FOCUS: Results for Evaluation Outcomes and Longitudinal Surveys

Northpointe Institute for Public Management, Inc.

William Oliver, Ph.D.
Marcus Breitenbach
William Dieterich, PhD.

Focus Report

The results from analyses of data on the FOCUS mentoring program are presented in this document. It is designed to be a part of a larger report, which will include the description of the intervention program. The results are organized as follows. The sample is described in the first section. A second section provides descriptive statistics on the participation of mentees in the program. A third section presents results on recidivism, and addresses the crucial question whether the incidence rate for participants in the FOCUS group is lower than for subjects in control group. Statistical tests on the initial differences between the FOCUS and control group are also included in this section. A fourth section presents results on mentor involvement. A fifth section presents descriptive statistics for the items of a Life Skills Matrix questionnaire. The responses to these items augment the previous results as they add to the characterization of the sample and provide data on the ways in which the mentors interacted with the mentees. A final section presents results on changes in responses to the Life Skills Matrix items. The sample size was too small to obtain definitive results for this last section, but as more data are collected the same analytical methods should reveal changes that should help explain why recidivism is lower for the FOCUS sample.

Description of the Sample

The sample consisted of 22 subjects in the Experimental, or FOCUS, group who answered the main test instrument (Life Skills Matrix) at least once and an additional 22 subjects who served as a Control Group. Therefore, the total number of subjects in the sample was 44. All subjects were drawn from a jail population in Boulder County, Colorado. The FOCUS group was given an intervention in the form of mentoring [described at length elsewhere in the report] and the control group consisted of comparable subjects who applied to participate in the FOCUS program but could not be taken into the program because mentors were not available at the time.

Information about the control group came from official records in a Management Information System (MIS data) for Colorado that provided information about age, gender, and ethnicity along with information about charges, arrests, and sentences. Information about the FOCUS group came from the same MIS data and from two questionnaires: an initial needs assessment given to 11 of the 22 subjects when they first entered the study and a Life Skills Matrix questionnaire given one or more times to all 22 subjects during their participation in the study.

The average age for the control group was 33.5 years (SD = 11.0). Of the 22 control group subjects, 12 were female (54%) and 10 were male (46%). Information about ethnicity also appeared in the MIS data: 20 were white (91%) and 2 were Hispanic/White (9%).

The average age for the FOCUS group was 38.5 years (SD = 11.8).  Of the 22 FOCUS group subjects, 17 were female (77%) and 5 were male (23%); 17 were white (77%) and 2 were Hispanic (9%), 2 were black (9%), and 1 was other (5%). Table 2 displays some of these descriptive statistics along with statistical tests for differences between the groups.

The marital status of the 11 FOCUS group subjects who answered the initial needs matrix was 2 co-habiting, 2 divorced, 1 married and 6 single. Thus, 72% of the subjects did not have a partner. The highest education levels attained for these subjects were as follows: 3 subjects had some high school education, 1 subject had a GED, 2 subjects had a high school diploma, 2 subjects had less than two years of college, 2 subjects had more than two years of college, and 1 subject had a graduate degree Thus, the range of educational attainment was fairly broad; 27% of the subjects did not attain a high school degree (or a GED), 27% of the subjects graduated high school (or had a GED), and 45% had some college experience. The answers given to Life Skills Matrix questionnaire provide a more complete overview of the FOCUS group and are discussed in a later section of this report

Participation in the FOCUS Program

The average number of months that the mentees were in the program was 10.2 (SD = 8.80). Of these mentees, 6 completed the program, 8 were still participating in the program at the time the final data were collected (end of study for these analyses), and 8 voluntarily withdrew from the program. Hence, the attrition rate for this sample was 36%. The 8 mentees may have left for reasons beyond their control (e.g., moved elsewhere), not because they found the program inadequate. Table 1 displays summary statistics for the mentees broken down by program status at the end of the study. Note that several of the mentees who withdrew had been in the program for an appreciable length of time—the number of months in the program for these mentees ranged from 1 month to 10 months.

Table 1 Number of months in the FOCUS program as a function of program status at the end of the study.

 

Program
status
n

Months (mean)

SD

Completed 6

19.8

11.6

Continuing 8

8.1

3.3

Voluntarily Withdrew 8

3.3

3.4

 

Analysis of New Arrests—Does the FOCUS Intervention Reduce Recidivism?

The MIS records allowed the subjects to be followed from the day they entered the program until approximately the end of December 2011. These records included date-stamped entries on new charges and the disposition of those charges—whether the person was found guilty and the specific sentence. These data allowed us to count offenses that resulted in arrests that occurred after the person entered the program. The counts of these arrests for each subject provide outcome data to address the question of whether the FOCUS intervention leads to fewer arrests for the mentees than for subjects in a comparison group. The number of new arrests is considered an important measure of recidivism and is often preferred over other measures (Maltz, 1986).

When counting new arrests, it was important not to assume that each new charge corresponded to an arrest. In some instances, there were multiple charges for the same offense that resulted in an arrest. In these cases, the multiple charges were given the same dates in the database, and it was assumed that only one arrest had occurred.

A natural control group was provided by people who applied to be in the FOCUS program, but could not be taken on because mentors were not available. The subjects in the two groups—those people who entered the FOCUS program and those who did not—were not selected to be different on any other variables. Therefore, although random assignment to group was not explicitly carried out, the assignment process was essentially random.

Data from county records were also available for the subjects in the control group. The numbers of new arrests that occurred between when they were considered for entry into the FOCUS program and the end of December could be computed for the subjects in the control group in exactly the same way that they were computed for the mentees in the FOCUS group.

For both groups of subjects, it was not possible to calculate the time that subjects were constrained from committing offenses that would lead to new charges. For example, the weekly progress reports by mentors indicate that some mentees were in jail for periods of time. Additional records on the movements of the subjects would allow us to control for “street time”. In the absence of these records, we treated street time as a random variable that would have approximately the same distribution for both groups.

It is important to note that both groups of subjects were followed for comparable lengths of time. In some instances, as noted previously, mentees in the FOCUS group left the program before the end of December 2011. For the analysis of outcomes, we nonetheless counted new charges on these subjects that occurred after they had dropped out and before the end of the study. All new charges for each mentee was counted between the FOCUS referral date and the end of December, regardless of whether the mentee had left the program before the end date. This method of counting was done in part to keep the groups comparable, but, more importantly, because an intervention such as FOCUS is designed to have positive effects after the intervention is stopped. These positive effects could come, for example, from improved attitudes on the part of the mentees and improved living conditions, especially if the mentors have helped the mentees find jobs or places to live. Thus, if FOCUS is an effective intervention, there should be fewer new arrests for the mentees in the FOCUS group than for the people in the control group.

There were initially 24 subjects who were identified to be in the control group. Two of these subjects were excluded because their referral dates were earlier by several years than the other subjects. The control group sample, therefore, consisted of 22 subjects.

Initial differences between groups.

Given the method of assignment of subjects to groups, there should be no important differences between the control and FOCUS groups on variables known to be predictors of recidivism. We follow the standard practice to look at possible differences between groups before carrying out further analyses.

Table 2 displays statistics for several important variables. The follow-up time variable represents for each subject the time in days between the FOCUS referral date and the final date of the study. The numbers of previous charges were available for all subjects and represent all charges that were filed for each person, including, in some cases, multiple charges for the same offense. The previous charges variable was extremely skewed, so a log-transform was applied before testing for differences between the groups. The means for ‘previous charges’ that appear in the table are reverse-transformed values of the means of the logged-variable and are more appropriate measures of central-tendency for the two groups than the means of the untransformed variables. To test for statistical differences between the groups, t-tests were carried out, except for the variable ‘proportion male’, for which a Chi-squared test was carried out instead. The number of cases (n) for all tests was 44.

Table 2 Proportion male and means for age, follow-up time (days) and number of previous charges for the control and FOCUS groups are displayed. The differences between groups are tested with t-tests except for proportion male, for
which a Chi-squared test was performed. The estimated measure of central tendency for previous charges was computed by reverse-transforming the means of the log-transformed variable.

