Research Reports

Case Management Strategies Survey

1991, No. R-08

Prepared By: D.A. Andrews
R.D. Hodge
D. Robinson
F.J. Andrews

Laboratory for Research on Assessment
and Evaluation in the Human Services

Research and Statistics Branch
Correctional Service of Canada

February 1990

TABLE OF CONTENTS

Introduction

Method

Results

Summary and Conclusions

ABSTRACT

Questionnaires concerning initial training in CMS and CMS in the workplace were sent to 966 CMS users and 528 (55%) were returned in time for the analyses reported here.

Overall evaluations of initial training in CMS were evenly distributed across the satisfied, dissatisfied and in-between categories. Some variation in the evaluations of training in CMS was traced to respondent, organizational and training variables. In brief, positive evaluations of initial training were associated with a positive pretraining attitude toward CMS, number of days of training received, having been trained by a trainer with links to NCCD, being institution-based, working in the Atlantic region, and not having pursued formal studies in criminology. Evaluations of training were more positive regarding the basics of the scripted interview than they were regarding CMS in the overall case management context.

Great variability in opinions regarding CMS in the workplace was found. However, on average the evaluations were negative. The sources of the variation in opinions regarding CMS in the workplace generally replicated those found for opinions regarding training in CMS.

Major efforts are indicated for the future in clarifying and perhaps rethinking the links between CMS and the overall responsibilities of case management officers.

LIST OF TABLES

TABLE 1 Demographic Data for the Respondents (Total N = 528)

TABLE 2 CMS Trainers of Respondents (n = 528)

TABLE 3 Use of CMS: Percentage of Caseload (N = 498) and Time Between Case Assignment and CMS Use (N = 466)

TABLE 4 Proportion of Time CMS Devoted to Various Activities: Percent Distributions

TABLE 5 Rated Usefulness of CMS Training Relative to Other Courses and Workshops and Satisfaction with CMS Training: Percent Distribution and Means/SDs

TABLE 6 Overall Satisfaction With Initial Training in CMS by Background, Position, Setting, and Region (Based on a 5-point Scale With 1 = Very Dissatisfied and 5 = Very Satisfied)

TABLE 7 Rated Attitude to Prospect of Implementing CMS and Anticipated Changes in Amount of Time Required to Manage Individual Cases: Percent Distributions and Means/SDs

TABLE 8 Mean Satisfaction with Initial Training by Type of Trainer, Setting, Major/Concentration, Region and Position; Means and Etas with and without Adjustments for Other Factors and Covariates

TABLE 9 Rated Quality of Initial Training in CMS in Specific Areas: Percent Distribution and Means/SDs

TABLE 10 Overall Attitude Toward Implementation of CMS: % Distribution and Means/SDs

TABLE 11 Overall Attitude Toward Implementation of CMS by Background, Position, Setting, and Region (Based on a 5-point Scale with 1 = Not At All Worth the Effort and 5 = Very Much Worth the Effort)

TABLE 12 Mean General Evaluation of CMS in Workplace by Type of Trainer, Setting, Major/Concentration, Region and Position: Means and Etas with and without Adjustments for Other Factors and Covariates

TABLE 13 Rated Effect of CMS on Current Workload

TABLE 14 (A) Extent of Satisfaction With CMS at Various Stages of the Case Management Process and (B) Extent to Which CMS Represents An Improvement Over Prior Methods: Percent Distributions and Means/SDs

TABLE 15 Rated Extent of Assistance in Making Recommendations to Decision-Makers: Percent Distribution and Means/SDs

TABLE 16 Rated Extent to Which CMS Facilitates Communication With Others: Percent Distributions and Means/SDs

TABLE 17 Rated Value of CMS in Various Areas: Percent Distributions and Means/SDs

TABLE 18 Evaluations of the Reliability, Accuracy and Appropriateness of CMS Interview and CMS Strategy Types

TABLE 19 Assessment of the Scripted Interview: Percent Distributions and Means/SDs

TABLE 20 Assessment of the Force-Field Analysis of Needs: Percent Distributions and Means/SDs

TABLE 21 Assessment of the Individual Correctional Treatment Planning Process: Percent Distributions and Means/SDs

TABLE 22 Contributions of CMS to Aspects of Case Management: Percent Distributions and Means/SDs (N = 486 to 493)

TABLE 23 Rated Utility of CMS in Monitoring CM Officers in Various Areas: Percent Distributions and Means/SDs

TABLE 24 Ratings of (A) Other Staffs' Perceptions of CMS and (B) Other Staffs' Familiarity with CMS: Percent Distributions and Means/SDs

TABLE 25 Rated Need for Additional Information Training in CMS by Specific Areas: Percent Distributions and Means/SDs

TABLE 26 Priority Ratings Regarding Planning for the Refinement and Future Applications of CMS: Percent Distributions and Means/SDs

TABLE 27 Priority Ratings For Areas of Further Training: Percent Distributions and Means/SDs

INTRODUCTION

The Case Management Strategies (CMS) Survey was a response to the requests of Correctional Services Canada (CSC) senior management for a review of the implementation of CMS. The survey is one step in the evaluation process recommended by the Research Advisory Group appointed to conduct a review of CMS. The survey results describe the extent and nature of use of CMS among CSC personnel. The survey also provided CMS users with the opportunity to evaluate their CMS training, express opinions regarding the overall utility of the system, and to make recommendations pertaining to the future refinements of the system.

This report provides an examination of data collected to date from a total of 528 survey respondents. Descriptive information regarding the use of CMS and evaluations of its utility are reported. In addition, satisfaction with the CMS system is examined in relation to basic demographic, CSC organizational, and training variables. The presentation of the findings follows the structure of the CMS questionnaire. Demographic information and a description of the training received by respondents is presented first. Evaluations of the training process are examined and demographic and organizational correlates of training satisfaction are reviewed. This section is followed by an examination of the data on general opinions about CMS in the workplace and their correlates. The next section deals with evaluations of the individual components of the CMS. Finally, respondents' recommendations for future refinements of CMS are detailed.

METHOD

The Case Management Strategies Survey was designed by the Laboratory for Research on Assessment and Evaluation in the Human Services in consultation with the Correctional Services Canada advisory committee. The latter group was composed of members from the regionally-represented Research Advisory Group responsible for the review of the implementation of the Case Management Strategies. Correctional Services Canada provided the principal investigators with a list of content areas reflecting the issues and components of CMS that were to be subjected to evaluation. Demographic characteristics of the CMS users considered relevant to the evaluation were also specified. The principal investigators were asked to design a questionnaire that incorporated all of the relevant variables in a format that would be easily self-administered and not too demanding of respondents' time. The questionnaire was to include multiple-choice questions with sufficient space available for open-ended comments. The questionnaire design involved several drafts and extensive consultations with the advisory committee before the final questionnaire was approved in March 1988.

Procedure

The Case Management Strategies Survey was mailed to all Correctional Services Canada personnel using the CMS. A memorandum dated 31 March 1988 from the Deputy Commissioner of Offender Policy and Program Development accompanied the questionnaire. The memorandum outlined the purpose of the survey, solicited the cooperation of CMS users in carrying out the evaluation, and assured respondents of the confidentiality with which their responses would be treated. Each questionnaire package was individually addressed, and regional offices managed the distribution of those questionnaires to the individuals involved. Respondents were asked to complete the questionnaires promptly and forward them to Carleton University using pre-addressed envelopes. In an attempt to augment the response rate, reminder letters were mailed to all respondents approximately two weeks after the questionnaire was initially distributed.

The Survey Instrument

The CMS Survey questionnaire consists of five sections: 1) Personal History; 2) Training in CMS; 3) General Opinions Regarding CMS in the Workplace; 4) Assessment of Specific Elements of CMS; 5) Future Refinements of CMS. English and French versions of the questionnaire are appended (Appendix 1).

1)   Personal History. The personal history section surveyed basic demographic information and CSC organizational variables. Demographic information included age, gender, educational attainment, and educational specializations. CSC organizational data included current and past CSC positions, duration of job tenures, and total years of CSC service. Respondents were asked to indicate the region in which they were employed (Atlantic, Quebec, Ontario, Prairies, Pacific) and the type of setting where they conducted their work (type of institution or community setting).

2)   Training in CMS. The questionnaire requested information about duration of training in CMS, training dates, date of implementation of CMS by the respondent, and details about any refresher training that may have been received. Respondents were asked to identify their trainers in terms of the following categories: NCCD trainers, CSC trainers trained by NCCD staff, CSC trainers not trained by NCCD staff. Training evaluation items consisted of attitudes toward CMS before and after training with respect to CMS implementation and how CMS would alter the time required to manage individual cases. Respondents were also asked to identify three courses or workshops they had recently attended and to indicate the usefulness of their CMS training relative to these non-CMS training experiences. A final section solicited an overall training satisfaction rating and space was provided for open-ended comments on CMS training.

