Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 



 

An examination of Hellison’s (2003) Teaching Personal and Social 

Responsibility model, its validity and effectiveness for primary 

school aged children in New Zealand. 

 

A thesis in partial fulfilment of the requirements for the degree of 

Doctor of Philosophy 

in 

Psychology 

At Massey University, Albany, 

New Zealand. 

 

Michael Hayes Smith 

2014 



COPYRIGHT NOTICE 

Copyright resides with the author, and is governed by the Copyright Act1994. 

No permission, or provision, is given in any manner for the sharing, copying, 

or transmitting of this document either in part or in its entirety. 



 ii 

Abstract 

The Hellison (2003) model was developed with a goal to enable any participant in its 

application to develop their personal and social responsibility. The literature 

demonstrates that the model has been typically made available for youth, as an 

after-school programme option, in which participation is voluntary. 

This thesis examined a longitudinal intervention over one scholastic year of an 

application of Hellison’s (2003) model as part of the curriculum in a New Zealand 

primary school. Within the model, the five levels of personal and social responsibility, 

1) Respect, 2) Participation, 3) Self-direction, 4) Caring, and 5) Outside the training 

venue, are already translated into five operationalised goals; 1) Self-control and 

respect, 2) Self-motivation, 3) On-task independence, 4) Sensitivity and 

responsiveness, and 5) Trying these ideas in other areas of life. The intervention, a 

forty week training programme using a ten lesson plan format, repeated four times, 

utilised the goal structure of the model to formulate the individual lesson plans. The 

lesson plans, to convey the five social goals of the model in accordance with 

Hellison’s (2003) guidelines, included a karate skill set as the physical activity 

component of the training sessions. Two cohorts were recruited into the study, from 

two scholastically equivalent schools, one to receive the intervention (n = 36), and 

one to act as control (n = 49). The control group completed the measures only and 

did not receive the intervention. 

Each goal of the model was empirically assessed to determine change over time 

whilst participating in the current application of the model. A time series empirical 

approach was used and psychometrically reliable and valid instruments were 



 iii 

administered at six equal-distant intervals. At each of the six intervals, four self-

reporting measures were completed by the child participants. Rosenberg’s Self-

Efficacy Scale (S-ES), Dishman’s Self-Motivation Inventory for Children (SMI-C9), 

Muris’s Self-Efficacy Questionnaire for Children (SEQ-C), and Caprara and 

Pastorelli’s Prosocial Behavior Scale. A separate measure, Polit’s Positive Behavior 

Scale, was completed each by the participant’s parent, and the participant’s class 

teacher, at each of the six intervals.  

Statistical analysis using regression model fitting on the data obtained from the 

empirical measures demonstrated that scores increased for the intervention 

participants on each measure over the course of the current study. A linear model 

was evident from the analysis. Multivariate repeated analysis of the four child self-

reporting measures demonstrated that the mean positive change, on goals 1 - 4, was 

greatest for the intervention group. Equality of means analysis also confirmed that 

the intervention group had the highest level of improvement in positive behaviour, as 

reported by the parents and teachers, on goal 5. 

Secondary analysis using Guttman (1947, 1950) scaling enabled a detailed 

examination of the model’s stage-like progression premise. The intention of the 

analysis was to determine if any participant could progress through the five stages, 

each in succession, as the model posits that they should. Scalograms were 

constructed at each of the six intervals to produce a Coefficient of Reproducibility, 

one each for the intervention and control groups. From the twelve coefficients, a 

mean Coefficient of Reproducibility ≥ .90 was acceptable evidence of temporal 

reliability of the scalograms in the current study. 



 iv 

Guttman (1950) Scalogram Analysis demonstrated that a four-item scalogram, for 

the child self-reporting measures, and a five-item quasi-scalogram, combining the 

child self-reporting measures with a mean score of the parent and teacher measure, 

was found in the current study. In a research first, the Teaching Personal and Social 

Responsibility model was demonstrated to possess stage progression in its goal 

structure. 

Further, participants in the intervention group demonstrated that they progressed 

along a single continuum, one stage after another, as the Hellison (2003) model 

claimed. However, the goals in the current study developed in a differing order than 

those proposed by the Hellison (2003) model. Specifically, the stage sequence 

discovered was firstly goal 2) Self-motivation, followed by goal 3) On-task 

independence, before goal 1) Self-control and respect, which was third. Goal 4) 

Sensitivity and responsiveness, and 5) Trying these ideas in other areas of life, 

emerged in the sequence as the model predicted. A repeated, between groups, 

analysis demonstrated that a significant difference between the intervention group 

and control group existed from sixteen weeks and beyond. 

The findings from the current study make a significant contribution to the literature. A 

detailed, empirical protocol, a research first of its type, was demonstrated to be 

reliable for the assessment of participant development whilst engaged in an 

application of the model. A revised Hellison (2003) model was valid, and effective, 

for the intervention participants as the model posits. A wider application of the 

Hellison model is warranted and continued research is recommended. A replication 

of the current study, as well as research of a school-wide application, of the revised 

Hellison (2003) model is suggested. 



 v 

Acknowledgements 

Firstly, I would like to thank the research participants that took part in this study. 

Conducting research centred on children is a sensitive undertaking. Without the 

participation of the two schools, their principals, teachers, support staff and most 

particularly, the students themselves, this study would not have been possible. Also 

to the parents of the students, a particular mention here. Without the unequivocal 

support from the parents of the students, equally this study could not have been 

conducted. In being allowed the privilege of delivering this study, I was solemnly 

aware that I was equally responsible for their pride and joy whilst participating, 

namely their children. In many respects, a collaborated effort was achieved, for 

which my mere thanks seem insufficient. To each of them however, your efforts 

sincerely appreciated, thank you. 

