National Coalition For Child Protection Reform - Responds to Statements From National CASA Association and Caliber Associates

National Coalition For Child Protection Reform - Responds to Statements From National CASA Association and Caliber Associates
Posted: 31 Dec 2009 09:52 PM PST
NATIONAL COALITION FOR CHILD PROTECTION REFORM

53 Skyhill Road (Suite 202) / Alexandria, Virginia, 22314
Phone and Fax: (703) 212-2006 / e-mail: info@nccpr.org / www.nccpr.org
For release: For further information,
contact:
Immediate Richard Wexler, Executive
Director (703) 212-2006 /
www.ncpr.org

NCCPR RESPONDS TO STATEMENTS FROM NATIONAL CASA
ASSOCIATION AND CALIBER ASSOCIATES

ALEXANDRIA VA. (June 23) – Richard Wexler, Executive Director of
the National Coalition for Child Protection Reform, issued the following
statement today in response to documents released by the National CASA
Association and Caliber Associates:

The National CASA Association has launched a desperate campaign of damage control in an effort to spin a study commissioned by the group itself. But no matter how desperately National CASA tries to divert attention, the organization can’t evade the key findings from its own study:

Children with CASAs were nearly five times more likely to be in foster care than children without CASAs.

Yet children without CASAs were found to be just as well off – and just as safe – as children with CASAs.

Thus, we conclude that the only real accomplishment of CASA is to encourage the needless removal of children from their homes.

BACKGROUND:

At its annual convention earlier this month the National CASA Association issued a summary of the most comprehensive study ever done of the program. It was commissioned by National CASA itself and conducted by Caliber Associates.

NCCPR put out a press release contending that while CASA volunteers are dedicated and mean well, in general the CASA program does no good and may well do harm. The Pittsburgh Post-Gazette wrote a story about the study.

National CASA subsequently released the full study, though it is hard to find on CASAs website. National CASA also put out a statement and Caliber put out a response to the NCCPR press release. Links to both were included in a weekly news summary on the website of the National Center for Adoption Law and Policy. NCALP broke with its longstanding practice of using these e-mails solely to send links to news stories in order to rush to CASAs defense.

THE BEST WAY TO EVALUATE THE CASA STUDY: READ IT.

Before getting into specifics about the comments from National CASA and the researchers, however, NCCPR suggests a simple way to settle the matter: People should read the full study for themselves and draw their own conclusions. National CASA has posted the study on one of its websites – though it requires an extremely diligent search to find it. (If the study is as favorable, or at least as neutral, as CASA claims, it’s hard to understand why CASA doesn’t make the study easy to find by posting it prominently on both of its websites). At the moment, the study can be reached directly via this link:

http://www.casanet.org/download/casasurveys/caliber_casa_report_representation.pdf

RESPONSE TO THE STATEMENT FROM CALIBER

The firm hired by National CASA to do the study, Caliber Associates, has prepared a “response” to a press release from NCCPR.

In the document, Caliber responds to our contention that the study does nothing to improve the lives of children and may do harm, by declaring that “as far as we are concerned the most critical facts about the well being of children with and without CASA volunteers” are that the children most in need of a CASA got one and that these children are at very high risk of poor outcomes “when they first enter the program.”

NCCPR’S RESPONSE: We don’t dispute either of these points. But neither tells us if the CASA, once assigned, does any good. We believe that other study findings indicate that, in general, CASA does no good and may do harm. Caliber complains that NCCPR didn’t note that children and parents who received CASAs got more services.

NCCPR’s RESPONSE: There is no indication that the increased services actually improved outcomes. Indeed, the same study also asked caseworkers to assess the percentage of parent and child needs met in cases with and without CASAs. They found “no significant difference,” suggesting that the additional services didn’t do any good. Caliber omits this from its response to NCCPR.

The researchers quote the following from NCCPR’s original press release. “Most important, there was no difference in ‘exposure to violence and maltreatment.” In “response” Caliber takes this quote out of context, implying that it is a reference to the children’s backgrounds before any intervention. In fact, this statement specifically refers to the children’s status after intervention,
comparing children whose intervention included a CASA and those whose intervention did not include a CASA. The results are shown in the study itself, in Table 26, and they do, indeed, show no difference in exposure to violence and maltreatment.

Caliber’s comments here are related to the straw CASA has been trying to grasp since the report’s findings became public: the claim that all the differences are due to the fact that CASAs handle more difficult cases. But Caliber went to enormous lengths to adjust for this in order to come up with an apples-to-apples comparison. On page 40, the report lists eight separate variables for which they adjusted. The report then states: “Inclusion of these variables means that the percentages and mean scores presented in this section indicate the outcomes that would be expected of children who did and did not have a CASA volunteer if the two groups had similar demographic characteristics and prior experiences.” [Emphasis added].

