                                                                               ii

Executive Summary                                                                                                                 1

Background                                                                                                                                4

Audit Objectives                                                                                                                        6

Chapter 1:           Training and services provided to the AFDC
                          recipients while enrolled in JTPA Title II-A.                             7

Chapter 2:           Program outcomes for AFDC recipients.                                        13

Chapter 3:         Characteristics of JTPA's AFDC participants                                23

Appendix I -        Scope and Methodology                                                                      29

                          Statistical Sampling Methodology                                                      30

Appendix II -       ETA's Response to Draft Report                                                       31
 
 

We performed an audit of the Job Training Partnership Act (JTPA) Title II-A program's efforts to serve the Aid to Families with Dependent Children (AFDC) population by evaluating services received and outcomes obtained for AFDC recipients who terminated from the JTPA program during the period July 1, 1995 through June 30, 1996.

Our purpose was to establish some benchmarks which the Employment and Training Administration (ETA), Office of Inspector General (OIG), and others could use to evaluate future programs' progress in serving AFDC participants.  We:

• identified the characteristics of the AFDC recipients enrolled in JTPA Title II-A;
• identified the type of services JTPA provided to AFDC recipients; and
• determined program outcomes for AFDC recipients.

Services provided to AFDC recipients (See chapter 1.)

Of the 76,246 participants:

• 42,488 (56 percent) received JTPA-funded training (occupation, nonoccupational, or both); 
  
  
• 42 percent (32,238) occupational training;
• 14 percent (10,250) nonoccupational training; 
  
  
• 19,005 (25 percent) appeared to have received some type of training (occupational, nonoccupational, or both) from other funding sources (e.g., JOBS, Pell grants);
  
• 14,753 (19 percent) received no training from any source; 
  
  
• 774 (5 percent) received some JTPA-funded supportive services;
• 13,979 (18 percent) received only objective assessment.

AFDC status:  Of the 76,246 participants, 52,238 (69 percent) were not receiving AFDC at the time of our audit fieldwork. However, the reason these participants' AFDC was stopped does not appear to be because of the participants' post-JTPA earnings. Of the 52,238 who were no longer receiving AFDC, 15 percent had no earnings, and 44 percent had no earnings or earnings of less than $5,000 for the four quarters following termination.

Earnings capacity:  For the four quarters following the quarter of JTPA program termination, 23 percent of all AFDC recipients had no earnings, 40 percent had no earnings or earned less than $2,500, and 54 percent had no earnings or earned less than $5,000. Participants who received JTPA-funded occupational skills training tended to have more earnings in the year following the quarter of termination from the JTPA program than those participants who received only nonoccupational training or objective assessment. Furthermore, participants who received nonoccupational skills training to address reading and math skills deficiencies earned more than those who did not receive the training.

Participants' attachment to the labor market: Of the 76,246 participants, 29,875 had wages in all four quarters following JTPA program termination. Almost half (13,637) of these participants had wages with only one employer, and 66 percent (19,685) had wages with only one or two employers. Of these 19,685 participants, 73.5 percent received occupational training. These 19,685 participants represent 26 percent of the AFDC participants terminated during our audit period.

Placement rates into unsubsidized employment: Of the 76,246 participants, the Service Delivery Areas (SDA) claimed placements for 38,364 participants (50 percent). We found 69 percent of the participants who received occupational skills training were placed, 39 percent of those who received only nonoccupational skills training were placed, and 4 percent of those who received no training were placed.

Characteristics of AFDC recipients (See chapter 3.)

The typical AFDC recipient enrolled in the JTPA program was an unemployed, single mother under 30 years of age with one dependent, little or no previous work history, a 12^th grade education or GED, and receiving between $250 and $500 per month in AFDC benefits.

The vast majority (76 percent) of JTPA's AFDC recipients had a high school diploma/GED education or higher, although the reading and math skills for these participants did not reflect the higher level of education; i.e, of the high school graduate/GED education level participants, 45 percent had math skills and 14 percent had reading skills below the 9^th grade level.

Further, 9 percent of the participants had no barriers, and 71 percent had two or fewer barriers to employment documented in the SPIR or SDAs' management information systems (MIS). 


