Effectiveness of rational-emotive education: a quantitative meta-analytical study

Simona TRIP,
Oradea State University, Oradea, Romania
Ann VERNON, James McMAHON
Albert Ellis Institute, New-York, USA

Abstract
Research on Rational Emotive Education (REE) is not as prolific as in Rational Emotive Behavior Therapy (REBT), on which it is based. No quantitative meta-analytic studies of REE were found in the literature; in fact, we found only 6 reviews on REE. The objective of this study was to investigate the effectiveness of REE through a quantitative meta-analitical study. Twenty-six (26) published articles, which fit the inclusion criteria, were examined. Results demonstrated that REE had a powerful effect on lessening irrational beliefs and dysfunctional behaviors, plus a moderate effect concerning positive inference making and decreasing negative emotions. The efficiency of REE appeared to not be affected by the length of applied REE. Rather, the REE effect was strong when participants were concerned with their problems. Types of psychometric measure used for irrational beliefs evaluation affected the results. Effect sizes increased from medium to large when the subjects were children and adolescents compared to young adults.

Key words: Rational Emotive Education, quantitative meta-analysis, effectiveness.

Pages: 81-93

INTRODUCTION

Compared to research on Rational-Emotive Behavior Therapy (REBT), the research concerning Rational-Emotive Education (REE) seemed not to be nearly so ubiquitous. Searches of ERIC, EBSCO, and PsychInfo based on the key words rational emotive education enabled the writers to discern the following:

1) for the 1970s, twenty-one (21) published articles, 2 books and 10 dissertations were located;

2) in the 1980s, Rational-Emotive Education (REE) was the topic for thirteen (13) published articles, 2 book chapters, and eighteen (18) dissertations;

3) between 1990-2006, the search revealed thirty-one (31) published articles, 2 book chapters, and fourteen (14) dissertations.

No quantitative meta-analytic studies of REE were found in the literature. Only 6 reviews were found as qualitative meta-analytic studies (DiGiuseppe, Miller, & Trexler, 1977; DiGiuseppe & Bernard, 1990; Gossette & O’Brien, 1989; Gossette & O’Brien, 1993; Hajzler & Bernard, 1991; Watter, 1988). Four of them presented data (percentages) about the efficiency of REE. The review studies by Gossette and O’Brien (1993) were the most critical and their advice was that it seemed fruitless for anyone to undertake REE research.

DiGiuseppe, Miller, and Trexler (1977) reviewed Maultsby and his other studies (Maultsby, 1974; Maultsby, Knipping & Carpenter, 1974; Maultsby, Costello & Carpenter) that argued that Rational-Emotive Education was an efficient prophylactic against mental deterioration among non-clinical populations of children and adolescents. Other studies reviewed by them asserted that the children involved in an REE program were able to learn the REBT assumptions, to modify their irrational beliefs, and to have more functional emotions and behavior than they had before REE.

Watter (1988) analyzed the research that had been done after the dates of those cited in the previous paragraph on Rational-Emotive Education. Watter concluded that elementary school pupils who attended REE had modified their anxiety levels, increased self-esteem, and raised low frustration tolerance (LFT) toward high frustration tolerance (HFT). Generally, such students became more skilled at coping with emotionally loaded situations. Compared to an educational program based on elements of Freudian theory as well as with a sex education program, REE was helpful for students to decrease irrational beliefs and dysfunctional emotions.

Gossette and O’Brien (1989; followed by Gossette & O’Brien, 1993) judged that the studies that had been conducted on REE did not offer enough data to support the possibility of efficiency with school populations. The major effect of REE was on irrational beliefs, as was expected. Their judgments were not surprising because the content of irrational beliefs measuring scales was identical with the content of the REE curriculum. Minor modifications were found on behaviors: students with problems not covered in the curriculum were less receptive to REE than students who at the outset reported no problems.

DiGiuseppe and Bernard (1990) found that more then 90% of the studies they reviewed supported REE efficiency in diminishing irrational beliefs. More then 50% of the studies sampled recorded behavior modification, locus of control internalization, changes for various personality styles and concerning some development milestones. Fifty per cent of the studies surveyed by these researchers supported anxiety levels having decreased while self-esteem increased (which could further have supported an ipsilateral statistical artifact in that as one score decreased another increased). The idea that the emotional and behavior change was due to beliefs modification could not be inferred from the results of any single study. However, REE had a higher potential for changing adaptive functioning than for changing any single, targeted behavior.