 

Control

FOCUS

Statistic

df

prob

Proportion male

0.45

        0.23

1.62

1

0.203

Mean Age

33.5

38.5

-1.46

42

0.152

Follow-up time (days)

518

470

-0.488

42

0.628

Mean No. Prior Charges

11.6

11.4

-0.042

42

0.967

 

The FOCUS group included proportionally more males than did the control group. In addition, the FOCUS group had a lower mean age and slightly fewer previous charges on average. In addition, the subjects in the control group were followed on average for more days than were the subjects in the FOCUS group. The directions of these trends might lead one to expect that the subjects in the control group would recidivate at higher rate than the subjects in the FOCUS group because high rates of recidivism are often associated with young, male offenders who have many prior charges. Note, however, that none of the differences between groups are statistically significant at the conventional significance level of 0.05. The observed differences could easily arise from sampling variation. We will nonetheless try to control for the differences in the analysis of recidivism.

Results for new arrests.

The number of new arrests during the follow-up period was much larger for the control group than for the FOCUS group. There were totals of 26 and 9 arrests for the control and FOCUS groups, respectively.

All of the arrests for the FOCUS group were for misdemeanors or petty offenses (0%). In contrast, for the control group, 7 of the arrests were for felony offenses (27%). The differences in these proportions was not significant by an exact Fisher’s test, p = 0.153.

The numbers of arrests for each subject were analyzed in several ways. The mean subject frequency of arrest was 1.18 and 0.41 for the control and FOCUS groups, respectively. A simple t-test on the square-root transformed counts indicated a significant difference between the groups, t(42) = 2.08, p = 0.043. The square-root transformation is an appropriate transform for the count data in this study. The groups also differed on a nonparametric test, the Wilcoxon rank sum test, W = 318, p = 0.035.

These statistical tests do not take into account the follow-up time for each subject. One would expect subjects with longer follow-up times to have more opportunity to be re-arrested than would subjects with shorter follow-up times. Also, these tests are not designed specifically to deal with count data.

Poisson regression and negative binomial regression provide statistical methods that are appropriate for analyzing count data. In these types of regression, the (logged) counts are accounted for by other predictor variables while making reasonable assumptions about the underlying distribution for the counts.

As in the case of standard linear regression, independent variables in either Poisson or negative binomial models can be tested for statistical significance. The (logged) follow-up time can also be included in the model to account appropriately for the varying opportunity to be re-arrested. The model predicts more arrests for people who have been followed for longer periods of time than for people who have been followed for shorter periods of time. For more information on Poisson and negative binomial regression see, for example, Cameron & Trivedi (1998). For an example of their use in a study of recidivism see Markman et al. (2010).

A Poisson model was carried out on the arrest outcome data. The model included predictor variables for condition and (log transformed) follow-up time. Analysis of the deviance for the model indicated the presence of over-dispersion, c2(42)
= 90.0, p < .001. That is, the assumptions made by the Poisson model about the distribution of the errors were not met. Therefore, we applied the negative binomial regression model to the count data instead, as it is specifically designed to handle this type of data.

The assumptions about the distribution of errors for the negative binomial regression mode were met for the arrest data, c2(42) = 32.4, p =0.855. The model coefficient for condition (b = 1.36) indicated that the incidence rate of new arrests in the FOCUS group is 0.257 times the incidence rate in the control group, p = 0.034. Alternatively, the incidence rate of arrests in the control group is 3.89 times the incidence rate in the FOCUS group. In any case, the effect of the intervention is large.

An additional analysis was carried out that added the variables log-transformed gender, age, and log-transformed number of prior charges. Table 3 displays the model coefficients for this expanded model. With these control variables included, the effect of condition actually increased.  The model coefficient for condition indicates that the incidence rate for new arrests in the FOCUS group is 0.202 times the incidence rate in the control group. Reversing the comparison, the incidence rate for new arrests in the control group is 4.94 times the incidence rate for new arrests in the FOCUS group.

Table 3 Model coefficients for a negative binomial regression model of new arrest counts. See text for details.

 

Estimate

Std. Error

z value

Pr(>|z|)

(Intercept)

-8.476

3.552

-2.39

0.017

log(age)

0.515

1.049

0.49

0.624

gender (female)

-0.969

0.719

-1.35

0.177

log(priors+1)

0.465

0.304

1.53

0.127

Condition (FOCUS)

-1.598

0.652

-2.45

0.014

To make this effect of condition concrete, consider a hypothetical group of subjects who are randomly selected from this same population of subjects, all followed for the same period of time and not given the FOCUS intervention. If there were 100 arrests for these subjects, one would expect only 20 arrests if one could repeat the experiment with all of the subjects receiving the FOCUS intervention.

We did not attempt to break down the different types of arrests for further analysis because of lack of power. We would need more observations or data with a greater prevalence of arrests (larger effective sample size) to carry out such analyses. It is, however, striking that all of the felony arrests for the sample occurred in the control group, which is consistent with the results that show the FOCUS intervention leads to a lower overall new arrest rate.

Mentor Involvement

A questionnaire called the Weekly Progress Report was administered repeatedly to 20 of the 22 FOCUS group mentees over the period that the mentor and mentee were in contact. The date of eachadministration was recorded along with how many hours the mentee and mentor had spent together during the previous week. The mentor and the mentee gave independent estimates of the time they spent together. The estimates of hours spent together were answered as ranges (e.g., 1 to 3 hours), which were then converted to hours (e.g., 2) for the analysis.

Many of the questions in the Weekly Progress Report addressed the appearance and state of mind of the mentees. These questions often had many missing answers or were answered with little variation. For example, all responses were “Clean” to the item “How would you characterize your mentee’s physical state this week? Check all that apply.” These particular questions were mainly of interest to the FOCUS staff, and are not analyzed in this report.

The Weekly Progress Report was done on average 13.0 times for each mentee (SD = 11.5). The mentees participated in the FOCUS program for varying lengths of time, so a more meaningful measure is the average number of days in the program per Weekly Progress Report across mentees—the value for this estimate was 63.0 days per report. There were very large outliers, however, so the median value of 26.9 days per report provides a better estimate for the period between administrations of the questionnaire.

The average number of hours that the mentors reported spending with the mentee during the previous week was 2.2 hours (SD = 1.2). The average number hours that the mentees reported spending with their mentors was 0.50 hours (SD = 0.39).  These means differed significantly, t(19) = 5.57, p < .001. In addition, the correlation between the hourly estimates was negative (r = -0.26). Many of the estimates given by the mentees were either 0 or ‘less than an hour”, so it seems likely that they did not understand how the hours should be counted.

The average total number of hours that the mentors reported spending with each mentee was 26.5 (SD = 23.2). Across mentees, the values ranged from 3 to 79 hours with a median of 17 hours. These numbers of hours represent large commitments of time on the part of the mentors, especially when one considers that it is unlikely that all of the hours that the mentees and mentors spent together were recorded in the Weekly Progress Reports. The median value of 17 mentor hours per mentee is likely biased low and the true value could be much larger.

The amount of time that a mentor spends with a mentee could possibly predict positive outcomes for the mentee and reflect one way in which the FOCUS intervention is effective. In fact, the mean number of hours that mentors spent with the 16 mentees that were not re-arrested at least once was much larger (mean = 29.5) than the number of hours that the mentors spent with the 4 mentees who were re-arrested at least once (mean = 9.7). Because of the small sample size and the imbalance in the numbers of subjects in these two groups, this difference was not statistically significant, t(18) = 1.4, p = .179.

Life Skills Matrix

A survey entitled “Life Skills Matrix” was administered repeatedly to mentees beginning in January of 2011. The goal was to have the mentors and mentees fill out the survey every 2 months, so that changes in responses could be monitored. Figure 1 displays the distribution in the number of times the survey was administered. Of the 22 subjects, 6 filled out the survey once, 10 filled out the survey twice, and the remaining 6 subjects filled out the survey 3 or more times. This distribution of the administration of the survey makes it somewhat difficult to assess change in the responses. Of course, an assessment of change is not possible for the 6 subjects who filled out the survey once. For the other subjects, the 6 subjects who filled out the survey 3 more times would tend dominate the pattern of change.

