Pilot Study Results
1040 Students Involved in Year-Long Esteem Builder Study
Research Shows Esteem Builder Program Significantly Decreases Student Aggression & Increases Academic and Prosocial Behaviors.


Effectiveness of Implementing the Esteem Builders Program School-wide on Elementary Students’ Behavior and Academic Self-concept
A Research Summary Written By
Michele Borba, Ed.D., Craig Borba, Ed.D., & Robert Reasoner


While many educators are convinced of the value of implementing programs emphasizing self-esteem, there has been little published recent research documenting the impact of such programs on student behavior or the relative effectiveness of a school-wide program with diverse populations. In fact, the literature continues to expound upon the failure of most programs to help our failing youth and offers few insights and hard data as to what works.

The major purpose of this study was to determine whether a skill-based self-esteem program implemented school-wide at the elementary level three times weekly during one academic year could increase student behaviors identified as productive to school academic performance as well as reduce aggressive behaviors. Using the program Esteem Builders: A K-8 Self-Esteem Curriculum (Torrance, CA: Jalmar Press, 1989), we sought to determine whether the effects of this self-esteem program could positively impact students identified as “lowest-self-esteem” and potentially at a greater risk for failure at school and in life.

Three elementary public schools in North America who agreed to the study requirements below were selected as the pilot sites. The schools represented a broad spectrum of communities, socioeconomic status and racial and ethnic composition. The student subjects who participated in the study represented a fairly large sample (n = 1040) from various grade levels (K-6) who were different in both abilities and needs. The study population included identified gifted, regular classroom and special education students. Site demographic diversities included rural, suburban and urban school sites. Qualitative and quantitative measures were used to determine program effectiveness including student aggression reports, anecdotal comments, teacher surveys, and BASE, a norm-referenced instrument. Data analysis was compiled by Wright State University.

Study Procedure

Site Selections and Participation Criterion. To maintain control in the study, six requirements were created for study participation:

  1. The self-esteem curriculum, Esteem Builders, would be the only self-esteem program used throughout the eight-month study duration.
  2. The teachers at each study site would implement a minimum of three activities a week in their classrooms directly from the chosen self-esteem program and log all activities implemented for later data collection. The log would list each self-esteem activity implemented, the date it was used and the total activity length. Only those pre and post student BASE protocols from teachers who implemented the specified activity minimum were collected for analysis.
  3. All students exposed to the self-esteem activities for the full duration of the study would be pre- and post-assessed by their respective teachers using a standardized self-concept rating scale, Behavioral Academic Self-Esteem (BASE by Stanley Coopersmith).
  4. Esteem Builder Themes. Each staff would choose a different esteem-building theme to work on each month both in their
    classrooms and school-wide. The themes were required to address one of the five acquired building blocks of self-esteem (as specified from the program, Esteem Builders) and were to be implemented in sequence of the building block model.
  5. Staff Planning Team. Each site would create a cadre of teachers who meet at least once a month to plan school-wide esteem-building activities based on the staff-selected esteem-building themes. The activities would be shared with the entire staff in a short faculty meeting so that all staff members would be reinforcing the same esteem-building skills school-wide.
  6. Questionnaires. Each participating staff member would agree to complete a short questionnaire regarding student behavior changes throughout the study.

Experimental Treatments. The self-esteem program used during the study was Esteem Builders (Borba, 1989), a skill-based curriculum theoretically based on Branden’s (1992) definition of self-esteem: “Self-esteem is the disposition to experience oneself as competent to cope with the challenges of life and as deserving of happiness.” It uses as its framework the five acquired components of authentic self-esteem derived from Reasoner's (1982) exhaustive review of self-esteem theory. These five building blocks of self-esteem are: security, the feeling of strong assuredness; self hood, the feeling of self-worth and accurate identity; affiliation, the feeling of belonging and social acceptance; mission, the feeling of purpose; and competence, the feeling of self-empowerment and efficacy. Although authentic self-esteem can be acquired with less than all five of the building blocks intact, in general, the higher the number of building blocks possessed, the higher the overall self-esteem (Borba, 1989).