Hajzler and Bernard (1991) asserted that irrational beliefs decreased in 88% of the studies they surveyed, while locus of control internalized in 71% of the studied undertaken with students who displayed learning problems. They reported that anxiety waned in 80% of the studies surveyed, while self-esteem and general adaptation (functional behaviors) improved in 50% of cases.

Gossette and O’Brien (1993) analyzed thirty-three dissertations and 2 published articles concerning REE. That survey showed some greater efficiency in managing irrational beliefs (53%), with irrational beliefs modification (43%) for managing neuroticism, and less than 30% of subjects for managing emotional and behavioral consequences. Based on these results the authors concluded that perhaps the basic principle of the REE model was troubling and that, “. . .the continued use of REE in the classroom with normal school children is unjustified and in fact contraindicated” (Gossette & O’Brien, 1993, p.23). David et al., (2005) classified this study as being biased and its conclusions wrong – the fact that Rational -Emotive Education was 30% less efficient than the other treatments (including group, placebo group, and other interventions) was a misinterpretation of the data; the correct conclusion would be that Rational-Emotive Education at least as efficient as other treatments and more efficient than other treatmentas about 30% of the time. The objective of this study was to investigate the effectiveness of REE through a quatitative meta-analitical study.

METHOD

Selection of Studies

The selection of studies was done by searching ERIC, EBSCO, and PsychInfo from 1970 to 2006, using the key words rational emotive education. The reference list of articles included in previous reviews was also used. To be included in the meta-analysis, each study had to fulfill the following criteria:

1. it had to be published in a specialty journal, which meant they were at least edited or edited-peer reviewed (therefore, dissertations were not included);

2. at least one study group had to involve REE and that program had to be clearly described;

3. the selected article had to offer pretest-posttest or REE – control, placebo, or other intervention group comparisons;

4. the article had to have statistical data to support the main effects;

5. the number of subjects for each group studied had to be specified.

A total of 26 studies, which met the above criteria, were identified and

included in the present quatitative meta-analysis

Main Effect Estimations

Differences between means evidenced in transformed scores were calculated, using Cohen’s „d” estimate for the effect size. Different people offer different advice regarding how to interpret the resultant effect size, but the most accepted opinion is that of Cohen, where 0.2 is indicative of a small/low effect, 0.5 a medium, and 0.8 a large/powerful/strong effect size. The formulae used for d calculation were: (1) d = the mean of difference scores divided by the standard deviation of difference scores (related samples), and (2) d = (experimental group mean – control group mean) / intra-group standard deviation (independent samples). The intra-group standard deviation was derived from the intra-group variance, that formula supported by Fisher (Hunter & Schmidt, 1990). Where the means and standard deviations were not offered, d was calculated based on t, r, F or χ² scores, that procedure having been supported by Hunter and Schmidt (1990).

Coding System

After having been selected, the studies were analyzed following these categories of variables: independent, moderators, and dependent, using the coding system proposed by Smith et al., (1980, as cited in Hunter & Schmidt, 1990). The independent variable was the treatment offered: Rational-Emotive Education was compared with the baseline level, control group (e.g., no treatment, waiting-list), placebo group, or other intervention (i.e., self-instructional training, human relationships, experiential training, and relaxation). Included in the moderator categories were the following variables: age (i.e., children, adolescents, students), measures, intervention length (i.e., short, medium, long), as well as subject diagnosis (i.e., no problems, academic problems, behavior disorders, anxiety). The dependent variables were: irrational beliefs, inferential beliefs (i.e., cognitive distortions – cold cognitions), emotions, and behaviors (see David, 2003 for details).

RESULTS AND DISCUSSION

REE was analyzed for its effect on different dependent variables. Table 1 shows the values of the effect size of REE on irrational beliefs, inferential beliefs, emotions, and behaviors. Significant differences were found between groups: F (3, 202)=2.85, p<.05, as REE seemed more efficient with irrational beliefs and behavior modification. The effect size seemed powerful concerning decrease in dysfunctional behaviors, and partially powerful with irrational beliefs modification, and adequately powerful statistically concerning inferential beliefs and changing emotions.