Many of questions in the Life Skills Matrix instrument were adapted from questions in a Case Supervision Tool in the COMPAS software. The questions were grouped into the following assessment categories: Housing, Food, Identification, Transportation, Medical, Substance Use, Family Reunification and Relationships, Finances, Employment, Education/Skills, Social Support/ Personal interests/ Spirituality, Personal Goals and Values, Arrest/Recidivism, Mentor Involvement.

Tables for each item in the Life Skills Matrix appear below. The tables include the question and responses to the questions with the frequencies and proportions for each response. Each item has a short name given in the caption, so that it can be referenced in subsequent analyses.  The frequencies and proportions for the responses are broken down by session. Different subjects contributed data at the different times that the instrument was administered (sessions), so only the results for the first session provide accurate descriptive statistics for the entire sample. In the brief discussions for each assessment category below, the emphasis is placed on the frequencies for the first session.

Housing

The mentees were mostly located in Jail, Other, or Probation. Less than 30% of the mentees changed their residence while followed.  The living conditions remained fairly stable and conducive for success for most of the mentees; across sessions 20% of responses indicated unstable living conditions and 19% of responses indicated that conditions remained poor or were worsening. In general, the responses indicate that about 20% of the mentees perceived themselves as being housed in poor conditions.

Table 4 Item table for mentee location

Current location:

Day
Reporting

DOC

Halfway
House

Jail

Other

Probation

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

0

0.00

0

0.00

1

0.05

8

0.36

8

0.36

5

0.23

2

2

0.12

1

0.06

0

0.00

3

0.19

5

0.31

5

0.31

3

0

0.00

0

0.00

0

0.00

2

0.33

2

0.33

2

0.33

4

0

0.00

0

0.00

0

0.00

1

0.33

1

0.33

1

0.33

5

0

0.00

0

0.00

0

0.00

1

0.50

1

0.50

0

0.00

6

0

0.00

0

0.00

0

0.00

0

0.00

1

0.50

1

0.50

All

2

0.04

1

0.02

1

0.02

15

0.29

18

0.35

14

0.27

Table 5 Item table for change in residence

How often has the mentee changed his/her residence in the last two months (since last review)?
0 1 2+
Session Freq Prop Freq Prop Freq Prop
1 17 0.77 5 0.23 0 0.00
2 8 0.50 5 0.31 3 0.19
3 5 0.83 1 0.17 0 0.00
4 3 1.00 0 0.00 0 0.00
5 1 0.50 0 0.00 1 0.50
6 1 0.50 0 0.00 1 0.50
All 35 0.69 11 0.22 5 0.10

Table 6 Item table for stable living

Does the person have a stable living situation that is helpful for them to succeed?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

4

0.19

14

0.67

2

2

0.13

3

0.20

10

0.67

3

3

0.50

1

0.17

2

0.33

4

0

0.00

0

0.00

3

1.00

5

1

0.50

1

0.50

0

0.00

6

1

0.50

0

0.00

1

0.50

All

10

0.20

9

0.18

30

0.61

Table 7 Item table for improvement of stable residence

Has the mentee improved the stability of their residence in the last two months (since last review?)

Same/Negative

Worse

Improvement

Same/Stable

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.06

1

0.06

3

0.17

13

0.72

2

2

0.13

2

0.13

3

0.20

8

0.53

3

1

0.20

1

0.20

1

0.20

2

0.40

4

0

0.00

0

0.00

0

0.00

3

1.00

5

1

0.50

0

0.00

0

0.00

1

0.50

6

0

0.00

0

0.00

1

0.50

1

0.50

All

5

0.11

4

0.09

8

0.18

28

0.62

 

Food

Access to healthy food and the means to prepare the food, as indicated by responses to items in this category, was a problem for about 15% of the mentees.

Table 8 Item table for sufficient access to food

Does the mentee have sufficient access to food?
Not enough food Enough food but not healthy or fresh Acceptable (i.e. not rotten) and
enough food
Able to make choices about the type
and quality of food they are eating
On-going reliable access to good food
Session Freq Prop Freq Prop Freq Prop Freq Prop Freq Prop
1

0

0.00

5

0.24

3

0.14

7

0.33

6

0.29

2

1

0.06

1

0.06

3

0.19

2

0.12

9

0.56

3

1

0.17

0

0.00

2

0.33

1

0.17

2

0.33

4

0

0.00

0

0.00

1

0.33

0

0.00

2

0.67

5

0

0.00

0

0.00

0

0.00

0

0.00

2

1.00

6

0

0.00

0

0.00

1

0.50

1

0.50

0

0.00

All

2

0.04

6

0.12

10

0.20

11

0.22

21

0.42

Table 9 Item table for planning to buy food

If the mentee chooses, are they able to plan what to buy and prepare for food?

Never

Sometimes

Always

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

5

0.28

6

0.33

7

0.39

2

2

0.13

9

0.60

4

0.27

3

1

0.25

1

0.25

2

0.50

4

0

0.00

2

0.67

1

0.33

5

0

0.00

2

1.00

0

0.00

6

0

0.00

1

0.50

1

0.50

All

8

0.18

21

0.48

15

0.34

Table 10 Item table for able eat as prescribed

Is the mentee able to eat as prescribed for his/her health situation (e.g., dietary restrictions)?

No

Unsure

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

5

0.24

7

0.33

6

0.29

2

1

0.07

4

0.27

5

0.33

5

0.33

3

0

0.00

1

0.17

2

0.33

3

0.50

4

0

0.00

2

0.67

1

0.33

0

0.00

5

0

0.00

1

0.50

1

0.50

0

0.00

6

0

0.00

0

0.00

1

0.50

1

0.50

All

4

0.08

13

0.27

17

0.35

15

0.31

 

Identification

Only 4 of 20 subjects reported that they did not have adequate ID at the first session. Virtually all subjects reported that they were prepared to work legally (91%).

Table 11 Item table for current id

Current identification:
Has
no form of ID or has major limitations in obtaining ID
Has
some ID cards and papers, but needs to apply to get more
Has
all the necessary ID cards and papers for housing, employment, medical needs
and benefits
Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.15

1

0.05

16

0.80

2

1

0.06

2

0.12

13

0.81

3

2

0.40

2

0.40

1

0.20

4

0

0.00

0

0.00

3

1.00

5

0

0.00

0

0.00

2

1.00

6

0

0.00

0

0.00

2

1.00

All

6

0.12

5

0.10

37

0.77

 

Table 12 Item table for prepared legal work

Is the mentee prepared to work (legally)?

No

Unsure

Yes

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

1

0.05

20

0.91

2

0

0.00

0

0.00

16

1.00

3

0

0.00

1

0.17

5

0.83

4

0

0.00

0

0.00

3

1.00

5

0

0.00

1

0.50

1

0.50

6

0

0.00

0

0.00

2

1.00

All

1

0.02

3

0.06

47

0.92

Transportation
About 10% of the subjects initially did not have access to transportation and another 15% did not have access to reliable transportation. In later sessions, larger proportions of people reported having limited access to transportation.

Table 13 Item table for access to transportation

What is your mentee’s accessibility to transportation?
No
access to transportation and unable to afford
No
access to transportation, but able to afford
Limited
or unreliable access to transportation
Ongoing,
reliable access to transportation

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.10

0

0.00

3

0.15

15

0.75

2

2

0.12

2

0.12

9

0.56

3

0.19

3

3

0.50

0

0.00

0

0.00

3

0.50

4

0

0.00

1

0.33

1

0.33

1

0.33

5

0

0.00

0

0.00

0

0.00

2

1.00

6

0

0.00

0

0.00

1

0.50

1

0.50

All

7

0.14

3

0.06

14

0.29

25

0.51

 

Medical

A large proportion (55%) of the subjects reported that they had medical needs requiring constant medical services. About half of the subjects reported having some, but not complete, access to medical services. About 20% of the subjects reported problems with attending appointments, taking prescribed medications as directed. Many of the subjects received mental health treatment  (81%). About half of the subjects had problems with affording medical services. It is clear
that the FOCUS sample had medical problems, and that they would benefit from help with these problems.