The curriculum is designed for use in kindergarten through eighth grade and contains over 250 strategies and activities for students. Each self-esteem building activity is correlated with particular subject areas as well as grade levels. Hence self-esteem enhancement can be easily integrated into existing curriculum. Teachers were instructed that they could implement any of the self-esteem enhancing activities as long as they addressed the targeted esteem theme and/or skill the staff had agreed upon for the month.

Since the activities nurturing each acquired self-esteem building block are presented in varying formats (including for journal writing, counseling, literature integration, school-wide involvement, concept circles, and activity projects), each teacher was encouraged to choose activities that best fit their teaching style and the needs of their students. Activities could be used with individual students, cooperative groups, a total class, or school-wide. In all, 1040 students from kindergarten through sixth grade were exposed to three weekly activities from the self-esteem program, Esteem Builders.

In addition to classroom implementation of self-esteem building activities, each site chose to emphasize different esteem-building concepts or skills school-wide each month of the study. All chosen esteem themes (such as teamwork, perseverance, feelings, listening, problem solving, caring) addressed the five acquired building blocks of self-esteem and targeted specific esteem-enhancing skills. Each new monthly theme was generally introduced to students by staff members at a school-wide assembly and further reinforced through school-wide activities, common staff language, posters, and visual aids. The monthly theme and the chosen esteem-building skills were also emphasized throughout the month by individual classroom teachers within their home-rooms.

Evaluation Measures and Data Collection. Multiple measures were used on data gathered over the one-year study period to determine program effectiveness, including: changes in school detentions, behavior referrals for verbal altercations, behavior referrals for physical altercations, teacher surveys, anecdotal teacher comments, and standardized measures of observed student academic self-concept behaviors.

Behavioral Results
The following results were found, based on both quantitative and qualitative data, after eight months of self-esteem program implementation at all three sites.

Student Behavior Changes. Three weeks prior to close of the school year, a final behavior survey developed by the researchers was distributed to all participating teachers consisting of 13 questions regarding types of observable student behavior commonly cited by educators as “highly desirable.” Respondents were asked to reflect upon their students’ behavior during the duration of the study and to answer yes or no to each question as to whether they noticed a positive change in that specified behavior. Survey results (n = 56) found eighty percent or greater of teachers at all sites perceived significant changes in students’ pro social behaviors as well as an increase in compliance with school rules and a reduction in physical aggression following eight months of using the skill-based program.

The majority of teachers perceived the greatest behavior increases following eight months of receiving self-esteem enhancement activities occurred in student pro social behaviors: (a) students spoke more positively, 100%; (b) students were more respectful and tolerant of each other, 95%; (c) students were more caring and supportive of each other, 93%; (d) students were more cooperative, 93%; and (e) students were more courteous, 91%. Teachers also perceived a significant increase in their students’ abilities to solve problems and conflicts, 89%; and found greater proficiency in friendship-making skills, 91%.

The majority of staff members at all sites also perceived significant changes in school climate and appropriate behaviors following program implementation, reporting (a) the general school climate is more positive, 98%; (b) students have a greater ability to comply with school rules and expectations, 88%; and (c) there were less serious incidents and fewer incidents of physical aggression, 82%.

Statements made by teachers, copied verbatim from comments they added to the survey, also supported the findings that there was an increase in positive student behaviors. The prompt posed to respondents stated, “Have you noticed any positive changes in your students this year that you have not seen in past years?” Since surveys were to be completed anonymously, only the respondent’s grade level and school site were known. A few staff comments that tend to support the behavior changes teachers perceived in students following program implementation are as follows:

  • “Students now try to work out problems among themselves without always involving the teacher” (Kansas site; grade 4).
  • “Students are friendlier and more complimentary” (Kansas site; grade 2).
  • “Students became more positive with each other and more accepting of differences” (Canada site; grade 6).
  • “Students seem to care more for each other. They are also more concerned that all students are included in all areas” (Canada site; grade 3).
  • “The majority of the students are more polite and well-mannered. They treat each other with respect. They also treat adults more respectfully” (Canada site; grade 5).
  • “There is a marked increase in students’ understanding of the feelings of others” (Minnesota site; grade 4).
  • “Students are working out their conflicts on their own. They’re also owning up to their choices and behaviors much more so than in the past” (Minnesota site; grade 6).

