Research Results

STUDY 1
One of three middle schools (grades 6-8) in a city of about 80,000 was randomly assigned to receive Say It Straight Training and two schools served as controls. The two control schools received their own information-oriented drug-prevention program. The school receiving SIS was compared with the two control schools on alcohol/drug related school suspensions and referrals during the whole academic year. A total of 509 students participated in SIS. Of these, 202 were sixth graders, 215 were seventh graders, and 92 were eighth graders. In the two control schools, 1,539 students were measured on alcohol and drug-related school suspensions. In addition, the Say It Straight Social Skills Questionnaire was administered to students in the treatment school and to a small subset of students in one of the control schools before and after their respective trainings. The SIS Social Skills Questionnaire measures student willingness to implement constructive decisions in difficult situations and how they feel doing so (an increase in the SIS Social Skills Questionnaire indicates a greater willingness to implement constructive decisions in difficult situations and feel more at ease doing so). The subset in the control school consisted of 21 sixth graders, 35 seventh graders, and 28 eighth graders of the total 714 students. A Subjective Feedback Questionnaire was administered in the treatment school after SIS. This questionnaire gives students the opportunity to write in their own words what they learned and what was useful to them in the training.
Students in the treatment school were 2.4 times less likely to have alcohol and other drug related school suspensions and referrals during the academic year compared to students in the two control schools. This is a highly statistically significant result. It is interesting to note that of the 12 suspensions and referrals that occurred in the treatment school, not a single one occurred after SIS. Students trained in the treatment school were followed during the next whole academic year. Not a single “new user” (defined by alcohol and other drug related school suspensions) was identified among them. In addition, only students in the treatment school showed statistically significant positive shifts on the SIS Social Skills Questionnaire that indicated a greater willingness to implement constructive decisions in difficult situations and feel more at ease doing so after SIS. These shifts were highly statistically significant. Finally, on the Subjective Feedback Questionnaire, trained students indicated that they had learned and were motivated to use the skills and knowledge taught by the program when faced with a difficult situation in real life.

STUDY 2
Because of the success of SIS in the treatment school in Study 1, SIS was requested by one of the previous year’s control schools. Furthermore, the original treatment school asked Dr. Englander-Golden to do SIS with all the incoming 6^th graders and do SIS reinforcement with 7^th and 8^th graders who had been trained as 6^th and 7^th graders the year before. Finally, one elementary school asked that their 5^th graders also be offered SIS. Alcohol/drug related school suspensions were monitored for all 5^th-9^th grade students in the city during the whole academic year. This included the 9^th graders trained as 8^th graders in Study #1. The SIS Social Skills Questionnaire was administered before and after training to the 5^th graders, the incoming 6^th graders in the previous year’s treatment school and to 6^th-8^th graders in one of the previous year’s control schools. The Subjective Feedback Questionnaire was administered after SIS.
Alcohol/drug related school suspensions were significantly lower among the 1483 6th – 9th graders who received SIS compared to 1295 students who did not have SIS. Trained students were 3.7 times less likely to have an alcohol/drug related school suspension compared to untrained students during the academic year. The 5^th graders were excluded from this analysis because no such suspensions were incurred either by trained or not trained 5^th graders. Scores on the SIS Social Skills Questionnaire showed highly statistically significant increases for 5th-8th graders after SIS. A subjective feedback questionnaire was also administered to these students. As in Study #1, in every grade students reported on the Subjective Feedback Questionnaire that they had learned and were motivated to use the skills learned.
The graph below shows significantly lower alcohol/drug related school suspensions per 1,000 students during the 1983-84 school year among 1,483 trained and 1,295 untrained 6th-9th graders in a town with a population of about 80,000 (Chi Square = 4.98, df = 1, p < 0.03). There were four suspensions among the trained students and 13 among the untrained students. About 30% of the trained students were trained during the previous school year. Thus, we are seeing long term effects of SIS training.