Table 14 Item table for special medical needs

Does the mentee have special medical needs (mental health, substance abuse, anything that requires constant medical services)?

No

Yes

Session

Freq

Prop

Freq

Prop

1

10

0.45

12

0.55

2

7

0.50

7

0.50

3

4

0.67

2

0.33

4

2

0.67

1

0.33

5

0

0.00

2

1.00

6

1

0.50

1

0.50

All

24

0.49

25

0.51

Table 15 Item table for access to medical services

Does the mentee have access to medical services?

None

Some

All

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

0

0.00

13

0.59

6

0.27

3

0.14

2

0

0.00

13

0.87

2

0.13

0

0.00

3

0

0.00

3

0.50

0

0.00

3

0.50

4

1

0.33

2

0.67

0

0.00

0

0.00

5

0

0.00

0

0.00

2

1.00

0

0.00

6

1

0.50

0

0.00

1

0.50

0

0.00

All

2

0.04

31

0.62

11

0.22

6

0.12

 

Table 16 Item table for attending appointments

Does the mentee attend necessary appointments?

None

Some

All

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

0

0.00

3

0.14

12

0.55

7

0.32

2

0

0.00

4

0.25

11

0.69

1

0.06

3

1

0.17

1

0.17

2

0.33

2

0.33

4

0

0.00

0

0.00

2

0.67

1

0.33

5

0

0.00

1

0.50

1

0.50

0

0.00

6

0

0.00

1

0.50

1

0.50

0

0.00

All

1

0.02

10

0.20

29

0.57

11

0.22

Table 17 Item table for take prescribed medicine

Does the mentee take prescribed medications as directed?

None

Some

All

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

3

0.14

5

0.24

12

0.57

2

1

0.07

1

0.07

2

0.13

11

0.73

3

1

0.17

0

0.00

1

0.17

4

0.67

4

1

0.33

0

0.00

0

0.00

2

0.67

5

0

0.00

0

0.00

1

0.50

1

0.50

6

0

0.00

0

0.00

1

0.50

1

0.50

All

4

0.08

4

0.08

10

0.20

31

0.63

Table 18 Item table for interest in pursuing medical services

Is the mentee interested in pursuing medical services (medical, dental, reproductive or general health care)?

No

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3.00

0.14

11.00

0.52

7.00

0.33

2

0.00

0.00

9.00

0.64

5.00

0.36

3

0.00

0.00

3.00

0.50

3.00

0.50

4

0.00

0.00

2.00

0.67

1.00

0.33

5

0.00

0.00

2.00

1.00

0.00

0.00

6

0.00

0.00

2.00

1.00

0.00

0.00

1

3.00

0.14

11.00

0.52

7.00

0.33

Table 19 Item table for affording medical services

Is the mentee able to afford the medical services he/she wants or needs?

No

Some

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

7

0.33

6

0.29

5

0.24

3

0.14

2

5

0.36

5

0.36

2

0.14

2

0.14

3

0

0.00

2

0.33

1

0.17

3

0.50

4

1

0.33

1

0.33

0

0.00

1

0.33

5

1

0.50

1

0.50

0

0.00

0

0.00

6

1

0.50

0

0.00

1

0.50

0

0.00

All

15

0.31

15

0.31

9

0.19

9

0.19

Table 20 Item table for receiving medical treatment

Is the mentee receiving mental health treatment?

No

Yes

No Need

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.19

13

0.81

0

0.00

2

3

0.23

10

0.77

0

0.00

3

2

1.00

0

0.00

0

0.00

4

1

1.00

0

0.00

0

0.00

5

0

0.00

2

1.00

0

0.00

6

1

1.00

0

0.00

0

0.00

All

10

0.29

25

0.71

0

0.00

Table 21 Item table for regular participation in mental health treatment

Is the mentee participating regularly in mental health treatment?

No

Yes

No Need

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

6

0.35

11

0.65

0

0.00

2

2

0.17

10

0.83

0

0.00

3

2

1.00

0

0.00

0

0.00

4

2

1.00

0

0.00

0

0.00

5

0

0.00

2

1.00

0

0.00

6

1

1.00

0

0.00

0

0.00

All

13

0.36

23

0.64

0

0.00

Table 22 Item table for interest in mental health treatment

Has the mentee expressed an interest in mental health treatment?

Not
interested

Interested,
but not seeking help

Looking
for treatment

Receiving
treatment

N/A

No
Need

Session Freq Prop Freq Prop Freq Prop Freq Prop Freq Prop
1

3

0.14

0

0.00

1

0.05

12

0.55

6

0.27

2

1

0.06

3

0.19

1

0.06

9

0.56

2

0.12

3

2

0.40

1

0.20

0

0.00

0

0.00

2

0.40

4

1

0.33

0

0.00

0

0.00

0

0.00

2

0.67

5

0

0.00

0

0.00

0

0.00

2

1.00

0

0.00

6

0

0.00

0

0.00

1

0.50

0

0.00

1

0.50

All

7

0.14

4

0.08

3

0.06

23

0.46

13

0.26

 

Substance Abuse. A small number of subjects were required to abstain from alcohol. Virtually all subjects reported that they had been drinking in the last two months, although only 70% reported that they had been drinking three or more times. No subjects admitted to having used illegal drugs in the last two months. A third of the subjects were receiving substance abuse treatment. Approximately, two-thirds of the subjects expressed an interest in substance abuse treatment and were open to receiving it.

Table 23 Item table for required abstain alchohol

Is the mentee required to abstain from alcohol?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

5

0.23

2

0.09

15

0.68

2

6

0.38

1

0.06

9

0.56

3

2

0.33

1

0.17

3

0.50

4

1

0.33

1

0.33

1

0.33

5

0

0.00

1

0.50

1

0.50

6

0

0.00

0

0.00

2

1.00

All

14

0.27

6

0.12

31

0.61

Table 24 Item table for alcohol in the past two months

How often has the mentee been drinking alcohol in the last two months (since the last review)?

Never

Once or twice

Three or more

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

5

0.25

14

0.70

2

2

0.13

1

0.07

12

0.80

3

1

0.17

1

0.17

4

0.67

4

0

0.00

0

0.00

3

1.00

5

0

0.00

1

0.50

1

0.50

6

0

0.00

1

0.50

1

0.50

All

4

0.08

9

0.19

35

0.73

Table 25 Item table for use of illegal drugs

How often has the mentee been using illegal drugs in the last two months (since the last review)?

Never

Once or twice

Three or more

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

0

0.00

0

0.00

19

1.0

2

0

0.00

0

0.00

12

1.0

3

0

0.00

0

0.00

5

1.0

4

0

0.00

0

0.00

3

1.0

5

0

0.00

0

0.00

2

1.0

6

0

0.00

0

0.00

2

1.0

All

0

0.00

0

0.00

43

1.0

Table 26 Item table for receiving substance abuse treatment

Is the mentee receiving substance abuse treatment?

No

Yes

No Need

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

12

0.67

6

0.33

0

0.00

2

7

0.64

4

0.36

0

0.00

3

4

1.00

0

0.00

0

0.00

4

1

0.50

1

0.50

0

0.00

5

1

0.50

1

0.50

0

0.00

6

1

0.50

1

0.50

0

0.00

All

26

0.67

13

0.33

0

0.00

Table 27 Item table for regular participation in substance abuse treatment

Is the mentee participating regularly in substance abuse treatment?