3)   General Opinions Regarding CMS in the Workplace. This section involved descriptive information about the extent of use of CMS by the respondent, how long it had been employed, and when it was completed in relation to case assignment. A number of CMS functions were evaluated through multiple-choice items in this section: the ability of CMS to provide an understanding of the offender at different stages in the management of the sentence, the perceptions and familiarity with CMS of other staff, the usefulness of CMS in aiding various decisions about the offender (e.g. temporary absences, psychiatric referrals), and the usefulness of CMS in facilitating communication with other personnel. CMS was also evaluated in terms of its utility in preparing court testimony, its contribution to the effective utilization of program resources, and its ability to identify new programming directions.

4)   Assessment of Specific Elements of CMS. In this section CMS users provided information about the utility of the scripted interview, force field analysis, treatment planning process, and the offender strategy types. Multiple-choice items were employed in evaluating the various elements. Other information solicited included the usefulness of CMS in addressing a variety of objectives related to risk assessment and promoting positive change in the offender. The potential usefulness of the CMS in supervising case management officers was also explored. Descriptive information about the proportions of time CMS users devote to conducting different CMS procedures and tasks was also recorded. Space for comments on specific elements of the CMS was available at the end of the section.

5)   Future Refinements of CMS. Multiple-choice items were used to seek information about the need for refinements of the various components of CMS and CMS training in specific areas. Respondents also indicated the level of priority which should be assigned to future training in non-CMS correctional areas. Open-ended comments on future refinements of the CMS were invited at the end of the section. A final open-ended section is available to allow respondents to raise issues not covered in other sections of the questionnaire.

Respondents

The Case Management Strategies Survey was sent to all CMS users in Correctional Services Canada. Approximately 966 questionnaires were mailed with a total of 528 received by the Laboratory for Research on Assessment and Evaluation in the Human Services by mid-June 1988. This represents a response rate of 55%. (Since mid-June and up to June 26th, an additional 32 questionnaires have been received by the Laboratory. This report is based on the N of 528.) The response rate is reasonably high for a survey conducted by mail. However, inspection of the bottom portion of Table 1 shows that there was regional variation in response rates.

Some basic descriptive information on the sample is presented in Table 1. The mean age of respondents was 37 years (SD = 7.9) and 69% were men. Nineteen percent of respondents had received graduate training, 63% had received Bachelor's degrees or some university education, and 9% had completed some course of study at a community college. Of those who completed some postsecondary training, criminology was the modal specialization (35%). An additional 23% reported concentrations in Psychology, 14% in Sociology, and 12% in Social Work.

Table 1 also presents the percent distribution for CSC current position. Seventy-five percent of the survey respondents were case management officers. Over one third of the respondents had been employed in their current positions for at least 3 years (mean = 6 years, SD. = 4.5). The average number of years the respondents were employed by CSC was 9.5 years (SD = 5.6). This figure includes current and previous CSC positions.

The breakdown of respondents by region is also displayed in Table 1. Not surprisingly, Quebec and Ontario respondents represent the largest groups in the sample. Further inspection of Table 1 reveals that 42% of the respondents were community-based.

RESULTS

The results are presented in several sections: descriptive information on training, descriptive information on the use of CMS, opinions regarding initial training in CMS, general opinions regarding CMS in the workplace, opinions regarding CMS in relation to some specific elements of case management, perceptions of how other CSC staff evaluate CMS, assessments of specific elements of CMS, and opinions regarding refinements of CMS and priority areas for action in the future.

The results section includes quantitative analyses of ratings gathered from the survey as well as qualitative summaries of the open-ended comments of respondents. For the qualitative analyses, all comments on all 528 questionnaires were read by at least one of two readers. These readings were preceded by a detailed quantification of the themes evident in the narratives provided on the first 114 questionnaires received. A total of 148 questionnaires were selected for more detailed review by an experienced qualitative researcher (F.J.A.). The selection reflected representativeness of content (as estimated from analyses of the first 114 returns), clarity of articulation, as well as judged centrality of the comments to case management issues.

Descriptive Information on Training

The CMS Survey instrument included three training measures. One measure was a simple report on the number of days of initial training in CMS received (question 2.1). Responses to this question ranged from zero days (14 respondents/2.7%) through the most frequent response of five days (335 respondents/66%) through to 10 or more days (33 respondents / 6%). The mean of Days in Training was 4.8, SD = 3.4, N = 514. Of those who were trained, the mean number of days training was 4.9 (SD = 3.4). The second training variable was Type of Trainer (question 2.5): a National Council on Crime and Delinquency (NCCD) trainer, a CSC trainer known to have been trained by an NCCD trainer, and a CSC trainer not known or thought to have been trained by NCCD. These data are presented in Table 2 where it is apparent that 53% were unsure of the training that their trainer had received. Finally, only 13% of the sample had received any formal refresher training. Of those who received refresher training, the mean number of days was 1.6 (SD = 1.1).

Table 1 - Demographic Data for the Respondents Total (N = 528)

Descriptive Information	Data		N
1. Age	Mean = 37 years, S.D. = 8, Range = 22 to 68		516
2. Gender	31% female, 69% male		521
3. Highest Earned Degree	Less than high school	0.8%	521
	Secondary or high school	8.3%
	Community college	9.0%
	Bachelors	62.2%
	Masters	18.2%
	Doctoral	0.6%
	B.S.W.	0.6%
	Police College	0.2%
4. Major/ Concentration	Criminology	35.0%	429
	Social Work	11.9%
	Psychology	22.6%
	sociology	14.0%
	Other	16.5%
5. Present CSC Position	CMO/Community	38.6%	521
	CMO/Instutution	36.3%
	Supervisor of CMO	8.4%
	Living Unit Officer	12.5%
	Other	4.2%
6. Type of Setting Maximum Institution	Maxium Institution	14.9%	524
	Medium Institution	27.7%
	Minimum Institution	7.3%
	Multilevel institution	6.5%
	Regional Psychiatric	0.4%
	Regional Reception	0.4%
	Community Centre	6.1%
	Parole Officer	36.3%
	Other	0.4%
7. Total Years of Service in CSC	Mean = 9.52 years, S.D. = 5.62, Range = 1 to 28		520
8. Region	Atlantic	10.0%	528
	Quebec	27.3%
	Ontario	24.2%
	Praries	23.3%
	Pacific	15.2%
9. Response Rate by Region	Atlantic	31%	53/170
	Quebec	67%	144/214
	Ontario	68%	128/187
	Praries	60%	123/205
	Pacific	42%	80/190

Table 2 - CMS Trainers of Respondents (n = 528)

Trainer	f	%
National Council on Crime an Delinquency Trainers	33	6.3
CSC Trainers Trained by NCCD	164	31.1
CSC Trainers not Trained By NCCD	35	6.6
Source of Trainer’s Training Unknown	282	53.6
Respondents Received No Formal Training	14	2.7

Descriptive Information on Use of CMS

Most respondents (59%) received their initial training in 1987 and began to use CMS in that year (74%). For most CMS users (82%) there was a gap of at least one month between the period of training and implementing the system in the workplace. In 25% of the cases the gap between training and implementation was five months or more. On average, the respondents had been employing CMS for a mean of 11 months (SD = 4.8) by the time of this survey. Thirty-one percent had been using the system for one year or longer.

Thirteen percent of the respondents reported that they did not use the CMS system, 13% applied it in ten percent or less of their cases, 23% used it with eleven to fifty percent of their caseload, and 22% used it with fifty-one to ninety-nine percent of their cases. A full 28% of the CMS users reported that it was applied to one hundred percent of their caseloads. Table 3 presents more complete information on the frequency of use of CMS by respondents.

Most respondents (85%) reported that they begin working with CMS sixteen days or more after having a client assigned to their caseload. In ten percent of the cases the gap between case assignment and CMS administration was over sixty days. The figures for the temporal gap between case assignment and CMS use are tabulated in more detail in Table 3.

Table 3 - Use of CMS: Percentage of Caseload (N = 498) and Time Between Case Assignment and CMS Use (N = 466)

CMS (Percentage of Caseload)	f	%
0	65	13.3
1 - 10	63	12.9
11 - 20	27	5.5
21 - 30	26	5.4
31 - 40	17	3.4
41 - 50	43	8.8
51 - 60	7	1.4
61 - 70	11	2.2
71 - 80	48	9.9
81 - 90	35	7.1
91 - 99	9	1.8
100	137	28.1
Time Period	f	%
Within 15 Days	66	14.8
Within 16 to 30 Days	171	38.3
Within 31 to 60 Days	163	36.5
Beyond 60 Days	46	10.3

Finally, Table 4 presents the proportion of case management time devoted to various activities. Only 11% of the respondents reported spending more than 60% of their time actually counselling offenders.