STUDY 3
This study demonstrates the long-range effectiveness of SIS in reducing the number of juvenile police offenders among 9^th-12^th graders. A six month pre-training base line period was used to monitor juvenile police offenders among all 740 9^th-12^th graders in a city with a population of about 5,000. The most rigorous design was applied to the 9^th grade. SIS was offered to all 9^th graders and 186 out of 208 participated. The number of juvenile offenders among this trained group of students was compared to offenders among 211 9^th graders who were not trained during the following school year and thus provided a control group. Comparisons for these two groups were done during equivalent time frames (February 1 through August 24) of two consecutive years. SIS was also conducted with 108 out of 221 10^th graders, 35 out of 166 11^th graders and 28 out of 145 12 graders. In addition to measuring juvenile police offenses, participants in the program were administered the SIS Social Skills Questionnaire before and after SIS and the Subjective Feedback Questionnaire after SIS.

The Police Department monitored juvenile police offenses for all 9^th-12^th graders for a base-line period of 6 months before SIS and 19 months after SIS. During the pre-training period, the number of offenders among the 357 students who would later receive SIS did not differ statistically from the number of offenders among the 383 students who would not receive SIS. During 19 months after SIS, there were statistically significantly fewer juvenile police offenders (behaviors ranging from aggravated assaults, vandalism and burglaries to traffic violations) among 357 9^th-12^th graders who received SIS compared to the 594 students who did not receive SIS (this number includes the 211 9th graders who entered the school one year after the study began). Looking at criminal juvenile police offenses, students who did not receive SIS had about 450% more such offenses compared to program participants, and their offenses were more severe as ranked by the Police Department, as shown below.

 

The breakdown of criminal offenses among the 357 trained and 594 untrained 9th-12th graders during the 19 months of follow-up after training is shown in order of severity as ranked by the police department is in the following table. Traffic violations also appear in the table.

* Severity not Ranked by Police Department.

As stated above, we were able to do a rigorous experimental design with 9th graders of two consecutive school years. A comparison was made between the trained 9th graders in the first year of the study and the untrained 9th graders in the following school year. During an equivalent 5-month pre-training period there were 5 juvenile police offenders among students who would later be trained, while there were no offenders among those who would not be trained. However, during an equivalent 7-month post-training period, there were 19 juvenile offenders among the 211 untrained students compared to 5 juvenile offenders among the 186 trained students. This difference is statistically significant (Chi Square = 5.877, df = 1, p = 0.015).
The table below shows juvenile police offenses among 5 trained and 19 untrained offenders during the 7-month period (February through August) of two consecutive years. The first 9 categories are considered criminal offenses in Oklahoma; the 1^st 8 are listed in order of severity (the 9^th was not ranked by the police department); the last two are traffic violations.

Scores on the SIS Social Skills Questionnaire showed highly statistically significant increases for 9th, 10th and 12th graders after SIS and were not significant for 11^th graders after SIS. Students in all grades reported on the Subjective Feedback Questionnaire that they had learned and were motivated to use the skills learned.