 

ETA's response to the draft report did not refute any of the facts presented in the report. However, ETA did suggest some changes to report presentation including a better differentiation between the welfare participants served by the JTPA and the Welfare-to-Work (WtW) programs. ETA believes that because of the different types of activities allowed by the JTPA and WtW programs and because WtW is supposed to serve the hard-to-serve welfare recipients, the data presented in our report may not be adequate benchmarks to use to evaluate the WtW program's performance. ETA also suggested eliminating subjective statements from the report to allow the readers to draw their own conclusions.

We have considered ETA's response and have made some changes in the report based on their comments.

ETA's entire response to the draft report is presented at appendix II, page 31.
 
 

On August 22, 1996, the President signed the "Personal Responsibility and Work Opportunity Reconciliation Act" (PRWORA), a comprehensive welfare reform bill. PRWORA established the TANF program to supersede three programs:

• Aid to Families with Dependent Children (AFDC)
• Job Opportunities and Basic Skills (JOBS) Training Program
• Emergency Assistance (EA) Program

Pursuant to Secretary of Labor's Order No. 4-75, the Assistant Secretary for Employment and Training has been delegated the responsibility to carry out the WtW policies, program, and activities for the Secretary.

Congress has authorized for Fiscal Years (FY) 1998 and 1999, a total of $3 billion for the WtW program. Consequently, ETA's FY 1998 budget includes approximately $1.5 billion for the WtW; approximately $1.1 billion for formula grants to states, with the remainder being awarded through competitive grants. A similar amount will be available for FY 1999.

Before the enactment of the WtW program, Title II-A of the JTPA was the primary DOL program used to provide training and services to disadvantaged adults, including AFDC recipients. Section 201 of the JTPA identifies the purposes and establishes a set of goals toward which programs should be designed.

It is the purpose of this Act to establish programs to prepare youth and adults facing serious barriers to employment for participation in the labor force by providing job training and other services that will result in increased employment and earnings, increased educational and occupational skills, and decreased welfare dependency, thereby improving the quality of the work force and enhancing the productivity and competitiveness of the Nation. [Emphasis added.]

The DOL has not established specific performance measures for directly evaluating the JTPA program's progress towards accomplishing the goal of "decreased welfare dependency"; however, employment and earnings measures at followup have been established for adult welfare participants. ETA's position is that employment and earnings are key to whether there is a decreased welfare dependency. This audit does not establish performance measures for evaluation purposes. Nor does the audit determine whether the Service Delivery Areas (SDA), states, or DOL's performance is good or bad in accomplishing this goal.

For reporting program accomplishments, DOL established a uniform management information system, the Standardized Program Information Record (SPIR), to track individual participant's results. The SPIR includes participant characteristics, types of training and services received, and participant outcomes for all participants terminated during the program year (PY). The SPIR began tracking participant data during PY 1994. PY 1995 was the first complete year; therefore, not all problems had been resolved with the new system for our audit period.

PY 1995 SPIR data indicated approximately 636,000 participants were reported terminated from all JTPA programs of whom over 223,000 were enrolled in the Title II-A program. Approximately 35 percent (77,000) of those were receiving AFDC at the time of their JTPA enrollment. Selections of our audit sites and participants' samples were based on this information. (See appendix I, page 30 for audit sampling information.)
 
 

This audit focused on training and supportive services and the impact those services had on AFDC recipients enrolled under JTPA Title II-A.

Our audit objective was to provide the Employment and Training Administration with information on the JTPA service providers' efforts in serving the AFDC population by establishing certain benchmarks which ETA could use to evaluate future programs' (both JTPA and WtW) progress in serving TANF recipients. We understand that the welfare participants to be served by the WtW program are supposed to be the hard-to-serve, whereas JTPA may serve any welfare participant. Also, JTPA can provide a variety of training and supportive services to participants before the participants find employment, whereas WtW's emphasis is on placing participants in jobs first, then providing services necessary to keep the participants employed. Nevertheless, we believe that the information presented in this report will provide some basis to evaluate not only future JTPA programs' efforts in serving welfare participants but also the WtW program's efforts.

We established these benchmarks by evaluating a sample of AFDC recipients who terminated from the JTPA program during the period July 1, 1995 through June 30, 1996. We:

• identified types of JTPA services provided to AFDC recipients (see chapter 1),
• determined program outcomes for AFDC participants (see chapter 2), and
• identified the characteristics of AFDC recipients enrolled in JTPA (see chapter 3).

For an explanation of the statistical sampling methodology, see appendix I, page 30.