Student Aggression Reduced with High At-Risk Population. At the Minnesota study site with a high population of “at-risk” and special education students, the incidents of student verbal aggression and physical aggression were tracked and recorded prior to and following one year of program implementation. A significant improvement in non-aggressive student behavior was found. A description of the specific findings in the reduction of student verbal and physical aggression and student detentions follows:

  • Verbal aggression reduced. Incidents of verbal aggression (insubordination, bullying, swearing, intimidation, threats, insults, and harassment) were reduced from 307 incidents in the year previous to program implementation to 186 incidents following one year of program implementation, producing a total 39% reduction in incidents of student verbal aggression.

  • Physical aggression reduced. Incidents of physical aggression (hitting, punching, pinching, tripping, biting, spitting, and throwing an object at a victim with intent to harm) were reduced from 410 incidents in the year previous to program implementation to 242 incidents following one year of program implementation, producing a 41% reduction in student physical aggression.

  • Student detentions reduced. The number of student detentions (students required to stay after school due to inappropriate school behavior as reported by a staff member) was reduced from 366 incidents in the year previous to program implementation to 196 incidents following one year of program implementation, producing a total 46% reduction in student detentions.

BASE Rating Scale Results. BASE (Behavioral Academic Self-Esteem Rating Scale by Stanley Coopersmith, 1967), a self-esteem rating scale that measures academic self-esteem by using direct observation of students’ classroom behaviors, was completed on each student four weeks after initial program implementation by each student’s respective homeroom teacher and again seven months following program implementation. The instrument addresses five factors found most revealing of children’s self-esteem as seen in their academic performance, and identifies low-self-esteem students by using the total BASE score. In addition, 16 student academic self-esteem behaviors determined to be most indicative of students with high academic self-esteem are rated by an observer according to behavior frequency. The protocols of 1040 students from all three elementary study sites were analyzed by Wright State University.

Students at all sites exposed to the self-esteem treatment for eight months showed significant gains in academic self-esteem at the <0.0001 level. No significant differences between scores in different grade levels were noted. Improvement was noted to be consistent throughout grade levels. Results indicated that students “lowest in self-esteem” (as identified by pre-BASE scores) at all sites and potentially highest at-risk for school failure showed significantly greater improvement in academic self-esteem gains than did those students categorized as not “lowest-self-esteem” (p-value <0.0001). Results also showed that 68% of the students categorized as “lowest-self-esteem” prior to program implementation no longer scored in the lowest category at the completion of the study.

Of the 16 academic self-esteem behaviors measured, 11 items showed significant improvement in all students: (a) willing to undertake new tasks, (b) able to make decisions and establish goals, (c) shows self-direction and independence in activities, (d) initiates new ideas relative to classroom activities, (e) asks questions when doesn’t understand, (f) deals with mistakes or failures easily and comfortably, (h) takes criticism or corrections in stride without overreacting, (i) company is sought by peers, (j) acts as leader in group situations, (k) refers to self in genuinely positive terms, and (l) readily expresses opinions. For “lowest-self-esteem” students, six behaviors showed the most significant gains: (a) initiating new ideas, (b) asking questions, (c) peers seek company, (d) leader in group situations, (e) refers to self positively, and (f) readily expresses opinions. There was less than a 1/10,000 chance that these differences would have just simply occurred without any cause other than program implementation. Specific results on the BASE rating scale are as follows:

  1. Academic self-esteem improvement noted in students at all three sites. A paired t-test of 1040 students in the three schools showed that the average increase in BASE total score of 6.2 is significant (df = 1039, p-value <0.0001). This indicates significant improvement between the pre-test and the post-test total BASE scores for students across all sites. The improvements seen at each school for all students (“lowest self-esteem” and “not lowest self-esteem”) are also significant.

  2. Significant improvement for “lowest-self-esteem” students in TOTAL BASE scores. Scores were analyzed separately for 228 students at all sites who were identified as “lowest-self-esteem” based on their pre-test total BASE score. Although this appears to be a high percentage of the total number of students in the study, it is a reliable figure when you consider 50% of the students at the Kansas site were identified by the State Department of Education of Kansas as being at-risk, and the Minnesota study site was identified one of the most at-risk sites in their school district.