STUDY 4
This study involves a large-scale replication done by 96 teachers, counselors, school nurses, school administrators, community volunteers and two project coordinators who were trained to facilitate SIS with students, parents and community in high risk environments. They trained 2,781 3rd -12th graders in classrooms, in student support groups, in a school within a chemical dependency treatment facility and in a school within a detention facility, as well as 227 parents and other adults. Two parent groups had Spanish speaking trainers and questionnaires were administered in Spanish. The SIS Social Skills Questionnaire was administered before and after SIS with students. Five additional questions pertaining to sexual behavior were added to the SIS Social Skills Questionnaire and were used by some of the trainers with 9th – 12th graders. The SIS Communications Questionnaire (measuring empowering and disempowering communications/behaviors), Quality of Life-Family Questionnaire (QLQ-F) and Quality of Life-Group Questionnaire (QLQ-G) measuring quality of life in the family and training group were administered to the adults before and after SIS. Both students and adults received the Subjective Feedback Questionnaire after SIS.
Although 2781 students were trained, data analysis is based on 2,695 students for whom paired pre and post-training questionnaires were available. Data loss occurred because of absences during pretest or posttest, lack of ID numbers to allow matching pre and posttests and two teachers who administered different pre and posttests making analysis impossible. All grades showed statistically significant increases on the SIS Social Skills Questionnaire, with 4th -12th grades showing highly significant results. Analysis by gender showed the same results.
Scores on the five added sexual behavior questions also showed highly statistically significant increases, even when analysis was done by gender. Students in all grades reported on the Subjective Feedback Questionnaire that they had learned and were motivated to use the skills learned. Students in detention and treatment showed similar results to students in regular school settings. In addition, a group of 8 students in the treatment facility was evaluated by their counselors on the Communication/Behavior Questionnaire before and after SIS. The counselors evaluated the students as being highly significantly more likely to use empowering communications/behaviors after SIS and engaging highly significantly less in disempowering communications/behaviors after SIS.
Adults reported highly statistically significant increases in empowering communications/behaviors and highly significant decreases in disempowering communications/behaviors as measured on the SIS Communication/Behavior Questionnaire, and highly significant increases on QLQ-F and QLQ-G after SIS. Adults reported on the Subjective Feedback Questionnaire that they had learned and were motivated to use the skills learned.

The next three graphs show highly statistically significant increases in willingness to implement constructive decisions in difficult situations among 3rd-12th graders after SIS training as measured on the SIS Social Skills Questionnaire which measures behavioral intentions to implement constructive decisions in difficult situations and the degree of comfort in doing so. These results were obtained by teachers, counselors, other school personnel and community volunteers.
Pre and post training means on the SIS Social Skills Questionnaire for 2,420 3rd-8th graders by grade (44 - 3rd, 148 - 4th, 426 - 5th, 446 - 6th, 245 - 7th and 1,111 - 8th graders) are presented below. Every grade moved significantly toward a greater willingness to make constructive decisions in difficult situations and felt more at ease doing so after training (higher scores after training), with p-values ranging from p < 0.048 for 3rd graders to p < 0.001 for 4th through 8th graders.

Eighty-eight of the 258 9th-12th graders in regular schools received five additional questions on the SIS Social Skills Situation Survey pertaining to sexual behavior. As can be seen on the following graph, the shift toward a more thoughtful approach to sexual behavior was highly significant with p < 0.002. This shift was highly significant for both girls (p < 0.002) and boys (p < 0.004).

 

ADDITIONAL RESULTS FOR YOUTH IN TREATMENT

Another group of six adolescents in treatment responded to the SIS Communication/Behavior Questionnaire before and after SIS training. The next graph shows pre and post training means on the SIS Communication/Behavior Questionnaire for these adolescents by factors of ineffective communications/behaviors, Placating (PL), Blaming (BL), Passive-Aggressive (PA), Irrelevant (IR), Super-Reasonable (SR) and effective communications/behaviors (SIS). The most desirable responses are low scores for ineffective communications/behaviors and high scores for effective communications/behaviors. As can be seen, all ineffective communications/behaviors are lower after training and the shifts are significant with p values from p < 0.033 to p < 0.001. The effective factor (straightforward communication) is higher after training and the shift is highly statistically significant with p < 0.001.

 

Five of these adolescents also responded to the SIS Quality of Life Questionnaire as it applied to their group (QLQ- Group) and their families (QLQ-Family). These results are shown in the following figure (with higher scores indicating higher quality of life). As can be seen, QLQ-Group and QLQ-Family are higher after training. Both results on a two tailed t-test for paired observations are highly statistically significant with t = 10.17, df = 4, p < 0.001 and t = 8.10, df = 4, p < 0.001, respectively.