Table 1. Dependent Variables.

Dependent variable	d	SD (standard deviation)	95% interval	No. of subjects	No. of comparisons
Irrational beliefs	0.73	0.03	0.67-0.78	2816	40
Inferential beliefs	0.50	0.20	0.10-0.89	1784	39
Emotions	0.60	0.26	0.09-1.10	2564	62
Behaviors	0.85	1.10	-1.30-3.00	2069	65

Table 2 shows no significant differences between the independent variables, F (3, 202)=1.06, p>.05. The effect size was medium when REE was compared with the baseline, control group and placebo group. The effect size was large when REE was linked to other interventions (i.e., relaxation, human relationship, self-instructional training, experiential therapies).

Table 2. Independent Variables.

Independent variable	d	SD	95% interval	No. of subjects	No. of comparisons
Pretest-Posttest	0.63	0.64	-0.62-1.88	1237	56
REE-Control group	0.64	0.55	-0.61-1.71	6187	104
REE-Placebo	0.60	0.10	0.40-0.79	308	6
REE-Other interventions	0.85	1.01	-1.12-2.82	1501	40

The results showed that the benefit of the REE participants maintained gains in the follow-up phase compared with the control group and other interventions (human relationship): d=2.69 [s.d.=2.17; 95% – (-1.56 – 6.94); 276 subjects; 9 comparisons].

The efficiency of REE seemed not to be influenced by length of training (see Table 3). A short length was considered less then 4 meetings (50 min. – 1 hour), while medium length was taken to mean 4-17 meetings (50 min.- 1 hour), and the long length was 17 – 85 lessons (30 min. – 1 hour). Even though the effect size became larger and larger as the length of training grew, the differences were not statistically significant (F (2, 43)=2.54, p>.05.

Table 3. Moderator Variables – Length of Training.

Training length	d	SD	95% interval	No. of subjects	No. of comparisons
Short	0.43	0.08	0.27-0.58	542	12
Mean	0.52	0.30	0.06-1.10	2304	20
Long	0.92	1.05	-1.13-2.97	635	14

Other moderator variables that seemed to influence the efficiency of REE were the diagnostic characteristics of the population involved. REE was an efficient program of secondary prevention with different subclinical problems manifested by children and adolescents. The problems studied in the research involved in this meta-analysis can be arranged in 3 categories: (1) academic problems including learning problems; school withdrawal; low academic performance; (2) behavioral problems; and (3) anxiety (test anxiety and state-trait anxiety). No differences (see Table 4) were found between the categories (F (2, 14)=2.44, p>.05). The effect size was the largest for academic problem solving. A powerful effect size was obtained concerning anxiety reduction. An average effect size was recorded with behavioral problems decreasing. The effect size of REE seems to have supported even a primary prevention program that aimed to reduce the strength of dependent variables (irrational beliefs, dysfunctional inferential beliefs, emotions, and behaviors) in a normal population (i.e., no problems).

Table 4. Moderator Variables – Diagnostic Characteristics of Population.

Diagnostic characteristics	d	SD	95% interval	No. of subjects	No. of comparisons
No problems	0.49	0.26	-0.01-0.99	3049	29
Problems	1.23	1.17	-1.06-3.52	432	17
Academic problems	2.10	1.67	-1.17-5.37	155	5
Behavior problems	0.56	0.24	0.08-1.03	72	2
Anxiety	0.80	0.15	0.50-1.09	205	10

An age effect was also revealed (see Table 5): REE was more efficient in working with children and adolescents compared to working with undergraduate or graduate students F (2, 204)=4, p<.05.

Table 5. Moderator Variables – Age of Subjects.