No

Yes

No Need

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

11

0.73

4

0.27

0

0.00

2

5

0.56

4

0.44

0

0.00

3

4

1.00

0

0.00

0

0.00

4

1

0.50

1

0.50

0

0.00

5

1

0.50

1

0.50

0

0.00

6

1

0.50

1

0.50

0

0.00

All

23

0.68

11

0.32

0

0.00

Table 28 Item table interest in substance abuse treatment

Has your mentee expressed interest in substance abuse treatment?
Not interested Interested, but not seeking help Looking for treatment Receiving treatment N/A; No need

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

5

0.24

3

0.14

3

0.14

7

0.33

2

4

0.27

2

0.13

1

0.07

5

0.33

3

0.20

3

1

0.17

1

0.17

1

0.17

0

0.00

3

0.50

4

0

0.00

0

0.00

1

0.33

0

0.00

2

0.67

5

0

0.00

2

1.00

0

0.00

0

0.00

0

0.00

6

0

0.00

1

0.50

0

0.00

1

0.50

0

0.00

All

8

0.16

11

0.22

6

0.12

9

0.18

15

0.31

Table 29 Item table for positive attitude towards substance abuse treatment

Does the mentee have a positive attitude towards substance abuse treatment?

Negative

Neutral

Positive

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

5

0.24

10

0.48

5

0.24

2

2

0.14

0

0.00

9

0.64

3

0.21

3

0

0.00

2

0.40

1

0.20

2

0.40

4

0

0.00

0

0.00

3

1.00

0

0.00

5

0

0.00

0

0.00

2

1.00

0

0.00

6

0

0.00

0

0.00

2

1.00

0

0.00

All

3

0.06

7

0.15

27

0.57

10

0.21

Table 30 Item table for interest in substance abuse treatment group

Has the mentee expressed an interest in a substance abuse support group?

Not
interested

Interested,
but not seeking help

Looking
for treatment

Receiving
treatment

N/A;
No need

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

5

0.24

3

0.14

3

0.14

7

0.33

2

4

0.27

2

0.13

1

0.07

5

0.33

3

0.20

3

1

0.17

1

0.17

1

0.17

0

0.00

3

0.50

4

0

0.00

0

0.00

1

0.33

0

0.00

2

0.67

5

0

0.00

2

1.00

0

0.00

0

0.00

0

0.00

6

0

0.00

1

0.50

0

0.00

1

0.50

0

0.00

All

8

0.16

11

0.22

6

0.12

9

0.18

15

0.31

Family reunification and relationships. Many subjects reported problems with family relations. On the other hand, 76% of the subjects reported that they have supportive family members and the answers given to several of the questions bear this out. Of twenty subjects, 4 (20%) reported no contact with family members at the first session. A fair number of subjects reported that they have unhealthy/abusive relationships with family members  (36%). Of 21 subjects, 8 (28%) reported that they were married or involved in relationships. Half of these people were satisfied with their relationships.

Of 22 subjects, 11 (5%) had children, and only two of these people reported that they provided some support for their children. However, 7 of the 11 people with children (64%) reported that they were significantly involved in the life of their children. Half of the people with children reported that they were significantly involved in the life the children. None of the people with children reported that they were unable to cope with the stress of parenting, although 24% of the respondents replied with ‘unsure’ to this question and 18% were seeking support to help with the stress of parenting.

Table 31 Item table for family members supportive substance abuse treatment

Does the mentee have family members that are supportive of treatment and re-socialization efforts?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.10

3

0.14

16

0.76

2

2

0.12

2

0.12

12

0.75

3

1

0.17

3

0.50

2

0.33

4

0

0.00

0

0.00

3

1.00

5

1

0.50

0

0.00

1

0.50

6

1

0.50

0

0.00

1

0.50

All

7

0.14

8

0.16

35

0.70

Table 32 Item table for regular contact with family members

Does the mentor have regular, personal contact with these family members?

No
contact

Monthly

Weekly

Daily

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.20

3

0.15

6

0.30

7

0.35

2

6

0.38

3

0.19

3

0.19

4

0.25

3

1

0.20

1

0.20

2

0.40

1

0.20

4

1

0.33

1

0.33

0

0.00

1

0.33

5

0

0.00

0

0.00

2

1.00

0

0.00

6

1

0.50

0

0.00

0

0.00

1

0.50

All

13

0.27

8

0.17

13

0.27

14

0.29

Table 33 Item table for abusive relationship with family members

Does the mentee have unhealthy/abusive relationships with some family members?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

9

0.41

5

0.23

8

0.36

2

9

0.56

3

0.19

4

0.25

3

0

0.00

4

0.67

2

0.33

4

1

0.33

1

0.33

1

0.33

5

0

0.00

1

0.50

1

0.50

6

1

0.50

1

0.50

0

0.00

All

20

0.39

15

0.29

16

0.31

Table 34 Item table for regular contact with abusive family members

Does the mentee have regular, personal contact with family members for which the relationship is unhealthy or abusive?

No contact

Monthly

Weekly

Daily

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

13

0.76

2

0.12

2

0.12

0

0.00

2

10

0.91

1

0.09

0

0.00

0

0.00

3

3

0.75

0

0.00

0

0.00

1

0.25

4

2

0.67

1

0.33

0

0.00

0

0.00

5

0

0.00

2

1.00

0

0.00

0

0.00

6

1

1.00

0

0.00

0

0.00

0

0.00

All

29

0.76

6

0.16

2

0.05

1

0.03

Table 35 Item table for family involvment in criminal alcohol drugs

Does the mentee have family members that are criminally involved or addicted to drugs/alcohol?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

12

0.57

5

0.24

4

0.19

2

8

0.50

6

0.38

2

0.12

­3

2

0.33

1

0.17

3

0.50

4

2

0.67

0

0.00

1

0.33

5

0

0.00

2

1.00

0

0.00

6

2

1.00

0

0.00

0

0.00

All

26

0.52

14

0.28

10

0.20

Table 36 Item table for regular contact with family involved in alcohol drugs

Does the mentee have regular, personal contact with family members that are criminally involved or addicted to drugs/alcohol?

Weekly

Monthly

No Contact

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.06

2

0.11

15

0.83

2

0

0.00

0

0.00

12

1.00

3

0

0.00

0

0.00

3

1.00

4

0

0.00

1

0.33

2

0.67

5

0

0.00

0

0.00

2

1.00

6

0

0.00

0

0.00

1

1.00

All

1

0.03

3

0.08

35

0.90

Table 37 Item table for married or involved

Is the mentee married or involved in a relationship?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

10

0.48

3

0.14

8

0.38

2

7

0.44

1

0.06

8

0.50

3

3

0.50

1

0.17

2

0.33

4

1

0.33

0

0.00

2

0.67

5

1

0.50

0

0.00

1

0.50

6

1

0.50

0

0.00

1

0.50

All

23

0.46

5

0.10

22

0.44

 Table 38 Item table for mentees satisfied with relationship

Is the mentee satisfied with this relationship?

No

Unsure

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.20

2

0.10

5

0.25

9

0.45

2

5

0.33

2

0.13

3

0.20

5

0.33

3

2

0.33

1

0.17

0

0.00

3

0.50

4

0

0.00

1

0.33

1

0.33

1

0.33

5

2

1.00

0

0.00

0

0.00

0

0.00

6

0

0.00

0

0.00

1

0.50

1

0.50

All

13

0.27

6

0.12

10

0.21

19

0.40

Table 39 Item table for number of children

Does the mentee have children?
No Yes

Session

Freq

Prop

Freq

Prop

1

10

0.45

12

0.55

2

6

0.38

10

0.62

3

2

0.40

3

0.60

4

1

0.33

2

0.67

5

1

0.50

1

0.50

6

1

0.50

1

0.50

All

21

0.42

29

0.58

Table 40 Item table for sole provider for children

Is the mentee the sole provider for these children?

No

Some, but not all

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

8

0.40

2

0.10

10

0.50

2

6

0.38

2

0.12

8

0.50

3

3

0.50

1

0.17

2

0.33

4

2

0.67

0

0.00

1

0.33

5

1

0.50

0

0.00

1

0.50

6

1

0.50

0

0.00

1

0.50

All

21

0.43

5

0.10

23

0.47

Table 41 Item table for involved with children

Is the mentee significantly involved in the life of his/her children?