RESULTS OF SIS TRAINING WITH PARENTS/COMMUNITY

The next graph shows pre and post training means on the SIS Communication/Behavior Questionnaire for 100 parents and other adults by factors of ineffective communications/behaviors, Placating (PL), Blaming (BL), Passive-Aggressive (PA), Irrelevant (IR), Super-Reasonable (SR) and empowering communications/behaviors (SIS). The most desirable responses are low scores for disempowering communications/behaviors and high scores for empowering communications/behaviors ( SIS ). As can be seen, all disempowering communications/behaviors are lower after training. The shifts are highly statistically significant. The SIS factor is higher after training and the shift is also highly statistically significant.

The following graph shows pre and post training means on the Quality of Life Questionnaire as it applies to the training groups (QLQ-Group) for 165 parents and other adults and pre and post training means on the Quality of Life Questionnaire as it applies to the family (QLQ-Family) for 117 parents and other adults. As can be seen, self-reports on quality of life both in the training groups and in the increase after training and the shifts are highly statistically significant, with p < 0.001 in both cases.

SIS WITH ADULTS IN TREATMENT

I. RESULTS WITH MOTHERS IN RESIDENTIAL CHEMICAL DEPENDENCY TREATMENT
STUDY 5
The effectiveness of SIS was evaluated with 36 mothers in residential chemical dependency treatment with their children. All were indigent, 32 had previous criminal offenses and 14 were on probation or parole at the time of training and 27 had dual diagnosis. Their ages ranged from 19 to 42. Average number of days in treatment before beginning SIS was 85.6, with a range of 10 to 223 days. Although it was not possible to assign women to treatment and control conditions, it was possible to compare women in treatment an average of 141 days before SIS to women in treatment an average of 40 days before SIS to evaluate the impact of SIS. The following questionnaires were administered before and after SIS: SIS Communication/Behavior Questionnaire, QLQ-F, QLQ-G and the Rosenberg Self-Esteem Questionnaire. The Subjective Feedback Questionnaire was administered after SIS.
For the whole group of women, disempowering communications/behaviors showed highly significant decreases after SIS. Empowering communications/behaviors, QLQ-F and QLQ-G showed significant or highly significant increases. Self-esteem, assessed with one group of eight mothers, also showed a highly significant increase after SIS. Statistically significant differences between mothers in treatment an average of 40 days compared to those in treatment an average of 141 days, observed prior to SIS, disappeared after 10-16 hours of SIS. Results indicate that SIS is an important addition to treatment and may have implication for shortened treatment time, increased treatment retention and reduction in relapse/recidivism. Self-reports on the Subjective Feedback Questionnaire regarding training effectiveness also were very positive. Some examples of what mothers reported follow: “Now I feel I can get my messages clearer to my children.” “I can speak from the heart how I truly feel.” “I am learning to say it straight to my mom and son. This is increasing my self-esteem.”
The results of this study have been published in the Journal of Offender Rehabilitation in 2002 and show SIS training is an important addition to treatment.

Average pre and post training scores for empowering behaviors (saying it straight) and disempowering behaviors (people pleasing, blaming or bullying, back-stabbing, lecturing, being disruptive or out of it) for 34 mothers are shown below (2 of the 36 mothers graduated from treatment before this questionnaire was administered). The results show highly statistically significant increases in empowering communications/behaviors and highly significant decreases in disempowering communications/behaviors.

Average pre and post training scores for 36 mothers on the Quality of Life Questionnaire both in the peer group and family are shown below. The results show significant or highly significant increases on both measures.