Table 4 - Proportion of Time CMS Devoted to Various Activities: Percent Distributions

Activities	Proportion of Time
	0 - 20%	21 -40%	41 - 60%	61 - 80%	81 -100%
Identitying Problems/Needs	33	33	20	11	3
Negotiationg Individual Treatment Plants	47	32	12	8	1
Monitoring Progress of Plans	46	31	15	7	1
Counselling Offenders	34	32	22	9	2

Opinions Regarding Initial Training in CMS

General satisfaction with initial training in CMS was assessed with a single question: "Overall, how would you rate your satisfaction with the initial training in CMS that you received?" (question 2.14). Responses were scored on a five point scale from '1' (Very Dissatisfied) to '5' (Very Satisfied). Satisfaction with initial training was distributed symmetrically about a mean of 3.02 (SD = 1.12, N = 512). Inspection of Table 5 reveals that the respondents were evenly distributed among the satisfied, dissatisfied and middle ranges of satisfaction.

In order to anchor evaluations of CMS training, respondents were also requested to evaluate CMS training relative to the most useful and most important courses or workshops that they had completed since completing their formal education (question 2.13). The responses were '1' (less useful/important) through '3' (equally useful/important) to '5' (more useful/important). As Table 5 reveals, 40% of the respondents thought that CMS training was as good or better than the best of the training that they had previously received (mean = 2.23, SD = 1.12, N = 464).

Table 5 - Rated Usefulness of CMS Training Relative to Other Courses and Workshops and Satisfaction with CMS Training: Percent Distribution and Means/SDs.

A. Rated Usefulness/Importance
Less useful/important		Equally useful/important	More useful/important
1	2	3	4	5	Mean	SD	N
33	27	27	9	4	2.23	1.12	464
B. Rated Satisfaction
Very Dissatisfied			Very Satisfied
1	2	3	4	5	Mean	SD	N
10	23	32	26	9	3.02	1.12	512

The findings reveal, on average, a moderate level of satisfaction with training, although a full third of the respondents expressed explicit dissatisfaction and a third expressed explicit satisfaction. In view of substantial variation in the evaluations of initial training, characteristics of respondents, work settings, and CMS training were examined in order to identify any factors that might distinguish between the less and more satisfied respondents. The characteristics of respondents surveyed included age, gender, level of education, major or area of concentration during postsecondary education, years of service with CSC, and position in CSC. The organizational variables were institutional vs. community setting and region. The training variables were type of trainer (level of NCCD involvement) and total days of initial training in CMS completed.

Personal and Organizational Correlates of Rated Satisfaction: Rated overall satisfaction with training was unrelated to age ( r = 0.02, N = 500), being male ( r = -.07, N = 505), level of education ( r = -.03, N = 505) and total years of CSC service ( r = -.06, N = 504). However, rated satisfaction with training did vary with postsecondary major/concentration ( p < .05), current position within CSC ( p < .06), type of work setting ( p < .02), and region ( p < .01). Inspection of Table 6 reveals that the lowest mean satisfaction ratings were reported by employees who majored or concentrated in criminology / law / criminal justice, by supervisors/managers and community-based CMSs, by community-based respondents generally, and by respondents from the Quebec region. The highest mean satisfaction ratings were reported by employees who majored in nontraditional areas and in psychology / sociology, by institution-based employees, and by respondents from the Atlantic region.

Training-based correlates: Both Days in Training and Type of Trainer were associated with overall satisfaction ratings. The greater the number of training days reported, the more positive were the satisfaction ratings, r = .15, p < .000, N = 505. Inspection of Table 6 additionally reveals that mean rated satisfaction increased with the fact or perception that one's trainer was from NCCD or trained by NCCD (eta = .29, p,.000).

Pretraining attitude toward CMS: A positive pretraining attitude toward CMS was associated with relatively high levels of satisfaction with CMS training: r = .24, p < .00, N = 507. Pretraining attitudes toward CMS were, of course, assessed retrospectively on the CMS survey instrument (question 2.6: the distribution of responses is presented in Table 7). A second pretraining attitudinal variable was anticipated effect of CMS on the amount of time required to manage individual cases (question 2.9: see Table 7). This retrospectively assessed pretraining variable was unrelated to rated satisfaction with initial training ( r = .04, N = 509).

Multiple correlations with overall satisfaction: A series of analyses of variance in rated satisfaction were conducted in order to estimate the multiple correlation as well as explore potential sources of some of the above-noted simple correlations with satisfaction scores. For example, it is possible that the apparent effects of region and/or being a community-based employee may reflect systematic differences in the number of days of training received or differences in pretraining attitudes. Our strategy was to first conduct an analysis of variance with the following variables entered as factors: Type of Trainer (NCCD trainer - CSC trainer trained by NCCD - CSC trainer not known to have been trained by NCCD), Setting (Institution - Community), Position (CMO - Others), Major/Concentration (Criminology - Others) and Region (Atlantic Ontario/Prairies/Pacific - Quebec). Subsequent analyses involved introducing Days in Training and Pretraining Attitude Toward CMS as covariates. The covariates were entered separately first and then combined. This approach yields a) an estimate of the overall multiple correlation with rated satisfaction, b) information on those correlates of satisfaction when other potential predictors are statistically controlled and c) suggestions regarding the sources of some of the simple correlations that were reviewed above. Findings regarding each of these three issues follow:

a)  The multiple correlation of the total set of correlates with rated satisfaction was .44, F(28/445) = 3.86, p < .000. This estimate includes all possible first order interactions among the four factors entered, but these interactions were statistically nonsignificant ( p > 0.20). In other words, there was no reason to believe that consideration of particular combinations of Trainer and Region (or Setting and Position etc.) improved upon the prediction of rated satisfaction achieved by simply considering the contribution of each factor and covariate in the manner illustrated in Table 8.

b)  Inspection of Table 8 reveals that, with the exception of Position, each factor was significantly associated with rated satisfaction when examined in isolation from each of the other factors. The eta values are measures of the magnitude of that association found when other factors are not statistically controlled. The beta values in the final column reflect the magnitude of association between each factor and satisfaction scores with each of the other three factors and two covariates statistically controlled. Thus, each of the following variables was making a statistically significant ( p < .05) contribution to the multiple correlation with rated satisfaction with initial training: Type of Trainer (Beta = 0.25), Setting (Beta = .14), Days in Training (Beta = .12) and Pretraining Attitude (Beta = .21).

Table 6 - Overall Satisfaction With Initial Training in CMS by Background, Position, Setting, and Region (Based on a 5-point Scale With 1 = Very Dissatisfied and 5 = Very Satisfied)

	Mean	SD	N	Multiple Range Tests
Major concentration
1. Criminoligy	2.79	1.05	157	Group 1 differs significantly from Groups 5 & 6: p<.05
2. Other	2.98	1.32	46
3. Social Work	3.08	1.16	51
4. Sociology	3.13	1.10	61
5. Psychology	3.13	1.12	120
6. Unknown	3.17	1.06	77
Position
1. Supervisor/Manager	2.79	1.11	52	Groups 1 & 2 differ significantly from Group 4 p<.05
2. Case Management Officer (Community)	2.91	1.06	196
3. LUOs	3.02	0.99	65
4. Case Management Officer (Institution)	3.16	1.89	184
Setting
1. Community	2.82	1.11	218	Group 1 differs significantly from all other groups: p<.05
2. Medium Institutions	3.13	1.06	142
3. Maximum	3.15	1.13	73
4. Minimum	3.20	1.21	35
5. Other Institutions	3.24	1.34	37
Region
1. Quebec	2.70	0.97	136	Group 1 differs significantly from groups 3, 4, 5,: p< .05
2. Pacific	2.95	1.13	77
3. Prairies	3.16	1.23	12
4. Ontario	3.20	1.21	35
5. Atlantic	3.19	1.34	37
Trainer
1. CSC Staff/Not NCCD Trained	2.73	1.12	33	Groups 1 & 2 differs significantly from Groups 3 & 4, and Group 3 differs from Group 4: p<0.05
2. CSC Staff/Bsckground Not Known	2.79	1.07	268
3. CSC Staff/ NCCD Trained	3.20	1.00	163
4. NCCD Trainer	4.00	1.32	33

Table 7 - Rated Attitude to Prospect of Implementing CMS and Anticipated Changes in Amount of Time Required to Manage Individual Cases: Percent Distributions and Means/SDs

A. Attitude Towards Implementation of CMS
Negative Attitude			Positive Attitude
1	2	3	4	5	Mean	SD	N
8	12	45	21	13	3.19	1.08	516
B. Anticipated Change in Amount of Time Required for Individual Cases
Increased Time			Decreased Time
1	2	3	4	5	Mean	SD	N
33	18	31	8	10	2.44	1.30	514

c)  Reinspection of Table 8 reveals that the introduction of controls for other factors and the covariates had little effect on the estimated magnitude of the correlation between Type of Trainer and rated satisfaction with training. An original eta of .28 was reduced minimally to a beta of .25 with all other variables controlled. Similarly, the predictive contribution of Setting was minimally influence by the introduction of statistical controls for the other factors and the two covariates (original eta = .15, final beta = .14)./ In contrast, the contribution of Major/Concentration was reduced form a value of .15 to a nonsignificant .06 immediately upon introduction of controls for the other three factors. This reduction was traced to an over-representation of criminology majors in community as opposed to institutional settings ( r = .09). The reduction of the contribution of Region to the multiple correlation with rated satisfaction followed two paths. First, the reduction of an eta of 0.20 to a beta of .12 was traced to the fact that respondents from the Atlantic region included an over-representation of employees whose trainers were involved with NCCD ( r = .12) and employees who had not completed studies in criminology ( r =.48). Second, the effect of region was reduced to nonsignificant levels once controls for Pretraining Attitudes toward CMS were introduced (from 0.12 to .08). Respondents from the Atlantic region were more positive in their attitudes to CMS prior to training that were respondents from other regions ( r = .30).