No

Unsure

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

2

0.09

7

0.32

10

0.45

2

4

0.25

0

0.00

6

0.38

6

0.38

3

1

0.17

1

0.17

1

0.17

3

0.50

4

1

0.33

0

0.00

1

0.33

1

0.33

5

1

0.50

0

0.00

0

0.00

1

0.50

6

1

0.50

0

0.00

0

0.00

1

0.50

All

11

0.22

3

0.06

15

0.29

22

0.43

Table 42 Item table for cope with stress parenting

Is the mentee able to cope with the stress of parenting?

No

Unsure

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

0

0.00

5

0.24

5

0.24

11

0.52

2

1

0.06

4

0.25

4

0.25

7

0.44

3

0

0.00

0

0.00

2

0.33

4

0.67

4

0

0.00

1

0.33

0

0.00

2

0.67

5

0

0.00

1

0.50

0

0.00

1

0.50

6

0

0.00

1

0.50

0

0.00

1

0.50

All

1

0.02

12

0.24

11

0.22

26

0.52

Table 43 Item table for seeking support with stress

Is the mentee seeking support to help with the stress of parenting?

No

Unsure

Yes

No Need

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.18

2

0.09

4

0.18

12

0.55

2

3

0.20

2

0.13

2

0.13

8

0.53

3

0

0.00

0

0.00

0

0.00

6

1.00

4

1

0.33

0

0.00

0

0.00

2

0.67

5

0

0.00

0

0.00

0

0.00

2

1.00

6

1

0.50

0

0.00

0

0.00

1

0.50

All

9

0.18

4

0.08

6

0.12

31

0.62

 

Finances.
More than half of the subjects had problems with paying of significant debts, but most of these people had plans for paying off those debts. About half of the people said that they were having conflicts around finance. About a third of the people were able to save money.

Table 44 Item table for significant debts

Does the mentee have significant debts that are difficult to pay off?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.18

5

0.23

13

0.59

2

4

0.25

1

0.06

11

0.69

3

0

0.00

1

0.20

4

0.80

4

2

0.67

1

0.33

0

0.00

5

1

0.50

0

0.00

1

0.50

6

0

0.00

1

0.50

1

0.50

All

11

0.22

9

0.18

30

0.60

Table 45 Item table for plan for debts

Does the mentee have a plan for paying off debts?

No

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

5

0.23

12

0.55

5

0.23

2

8

0.50

6

0.38

2

0.12

3

2

0.40

3

0.60

0

0.00

4

0

0.00

1

0.33

2

0.67

5

0

0.00

1

0.50

1

0.50

6

0

0.00

1

0.50

1

0.50

All

15

0.30

24

0.48

11

0.22

Table 46 Item table for conflict around finances

Does the mentee experience conflict around finances?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

8

0.36

11

0.50

2

1

0.06

4

0.25

11

0.69

3

0

0.00

2

0.33

4

0.67

4

1

0.33

2

0.67

0

0.00

5

0

0.00

2

1.00

0

0.00

6

1

0.50

1

0.50

0

0.00

All

6

0.12

19

0.37

26

0.51

Table 47 Item table for trouble paying bills

Does the mentee have trouble paying bills?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

5

0.24

9

0.43

7

0.33

2

3

0.19

5

0.31

8

0.50

3

0

0.00

2

0.33

4

0.67

4

1

0.33

2

0.67

0

0.00

5

0

0.00

2

1.00

0

0.00

6

1

0.50

1

0.50

0

0.00

All

10

0.20

21

0.42

19

0.38

Table 48 Item table for able to save money

Is the mentee able to save money?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.09

12

0.55

8

0.36

2

9

0.56

5

0.31

2

0.12

3

4

0.67

1

0.17

1

0.17

4

2

0.67

1

0.33

0

0.00

5

1

0.50

1

0.50

0

0.00

6

1

0.50

0

0.00

1

0.50

All

19

0.37

20

0.39

12

0.24

 

Employment

About half of the subjects (45%) had jobs at the time of first session. Only 6 of 18 subjects (32%) reported that their job situation was stable. Of the people with jobs, about half reported that they were satisfied with their job situation. Only 4 subjects of 20 subjects who responded to the question were taking steps to find jobs around the time of the first session.

 

Table 49 Item table for currently have job

Does the mentee currently have a job?

No

Yes

Session Freq Prop Freq Prop
1

12

0.55

10

0.45

2

10

0.62

6

0.38

3

4

0.67

2

0.33

4

2

0.67

1

0.33

5

2

1.00

0

0.00

6

1

0.50

1

0.50

All

31

0.61

20

0.39

Table 50 Item table for stable job

If the mentee is employed, is their job situation stable?

No

Unsure

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

4

0.21

6

0.32

8

0.42

2

2

0.15

2

0.15

3

0.23

6

0.46

3

2

0.33

0

0.00

1

0.17

3

0.50

4

0

0.00

0

0.00

1

0.33

2

0.67

5

0

0.00

0

0.00

0

0.00

2

1.00

6

0

0.00

0

0.00

1

0.50

1

0.50

All

5

0.11

6

0.13

12

0.27

22

0.49

Table 51 Item table for consistant job change

How often did the mentee change jobs since the last review?

0

1

2 or more

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

17

0.85

3

0.15

0

0.00

2

9

0.75

3

0.25

0

0.00

3

4

0.80

0

0.00

1

0.20

4

3

1.00

0

0.00

0

0.00

5

2

1.00

0

0.00

0

0.00

6

2

1.00

0

0.00

0

0.00

All

37

0.84

6

0.14

1

0.02

Table 52 Item table for satisfied with job

Is the mentee satisfied with their job situation?

No

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

6

0.29

5

0.24

10

0.48

2

8

0.62

2

0.15

3

0.23

3

2

0.33

1

0.17

3

0.50

4

2

0.67

0

0.00

1

0.33

5

0

0.00

0

0.00

2

1.00

6

1

0.50

0

0.00

1

0.50

All

19

0.40

8

0.17

20

0.43

Table 53 Item table for finding a job

If unemployed, has mentee taken steps to find a job?

No

Some

Yes

N/A

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.10

1

0.05

3

0.15

14

0.70

2

1

0.07

2

0.14

3

0.21

8

0.57

3

1

0.17

2

0.33

1

0.17

2

0.33

4

0

0.00

1

0.33

0

0.00

2

0.67

5

1

0.50

0

0.00

0

0.00

1

0.50

6

1

0.50

0

0.00

0

0.00

1

0.50

All

6

0.13

6

0.13

7

0.15

28

0.60

 

Education/Skills.
Most of the subjects (82%) reported that they had qualifications that allow them to find a job that pays above minimum wage. A little under half of the respondents (41%) reported that the could obtain a job that would pay a living wage given their education and skills. A fairly large proportion (27%) of the respondents said that they had education and/or skills training goals but did not have a plan to achieve them. Of the 21 respondents, 7 (33%) reported that they were either pursuing their goals or had a plan to do so.

Table 54 Item table for qualifications above minimum

Does the mentee have qualifications that allow him/her to find a job that pays above the minimum wage?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

3

0.14

18

0.82

2

4

0.25

1

0.06

11

0.69

3

1

0.17

0

0.00

5

0.83

4

1

0.33

0

0.00

2

0.67

5

1

0.50

1

0.50

0

0.00

6

0

0.00

0

0.00

2

1.00

All

8

0.16

5

0.10

38

0.75

Table 55 Item table for educational skill goals

Does the mentee have any educational or skill training goals in the future?
No educational or skills training
goals
Has education and/or skills training
goals but does not have a plan to achieve them
Has education and/or skills training
goals and a plan to achieve them, but does not have the means
Pursuing education and skills training
goals and has the means
N/A (Can obtain job that will pay a
living wage with current education and/or skills)
Session Freq Prop Freq Prop Freq Prop Freq Prop Freq Prop

1

0

0.00

6

0.27

3

0.14

4

0.18

9

0.41

2

1

0.07

6

0.40

2

0.13

2

0.13

4

0.27

3

0

0.00

1

0.17

1

0.17

0

0.00

4

0.67

4

0

0.00

0

0.00

1

0.33

2

0.67

0

0.00

5

2

1.00

0

0.00

0

0.00

0

0.00

0

0.00

6

0

0.00

0

0.00

1

0.50

1

0.50

0

0.00

All

3

0.06

13

0.26

8

0.16

9

0.18

17

0.34

Social Support/ Personal Interests / Spirituality.