Age	d	SD	95% interval	No. of subjects	No. of comparisons
Children	0.70	0.03	0.64-0.75	3662	54
Adolescents	0.76	0.88	-0.96-2.48	3603	100
Youth – students	0.47	0.04	0.39-0.54	2055	53

Furthermore, the age moderator effect on different dependent variable decreased through REE (see Table 6). Even though the effect size was large for children, no age effect was revealed for irrational beliefs F (2, 37)=2.29, p>.05. The same pattern of results was obtained for inferential beliefs F(2, 36)=1, p>.05, emotions F(2, 59)=1.72, p>.05, and behaviors F(2 ,62)=2.46, p>.05. The effect size was powerful for irrational beliefs and emotions modification with children, adequate for the inferential beliefs, and low for behaviors; however, the differences were not statistically significant (F (3, 49)=2.7, p>.05). For the adolescents, REE benefits were higher on behavior, but again no significant differences were found between the dependent variables (F (3, 96)=1.89, p>.05). The effect size value was adequate for irrational beliefs, inferential beliefs, and emotions. No variation of the effect size was recorded for young adults; the values of effect size for young adults were close to medium (F (3, 49)=3, p>.05).

Table 6. Influence of Age on REE Effects (dependent variables).

Age	d	SD	95% interval	No. of subjects	No. of comparisons
Irrational beliefs
Children	0.82	0.70	0.55-2.19	2010	19
Adolescents	0.52	0.07	0.38-0.65	639	16
Youth – students	0.41	0.14	0.27-0.54	167	5
Inferential beliefs
Children	0.46	0.26	-0.04-0.96	776	15
Adolescents	0.48	0.15	0.18-0.77	198	9
Youth – students	0.55	0.10	0.35-0.74	810	15
Emotions
Children	0.70	0.07	0.56-0.83	670	16
Adolescents	0.64	0.43	-0.20-1.48	1163	38
Youth – students	0.43	0.07	0.29-0.56	731	8
Behaviors
Children	0.24	0.18	-0.11-0.59	119	3
Adolescents	0.97	1.24	-1.46-3.40	1603	37
Youth – students	0.48	0.11	0.26-0.69	347	25

It seemed reasonable for the present writers to learn if the values of REE effect size on the dependent variables were different in function of the independent variables (see Table 7). As was expected, REE was more efficient than other interventions (i.e., relaxation, human relationship, self-instructional training, and experiential training) in irrational beliefs modification; the effect size was large. Significant differences between designs (e.g., REE vs baseline, REE vs control group, REE vs placebo, REE vs other interventions) were obtained F (2, 36)=39.20, p<.01. The effect size value was adequate when REE was compared with its baseline, or to a control, or to placebo.

The same values were recorded for dysfunctional inferential beliefs when REE was compared with the control and placebo groups. There were significant differences between the four modalities of independent variables: F (3, 35)=10, p<.01; the effect size was low when REE was linked to its baseline and other interventions.

No differences seemed evident for emotions (F (2, 59)=0.5, p>.05) and behaviors F (2, 62)=2.18, p>.05. The evolution of effect size values seemed relatively constant for emotions. REE seemed to be more efficient regarding modification of dysfunctional behaviors when it was compared with its baseline and other interventions.

Table 7. Influence of Independent Variables on REE Effects (dependent variables).

Independent variable	d	SD	95% interval	No. of subjects	No. of comparisons
Irrational beliefs
Pretest-Posttest	0.56	0.10	0.36-0.75	387	16
REE-Control group	0.67	0.58	-0.46-1.80	2162	20
REE-Other interventions	1.42	0.84	-0.22-3.06	268	4
Inferential beliefs
Pretest-Posttest	0.34	0.17	0.006-0.67	256	9
REE-Control group l	0.58	0.14	0.30-0.85	1200	26
REE-Placebo group	0.53	0.10	0.33-0.72	84	2
REE-Other interventions	0.30	0.17	-0.03-0.63	244	2
Emotions
Pretest-Posttest	0.58	0.20	0.18-0,97	422	12
REE-Control group	0.63	0.40	-0.15-1.41	1542	35
REE-Other interventions	0.53	0.08	0.37-0.68	600	15
Behaviors
Pretest-Posttest	1.34	1.50	-7.48-10.16	172	19
REE-Control group	0.64	0.75	-0.83-2.11	1508	27
REE-Other interventions	1.30	1.65	-1.93-4.53	389	19

Table 8 presents the impact of REE on irrational beliefs. A large effect size resulted for The Idea Inventory, Children’s Survey of Rational Beliefs, and The Adulat Irrational idea. For the rest of the instruments, the effect size was average. It therefore can be concluded that the results regarding REE efficiency were influenced by the quality of the instruments used.