Table 8 - Mean Satisfaction with Initial Training by Type of Trainer, Setting, Major/Concentration, Region and Position; Means and Etas with and without Adjustments for Other Factors and Covariates

	Unadjusted			Adjusted for
	N	M	E	FB	FDB	FAB	FDAM	B
Trainer
NCCD	31	3.94	.28	.27	.25	.25	3.78	.23
CSC-NCCD	154	3.20					3.17
CSC-?	289	2.79					2.83
Setting
Institution	265	3.14	.15	.14	.14	.14	3.13	.13
Community	209	2.82					2.83
Major
Criminology	146	2.76	.15	.06	.06	.06	2.90	.06
Other	328	3.11					3.05
Region
Atlantic	48	3.21	.20	.12	.11	.08	2.95	.08
Ont./Pr./Pac.	300	3.12					3.06
Quebec	126	2.63					2.86
Position
CMO	360	3.03	.04	.09	.09	.08	3.05	.08
Others	114	2.92					2.85
N: number of cases M: Mean E: Eta B: Beta F: Factors D: Days in Training (covariate) A: Attitude (covariate)

Quality of training by area of case management: Quality of initial training was also assessed in particular areas of case management. Percent distributions over the five-point quality of training ratings are presented in Table 9. The overall pattern of results is quite clear. While 88% of the respondents were at least moderately positive regarding training in the scripted interview, less than 50% of the respondents rated their training positively for purposes of making recommendations to decision makers or understanding where CMS stands in relation to the overall case management process. Quality ratings in the areas of developing treatment plans took an intermediate position.

Table 9 - Rated Quality of Initial Training In CMS In Specific Areas: Percent Distribution and Means/SDs

	Rated Quality
	Low		Moderate	High		M	(SD)	Missing (N)
	1	2	3	4	5
Scripted Interview	5	7	27	40	21	3.64	(1.06)	475 (53)
Apply Strategies	11	15	35	29	10	3.12	(1.22)	472 (56)
Force Field Analysis	13	16	34	25	12	3.07	(1.18)	472 (56)
Defining Problems/ Needs	10	19	36	25	5	3.08	(1.12)	470 (58)
Set Long-Term Goals	13	21	36	21	10	2.93	(1.15)	472 (56)
Set Short-Term Goals	13	19	36	23	9	2.95	(1.14)	473 (55)
Develop Action Plan	14	22	35	21	8	2.89	(1.14)	473 (56)
Develop Referral Plan	16	23	35	17	9	2.80	(1.17)	463 (65)
Negotiation With Offender	19	26	32	17	6	2.64	(1.15)	467 (61)
Recommendations to Instutions	33	24	27	11	4	2.30	(1.17)	454 (740)
Conditional Release Recommendations	36	25	24	11	4	2.23	(1.17)	458 (70)
CMS re Overall Responsibilities as CM officer	30	20	27	17	6	2.49	(1.25)	464 (64)
CMS re Overall Responsibilities as Supervisor	35	21	24	14	6	2.34	(1.25)	429 (99)

Open-ended comments regarding training. Of all the open-ended material, training received the most positive comments. In particular, trainers were praised even when comments concerning other aspects of training and CMS were negative:

Well-trained, lively trainers, who, with their attitude and approach, made CMS initially interesting, easily obliterating some of the trainee's skeptical cynicism.

Another fairly frequent set of comments concerned the amount of training. Either the duration of training was too short, or there was not enough training on specific sections of the CMS, such as forcefield analysis or in case planning. Notable among the criticisms of training was considerable resentment by Living Unit Officers whose rank apparently precluded extensive training. They argued that there was little relationship between extent of use of CMs and amount of training received. This issue is developed further in our discussion of inter-staff communication.

Summary of overall evaluations of initial training in CMS: The respondents were greatly divided regarding the overall quality of initial training that they received on CMS. Roughly, a third of the respondents fell in each of the dissatisfied, middle and satisfied categories. A relatively positive rated satisfaction with initial training was traced to a) having an NCCD trainer or a trainer who was thought to have been trained by NCCD staff, b) working in an institution, c) total number of days of initial training and d) a positive pretraining attitude toward implementation of CMS. Regional variation in satisfaction with training appeared to reflect type of trainer (links with NCCD), postsecondary area of study (not criminology), and positive pretraining attitudes toward CMS. According to the comments, training was enjoyable because the trainers were knowledgeable, interesting, and well-prepared.

General Opinions Regarding CMS in the Workplace

An overall assessment of opinions regarding CMS in the workplace was obtained from the following item (question 2.8): "Overall, how do you feel now regarding the implementation of CMS?" As displayed in Table 10, fewer than 20% (19%) of the respondents felt that the implementation of CMS was worth the effort, and 53% felt that it was not worth the effort. Thus, the average opinion regarding CMS implementation was negative (mean = 2.42).

Table 10 - Overall Attitude Toward Implementation of CMS: % Distribution and Means/SDs

Not At All Worth The Effort			Very Much Worth the Effort
1	2	3	4	5	Mean	SD	N
28	25	27	15	4	2.42	1.17	514

The open-ended comments revealed the strength of feeling which lies behind these rated evaluations of CMS. The comments also expand upon the reasons for the lack of endorsement of CMS. For example, there is a belief that CMS detracts from the professionalism of the officer, making a trained, experienced person into a "clerk" through removing discretion and discounting officer judgment. Many statements were made regarding the officers' ability to reach the same conclusions through less formal methods in a fraction of the time it takes to administer CMS. Often, a set of global remarks, such as the following, would conclude the questionnaire:

Does anybody actually believe that society is better protected because we have another assessment tool which is cumbersome, obsolete, ineffective, inappropriate, and time consuming? CSC has sold this model to society as a panacea for the problems that were outlined in the Ruygrok inquiry. In fact, I believe that society is less well protected because CSC is too busy protecting its own policies. Someone should do a study of what parole officers spend most of their time doing. They would be shocked to see how much time is wasted on bureaucratic BS.

This particular set of remarks is quoted extensively because it indicates the overall dissatisfaction with CMS, as well as some more specific areas of discord. The idea that is the officer's ultimate duty to protect the public was echoed in many of the comments. This idea was usually coupled with a statement that bureaucratic decisions are self-serving and intended to benefit administrators at the expense of officers, offenders and the public. Other comments implied less sinister intentions; however, many saw the instituting of CMS as one more recent instance of poor judgment based upon ignorance of the nature of the officer's job and the nature of the cases involved.

The following material examines some personal, organizational and training variables in relation to overall evaluations of CMS in the workplace.

Personal and organizational correlates of the general evaluation: Rated worthiness of the CMS effort was unrelated to age ( r = 0.00, N = 503), being male ( r = -.06, N = 507), level of education ( r = -.01, N = 506) and total years of CSC service ( r = -.01, N = 506) However, once again evaluations of CMS did vary with postsecondary major/concentration ( p < .025), current position within CSC ( p < 0.002) and region ( p < .000). Means and SDs are presented in Table 11, where it is found that the lowest mean evaluation scores were given by employees who majored or concentrated in criminology/law/criminal justice, by case management officers, and by respondents from the Quebec region. The highest mean evaluations of CMs were given by employees who majored in nontraditional areas, by supervisors of CMOs and Living Unit Officers, and by respondents from the Atlantic region. As was the case in rated satisfaction with training, community-based employees were tending to be less positive regarding CMs than were institution-based employees.