 

While 62,267 participants received some services, JTPA-funded services were provided to only 43,262 participants. Conversely, 32,984 participants (43 percent) received no JTPA-funded services.

Figure Figure 1A at left shows the training received by JTPA's AFDC recipients by funding source.

We were able to identify some non-JTPA funding sources (JOBS, Pell grants, etc.) for 8,974 of the 19,005 participants whose training is shown as non-JTPA- funded.
 

However, because some participant files were either not available or the participant files were incomplete, we were unable to determine the specific source of funding for the training 10,031 participants received. Consequently, we have classified these 10,031 participants' training as non-JTPA-funded.
 

As Figure 1A shows, 76 percent of the participants (15,098 + 17,140) who received JTPA- funded training did receive occupational skills training such as classroom (CRT), on-the-job (OJT) and work experience (WE), while some also received nonoccupational training. However, these participants represent only 42 percent (32,238 of 76,246) of the participant population.

JTPA-funded occupational skills training provided
The specific types of JTPA-funded occupational skills training in which the 32,238 AFDC participants were enrolled are identified on figure 1B below.
 

JTPA-funded nonoccupational skills training provided

JTPA funded nonoccupational training for 27,390 participants (10,250 nonoccupational only; 17,140 nonoccupational and occupational). Some of these 27,390 individuals received more than one type of nonoccupational skills training since 44,237 nonoccupational skills training activities were identified. The types of nonoccupational skills training activities provided are shown on the following table 1C by participants' educational levels.
 

During our audit fieldwork, we interviewed SDA officials and staff to determine what they perceived as problems in implementing the new WtW grants to serve TANF recipients. Many of the respondents thought that WtW would inherit most of the hard-to-serve welfare clients with barriers that the states' TANF agencies had failed to place into jobs. The most significant economic and social barriers identified by the majority of SDAs included:

• lack of transportation
• lack of child care
• drug/family abusers

• lack of job motivation, or desire to work
• lack of soft skills (ability to function in the work world and social skills)
• lack of basic skills (reading, writing, etc.)

Nonoccupational training given to address math/reading skills deficiencies

We have identified all participants who had math and reading skill levels below the 10^th grade level and determined if the nonoccupational training they received addressed these skill deficiencies. We found that 36,951 of the 76,246 participants (49 percent) had math skills deficiencies while 16,915 (22 percent) had reading skills deficiencies. We then analyzed the following specific and related nonoccupational training activities that could possibly address these skill deficiencies:

• reading skills training (specific)
• math skills training (specific)
• basic skills training (related -- may address both skill deficiencies)
• GED training (related -- may address both skill deficiencies)

As figure 1D shows, only 13,835 of the 36,951 participants (37.4 percent) with math skills deficiencies received nonoccupational skills training to address the deficiencies. In fact, only 1,680 (4.5 percent) got specific math skills training.
 
 
 
 
 
 
 
 
 
 
 
 

Nonoccupational training given to address certain barriers

In addition to addressing math and reading skill deficiencies, we identified participants with certain barriers which we believe may be addressed by nonoccupational skills training. In addition to those participants with barriers identified in the SPIR, we also included those participants with these barriers we identified through audit examination of participant records. We determined:

• Of the 1,731 participants with limited English speaking skills, 302 (17 percent) received English as a second language training.

 
• Of the 7,081 offenders, 1,049 (15 percent) received either job search assistance, pre-employment skills, or life skills training.

 
• Of the 6,391 displaced homemakers, 1,426 (22 percent) received job search assistance or pre-employment skills training.

Other nontraining services provided to JTPA's AFDC recipients

Of the 14,753 participants who did not receive any training from any source, 13,979 did not receive services other than objective assessment.
 

Furthermore, it does not appear that skills deficiencies/barriers to employment that may be addressed by nonoccupational skills training are being sufficiently addressed, particularly since SDAs indicate they see this as a major problem in the upcoming WtW program for which the goal is to move hard-to-serve TANF welfare recipients into and keep them in lasting unsubsidized employment. We recognize that training under WtW is restricted to post-placement services only. However, if WtW's goal is to move individuals into and keep individuals in lasting unsubsidized employment, basic skill deficiencies and barriers to employment must be addressed.
 

• AFDC status at a specific point in time after termination from the JTPA program, 
  
  
• post-JTPA earnings for the four quarters following the quarter of termination, and 
  
  
• attachment to the labor market (by showing number of employers with whom the participant had wages in the four quarters following termination).