Table 8. The REE Effect on Irrational Beliefs.

Irrational Beliefs measures with different psychometric instrument	d	SD	95% interval	No. of subjects	No. of comparisons
Idea Inventory	1.36	0.85	-0.30-3.02	561	14
Children’ Survey of Rational Beliefs	1.24	0.61	0.04-2.43	377	8
Common Beliefs Scale	0.36	0.42	-0.58-1.35	24	3
CASI	0.49	0.16	0.17-0.80	160	5
Irrational Beliefs Test	0.59	0.05	0.47-0.70	60	5
Common Beliefs Inventory for Students	0.42	0.05	0.32-0.51	1324	1
The Rational Behavior Inventory	0.21	0.17	-0.12-0.54	109	1
Self-Report of Rationality	0.34	0.20	-0.05-0.73	83	1
The Adult Irrational Idea	1.04	0.26	0.53-1.54	58	1
The Personal Orientation Inventory	0.54	0.25	0.05-1.03	60	1

The next table (see Table 9) presents the results of REE effect on different inferential beliefs. Locus of control is one of them, measured in all the involved studies with Nowicki – Strickland Internal – External Control Scale. Under the category of self-concept, self-esteem, self-description, and the self-concept were grouped. Personal orientation and self-control are representatives for the third category of inferential beliefs. The forth category is given by attributions and self-efficacy (they were variables only in one study) Significant differences were found between inferential beliefs categories F (3, 35)=14, p<.01. The effect size was higher for locus of control and self-concept then for personal orientation, self-control and attributions, self-efficacy. REE seemed to be more efficient in locus of control internalization and a positive self-concept development.

Table 9. The REE Effect on Inferential Beliefs Modification.

Inferential beliefs	d	SD	95% interval	No. of subjects	No. of comparisons
Locus of control	0.70	0.30	0.11-1.28	148	4
Self-concept	0.60	0.09	0.42-0.77	486	14
Personal orientation and self-control	0.48	0.17	0.14-0.81	1000	16
Attributions and Self-efficacy	0.17	0.14	0.03-0.30	150	5

Significant differences (see Table 10) were found for emotions F (5, 56)=32,25, p<.01. The effect size was large in modification of concern and emotionality related to test anxiety. Average to large effect sizes were observed in anger modification. Anxiety (i.e., test anxiety, state-trait anxiety) decreases revealed an average effect size. REE has alower effect size in lessening neuroticism and negative mental health indices.

Table 10. REE Effect on Emotional Indicators.

Emotions	d	SD	95% interval	No. of subjects	No. of comparisons
Anxiety	0.67	0.22	0.29-1.16	1114	29
Neuroticism and mental health	0.34	0.06	0.22-0.45	881	12
Depression	0.39	0.15	0.09-0.68	172	6
Concern	1.60	0.37	0.87-2.32	110	5
Anger	0.78	0.20	0.38-1.17	107	3
Emotionality & emotional reactivity	0.89	0.15	0.59-1.18	180	7

The differences between the effect size of REE on different behaviors were also significant F (6, 56)=45.03, p<.01. A very large effect size (see Table 11) was revealed on playing hookey from school, behavioral problems, GPA, and relationship with others. An average value of the main effect value for REE on academic performance was measured on standardized tests, academic interest, and achievement-motivation. The REE seemed not to be as efficient in assertiveness modification.

Table 11. The REE Effect on Dysfunctional Behaviors Modification.

Behaviors	d	SD	95% interval	No. of subjects	No. of. comparisons
Academic performance measured with standardized tests	0.56	0.37	-0.16-1.28	352	15
GPA	0.95	0.75	-0.50-2.40	316	7
Academic interest / achievement motivation / Academic effort and satisfaction	0.47	0.16	0.15-0.78	285	4
Behavioral problems	1.02	1.15	-1.23-3.27	468	8
Playing hookey from school	5.51	0.51	4.51-6.50	72	3
Assertive behavior	0.36	0.10	0.16-0.55	287	24
Relationship with others	0.80	0.17	0.46-1.13	120	2