Table 11 - Overall Attitude Toward Implementation of CMS by Background, Position, Setting, and Region (Based on a 5-point Scale with 1 = Not At All Worth the Effort and 5 = Very Much Worth the Effort)

	Mean	SD	M	Multiple Range Tests
Background
1. Criminoligy	2.22	1.09	160	Group 1 and 6 and 2 and 6 differ significantly; p<.05
2. Other	2.34	1.12	119
3. Social Work	2.43	1.19	61
4. Sociology	2.59	1.17	46
5. Psychology	2.62	1.32	50
6. Unknown	2.72	1.19	78
Position
1. Case Management Officer (Community)	2.28	1.11	194	Groups 1 differs significantly from Groups 3 and 4 as does Group 2 p < .05
2. Case Management Officer (Institution)	2.34	1.14	188
3. Supervisor/Manager	2.77	1.35	52
4. Other	2.78	1.02	65
Setting
Institution	2.82	1.11	218	No significant differences
Maximum Institutions	2.48	1.16	67
Medium	2.34	1.07	133
Minimum	2.39	1.22	33
Community	2.03	1.13	191
Region
1. Atlantic	3.30	1.20	53	Group 1 differs significantly from groups 2, 3, 4, 5: p< .05
2. Pacific	2.44	1.14	77
3. Ontario	2.42	1.13	123
4. Prairies	2.26	1.13	121
5. Quebec	2.21	1.08	140
Trainer
1. CSC Staff/ Not NCCD Trained	2.15	1.12	33	Groups 1 & 2 and 3 differ significantly from Groups 4 p < 0.05
2. CSC Staff/ No Knowledge	2.27	1.12	268
3. CSC Staff/ NCCD Trained	2.46	1.13	162
4. NCCD Trainer	3.58	1.00	33

Table 12 - Mean General Evaluation of CMS in Workplace by Type of Trainer, Setting, Major/Concentration, Region and Position: Means and Etas with and without Adjustments for Other Factors and Covariates

	Unadjusted			Adjusted for
	N	M	E	FB	FDB	FAB	FDMA	B
Trainer
NCCD	31	3.62	.29	.25	.20	.20	3.12	.17
CSC-NCCD	153	2.47					2.46
CSC-?	289	2.26					2.32
Setting
Institution	266	2.50	.08	.08	.08	.07	2.47	.07
Community	209	2.32					2.33
Major
Criminology	147	2.20	.13	.07	.06	.07	2.31	.06
Other	326	2.52					2.47
Region
Atlantic	48	3.30	.27	.22	.23	.17	3.21	.17
Ont./Pr./Pac.	298	2.39					2.46
Quebec	127	2.17					2.32
Position
CMO	359	2.31	.17	.16	.16	.18	2.30	.18
Others	114	2.76					2.78
N: number of cases M: Mean E: Eta B: Beta F: Factors D: Days in Training (covariate) A: Attitude (covariate)

Table 12 - Rated Effect of CMS on Current Workload

Increase		No Change	Decrease
1	2	3	4	5	Mean	SD	N
49	22	25	3	1	1.85	0.96	505

Training-based correlates of general evaluations of CMS: Both Days in Training and Type of Trainer were associated with the evaluations of CMS. The greater the number of training days reported, the more positive was the evaluation of CMS ( r = .16, p < .000, N = 507) and mean evaluations increased with the fact or perception that one's trainer was from NCCD or trained by NCCD (eta = .28, p,.000).

Pretraining attitude toward CMS>: A positive pretraining attitude toward CMS was associated with relatively positive evaluations of CMS in the workplace: r =.31, p < .000, N = 507.

Multiple correlations with overall evaluation of CMS: The same series of analyses of variance that were conducted on rated satisfaction with training were conducted on rated worthiness of the CMS implementation effort. The multiple correlation of the total set of correlates with evaluations of CMS was .51, F(28/473) = 5.99, p. < 0.000. None of the terms reflecting interactions among the factors were statistically significant. With the exception of Setting, each factor was significantly associated with evaluations of CMS when examined in isolation from each of the other factors. The eta values are presented in Table 12. Once again, the beta values in the final column reflect the magnitude of association between each factor and evaluation scores with each of the other three factors and two covariates statistically controlled. Each of the following variables was making a statistically significant ( p < 0.05) contribution to the multiple correlation with general evaluations of CMs: Type of Trainer (Beta =.17), Position (Beta = 0.18), Region (Beta =.20), Days in Training (Beta .14) and Pretraining Attitude (Beta = 0.29).

Workload Implications

A second general assessment of CMS in the workplace was conducted by way of an item on workload (question 2:11: see Table 13). Inspection of Table 13 shows that the vast majority of the respondents (71%) felt the amount of time required to manage individual cases had increased with the introduction of CMs. Only 25% reported no change and 4% reported a decrease in time required to manage a case. Interestingly, reports of decreased workload were significantly but only weakly correlated with the other general evaluation of CMS ( r = 0.19) and uncorrelated with general satisfaction with initial training( r = 0.01).

The burden of increased workload was the focus of many comments on the questionnaires. The workload increase was linked to increased paperwork, ritualistic and meaningless recording activity, and diversion of attention toward bureaucratic functions and away from time actually spent with offenders.

Some comments indicated that the implementation of CMs was a victim of unfortunate timing. It was one of a number of newly-instituted tasks which increased administrative work (e.g., Critical Cases). Coupled with cutbacks in staff, CMS is viewed to entail more work than a CMO can handle with appropriate facility:

The addition of CMS to a Case Management Officer's workload has put a tremendous strain on that workload to the point where many officers are feeling completely frustrated (anI have known more than one who actually left the service as a result). While we were led to believe that CMS would decrease our workload, in reality, it has almost doubled it. ... we have a group of CMOIs who are finding that we are now so busy trying to keep up with administrative requirements that actual counselling is almost non-existent and case monitoring has become almost mechanical. I believe that CMS could be highly effective, if caseloads were sufficiently small, however in our case, it has been nothing more than a complete irritation... .

Some More Specific Evaluations of CMS in the Workplace

Evaluations of CMS in the workplace were also conducted with reference to stage of the case management process, the contributions of CMS to recommendations made to decision-makers, and the facilitation of other objectives such as improved staff communication and use of program resources. On average, the respondents were moderately satisfied with the contributions made by CMS to understanding the offender immediately after admission (mean = 3.04, SD = 1.13) but their reported satisfaction levels were notably lower at the intermediate phase of incarceration (2.57/1.05) and at the stage of community-based supervision (2.44/1.17: see Table 14). Inspection of the bottom portion of Table 14 reveals that 39% of the respondents thought that CMS was an improvement over prior case management methods in use immediately after admission, and 82% thought CMS was at least as good as prior methods at this early stage of case management. However, nearly 30% of the respondents felt that CMS was worse than prior methods of understanding cases at the late stages of case management.

The perceived lack of utility at the later stages of case management was linked to program resources in many of the open-ended comments. In particular, respondents noted the lack of resources to follow through on case plans:

I have found no change in assessment of treatment through using CMS. How could there be? It adds no resources...

Table 14(A) - Extent of Satisfaction With CMS at Various Stages of the Case Management Process: Percent Distributions and Means/SDs

Producing a clear understanding of offender:	Very Dissatisfied			Very Satisfied
	1	2	3	4	5	Mean	SD	N
Immediatly after admission	13	16	32	32	7	3.04	(1.13)	441
Intermediate Phase	20	23	39	15	2	2.57	(1.05)	429
for release planning and parole recommendation	26	25	30	16	3	2.44	(1.13)	450
for community - based supervision	29	22	30	15	4	2.44	(1.17)	429

Table 14(B) - Extent to Which CMS Represents An Improvement Over Prior Methods: Percent Distributions and Means/SDs

Producing a clear understanding of offender:	Worse			Better
	1	2	3	4	5	Mean	SD	N
Immediatly after admission	7	11	43	29	11	3.25	(1.00)	426
Intermediate Phase	10	17	55	15	3	2.86	(0.91)	416
for release planning and parole recommendation	15	15	52	14	3	2.75	(0.99)	423
for community - based supervision	14	15	52	13	6	2.81	(1.02)	411

Inspection of Table 15 reveals that mean rated extent of assistance in making recommendations to decision-makers were all below 3.00 on a five-point assistance scale. However, relative to prior methods of case management, CMS was judges as good or better by over 80% of the respondents for purposes of making program and mental health referrals. Generally, while CMS was given mean ratings well below "3" on extent of assistance, it was not, on average, judged to be significantly worse than prior methods in this regard.

Table 15 - Rated Extent of Assistance in Making Recommendations to Decision-Makers: Percent Distribution and Means/SDs

	Very Low			Very High
	1	2	3	4	5	Mean	SD	N
Program Referrals	23	14	29	28	6	2.81	1.25	435
Referrals: Psychological	33	15	27	21	4	2.48	1.25	439
Referrals: Psychiatric	33	16	26	20	4	2.44	1.24	439
Work Placement	32	21	31	14	3	2.36	1.15	419
Temporary Absence	33	24	28	12	3	2.28	1.13	409
Transfer	33	23	28	13	2	2.27	1.12	411
Release Plans	29	17	32	18	4	2.49	1.19	445
Conditional Release	32	19	31	14	3	2.36	1.15	436
Community Program Referrals	25	20	34	17	4	2.54	1.14	454
Community Supervision	34	21	29	13	3	2.29	1.15	441
Suspension/ Revocation	48	19	24	7	2	1.96	1.07	436

One of the desirable characteristics of a shared case management system is presumed to be the extent to which it will facilitate communication among staff and supervisors. Inspection of Table 16 shows that CMS received very negative evaluations regarding the facilitation of communication about individual cases. The mean ratings were uniformly well below 3.00 and frequently below a mean of 2.00.