Finally, we will compare earnings of those participants whose math and reading skill deficiencies were addressed by nonoccupational training with those whose needs were not addressed.

Participants' post-JTPA AFDC status

During our audit fieldwork between October and December 1997, we confirmed the welfare status of the 76,246 participants. We found that 52,238 of the 76,246 participants (69 percent) were not receiving AFDC at the time of our confirmation process. Our confirmations occurred between 15 to 18 months after the end of PY 1995. Therefore, our results do not show any participants who may have stopped receiving AFDC after JTPA termination but who were again receiving AFDC at the time of our audit, nor do the results show those participants who were off AFDC at the time of our confirmation but later may have started receiving AFDC again.

We were not able to determine why the participants were no longer receiving AFDC, only that they were not. There could be many reasons why the AFDC was terminated. However, it does not appear the AFDC was terminated because of the participants' post-JTPA earnings. Of the 52,238 who were no longer receiving AFDC, 15 percent had no wage record for the four quarters following termination; 31 percent had no earnings or earnings of less than $2,500; and 44 percent had no earnings or earnings of less than $5,000.

The following discussion on JTPA's AFDC recipients' post-JTPA earnings is further evidence that earnings do not appear to be the reason these participants stopped receiving AFDC.
 

Our audit results show that for the four quarters following the quarter of termination, 23 percent of all AFDC recipients enrolled in JTPA had no earnings at all, 40 percent had no earnings or earned less than $2,500, and 54 percent had no earnings or earned less than $5,000 (see table 2A). Our audit results also revealed that participants who received JTPA-funded occupational skills training tended to have more earnings in the year following the quarter of termination from the JTPA program than those participants who received only nonoccupational training or objective assessment.

The following table 2A shows the individual levels of earnings for all participants. Subsequent tables show earnings levels by type of JTPA services received -- occupational training (table 2B); nonoccupational training (table 2C); and objective assessment only (table 2D).

As shown on table 2B above, approximately 19 percent of AFDC participants who received JTPA-funded occupational training had no reported earnings for the year immediately following their termination from the JTPA program, 30 percent had no earnings or annual earnings under $2,500, and 42 percent had no earnings or annual earnings under $5,000.
 

As discussed in chapter 1, the JTPA program did not always provide nonoccupational skills training to address math and reading skills deficiencies. We identified 36,951 participants who had math skills deficiencies of whom 13,835 received direct or related nonoccupational skills training to address the deficiencies. We also identified 16,915 participants with reading skills deficiencies of whom 5,814 received direct or related training to address the skills deficiencies. Figures 2F and 2G demonstrate that, where nonoccupational training was provided to address the math and reading skills deficiencies, the participants' earning capacity was higher than those who did not receive the training.
 

As figure 2F shows, 30 percent of those who received math-related training earned over $10,000 as opposed to 22 percent for those who did not receive such training; 47 percent of those who received the training earned more than $5,000 as opposed to 39 percent for those who did not.
 
 
 
 
 
 

The ultimate goal of the WtW program is to move individuals into and keep individuals in lasting unsubsidized employment. To evaluate the JTPA Title II-A program's success in accomplishing this goal for AFDC recipients, we analyzed the number of different employers the AFDC recipients had wages with in the four quarters following JTPA program termination. The following table 2H displays the results for the 58,878 participants (of 76,246) who had wages in the four quarters following termination:

The following table 2J breaks down those 19,685 participants who had only one or two employers (from table 2I) by type of services the JTPA program provided to attempt to identify the types of services that positively impact employer attachment.
 

Participant placement rates into unsubsidized employment

We evaluated the SDAs' reported participant placements to compare participant placement rates by type of services received. Of the 76,246 participants, the SDAs reported placements for 38,364 participants (50 percent). We found that 31,689 of these placements (83 percent) were for participants who received occupational skills training, 5,918 placements (15 percent) were for those who received only nonoccupational skills training, and 577 (2 percent) were for those who received no training. 

• family status,
• educational/skills levels,
• previous employment history/labor force status,
• documented employment barriers, and
• amount of AFDC benefits.

The vast majority (76 percent) of JTPA's AFDC recipients had a high school diploma/GED education or higher, although the reading and math skills do not appear to reflect the higher level of education. Further, one would expect the hard-to-serve to face multiple barriers to employment, yet 9 percent of the participants had no barriers, 33 percent had one or no barriers, and 71 percent had two or less barriers to employment documented in the SDA's MIS or the SPIR.