CONCLUSIONS AND DISCUSSION

The results of this quatitative meta-analysis generally supported the efficacy and effectiveness of REE. Unlike some previous studies (e.g., Gossette & O’Brien, 1993) the results encouraged even further research in REE. Gosette and O’Brien’s comments seemed antagonistic toward psychoeducational research, and their reasoning seemed shoddy based on 35 particular cases (until 1993 there were more then 55 unpublished dissertations and published research articles). Thus, they underestimated possible future REE research potentials through their biases. They also ignored the research on REBT, arguing that the principle of the model was wrong. Finally, they totally ignored the notion of primary prevention. This meta-analysis also sustained the argument made by David et al., (2005) concerning the biases of the Gossette and O’Brien’ studies.

REE had a powerful effect on lessening irrational beliefs and dysfunctional behaviors, plus a moderate effect concerning inference and decreasing negative emotions. These results were similar to conclusions made by DiGiuseppe & Bernard (1990) who noted that REE was more efficient then other interventions concerning irrational belief modifications. DiGiuseppe and Bernard (1990), as well as Hajzler and Bernard (1991), found reasonable support for REE’s efficacy on anxiety. Our results also supported a medium to low REE effect on neuroticism.

In conclusion, based on this first quantitative meta-analysis on REE, REE does have significant effects on a number of variables. It can be concluded that REE is a viable approach that can affect negative emotions and behaviors, as well as lessen irrational beliefs and dysfunctional behaviors. However, to strengthen this conclusion and to fully explore the potential of REE, shortcomings of REE research need to be corrected, and high-quality studies should be implemented and better promoted. More attention should be paid to generic methodological criteria such as (a) formal clinical assessment of psychopathology; (b) adherence to/adequacy of REE protocols; (c) measures of the clinical significance of change; (d) collection of follow-up data; (e) subject attrition. Finally, careful attention needs to be paid to the instruments used, as those will affect outcome research.

REFERENCES

David, D., Szentagotai, A., Kallay, E., & Macavei, B. (2005). A synopsis of rational – emotive behavior therapy; Basic fundamental and applied research. Journal of Rational-Emotive and Cognitive-Behavior Therapy, 3, 175-221.

DiGiuseppe, A.R., Miller, J.N., & Trexler, D.L. (1977). A review of rationale emotive psychotherapy outcome studies. The Counseling Psychologist, 7, 64-72.

DiGiuseppe, R., & Bernard, M. (1990). The application of rational-emotive theory and therapy to school-aged children, School Psychology Review, 19, 268-286.

Gossette, L.R., & O’Brien, M.R. (1993). Efficacy of rational emotive therapy with children: A critical reappraisal, Journal of Behavior Therapy and Experimental Psychiatry, 24, 15-25.

Hajzler, J.D., & Bernard, E.M (1991). A review of rational-emotive education outcome studies. School Psychology Quarterly. 6, 27-49.

Hunter, E.J., & Schmidt, L.F. (1990). Methods of Meta-Analysis. Correcting errors and Bias in Research Findings, London: Sage Publications.

Maultsby, C.M. (1974).The classroom as an emotional health center, The Educational Magazine, 31, 8-11.

Maultsby, C.M., Knipping, P., & Carpenter, L (1974). Teaching self help in the classroom with rational self counseling, Journal of School Health, 44, 445-448.

Maultsby, C.M., Costello, T.R., & Carpenter, L.L. (1976), Classroom Emotional Education and Optimum Health. The Journal of the International Academy of Preventive Medicine, 12, 24-31.

Watter, N.D. (1988). Rational- Emotive Education: A Review of the Literature, Journal of Rational-Emotive and Cognitive-Behavior Therapy, 6, 139 – 145.

The Articles Providing Data for This Analysis

Baither, C.R., & Godsey, R. (1979). Rational emotive education and relaxation training in large group treatment of test anxiety. Psychological Reports, 45, 326.

Block, J (1978). Effects of a rational – emotive mental health program on poorly achieving, disruptive high school students. Journal of Counseling Psychology, 25, 61-65.

Buffington, P.W., & Stillwell, W.E. (1980). Self-control and affective education: A case of omission. Elementary School Guidance and Counseling, 15, 152-156.

Costello, T.R., & Dougherty, D. (1977). Rational Behavior Training in the Classroom, Rational Living, 12, 13-15.