Table 16 - Rated Extent to Which CMS Facilitates Communication With Others: Percent Distributions and Means/SDs

	Very Low			Very Much
	1	2	3	4	5	Mean	SD	N
Your Spervisor	41	20	24	12	2	2.14	(1.15)	482
CM Officers Under Your Supervision	54	14	20	11	4	1.93	(1.15)	238
Other Supervisiors	48	20	23	8	1	1.95	(1.07)	449
Other CM Officers	38	20	21	18	3	2.28	(1.24)	470
Correctional Officers	62	19	14	5	0	1.64	(0.93)	450
Psychologists	49	16	23	11	5	2.00	(1.12)	457
Community Counsellors	58	16	20	6	1	1.77	(1.03)	444
Community Agencies	60	14	19	6	1	1.75	(1.04)	451
Instructors/ Teachers	61	14	18	6	0	1.69	(0.98)	416
Others	33	7	23	20	17	2.80	(1.52)	30

The comments of respondents suggested that training issues were an important source of difficulty among staff, especially when Living Unit Officers are involved. The following comments, first from a CMO and second from a LUO are illustrative:

CMO: Their (LUOs) lack of training makes it just that much more work for us to supervise and assist them in this process...

LUO: It is an absolute necessity to provide LU officers as caseworkers a full 5 days training in the CMS process so that they may conduct the interview without the presence of a CMO.

Another LUO writes:

We, the Living Unit Officers, complete the CMSs and all the rest of the work. And we received only two days of courses. The Case Management Officers received 5 days of courses. The only work they do is verify the quality of our work. With all the other supervisors that we have, I cannot understand their reason for being, i.e., what do they do for a living?

If fact, the LUOs did report significantly fewer initial training days (Mean = 3.53, N = 66) than did the CMOs and Supervisors of CMOs (who averaged just under 5 days). Of the 13% of the respondents who reported some refresher training in CMS, not a single one was a LUO. Recall, however, that LUOs rated CMS more positively in the workplace than did CMOs, and that the contributions of Position and Training were making independent contributions to the prediction of evaluations of CMS (as reviewed in Table 12). (Appendix 2 presents a detailed comparison of the responses of CMOs, LUOs and Supervisors of CMOs. There it will be noted that the LUOs in our sample were from the Prairies and Pacific regions).

Finally, Table 17 presents the rated evaluations of CMS with regard to potential contributions to testimony in court, effective use of current program resources and the identification of new programming directions. Regarding court testimony, CMS was rated valuable by a scant 14%, while a full 42% saw CMS as not at all valuable in court. The ratings were slightly more positive with regard to use of current program resources, but still 51% of the respondents were clearly negative. Similarly, 43% were clearly negative regarding the contributions of CMS to program development.

Again, according to comments, resources - mainly staff - are not available to realize case plans. There was substantial mention of circumvention of directives regarding case plans and suggestions that CMS was not taken seriously: "I have yet to see a case where the plan has been completed". While comments indicated that a plan is certain to be in each file, the plan may have little to do with the needs of the case. Indeed, it was mentioned that some settings have four standard write-ups prepared that correspond to a CMS classification and one of these four set pieces are placed in case files.

Summary: The evaluations of CMS in the workplace were, on average, negative. However, the reports from the Atlantic region were more positive than those from other regions. Generally, the evaluations were more positive regarding contributions during the early phases of incarceration than when the offender is in the community. The next section of the report explores scientific elements of CMS.

Evaluations of Scientific Elements of CMS

Assessments of the accuracy and appropriateness of CMS offender descriptions and strategies were generally positive (Table 18). Over 80% of the respondents considered the descriptions of offenders to be at least somewhat accurate and the CMS strategies to be at least somewhat appropriate. However, 82% of the respondents expressed some lack of confidence regarding whether different officers would classify a particular offender in the same way with CMS. Written comments expressing doubts about reliability suggested that CMS communicates an illusion of objectivity, while, in reality, it elicits subjective judgments. Several respondents commented upon what they saw as an over-reliance on the self-reports of inmates.

Table 17 - Rated Value of CMS In Various Areas: Percent Distributions and Means/SDs

A. Value in court Testimoty and Public Inquiries
Not at All Valuable			Extremely Valuable
1	2	3	4	5	Mean	(SD)	N
42	18	26	13	1	2.13	1.14	487
B. Contribution of Effective Utilization for Current Program Resources
Makes No Contribution			Significant Contribution
1	2	3	4	5	Mean	(SD)	N
29	22	22	22	5	2.53	1.26	500
C. Contribution to Identification of New Programming Directions
Makes No Contribution			Significant Contribution
1	2	3	4	5	Mean	(SD)	N
23	19	24	24	9	2.77	1.30	499

When questioned regarding the ease of conducting and scoring the scripted interview, over 70% responded in an non-negative manner (Table 19). However, 37% were clearly negative when asked to compare the scripted interview with past practices.

There were a number of positive comments about the interview. Some officers noted that the interview "opened up" the offenders, eliciting information beyond merely responses to the questions. Experienced officers considered the interview to be particularly valuable to newer staff, because it ensured that relevant issues would be covered. Many comments were in praise of the standardization provided by the scripted interview. The interview was also praised as an "icebreaker":

The interview and the force field analysis forces you to take the time to get to know the offender, and upon completion, you have a lot of knowledge and insight which will in fact help you to deal with that offender.

Table 18 - Evaluations of the Reliability, Accuracy and Appropriateness of CMS Interview and CMS Strategy Types

4.2 Likelyhood of Agreement Between Officers: Percent Distributions and Mean/SD
	Very Likely			Very Unlikely
	1	2	3	4	5	Mean	(SD)	N
	3	15	22	40	20	3.59	1.06	503
4.3 Accuracy of Description of Offenders by Strategy Types: Percent Cistributions and Means/SD
	Not at All Accurate			Very Accurate
	1	2	3	4	5	Mean	(SD)	N
Selective Intervention	2	13	25	41	18	3.60	1.01	497
Casework Control	1	11	30	43	15	3.60	0.92	496
Environmental Structure	2	9	30	43	16	3.60	0.93	496
Limit Setting	2	10	25	44	19	3.67	0.96	497
4.4 Appropriateness of Strategies Associated With Each Type: Percent Distributions and Means/SD
	Not at All Appropriate			Very Appropriate
	1	2	3	4	5	Mean	(SD)	N
Selective Intervention	5	13	29	38	16	3.48	1.05	491
Casework Control	4	14	32	38	13	3.41	1.00	490
Environmental Structure	4	13	32	38	13	3.42	1.01	490
Limit Setting	6	13	29	36	16	3.43	1.09	492

Table 19 - Assessment of the Scripted Interview: Percent Distributions and Means/SDs

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)	N
Ease of Conducting	10	15	29	34	12	3.22	1.16	497
Ease of Scoring	14	16	29	28	13	3.10	1.22	496
Comprehensiveness of Content	9	23	27	31	9	3.08	1.13	493
Compared With Past Practice	17	20	33	21	9	2.86	1.19	482

On the other hand some were offended by the formality of a scripted interview:

I can find out more from a casual conversation than this structured interview. Inmates are more open when they don't think I'm trying to analyze them.

Some questioned the appropriateness of the interview for long term incarcerates, older offenders, Natives, and offenders with limited verbal skills. Others felt that the scripted interview was not detailed enough in the areas of substance abuse, sexual deviance and sexual orientation. Some respondents noted irritating mechanical problems with scoring the interview, complaining that it is difficult to line the template with the holes. Finally, several comments questioned the validity of adopting an instrument developed in the United States when home-grown possibilities existed.

Returning to the quantitative scales, the mean evaluations of the force field analysis of needs were relatively positive, with mean ratings all exceeding 3.00 (Table 20). However, once again, 30% of the respondents were clearly negative about CMS relative to past practices. Some respondents commented that a former system (NAP) was preferable to the force field analysis. Very specifically, there were several complaints about the amount of space provided for case recording when the instructions called for detailed commentary.

