1998 REPORT TO CONGRESS: JUVENILE MENTORING PROGRAM -- December 1998

It makes sense that mentoring should help youth develop into caring, self-sufficient adults. Intuitively, we believe that mentoring works. There is substantial anecdotal evidence that mentoring has an impact on positive youth development. Research supports the notion that children may develop resiliency, at least in part, because of the personal connections they are able to make with significant adults in their lives. Logically, if mentoring can support positive youth development, then it also should support the goals of the JUMP program, namely reduction in delinquency and gang involvement, improvement in academic performance, and reduction in school drop-out rates. The reality, however, is that historically there has been limited empirical research to verify the effectiveness of mentoring.

The recently completed BB/BS study has begun to fill this need for empirical information and has generated considerable excitement about the evidence that mentoring holds promise as an effective intervention for youth at risk for future difficulties. In this chapter we discuss aspects of the JUMP projects that currently can be assessed, including youth and mentor feelings of satisfaction with the mentoring relationship, and whether or not each perceives any benefit to the youth as a result of participating in the JUMP project. JUMP grantees obtained feedback from youth and mentors using a standardized format to ensure consistency in reporting across sites. The following discussion is based on responses from 962 youth and 792 mentors. These responses form a substantial subset of the total number of youth and mentors represented in this report and can provide important information about the effectiveness of mentoring on which to further build the ongoing national evaluation. One cautionary note is important. As with all client feedback, this information was obtained from youth and mentors who voluntarily responded. There is no assurance that the respondents are representative of youth or mentors who did not choose to respond.

While both youth and mentors were extremely positive when rating various aspects of their mentoring experiences (Exhibit 6:1), mentor and youth perceptions of their relationships did not completely correspond. The fact that youth responses generally are more positive than the mentor responses may reflect actual perceptions, or may be an indication of a self-presentation response bias frequently found in research involving young people. Participation in an evaluation process may skew a child’s responses positively, especially when the child is being asked to rate his or her relationship with an adult.

Exhibit 6:1 Youth and Mentor Report of Satisfaction

ITEM	YOUTH RESPONSES (%)
ITEM	Not Very Much	A Little	Pretty Much	A Lot
How much do you like your mentor?	2%	5%	23%	70%
How well do you get along with your mentor?	1%	5%	26%	68%
How well do you feel your mentor understood you?	3%	9%	37%	51%
How helpful do you feel your mentor was to you?	3%	7%	26%	64%

ITEM MENTOR RESPONSES (%)

Not Very Much A Little Pretty Much A Lot

How much do you like your mentee? 1% 1% 22% 76%

How well do you get along with your mentee? 1% 2% 35% 62%

How well do you feel you understood your mentee? 1% 12% 51% 36%

How helpful do you feel you were to your mentee? 3% 22% 43% 31%

In spite of this apparent mentor/mentee disparity, when the youth responses were compared with their corresponding mentor responses, weak but positive correlations were present between youth and mentor ratings on items specifying how much they liked each other, how well they got along, and how helpful the mentor was to the youth.

Youth and mentor perception of benefits also correlated with one another, indicating that both may have perceived similar improvements within the youth. The following table summarizes the percentage of youth and mentors that report seeing improvement in specified risk areas (Exhibit 6:2). Each was asked to indicate whether they believed having a mentor helped a little, a lot, or not at all with each risk behavior.

Exhibit 6:2 Perceptions of Benefits Received

	A Little Bit	A Lot	A Little Bit	A Lot
	Youth Perception of Benefit		Mentor Perception of Benefit
Getting better grades	42.0%	49.1%	56.0%	30.1%
Attending all classes	23.3%	64.3%	45.2%	36.5%
Staying away from alcohol	13.2%	69.9%	40.1%	41.3%
Staying away from drugs	12.2%	71.0%	39.2%	42.7%
Avoiding fights	27.3%	57.4%	43.3%	41.5%
Staying away from gangs	15.6%	67.6%	37.5%	44.0%
Not using knives or guns	11.1%	68.0%	31.9%	47.1%
Avoiding friends starting trouble	27.5%	51.0%	47.8%	34.4%
Getting along with family	24.4%	61.6%	48.3%	40.2%

There were 463 (265 female and 198 male) instances where, using assigned identification numbers, information received from mentors and youth regarding perceived benefits could be combined with corresponding demographic and match characteristics. From these cases, an indication of the perceived satisfaction and benefits could be analyzed with relation to some youth and mentor descriptive characteristics.