Cangelosi, A., Gressard, F.C., & Mines, A.R (1980). The Effects of a Rational Thinking Group on Self-Concepts in Adolescents. The School Counselor, 27, 357-361.

DiGiuseppe, R., & Kassionove, H. (1976). Effects of rational-emotive school mental health program on children’s emotional adjustment. Journal of Community Psychology, 4, 382-387.

Haynes, C. et al. 1983). Rational emotive counseling and self-instruction training for test anxious high school students, Canadian Counselor, 18, 31-38.

Hooper, R.S., & Layne, C.C (1985). Rational emotive as a short primary prevention technique. Techniques, 1, 264 – 269.

Kachman, J.D., & Mazer, E.G (1990). Effects of rational emotive education on the rationality, neuroticism and defense mechanisms of adolescents, Adolescence, 25, 131-144.

Knaus, W.J., & Bokor, S. (1975). The effects of rational -emotive education lessons on anxiety and self-concept in sixth grade students. Rational Living, 11, 25-28.

Jacobs, E., & Croake, J. (1976). Rational emotive theory applied to groups. Journal of College Student Personnel, 17, 127-129.

Laconte, A.M., Shaw, D., & Dunn, I. (1993). The effects of a rational-emotive program for high-risk middle school students, Psychology in the School, 30, 274-281.

Leaf, C. et al. (1986). Placebo -like effects of education about rational emotive therapy, Psychological Reports, 58, 351-370

Maultsby, C.M., Costello, T.R., & Carpenter, L.L. (1976), Classroom emotional education and optimum health. The Journal of the International Academy of Preventive Medicine, 12, 24-31

Morley, L.E., & Watkins, T.J. (1974). Locus of control and effectiveness of two rational – emotive therapy styles. Rational Living, 9, 22-24.

Miller, N., & Kassinove, H. (1978). Effects of behavior rehearsal, written homework, and level of intelligence on the efficacy of rational emotive school mental health program. Journal of Community Psychology, 6, 366-373.

Morris, G.B. (1993). A Rational – Emotive Treatment Program with Conduct Disorders and Attention -Deficit Hyperactivity Disorder Adolescents, Journal of Rational Rational Emotive and Cognitive – Behavior Therapy, 11, 122-133.

Omizio, M.M., Cubberly, W.E., & Omizio, S.A. (1985). The effects of Rational -Emotive Education groups on self-concept and locus of control among learning disabled children. Exceptional Child, 32, 13- 19.

Popa, S. (2004). Eficienţa unui program de educaţie raţional emotivă în modificarea cogniţiilor iraţionale & inferenţiale la copii. Romanian Journal of Cognitive and Behavioral Psychotherapies, 4, 53-67.

Rosenbaum, T., McMurray, E.N., & Campbell, M.I. (1991). The Effects of rational Emotive Education on Locus of Control, Rationality and Anxiety in Primary School Children, Australian Journal of Education, 35, 187-200.

Shannon, D.H., & Allen, W.T. (1998). The effectiveness of a REBT training program in increasing the performance of high school students in mathematics, Journal of Rational Emotive and Cognitive-Behavior Therapy, 16, 197-209.

Thorpe, L.G, Freedman, G.E., & McGalliard, W.D. (1984). Components of Rational – Emotive Imagery: Two Experiments with Nonassertive Students. Journal of Rational Emotive Therapy, 2, 11-20.

Warren, R., Deffenbacher, L.J., & Brading, P (1976). Rational – Emotive Therapy and the Reduction of Test Anxiety in Elementary School Students, Rational Living, 11, 26-29.

Wilde, K.J. (1996). The relationship between rational thinking and intelligence in children, Journal of Rational-Emotive and Cognitive-Behavior Therapy, 14, 187 – 192.

Wilde, K.J. (1994). The effects of the Let’s get rational board game on rational thinking, depression, and self-acceptance in adolescents. Journal of Rational-Emotive and Cognitive-Behavior Therapy, 12, 189 – 196.

Zelie, K., Dtone, I. C., & Lehr, E. (1980). Cognitive -Behavioral Intervention in School Discipline: A Preliminary Study. Personnel and Guidance Journal, 63, 80-83.