Table 20 - Assessment of the Force-Field Analysis of Needs: Percent Distributions and Means/SDs

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)	N
Logic and Rationale	8	13	40	32	8	3.18	1.02	483
Ease of Identifying Strengths/ Resources	7	13	38	35	7	3.22	0.99	485
Ease of Identifying Problems/ Weaknesses	7	13	36	37	8	3.25	1.01	484
Ease of Priorizing Needs	9	16	37	32	6	3.12	1.04	482
Compared With Past Practice	13	17	41	19	9	2.95	1.12	478

The mean evaluations of the individual correctional treatment planning process (Table 21) were less positive than evaluations of the scripted interview and force field analysis. Here the means fell below 3.00 and a full 43% of the respondents thought that CMS was less useful than past practice. Comments made here concentrated on the perception that the plan was highly theoretical in view of heavy workloads. One officer noted that planning depended more upon available resources than upon assessments of need:

my plans are often built as a function of the available resources. This means that in a case where a non-existent resource would meet the needs of the offender, I abstain from identifying (them in the plan).

Generally, comments suggested that even though CMS-informed planning may have merit, it is not applicable under the present workload and resource conditions.

With the exception of identifying criminogenic factors (see Table 22), the mean usefulness ratings of CMS for various aspects of case management fell below 3.00. The mean ratings were particularly low for identifying high risk (mean = 2.47) and violent offenders (mean = 2.34). Once again, the lowest ratings were obtained on linking case management to NPB decision-making policies (mean = 1.94).

Table 21 - Assessment of the Individual Correctional Treatment Planning Process: Percent Distributions and Means/SDs

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)	N
Logic and Rationale	13	21	39	24	4	2.85	1.04	469
Ease of Identifying Problem Areas	11	19	38	28	4	2.93	1.03	475
Ease of Identifying Long- Range Goals	15	22	39	21	4	2.76	1.06	473
Ease of Identifying Short-Term Goals	13	20	39	25	3	2.84	1.04	472
Ease of Preparing Offender Action Plan	15	21	40	21	4	2.78	1.06	472
Ease of Preparing Officer Referral Plan	16	21	40	19	4	2.73	1.06	465
Ease of Applying Interventions	20	24	39	15	3	2.57	1.04	464
Compared With Past Practice	23	23	38	13	4	2.53	1.10	458

Another theoretically desirable feature of a case management system is that it be of value/benefit in the supervision of case management officers, that is, in monitoring the timeliness, accuracy and completeness of the CMOs work. Across four items sampling supervision issues (see Table 23), approximately 50% of the respondents expressed the clear opinion that CMS was not useful for purposes of supervising CMOs. Our basic series of analyses of variance were conducted on the mean of responses to the four supervision items. Once again, positive evaluations of CMS were a significant ( p < .05) and incremental reflection of Days in Training, Positive Pretraining Attitude, Type of Trainer (trainer is linked with NCCD), Setting (being institution-based), Region (being in the Atlantic region), Position (not being a CMO), and Major/Concentration (not being a criminology graduate). Only the contributions of the latter variable were reduced to nonsignificant levels with the introduction of statistical controls for the other factors and covariates. With the interaction terms nonsignificant, the multiple correlation was .46, F (28/423) = 4.09, p < .000.

Table 22 - Contributions of CMS to Aspects of Case Management: Percent Distributions and Means/SDs (N = 486 to 493)

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)
Identifying Criminogenic factors	11	18	33	32	6	3.04	(1.09)
Proactive re Offender Problems	12	23	31	30	4	2.9	(1.08)
Focus on Positive Offender Change	12	26	34	24	4	2.82	(1.05)
Promote Cumulative Positive Change	15	28	35	20	6	2.64	(1.00)
Efficient Use of Energy/ Recources	24	26	28	19	3	2.51	(1.13)
Identifying Espectations	16	18	35	25	6	2.8	(1.13)
Increase Offender’s Ownership	29	27	25	16	3	2.37	(1.15)
Increace Offender’s Motivation	28	31	30	10	4	2.25	(1.00)
Structuring With Individuals	18	24	32	23	5	2.68	(1.08)
Documenting Responsible Activity	16	25	35	21	3	2.69	(1.06)
Identifying Risk Levels	22	28	31	16	2	2.47	(1.07)
Linking With Appropriate Program	13	24	35	26	2	2.80	(1.03)
Identifying Violent Offenders	31	23	29	15	2	2.34	(1.13)
Linking With NPB Policies	44	26	24	6	1	1.94	(0.98)

Summary: The scripted interview and its most immediate product - the strategy types received moderately positive evaluations. However, evaluations with reference to various aspects of case management and the supervision of case managers were notably more negative. Interestingly, the correlates of satisfaction with initial training, with CMS in the workplace and with CMS as a CMO supervision tool were very similar. Once again, positive evaluations tended to be forthcoming from the Atlantic region, from institution-based employees, from non-CMOs, from people who were not trained in criminology, from people trained by trainers with an NCCD link, from those who reported a greater number of days in training, and from those who reported having a positive attitude toward CMS prior to training.

Table 23 - Rated Utility of CMS in Monitoring CM Officers in Various Areas: Percent Distributions and Means/SDs

	Not at all Useful			Very Useful
	1	2	3	4	5	Mean	(SD)	N
CM Offecers Assesments of Offenders	26	25	27	19	3	2.48	(1.15)	487
Case Recording and Reporting	30	24	26	18	2	2.39	(1.15)	487
Treatment Planning	25	21	30	23	2	2.57	(1.15)	487
Programming Recommendations	26	21	29	22	2	2.53	(1.16)	487

Perception of How Other Staff View CMS

In social psychological terms, the behavioural significance of personal opinions and attitudes may be influenced by perceptions of the opinions and attitudes of significant others. In brief, personal opinions are more likely to be acted upon when people think that others share their opinions. Thus, the CMS survey requested ratings regarding how a variety of CSC employees evaluated CMS. Inspection of Table 24 reveals that the mean perceived evaluation of CMS was negative (substantially below a mean of 3.00) for all but three classes of CSC employees. The exceptional three were Case Management Supervisors, Line Management and Functional Management. This pattern of results suggests that the case management officers perceive high levels of consensus among line staff regarding low levels of utility for CMS. Interestingly, and perhaps reflecting the previously-noted difficulties between the CMOs and the LUOs surveyed, LUOs in fact were more positive regarding CMS than were CMOs. Overall, the respondents perceive a dramatic contrast between the opinions of management and the opinions of line staff.

The bottom portion of Table 24 presents the perceived familiarity of a variety of CSC staff with CMS. Clearly, the respondents - primarily composed of CMOs - consider themselves to be much more familiar with CMS than is management. In addition, it is notable that other key players such as security officers and psychologists are thought not to be familiar with a system that is intended to help guide the offender through the period of incarceration and back into the community.

Table 24(A) - Ratings of Other Staffs' Perceptions of CMS: Percent Distributions and Means/SDs

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)	N
Case Management Officers	27	34	24	13	2	2.29	(1.07)	494
Case Management Supervisors	12	25	31	21	11	2.94	(1.18)	473
Living Unit Officers	34	34	26	5	1	2.03	(0.93)	419
Security Officers	52	27	17	4	1	1.74	(0.92)	408
Psychologists	27	26	34	11	1	2.33	(1.04)	398
Others Programming Staff	23	29	38	9	2	2.39	(1.00)	397
CSC Line Management	13	15	29	31	12	3.12	(1.21)	419
CSC Functional Management	12	14	30	30	15	3.22	(1.21)	413

Table 24(B) - Other Staffs' Familiarity with CMS: Percent Distributions and Means/SDs

	Very Negative			Very Positive
	1	2	3	4	5	Mean	(SD)	N
Case Management Officers	1	25	3	11	40	4.22	(0.88)	493
Case Management Supervisors	4	10	24	32	30	3.74	(1.11)	483
Living Unit Officers	13	25	30	21	11	2.90	(1.19)	430
Security Officers	60	24	11	4	1	1.61	(0.88)	439
Psychologists	20	30	31	15	3	2.51	(1.08)	430
Others Programming Staff	37	28	25	8	2	2.09	(1.04)	423
CSC Line Management	25	22	31	16	7	2.58	(1.21)	438
CSC Functional Management	22	24	26	20	7	2.65	(1.22)	434

Table 25 - Rated Need for Additional Information Training in CMS by Specific Areas: Percent Distributions and Means/SDs

	None		Moderate	High
	1	2	3	4	5	Mean	(SD)	N (missing)
Scripted Interview	43	20	19	11	7	2.19	(1.29)	444(84)
Apply Strategies	27	18	21	21	13	2.74	(1.40)	448(80)
Force Field Analysis	30	16	19	23	12	2.70	(1.40)	449(79)
Defining Problems/Needs	27	17	26	20	10	2.69	(1.32)	447(81)
Set long-Term Goals	25	15	26	22	12	2.80	(1.35)	446(82)
Set Short-Term Goals	24	16	26	21	14	2.84	(1.36)	448(80)
Develop Action Plan	23	18	22	21	16	2.88	(1.39)	447(81)
Develop Referral Plan	26	14	27	21	13	2.77	(1.35)	439(89)
Negociation With Offender	28	15	26	19	14	2.79	(1.37)	443(85)
Recommendations to Institutions	28	13	24	20	15	2.81	(1.42)	430(98)
Conditional Release Recommendations	28	14	20	21	17	2.85	(1.47)	433(95)
CMS re Overall Respondibilities as CM Officer	28	15	19	19	19	2.87	(1.49)	437(91)
Responsibilities as Supervisor	28	15	19	18	21	2.90	(1.50)	403(125)

Opinions regarding Future Refinements of CMS

The CMS survey provided an opportunity for respondents to rate the need for additional information or training in a variety of areas involving CMS (Table 25). Mean rated need did not reach 3.00 in a single area. However, the highest rated areas were understanding the linkage between CMS and the supervision of CMOs (mean = 2.90), understanding CMS in the overall context of case management (mean = 2.87), and making recommendations to decision-makers (institutional: mean = 2.81: conditional release: mean = 2.85). Other areas in need of clarification or training include goal setting and treatment planning. Notably, over 40% of respondents indicated no need at all for additional information/training on most of the areas surveyed. This percentage is, of course, based only upon those respondents who completed ratings in this section of the questionnaire. Many of those who did not complete the ratings did not hesitate to provide open-ended comments that were negative (as will be reviewed below).