Family status

As table 3A below shows, 81 percent of the participants were single parents.

We found that the AFDC participants' educational levels achieved were not necessarily commensurate with math and reading skill levels.

Table 3B below shows that 76 percent of the AFDC participants had either a high school diploma/GED, or higher, educational level. In fact, approximately 32 percent had post-high school education.

As shown in table 3C and figure 3D, 45 percent of the high school graduate/GED educational level participants had math skills below the 9^th grade level; 71 percent were below the 12^th grade level. These participants fared somewhat better with reading skills; 14 percent read below the 9^th grade level, and 41 percent read below the 12^th grade level.
 
 
 

The audit looked at the participants' work history for the 3-year period prior to enrollment in the program (per audit review of participants' files) and labor force status at time of JTPA enrollment (per SPIR). The following graphic displays both. 

Documented participants' employment barriers

Although participants enrolled in the program were to have multiple barriers to employment, the SPIR and SDA documents did not always address what the individual barriers were. Of the 76,246 participants, there were no barriers identified for 6,605 (9 percent) even though many participants had no recent work history, had poor math and reading skills, and were not high school graduates. 


 

Amount of AFDC benefits

The following table profiles the JTPA AFDC recipients by amount of AFDC benefits the participant received at time of enrollment into JTPA.

The audit scope included Title II-A participants identified in the PY 1995 SPIR database as receiving AFDC benefits at the time of enrollment into JTPA and who terminated the JTPA program during the period July 1, 1995 through June 30, 1996. Only participants terminated from the II-A program were included as part of the universe, except participants in the II-A 8 percent and Section 204d (older worker) programs were excluded. SDAs included in the audit sample were:
 
 

• Analysis of the SPIR database for the following purposes: 
  
  
• determining the SPIR's accuracy, and
• establishing the audit universe and selecting participant samples. 
  
  
• Selection of a random sample of 97 participants at each SDA. 
  
  
• Reviews of sampled participant case files maintained at the SDA offices. 
  
  
• Interviews of SDA staff. 
  
  
• Analysis of sampled participants' incomes 12-15 months after JTPA program termination. 
  
  
• Analysis of participants' welfare status at time of fieldwork. 
  
  
• Interviews with reported placement employers verifying placement information, current employment status.

Sampling Frame: The universe for this audit covers PY 1995 (July 1, 1995 to June 30, 1996), and includes ETA's SPIR records of AFDC recipients who terminated within this program year. Data from the SPIR indicated there were 640 SDAs reporting information for PY 1995. We excluded Alaska, Hawaii, Puerto Rico, and all trust territories from the audit universe. Therefore, the universe consisted of SDAs within the 48 contiguous states plus the District of Columbia. After the exclusion, only 619 SDAs remained.

The audit universe consisted of 77,026 participants identified in the SPIR as being AFDC recipients. We selected a random sample of 1,164 participants. We found 10 duplicate records and 9 non-AFDC participants in our sample. These 19 invalid AFDC records represented 780 participant records in the SPIR. Therefore, our revised 1,145 participant sample results were projected to a revised population of 76,246 JTPA Title II-A AFDC participants. The statistical data presented in this report relate to the revised population of 76,246 participants.

Sampling Plan: We used stratified, two-stage cluster sampling to draw the sampling units from the sampling frame keeping in view the availability of the limited resources and time constraints. In order to draw a representative sample, the universe of ETA regions (1 through 10) was separated into four strata.

Strata No.         ETA Regions         No. of SDAs

1                         1, 2, and 3                 141
2                         4, 6, and 8                 220
3                         5 and 7                      168
4                         9 and 10                     90

Each of these four strata was then divided into two sub-strata according to the number of AFDC recipients reported. SDAs having more than 500 AFDC recipients formed sub-strata 1, and SDAs having less than 500 formed sub-strata 2.

Sampling was a two-stage process. In the first stage, a sample of SDAs was selected from the two sub-strata created earlier, using probability proportional to size sampling. It was decided by the management team to select a total of 12 SDAs. Two SDAs were drawn from sub-strata 1 and one from sub-strata 2. A total of three SDAs from each of the four strata were selected. The first sub-strata included 16 SDAs while the second sub-strata included 603 SDAs.