    Paired t-tests of the 228 “lowest-self-esteem” students at all three sites showed that an average increase in BASE total score of 12.0 is significant (df = 227, p-value <0.0001) with the p-value <0.0001. There are significant improvements between the pre-test and post-test total BASE scores for “lowest-self-esteem” students at all sites. The improvements seen at each school for the “lowest-self-esteem” students are also significant.

    In addition, of the 228 students categorized as “lowest-self-esteem” based on the BASE pre-test score, 68% (154 of the 228 students) were no longer in the “lowest-self-esteem” category at the post-test evaluation. This result is further substantiated by teacher annotations. In fact, the “lowest self-esteem” or most-at-risk students showed significantly greater improvement than did other students.

  3. Significant improvement in “lowest-self-esteem” students BASE SUBSCORES at all sites. Paired t-tests showed that there were significant improvements (p-values <0.0001) between the pre-test and post-test in the five BASE subscores for the “lowest-self-esteem” group at all three schools. There are also many significant improvements in the subscores seen at separate schools.

    Since over half of the student population at the Kansas site was identified by their state as at- risk, additional analysis was conducted to ascertain program effectiveness on those students whose total scores on BASE fell in the “lowest-self-esteem” range. The results for these students are surprising, revealing significant growth at the .05 level in all areas measured by the BASE instrument including: (a) Student Initiative, (b) Success/Failure, (c) Social Attraction, and (d) Self-confidence.

  4. Significant improvement in all students in all sites in the BASE SUBSCORES. Paired t-tests show that there are significant improvements (p-values <0.0001) not only in the total BASE score, but also between the pre-test and post-test BASE subscores for students across all three schools. There are also many significant improvements noticed at separate schools. These tables show the significant gains made by the Canadian and Minnesota sites in all subscores. The figures for the Kansas schools were not as significant for the Self-confidence Score (p-value <0.0002). It should be noted that while all five subscores in BASE showed significant improvements, the greatest gains were in Student Initiative, Social Attention and Success/Failure. These findings were true for all students in all three sites.

  5. Significant improvement in 11 academic self-concept behaviors for all students at all sites. In addition, 16 individual academic self-esteem behaviors, as identified on BASE, were analyzed for all students at all sites. Multiple paired t-tests show that there are significant improvements (p-values <0.0001) between the 16 pre-test and post-test BASE behavior scores for all students across all three schools. There are also many significant improvements seen at separate schools.

    Of the 16 individual academic self-esteem behaviors, the following 11 behaviors showed significantly greater improvement in all students than the other five behaviors: (a) willing to undertake new tasks, (b) able to make decisions and establish goals, (c) shows self-direction and independence in activities, (d) initiates new ideas relative to classroom activities, (e) asks questions when doesn’t understand, (f) deals with mistakes or failures easily and comfortably, (g) takes criticism or corrections in stride without overreacting, (h) company sought by peers, (i) acts as leader in group situations, (j) refers to self in genuinely positive terms, and (k) readily expresses opinions.

  6. Significant improvement in six academic self-concept behaviors for “lowest-self-esteem” students at all sites. Multiple paired t-tests show that there are significant improvements (p-values <0.0001) between the 16 pre-test and post-test BASE behavior scores for “lowest-self-esteem” students at all three schools. There are also many significant improvements seen at separate schools. Of the 16 individual academic self-esteem behaviors assessed, the following six behaviors showed significantly greater improvement in the “lowest-self-esteem” students than the other 11 behaviors: (a) initiates new ideas, (b) asks questions when doesn’t understand, (c) company sought by peers, (d) acts as leader in group situations, (e) refers to self in genuinely positive terms, and (f) and readily expresses opinions.

Findings
The results document that using Esteem Builders, an authentic skill-based school-wide self-esteem program based on the five acquired building blocks of self-esteem, can have a positive impact on elementary-aged students’ academic self-concept and behavior when implemented using a minimum of three activities weekly by classroom teachers over an eight-month period. Positive academic self-concept and positive behavior changes were achieved with students representing different geographic areas (rural, suburban, and urban school sites), abilities (gifted, regular classroom, and special needs: speech impaired, emotionally behavior disordered, learning disabled, and educationally handicapped), and grade levels (kindergarten through grade six). The self-esteem treatment was found to be most effective in the following areas: (a) reducing incidents of verbal and physical aggression; (b) increasing 11 academic self-concept behaviors deemed important for school success; (c) increasing specific prosocial behaviors; and (d) for students identified as “lowest-self-esteem” or most-at-risk for failure in school or in life on the BASE rating scale.