Table 26 - Priority Ratings Regarding Planning for the Refinement and Future Applications of CMS: Percent Distributions and Means/SDs

	Low		High
	1	2	3	Mean	(SD)	N
Assessment of CMS Regarding:
Workload Implications	12	27	61	2.50	0.70	483
Objectivity/Reliability	15	40	45	2.30	0.72	484
Validity	17	32	51	2.34	0.75	476
Need for Better Quality Control:
Scripted Interview	41	40	19	1.78	0.74	477
Force Field Analysis	35	41	24	4.89	0.76	477
Correctional Treatment Planning	23	38	39	2.17	0.77	478
Increasing Knowledge and Understanding of Applications of CMS Through Research	25	35	40	2.15	0.79	479

Provided with seven areas to rate according to priority in planning for the refinement of CMS, the respondents assigned the highest rating to assessment of the workload implications of CMS (Table 26). Assessments of the reliability and validity of CMS were also assigned high priority ratings as were concerns with correctional treatment planning and research on applications of CMS. The lowest ratings were assigned quality control of the CMS interview and force field analysis.

Few written comments were provided in the space provided for suggestions concerning refinements of CMS. What there was in the way of commentary was overwhelmingly negative: In particular, "Scrap it!". Otherwise suggestions had to do with making CMS less time-consuming, essentially retaining assessment while dropping the rest of CMS. Some officers recommended bringing back elements of the pre-CMS system.

Respondents were asked to rate the priority of additional training in certain general areas of interest in correctional assessment and correctional counselling (Table 27). The highest ratings were assigned to prediction of recidivism and violence. The next highest ratings were assigned to principles of correctional counselling.

Table 27 - Priority Ratings For Areas of Further Training: Percent Distributions and Means/SDs

	Low		High
	1	2	3	Mean	(SD)	N
General Theory and Research on Criminal Behavior	30	40	30	2.00	0.78	478
Prediction of Recidivism	19	35	46	2.26	0.76	478
Prediction of Violence	10	24	66	2.56	0.67	477
General Principles of Assessment	30	49	21	1.90	0.71	480
Principles of Correctional Assessment	27	48	25	1.98	0.72	483
General Principles of Counselling	29	37	34	2.05	0.80	485
Principles of correctional counselling	24	37	39	2.14	0.78	483
Structuring and Directing Skills	30	46	24	1.94	0.73	478
Advocacy Skills	38	46	16	1.79	0.70	477
Negotiationg Skills	31	39	30	1.99	0.78	476
Using Community Resources	30	39	30	2.00	0.78	476

SUMMARY AND CONCLUSIONS

The study was conducted in order to survey a) users' evaluations of their initial training in CMS, b) opinions of CMS in the workplace, and c) opinions regarding future refinements of CMS and priority areas for future action. Fifty-five percent of those who were sent a questionnaire responded (528/966), although there was regional variation in response rates.

Opinions regarding initial training. The following opinions were voiced by the respondents:

      1)  The average evaluations of initial training in CMS were modestly positive, but respondents were greatly divided in their evaluations. Approximately a third were clearly positive, a third were clearly negative, and the remaining third took a middle position.

      2)  The ratings of initial training were much more positive with regard to the basics of the scripted interview, analysis of needs and strategy types than with regard to CSC's extension of these basics into a total case management system. It appears that the links between CMS and the day-to-day matters of treatment planning, making recommendations to decision-makers, communication with other staff and supervisors were not well-established in the initial round of training. (In fact, as was found in the evaluations of CMS in the workplace, a significant proportion of respondents do not believe that there is any facilitative linkage between CMS and the basic responsibilities of a CMO beyond the early period of incarceration).

      3)  Variation in evaluations of initial training was traced to regions, settings, field in which postsecondary studies where pursued, trainers link with NCCD, number of days of training in CMS, and pretraining attitudes toward CMS. Overall, the findings suggested that the Atlantic region was most favourable in evaluations of training by virtue of positive correlations with the other covariates of a positive evaluation of training. These other correlates were having a trainer who was linked with NCCD, not having concentrated in academic criminology, being institution-based and having a positive pretraining attitude to CMS.

      4)  Qualitative analyses of the comments of respondents revealed a fairly uniform appreciation for the efforts made by trainers.

      Evaluations of CMS in the workplace. Analyses revealed the following:

      1)  A large majority (over 70%) of the respondents reported that CMS has increased the amount of time required to manage individual cases. The workload implications of CMS were a major focus of negative comments.

      2)  The overall average evaluations of CMS in the workplace were negative. Only 46% of the respondents reported that the implementation of CMS was worth the effort.

      3)  Negative evaluations of CMS were particularly notable outside of the Atlantic region, among community-based respondents, among case managers themselves, and among staff with a format educational background in criminology. The negative effects of having a trainer not linked with NCCD that were found in the evaluation of training persisted in evaluations of CMS in the workplace.

      4)  Analyses of specific aspects of the case management role and of specific elements of CMS converged in revealing major problems at the levels of CMS in relation to the overall case management context and the supervision of case managers. Negative evaluations were assigned treatment planning, communication among staff, making recommendations to decision makers, and community-based case management. Evaluations of CMS in the areas most closely linked with the scripted interview and needs analysis were more positive than evaluations of CMS relative to the overall case management process.

      Concerns for the future. The study established the importance of the following points for purposes of planning regarding the future of CMS:

      1)  CMS users are greatly divided regarding the utility of CMS, although the average opinion is negative. Those holding negative opinions feel very strongly that the implementation of CMS was a mistake, and "refinements" of CMS or additional training in CMS are not considered valuable areas for future action.

      2)  Divisions in opinions and perceptions of opinions regarding CMS are evident between LUOs and CMOs, and between line staff and management. The LUOs were open to additional training in CMS.

      3)  A rethinking and clarification of the place of CMS in the overall case management process is required. Facilitative linkages with case management beyond the early phases of incarceration were not apparent to a majority of the respondents.

      4)  The workload implications of CMS were assigned the highest priority rating for purposes of planning and refinements of CMS. Similarly, many respondents saw potential value in basic research on the reliability, validity and applications of CMS.

      5)  Professionalism was a major theme in open-ended comments, and training in the professional areas of prediction and counselling was endorsed more frequently than was training specific to CMS.

      6)  Perhaps, the CMS interview and scoring systems, once subjected to basic research on objectivity and validity, will be found valuable as an aid to professional case management. However, the qualitative analyses of comments revealed that many officers continue to believe that systematic assessment is somehow incompatible with professionalism. Any training on prediction must include exposure to the wealth of data regarding the superiority of systematic empirical methods over clinical prediction.

      7)  At the present time, CMS should not be presented as a representation of case management policy in CSC. In brief, it is not at all obvious that knowledge of strategy types does relate in any direct way to the day-to-day practice of case management outside of providing some suggestions regarding counselling style and some contribution to selection of targets of counselling. The very strong CSC policy statements regarding the importance of the identification of criminogenic needs and treatment planning and delivery may be being weakened in implementation. It appears that a strong and positive policy is being confused with CMS scripted interviews, strategies and time consuming structured recording.

      8)  The professionals surveyed appear to strongly endorse CSC's commitment to protection of the public, and yet they feel that the implementation of CMS emphasized protection of CSC. They are asking for increased resources to enact treatment plans, opportunities to enhance professional skills, and respect for their professionalism.

      9)  The opinions described in this report cannot be claimed to be representative of the opinions of all users of CMS. (The overall rate of questionnaire returns was a respectable 55%, but there was regional variation in response rate). The report is an accurate reflection of the opinions of those who did respond, and many of those opinions were expressed strongly.