Two of the mentees (9%) reported that they had pro-social friends. Of the remaining 20 mentees only three reported that they had no pro-social friends. Of the 22 subjects, 7 (32%) participate in a spiritual community. Only 3 of the 22 respondents (14%) say that they are keeping in touch with friends that are criminally involved or drug addicted.

Table 56 Item table for prosocial friends

Does the mentee have pro-social friends?

None

Some

Many

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

3

0.14

17

0.77

2

0.09

2

4

0.27

10

0.67

1

0.07

3

1

0.17

5

0.83

0

0.00

4

1

0.33

2

0.67

0

0.00

5

1

0.50

1

0.50

0

0.00

6

1

0.50

0

0.00

1

0.50

All

11

0.22

35

0.70

4

0.08

Table 57 Item table for participating in a  spiritual community

Does the mentee participate in a spiritual community?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

10

0.45

5

0.23

7

0.32

2

8

0.50

4

0.25

4

0.25

3

2

0.33

4

0.67

0

0.00

4

1

0.33

0

0.00

2

0.67

5

0

0.00

1

0.50

1

0.50

6

1

0.50

0

0.00

1

0.50

All

22

0.43

14

0.27

15

0.29

Table 58 Item table intouch criminal friends

Is the mentee keeping in touch with friends that are criminally involved or drug addicted?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

11

0.52

7

0.33

3

0.14

2

5

0.31

8

0.50

3

0.19

3

3

0.50

2

0.33

1

0.17

4

2

0.67

0

0.00

1

0.33

5

0

0.00

0

0.00

2

1.00

6

2

1.00

0

0.00

0

0.00

All

23

0.46

17

0.34

10

0.20

 

Personal Goals Values. All but one person of the 22 mentees recognized that they must change their lifestyle away from crime. Most of the subjects (86%) had outlined goals that would move them to self-sufficiency and a large percentage (68%) were pursuing these goals. A smaller percentage (59%) had reached some of their goals and 41% were concerned about sliding back into bad habits. Most of the subjects (68%) had taken steps to prevent a full-blown relapse. Finally, 16 of 22 subjects (73%) report that they recognize the consequences of their actions and do not externalize blame.

 

Table 59 Item table recognize a need for change

Does the mentee recognize that he/she must change their lifestyle away from crime?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

0

0.00

21

0.95

2

0

0.00

2

0.13

13

0.87

3

0

0.00

0

0.00

6

1.00

4

0

0.00

0

0.00

3

1.00

5

0

0.00

0

0.00

2

1.00

6

0

0.00

1

0.50

1

0.50

All

1

0.02

3

0.06

46

0.92

Table 60 Item table positive goals for self sufficiency

Does the mentee have positive goals outlined that will move them towards self-sufficiency?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

2

0.09

19

0.86

2

1

0.06

4

0.25

11

0.69

3

0

0.00

2

0.33

4

0.67

4

0

0.00

0

0.00

3

1.00

5

0

0.00

2

1.00

0

0.00

6

0

0.00

1

0.50

1

0.50

All

2

0.04

11

0.22

38

0.75

Table 61 Item table pursuing positive goals for self sufficiency

Is the mentee pursuing positive goals that will move them towards self-sufficiency?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.09

5

0.23

15

0.68

2

3

0.19

5

0.31

8

0.50

3

0

0.00

2

0.33

4

0.67

4

0

0.00

0

0.00

3

1.00

5

0

0.00

1

0.50

1

0.50

6

1

0.50

0

0.00

1

0.50

All

6

0.12

13

0.25

32

0.63

Table 62 Item table reached some positive goals

Has the mentee reached some of the positive goals he/she has set forth?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.18

5

0.23

13

0.59

2

4

0.27

1

0.07

10

0.67

3

0

0.00

3

0.50

3

0.50

4

0

0.00

2

0.67

1

0.33

5

1

0.50

0

0.00

1

0.50

6

1

0.50

0

0.00

1

0.50

All

10

0.20

11

0.22

29

0.58

Table 63 Item table for taken steps toward positive goals

If the mentee has taken steps towards positive goals, is he/she worried about slipping back into old habits?

No

Unsure

Yes

N/A (No goals)

Session

Freq

Prop

Freq

Prop

Freq

Prop

Freq

Prop

1

4

0.18

7

0.32

9

0.41

2

0.09

2

1

0.07

3

0.21

8

0.57

2

0.14

3

1

0.17

3

0.50

2

0.33

0

0.00

4

1

0.33

1

0.33

1

0.33

0

0.00

5

0

0.00

0

0.00

2

1.00

0

0.00

6

1

0.50

0

0.00

0

0.00

1

0.50

All

8

0.16

14

0.29

22

0.45

5

0.10

Table 64 Item table for taken steps to prevent relapse

Has the mentee taken steps to prevent a full-blown relapse?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.09

5

0.23

15

0.68

2

3

0.20

4

0.27

8

0.53

3

1

0.17

4

0.67

1

0.17

4

0

0.00

1

0.33

2

0.67

5

1

0.50

0

0.00

1

0.50

6

1

0.50

0

0.00

1

0.50

All

8

0.16

14

0.28

28

0.56

Table 65 Item table for recognizing consequences of actions

Does the mentee recognize the consequences of their actions rather than externalizing blame (e.g. blaming victims, the criminal justice system)?

No

Unsure

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

5

0.23

16

0.73

2

2

0.12

4

0.25

10

0.62

3

0

0.00

2

0.33

4

0.67

4

0

0.00

0

0.00

3

1.00

5

0

0.00

0

0.00

2

1.00

6

0

0.00

1

0.50

1

0.50

All

3

0.06

12

0.24

36

0.71

 

Arrest/Recidivism.
Across all sessions, there were very few re-arrests with charges or failures to comply with the terms of release, as indicated by self-report to Item 66. There were two new offenses and no technical violations. Both new offenses occurred before the first session. Separate data on arrests were collected from the MIS data, and these data were used for further analysis, as they would likely be more accurate.

 

Table 66 Item table for charged or arrested since last

66. Has the mentee been arrested and charged for a new offense or failure to comply with term of release since
the last review?

New Offense

New Technical Violation

No

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

2

0.10

0

0.00

18

0.90

2

0

0.00

0

0.00

14

1.00

3

0

0.00

0

0.00

4

1.00

4

0

0.00

0

0.00

3

1.00

5

0

0.00

0

0.00

1

1.00

6

0

0.00

0

0.00

2

1.00

All

2

0.05

0

0.00

42

0.95

 

Mentor Support

The table below breaks down the ways in which the mentors supported the mentees in the previous two months. The mentors were involved in many activities to help the mentees. The final question asked whether the mentee thought that his or her life and chances for success had improved since the last review. Only 1 subject in the first session said ‘No’, whereas 10 subjects said ‘Yes’ (50%). The remaining 9 respondents (45%) were ‘undecided’.

Table 67 Table combining responses to items coding the type of support provided by mentors.