The finding that an average 40% reduction of staff-reported aggressive physical and verbal student behavior following eight months of program treatment was substantiated on school-wide behavior logs, teacher questionnaires and by anecdotal staff comments. Additionally, responses on the final study questionnaire completed by all participating teachers supported the reduction in student aggressive behaviors. A large percentage of staff members reported observing noticeable changes in student behaviors following one year of program implementation: (a) less serious incidents of physical aggression causing injury, 82%; (b) a decrease in student name calling and verbal “put downs”, 77%; and (c) fewer incidents of threats and verbal intimidation, 73%. A 46% reduction in the number of student after-school detentions further substantiates that aggressive student behaviors declined.

Teacher responses and anecdotal comments (on final survey questionnaires) perceiving an increase in prosocial behaviors was an unexpected and welcome treatment outcome. The fact that 100% of all teachers at all sites perceived “positive student changes” following eight months of program implementation was particularly noteworthy. Ninety-three percent of the teachers at all study sites perceived students as (a) friendlier and more complimentary; (b) demonstrating the ability to cooperate better with one another; (c) more courteous; and (d) more tolerant and respectful.

The finding that the students at all sites identified as those with “lowest-self-esteem” showed significantly greater improvement on BASE post-test scores when compared to pre-test scores suggests that the Esteem Builders treatment can be an effective intervention for elementary-aged students with recognized emotional, social and behavioral needs. The self-esteem treatment effectiveness on this population was further substantiated by subtests at the Minnesota and Kansas study sites with the highest at-risk populations. It is important to note that though the raters were the same for both the pre- and post-testing periods, none of the teachers were informed as to which students were identified as “low-self-esteem” following the first rating session in late September. All students, regardless of their BASE-identified academic self-esteem level, continued to receive the same self-esteem lessons in frequency and type within their regular classrooms with their regular homeroom teacher with no additional interventions employed. A comparison of pre- and post-test subtest scores for the 26 students on the Kansas site identified as “lowest-self-esteem” also showed significant improvements in all five BASE subscores.

Summary
Based on data findings, the researchers propose the following recommendations for educators particularly concerned about students lowest in self-esteem as well as their “at-risk” student population. First, that self-esteem enhancement for students should be nurtured early, comprehensively and systematically to help the steady decline in adolescent self-esteem of both boys and girls (Harper & Purkey, 1993; Orenstein, P. 1994; Greenberg & Lake, 1991). Second, self-esteem enhancement is suggested to avoid the noted increase in student aggression and deviant behaviors (Garabino, 1999; Pollack, 1999). Early intervention has been suggested as most suitable to break these troubling trends (Yoshikawa, 1994; Garabino, 1999). Third, we recommend using skill-based activities such as those in Esteem Builders that are based on the five building blocks of self-esteem to enhance self-esteem and prosocial behaviors. Implementing esteem-building activities for fifteen minute sessions three times weekly can be effective in reducing student aggression and increasing academic self-concept behaviors and does not have to take away from too much class time, especially when activities are used that are curriculum-content and subject-level correlated.

It is noteworthy that all of the study sites implemented self-esteem lessons as a site-based program. Since all students at all sites were exposed to the same concepts, skills and strategies, staff members were able to apply these to real-life situations with all students, whether in the lunchroom, hallway, playground, or classroom. This comprehensive site-based approach may have enhanced the positive study results. Future studies may want to evaluate the effectiveness of implementing a self-esteem treatment based on the five building blocks as a single-class model as compared to the results of a total-staff model as well as following subjects over a longer period of time to assess whether self-esteem and behavior gains are maintained.

Certainly there are more gaps to be filled in our understanding of the long-term impact of increasing students’ academic self-concept and prosocial behaviors. With the steady increase in the at-risk student population and an alarming rise in student depression, aggression, deviant behavior, suicide, and over-all emotional needs, researchers might be well-advised to rigorously explore further the most effective and realistic ways to enhance students’ authentic self-esteem in elementary school settings.


Home | Articles | Bookstore | Featured Author | Links | Contact Us | Privacy Statement

All Contents Used By Permission © Copyright 1999-2001
ParentingBookmark.com