Results on Quality of Life Questionnaire (10-point scale)

II. RESULTS FOR MEN AND WOMEN IN RESIDENTIAL TREATMENT FOR CHEMICAL DEPENDENCY, EATING DISORDERS, SEXUAL BEHAVIOR PROBLEMS AND DUAL DIAGNOSES

STUDY 6
This study evaluated SIS in a residential treatment facility (chemical dependency, sexual compulsivity, compulsive gambling, eating disorders as well as psychiatric diagnoses) with 26 men and women. Effectiveness of SIS was measured by comparing leaves from the treatment facility against medical advice before SIS and during SIS, as well as with questionnaires used in Study 5. Another questionnaire was used to evaluate anomie (the sense of alienation) before and after SIS.
Self-reports of empowering behaviors, quality of family and group life and self-esteem showed highly significant increases following SIS. Self-reports of disempowering behaviors showed highly significant decreases following SIS and anomie (a person's perceived level of alienation from self and from others) showed a significant decrease. Subjective reports regarding training effectiveness were also very positive. One resident reported, “finding his voice” as a result of SIS and how this was important for him in coping with triggers for relapse. Treatment staff reported that SIS resulted in a prevalent attitude of increased concern among residents for one another’s recovery and willingness on the part of residents to go out of their way to assist staff in forestalling residents leaving against medical advice as well as intervene informally on their own with greater frequency. Leaves against medical advice were significantly reduced (odds ratio of 2.1). This is a large effect size showing that SIS leads to an increase in treatment retention and completion.

Average pre and post training scores for 26 residents in empowering behaviors (saying it straight) and disempowering behaviors (people pleasing, blaming or bullying, back-stabbing, lecturing, being disruptive or out of it) are shown below. The results show highly statistically significant increases in empowering communications/behaviors and highly significant decreases in disempowering communications/behaviors.

Results on SIS Communication/Behavior Questionnaire (6-point scale)

Average pre and post training scores for 26 residents on the Quality of Life Questionnaire both in the peer group and family are shown below. The results show highly statistically significant increases on both measures.

Results on Quality of Life Questionnaire (10-point scale)

Average pre and post training scores for 26 residents on self esteem and anomie are shown below. The results show highly statistically significant increases in self esteem and significant decreases in anomie.

Results on the Self Esteem and Anomie Questionnaire (9-point scale)

Selected Publications

Wood, Tom (2004) Reduction of Anomie through the Use of Say It Straight Training, Ph.D. Dissertation, University of North Texas.

Englander-Golden, P., Gitchel, E., Henderson, C.E., Golden, D.E and Hardy, R. (2002) Say It Straight Training with Mothers in Chemical Dependency Treatment, Journal of Offender Rehabilitation 35 (1) 1-22.

Englander-Golden, P., Golden, D.E. and McCoy, B.G. (1996). Transforming Breakdowns to Breakthroughs in Work Teams. Proceeding of the International Conference on Work Teams. Dallas, TX, 39-44.

Englander-Golden, P., and Golden, D.E. (1996). The impact of Virginia Satir on prevention of destructive behaviors and promotion of wellness. Journal of Couples Therapy, 6 (3/4), 69-93.

Englander-Golden, P., Golden, D.E., Brookshire, W., Snow, C.P., Haag, M.S., & Chang, T. S. (1996). Communication Skills Program for Prevention of Risky Behaviors with Students and Families. Journal of Substance Misuse for Nursing, Health and Social Care , 1, 38-46.

Englander-Golden, P., & Satir, V. (1991). Say It Straight: From Compulsions to Choices. CA.: Science & Behavior Books. Companion audiocassette of the same title also available from the publisher.

Englander-Golden, P., Jackson, J.E., Crane, K., Schwarzkopf, A.B., & Lyle, P. (1989). Communication skills and self-esteem in prevention of destructive behaviors. Adolescence , 24(94) 481-502.

Englander-Golden, P., Elconin, J., & Miller, K.J. (1986). Brief SAY IT STRAIGHT training and follow-up in adolescent substance abuse prevention. Journal of Primary Prevention , 6(4), 219-230.

Englander-Golden, P., Elconin, J., & Satir, V. (1986). Assertive/leveling communication and empathy in adolescent drug abuse prevention. Journal of Primary Prevention , 6(4), 231-243.

Englander-Golden, P., Elconin, J., & Miller, K.J. (1985). SAY IT STRAIGHT: Adolescent substance abuse prevention training. Academic Psychology Bulletin, 7, 65-79.