How have you supported your mentee since last review (in the last two Months)? What services were delivered? (check all that apply)

Items

Frequency

Proportion

Spoke in depth about serious
issues

40

0.78

Recreated (athletics, sporting events,
thearter, movie, played games/cards, etc)

5

0.10

Met for a meal, coffee, snack

32

0.63

Attended church services

 1

0.02

Helped navigate criminal justice
system

13

0.25

Helped obtain community resources
(government offices, non-profits)

14

0.27

Helped obtain community resources
(government offices, non-profits)

14

0.27

Crisis intervention (violence, arrest,
emotional breakdown, intoxication)

3

0.06

Met with mentee’s associates (friends,
family, employer, counselor)

13

0.25

 

[Continued] How have you supported your mentee since last review (in the last two Months)? What services were delivered? (check all that apply)

Items

Frequency

Proportion

 Gave a gift or made a special gesture

18

0.35

 Celebrated an event or achievement

12

0.24

Worked through an Action Plan or Mind
Map

12

0.24

Provided them with resources to
support their kids (e.g. clothing vouchers, money for Christmas
presents)

4

0.08

Used personal connections to provide
resources

10

0.20

Attended an appointment or meeting
with them related to getting resources (e.g. food assistance, section 8
housing)

8

0.16

Attended a hearing

10

0.20

Took them to a cultural event

1

0.02

Other

4

0.08

Table 68 Item table for think success improved

Does the mentee think that the mentee’s life and chances for success have improved since last review?

No

Undecided

Yes

Session

Freq

Prop

Freq

Prop

Freq

Prop

1

1

0.05

9

0.45

10

0.50

2

1

0.07

4

0.27

10

0.67

3

0

0.00

3

0.60

2

0.40

4

0

0.00

3

1.00

0

0.00

5

1

0.50

0

0.00

1

0.50

6

2

1.00

0

0.00

0

0.00

All

5

0.11

19

0.40

23

0.49

 

Analysis of Change

The Life Skills Matrix was given repeatedly to the subjects over the course of their participation in the study. This repeated administration should make it possible to assess whether the responses to the questions changed with time. This change could reflect the positive impact of the FOCUS intervention on the mentees and help explain the reduced recidivism for the FOCUS group when compared with the control group.

Unfortunately, the sample size for assessing change was quite small. Only 16 subjects, were administered the Life Skills Matrix questionnaire 2 or more times. In addition, many of the responses subjects gave were missing or non-informative (e.g., ‘unsure’, ‘not applicable’), which reduced the statistical power further.

We used several approaches to assess change in the responses to the Life Skills Matrix items. One approach involved coding the items so that numerical scales for a category (e.g., ‘Housing’, ‘Food’) could be computed for each subject. These scale scores were then analyzed across sessions to assess change (e.g., did the scale scores for ‘Housing’ improve?). To code the responses,  ‘good’ responses to items received low scores and ‘bad’ response received high scores. For example, the item stable.living was coded a 0 if the answer was ‘No’, a 1 if the answer was ‘Yes’, and missing if the answer was ‘Unsure’. This approach resulted in many missing values, which reduced the power the statistical tests, so that statistical differences could not be identified. With a larger sample size, the power would naturally increase and the missing values could possibly be imputed (filled in), increasing the power further. Another problem with this scaling approach was that we could not be sure that the items within a category were correlated because of the small sample size. When forming a summated scale, the items should be correlated so that the overall scale is reliable.

The scale values were entered into mixed linear regression models designed to assess change. The time between sessions was included in the models as a covariate and subjects were treated as a random effect. Improvement, in these models would be reflected as reduction in scores over time. The power for these tests was very low, in part because of the missing values, and statistical differences were not observed. We also applied this modeling approach for individual items that were coded numerically, but again the low statistical power was too low to find significant differences.

Another approach for assessing change involved focusing on the responses to individual items at the first and final sessions for each subject. Again, only the data for the 16 subjects who were administered the Life Skills Matrix on at least two occasions were analyzed. Table 69 displays the results of paired t-tests for several of the items that were encoded numerically. These items did not have non-informative responses, such as ‘unsure’.  The paired t-test determines whether the mean of the differences between the session 1 and 2 scores significantly differs from 0. There was a significant difference for CHANGE. RESIDENCE, indicating that the subjects changed residence more often around the time of the last session. There were no other significant differences.

Table 69 Means for numerically coded items for the first and last FOCUS sessions. The Life Skills Matrix was administered during these sessions. The number of pairs (n) and means differences between the pairs of scores also appears
along with the results of paired t-tests.

Item

Mean

Sess. 1

Mean

Sess. 2

Mean
diff.

n

t
value

Pr(>|t|)

change.residence

0.12

0.69

-0.56

16

-2.33

0.034

often.alcohol

2.21

2.21

0.00

14

0.00

1.000

often.drugs

2.07

2.07

0.00

14

0.00

1.000

often.change.job

2.20

2.20

0.00

10

0.00

1.000

 

Many of the items on the Life Skills Matrix, although non-numeric, could be coded to reflect improvement in the responses between the first and last test administrations (sessions). For example, if the response was ‘No’ and then later ‘Yes’ in sessions 1 and 2 to the question about whether the person had a stable living situation, the person would be coded as “Improve”. On the other hand, had the response been initially “Yes” and then “No”, the person would have been coded as “Worse”. The McNemar’s test provides a test of whether change of this type over a sample of subjects is statistically significant. Note that subjects who do not change—they respond either ‘Yes’ or ‘No’ on both occasions—do not contribute information to the test, thereby reducing the power of the test. Table 70 displays the proportions of subjects who improved or got worse between the first and last administrations of the Life Skills Matrix. Positive odds ratios indicate improvement. None of the differences were significant at the conventional level of 0.05. The most notable trend was for CURRENTLY.HAVE. JOB, for which more subjects got worse (44%) than became better (19%).

Table 70 Analysis of change for binary items. The odds ratio with results from McNemar tests also appear in the table.

Item Improve Worse Same Odds Ratio Prob. n
stable.living

0.00

0.11

0.89

0.00

1.00

9

Conflict.around.finances

0.25

0.25

0.50

1.00

1.00

4

Trouble.pay.bill

0.43

0.29

0.29

1.50

1.00

7

Currently.have.job

0.19

0.44

0.38

0.43

0.43

16

Participate.spiritual

0.22

0.11

0.67

2.00

1.00

9

In.touch.criminal.friends

0.20

0.40

0.40

2.00

1.00

5

 

If more data are collected for this study, the methods used to analyze change in this section should yield results that show the beneficial effects of mentoring on the subjects. Such results could help explain why the recidivism rate is lower in the FOCUS group than in the control group.

Summary of Result

The major result of this study is that the incidence rate for re-arrest rate for the FOCUS group was significantly lower than the incidence rate for a comparable group of control subjects. The incidence rates for the two groups differed by a factor of 4. Analysis of the time that the mentors and mentees spent together along with descriptive data on the various activities that the mentors and mentees carried out together indicates that the intervention was quite intense. The large time
commitments by the mentors and mentees no doubt contributed to the effectiveness of the intervention.

On the basis of our results, the recidivism rate as usually defined (re-arrest within a year) should be significantly lower for people who have received the FOCUS intervention than for people who have not. A reduction in the recidivism rate of a factor of 4 would be consistent with our findings, but additional data collected from a larger pool of offender/mentees is needed to confirm this finding. Hence, the recidivism rate of 67% in the jail population could be reduced with the FOCUS intervention to a rate as low as 17%. We are continuing to expand the number of offenders in the FOCUS program and to collect data through 2012 to verify this conclusion. (From a FOCUS conversation for clarification with B. Oliver, evaluator, and FOCUS staff post issuance of report. May 7,2012)

An extensive Life Skills Matrix questionnaire that was administered repeatedly to the mentees provides a rich description of their needs for re-entry into the community. For many of the mentees, these needs were in the areas of housing, employment and education. The answers to other questions indicated that many of the mentees had family problems, including problems with raising their children. With the collection of additional data, it should be possible to determine the ways in which the FOCUS mentors help the mentees with these various needs.

References

Cameron, A. C., & Trivedi (1998). Regression analysis of count data. NY: Cambridge University Press.

Maltz, M. D. (1984) Recidivism. New York: Academic Press.

Markman, J. A., Fontaine, J., Roman, J. K., & Nadeau, C. A. (2010). Evaluation of the Ohio Department of Rehabilitation and Correction and Corporation for Supportive Housing’s Pilot Program: Interim Re-Arrest Analysis. Washington, DC: Urban Institute Research Report.