In the second stage, the sampling units (AFDC recipients) were randomly selected from each of the selected SDAs (primary units). A sample of 97 recipients from each of these SDAs was selected randomly for a total sample size of 1,164 AFDC recipients.

This sample size was based upon 95 percent confidence level and +/- 5 percent sampling precision for the estimates, taking into consideration the design effect of this sampling plan.
 

Assistant Secretary for 

APR 2 8 1998
 
 

MEMORANDUM FOR:         JOHN J. GETEK

SUBJECT:                             Profiling JTPA's AFDC Participants Draft Audit
                                                  Report No. 06-98-002-03-340

This is in response to the subject report that was issued to my office on March 23, 1998.

Thank you for the opportunity to comment on the subject report. Also, we would like to express our appreciation for involving our staff in discussing the findings as you proceeded. The report itself provides some interesting highlights.

At the same time, I have the following suggestions and concerns:

1.  The methodology utilized is missing from the introduction.  It would be helpful for the reader to know and understand how samples were selected, pulled, etc. A brief up front section on methodology would be of assistance. Note: it is acknowledged that you make reference to the sampling methodology being in Appendix 1, on page 5.

2.  While we do not disagree with the conclusion reached on page two of the executive summary with regard to the participant's attachment to the labor force, it is suggested that you remove interpretative language in the last sentence of this paragraph, i.e., "While this ... participants." This would limit the executive summary to factual statements.

3. The individual folders of the participants should be reviewed to ascertain whether the finding of nine percent of the participants having no barriers is true. On the surface, it is difficult envisioning persons on AFDC without barriers. This seems like a documentation shortcoming. If you go to page 26, one learns that approximately 77 percent had multiple barriers.

4.  The AFDC recipients under JTPA and WtW are likely to be very different since the TANF recipients covered under the WtW initiative are supposed to be those most in need. Under the JTPA program, 76 percent have either diplomas or GEDs (see page 2). Beyond the characteristics, the remedies available under JTPA vs WtW are different. It may not be possible to utilize these findings as a baseline when considering the WtW recipients. In a similar manner on page five, reference is made to establishing certain benchmarks which ETA could use to evaluate future programs. To make the report more creditable, it is suggested that you acknowledge the differences between the two programs.

5.  The standards for the JTPA system are intended to serve as a proxy for decreased welfare dependency; however, the language on page four of the background statement takes issue with that
basic premise. We do not disagree that an index has not been established to serve as a measure, but employment and earnings are key to whether there is a decreased welfare dependency.

6.  In a similar manner, we are not in large disagreement with the shake down period which occurred with the program year 1994 data; nevertheless, the implication of the third paragraph on page four is non sequitur.

7.  With regard to questions posed to SDA officials on perceived problems under WtW grants, it would be helpful to either separate those out of the body of the report or eliminate them, since the report is factual for the most part. Statements such as the second paragraph on page 8 are out of context and represent opinions rather than facts.

8. It is unclear what the summary on page 11 is attempting to say. As you know, other sources of training and support that have traditionally been available will no longer be available. Many SDAs have operated the JOBS program for years; however, there is no JOBS program. Furthermore, the remedy of training
under WtW is restricted to post employment services only. The conclusion that skill deficiencies to employment are not being addressed and then relating the initial statement to the upcoming WtW program needs to be clarified.

9. On page 12, the report again moves to embrace perceptions vs facts, i.e., appears. Striking statements which are not completely factual; or forming factual statements would leave your report less subject to criticism from an outside reader. You may want to re-read the document for subjective statements. In contrast statements such as ... our audit report reveals that

participants who received JTPA-funded occupational skills..., give the reader a firm base for drawing their own conclusions.

10. On page 22, there is a subjective statement on "creaming" in the second paragraph. As noted earlier on the JOBS program, the AFDC population often passed through other programs consistent with figure IA. These individuals were on an upward bound track to continue their achievements. Many received their
high school diplomas as a precursor to JTPA. There is nothing wrong with an SDA moving them from a more elementary to a more skilled program. The academic levels of welfare recipients as a universe are 42 percent with less than high school and 58 percent with equivalency or beyond.

11.  At the same time, it is implied that the educational levels are irrelevant when you look at achievement levels, i.e., 45 percent of HSG/GED participants had math skills below the 9th grade and 71 percent below the 12th grade.

I hope these comments are helpful as you revise the report.