The authors of the report also include a section speculating that, because the results were so surprising they must not have done a good enough job in adjusting for severity. But this is circular reasoning. The argument, essentially, is that “the results are so bad for CASA that we must have measured wrong.” They also cite “tests” they performed in an effort to confirm this. Again, NCCPR suggests that readers examine the report for themselves to evaluate the plausibility of this argument.

We believe the more plausible explanation is that the results reflect racial and class bias built into the CASA model itself. And rather than suggesting flawed methodology, the size of the differences suggests the extent of the bias. Even if one believes that the comparison between the two groups is not perfect, is it implausible that such imperfection would account for the fact that the children with CASAs were nearly five times more likely to be in foster care than the
children without CASAs, and yet there was no difference in safety and well-being outcomes.

Furthermore, if Caliber’s explanation is to be believed, why doesn’t it apply to the so-called “good news” about CASA? Why isn’t the fact that children who have CASAs get more services also simply a function of their cases being more severe?

Please note that NCCPR included Caliber’s explanation, and our response, in our original press release. Caliber, in contrast, has taken NCCPR’s release out of context.

RESPONSE TO COMMENTS FROM NATIONAL CASA

In an unsigned statement, attributed to CASA CEO Michael Piraino, found via a link on the website of the National Center for Adoption Law and Policy, CASA relies almost exclusively on the claim that all negative or “neutral” findings can be attributed to the fact that CASAs handle more severe cases. The statement does not acknowledge the strenuous efforts of the researchers to control for this.

The statement also makes some interesting comments on some other issues:

Concerning the surprisingly low number of hours per month CASAs report spending on their cases, CASA claims that the volunteers actually spend more time, but don’t bother to write it down “…and the last thing we want to do is turn [volunteers] into data input people.”

That’s exactly what poorly- functioning child welfare agencies say when issues are raised about whether caseworkers actually performed required tasks – such as visiting children. “Oh, we’re sure they did it,” we are told. “They just didn’t have time to write it down.”

This also doesn’t explain why, according to the report “cases involving African- American children were associated with over an hour less volunteer time each month…” [emphasis added].

It seems unlikely that volunteers dealing with cases involving African-American children consistently were less likely to fill out time logs. Furthermore, this section of the CASA statement contains a significant factual error. The statement claims the researchers found only that workers spent less time per child in cases involving African-American children. It goes on to
speculate that maybe these families had more children. The CASA statement then poses the question: “Is there a difference in time spent per case?” [Emphasis in original].

But the study itself already supplies the answer, and the answer is yes. Indeed, the study refers to less time per child and less time per case in cases involving African-American children. For example, on page 22, the study says: “The mean number of hours spent on African-American children’s cases was 2.67 versus 4.30 for children of other races.” [Emphasis added]. The CASA statement also goes on to cite a different study, a “National CASA Consumer Satisfaction Survey.”

But the CASA statement leaves out some important findings from this survey, which is available here: http://www.casanet.org/download/casasurveys/CS-survey- final-report-09-03.pdf

· The study included a number of open-ended questions, in which
respondents had to write comments rather than check boxes. In their responses, so
many caseworkers singled out concerns about class and cultural bias in CASAs
that the researcher felt compelled to acknowledge it as a “theme” in survey
comments.

As the author acknowledges, this was not a study of a random cross-section of CASA programs; rather these programs volunteered. That makes it likely that these programs are better than typical CAS A programs.

People who have bad experiences with a program generally are less likely to respond to surveys about it. Thus, those who didn’t fill out the survey forms probably would have given CASA lower ratings. This is especially significant in light of the low response rate from parents, compared with other groups.

Even with all these factors biasing the study in favor of CASA, one of the two questions where, comparatively, CASA “consistently scored low” concerned CASAs’ objectivity.

Though it is not statistically significant, birth parents gave CASAs lower ratings on every single question except one. The one exception is that birth parents say CASAs have more influence on the court -- hardly a compliment.

If CASAs truly were objective, views of birth parent attorneys and children’s attorneys should be similar. In fact, birth parent attorneys gave CASAs lower ratings on every question except two – and most of the time, these differences are statistically significant.

The survey asks respondents to give an example of something a CASA did that was helpful. There is no request for an example of something a CASA did that was harmful. (Although it is interesting to note that on the “something helpful” question enough people specifically wrote the word “nothing” as opposed to just leaving it blank, for this to emerge as a “theme.”)

CONCLUSION

The two studies combined are a clear indictment of the CASA model. But rather than heed the warnings, CASA is burying its head in the sand. Of course some individual CASAs do some good for some children. There are undoubtedly some good CASA programs. And, as we emphasized in our original statement, we do not question the dedication or motivation of CASA
volunteers. But the two studies combined suggest that, on balance, CASA does more harm than good and needs radical reform. At a minimum, these studies should be a wake-up call for the constituency that, according to the “Consumer Satisfaction Survey,” appears least willing to show any skepticism about CASA: Judges. They need to be far more willing to question CASAs closely about how they reach their conclusions and far less prone to rubber-stamp CASA
recommendations.