It is interesting to note that the boys in the sample who were paired with female mentors reported both liking their mentor and feeling understood by their mentors equally with those paired with male mentors. However, those boys matched with a male mentor report greater benefits in some areas than those matched with female mentors. Specifically they report receiving greater benefit with respect to avoiding drugs and gangs. There also are marginally significant differences in reports that mentoring helped youth avoid alcohol and weapons. These preliminary results must be confirmed with more objective data, as the reports of perceived benefits by the youth cannot be taken as an undisputable measure of program effectiveness. However, it does suggest that there are differences between those boys matched with female as opposed to male mentors. There were only two cases of girls matched with male mentors within this smaller subset of data, so no similar analysis can be conducted for the girls.

As discussed before, about 40% of the youth are matched with a mentor of a different race. Given the paucity of scientific research on the mentoring relationship, it is not possible to predict whether race has any consistent bearing on the effect of the match or the possibilities for benefits. In the pilot data set, 170 youth had a mentor of a different race or ethnic identification, while 270 were of the same race or ethnicity. Again, youth assigned to multiple mentors were not included in this subset. On the self-reported description of the mentoring relationship, and the perceived benefits, there was only one item that showed any relationship with race-matching. Youth matched with a mentor of a different race reported liking their mentor more than those matched with a mentor of the same race.

While there was little difference among youth who were matched with same-gender or same-ethnicity mentors and the extent they reported liking their mentor or reported perceived benefits from the relationship, the same did not hold true for feedback from the mentors. Women paired with male mentees reported that they observed significantly less improvement in a number of areas than men paired with male mentees. The behaviors in which the greatest differences occurred between male and female mentors included:

Female mentors also perceived less improvement in the youth’s relationship with his/her family members.

When youth and mentors were of different race or ethnicity, the mentors also reported perceiving significantly less improvement in these same areas, and also in class attendance. Mentors paired with a youth of the same race or ethnicity reported that they believed they understood their mentee better than those involved in cross-race matches.

There are a number of complicating issues that must be considered when interpreting the race-match results presented above. The most important consideration should be the race of the youth. Recall that almost all white youth are paired with a white mentor. Thus, when splitting the cases into those with same versus cross-race matches, all but seven of the white youth are placed in the same-race category. Conversely, in this data set 52.2% of the non-white youth are in cross-race matches. Therefore, the differences observed may be the result of the stark demographic differences between those youth in same versus cross-race matches. It is therefore reasonable to repeat the race-match analyses presented above, excluding white youth. Such an analysis was conducted with results mirroring the results already presented. It would seem that our preliminary data indicate that cross-gender, and cross-race matches are perceived by the mentor as less beneficial to the youth. However, youth reports did not differ significantly regardless of gender and racial matching. Clearly, this will be an area for future research. It will be important to verify the self-reports of improvement with more objective measures. Further, more sophisticated research designs and analyses will be needed to control for potentially confounding variables.

Preliminary reports from youth and mentors leave us with great optimism for the promise of mentoring.

The preliminary reports from youth and mentors leave us with great optimism for the promise of mentoring. The challenge for OJJDP is to continue supporting the JUMP grant projects and the national evaluation of those projects long enough to learn which features of the projects and matching models are statistically related to successful outcomes. Our goal is to identify and share effective practices that can be replicated by community-based organizations across the country. Ultimately, through the addition of the protective factor of mentoring to support healthy youth development, we will be moving closer to reaching the JUMP program goals.

Previous	Contents	Next

1998 Report to Congress: Juvenile Mentoring Program		OJJDP Report December 1998