I would like to acknowledge my supervision team who have diligently watched over 

my career through this study, it has been a long and challenging process. Thank you 

to Dr. Richard Fletcher for his ongoing commitment to my efforts, particularly in his 

guidance in establishing this study and keeping it focused at each step along the 

way. Thank you to Professor Stuart Carr whom reawakened not only my passion for 

learning, but most importantly reminding me the benefits of remaining objective, to 

‘hold everything up to the light’. To each of them both, your scholarly direction and 

professional expertise provided, I shall remain forever grateful. I also make a special 

mention to the Office of the Assistant Vice Chancellor (Māori and Pasifika), Massey 

University, for the Pūrehuroa Awards for which I was a most grateful recipient. 

 



 vi 

I would like to express my gratitude to Professor Don Hellison for his Teaching 

Personal and Social Responsibility model. I also like to thank him, Dr. Paul Wright, 

and Dr. David Walsh for their initial correspondence and encouragement when I was 

first formulating ideas about this project. I am also grateful to the Morris Rosenberg 

Foundation, Dr. Rod Dishman, Professor Peter Muris, Professor Gian Vittorio 

Caprara, and Dr. Sylvia Epps for their permissions for the use of their respective 

measures in the current study. From all of these sources that enabled my research, 

without which I could not have undertaken this work, my thanks to you all. 

Last but certainly not least, I must thank my family who have provided their support 

and assistance over these years of my career. To my mother Kaye, my eternal 

gratitude, who has lived alongside this study, every step of the way. To my father 

David, a retired police officer who sadly passed away during this time, thank you 

most especially for all the esoteric discussions that influenced my thinking around 

child welfare and crime prevention. It will be my hope that I continue to be the person 

they have always hoped me to be. I also take this opportunity to thank the many 

friends and extended family who have given their ongoing understanding and 

support, despite my prolonged absence from their lives. Finally to Katie, if ever any 

one individual has invested so much for so little, I look forward to now being able to 

return her kindness, love, and support. 



 vii 

Table of Contents 

Abstract           ii 

Acknowledgements         vi 

Table of Contents          viii 

List of Tables          xv 

List of Figures          xviii 

Chapter One: 

Thesis Question and Critical Literature Review      1 

Thesis Question         1 

‘How’ does the model predict that a child will become progressively more 

personally and socially responsible?   2 

Details of the five levels in the model       4 

From levels to goals         5 

Context and the Teaching Personal and Social Responsibility model  11 

Country         11 

Socioeconomic        11 

Beyond sport         13 

Beyond school        15 

Age          15 



 viii 

The evolution of physical education, and an opportunity for the current study 16 

Research evidence on the Hellison (2003) model     17 

Steps versus Curves: A theoretical reappraisal of the Hellison (2003) model 25 

A role for stage change and the development of personal and 

social responsibility  28 

An argument for maturational change and the development of personal and 

social responsibility  32 

An acid test for choosing between Figures 1, 2, and 3.    33 

Summary          35 

Hypotheses          36 

Chapter Two: 

Methodology           37 

Participants and environment        37 

Recruitment          39 

Procedure          40 

The Lesson Plan        41 



 ix 

Activities as the basis of the Lesson Plan     43 

Matching the Lesson Plan to the Teaching Personal and 

Social Responsibility model      47 

Overview of a typical lesson in the current study    49 

The final activity session of the current study    52 

Measures used in the current study       53 

Rosenberg's Self-Esteem Scale (S-ES)     54 

Dishman's Self-Motivation Inventory for Children (SMI-C9)   55 

Muris's Self-Efficacy Questionnaire for Children (SEQ-C)   57 

Caprara’s Prosocial Behaviour Scale (PBS)     59 

Positive Behavior Scale (Parent and Teacher versions)   61 

Data collection and administration       63 

Study timelines         65 

Data analysis and rational        67 

Statistical measurement of individual change over time   68 



 x 

Statistical measurement of group change over time    71 

The application of Guttman scaling in the current study    72 

Chapter Three: 

Preliminary Results          73 

Reliability analysis         74 

Determination of Variance in the data      78 

Assessing the variance in the child measures    78 

Results from the control group on the child measures   80 

Results from the intervention group on the child measures   81 

Overview of the child measures       88 

Assessing the variance from the Positive Behavior Scale – 

parent and teacher measure     90 

Assessment of the Positive Behavior Scale by teachers (PBS (T))  91 

Results from the control group on the teachers measure   91 

Results from the intervention group on the teachers measure  92 



 xi 

Assessment of the Positive Behavior Scale by parents (PBS (P))  93 

Results from the control group on the parents measure   93 

Results from the intervention group on the parents measure  94 

Variety of trajectories in the Positive Behavior Scale    95 

Overview of the Positive Behavior Scale - parent and teacher measure  102 

Comparisons between the teacher and parent results   106 

Preliminary correlation analysis of all measures in the current study   108 

Chapter Four: 

Statistical analysis: Child, parent and teacher measures    115 

Multivariate repeated measures (child measures)     115 

Independent t-tests (parent and teacher measure)     123 

Chapter Five: 

Guttman Scalograms: Construction and Analysis     126 

Was there a scalogram?        126 

Did the participants in the Intervention group, but not in the Control, 

climb up the scalogram over time?   136 



 xii 

Conclusions          142 

Chapter Six: 

Discussion           143 

Revisiting the Hypotheses        144 

Theoretical implications from the hypotheses     146 

Practical implications from the hypotheses     149 

Contribution to Research        152 

The choice of medium in the application of the model   153 

Empirical assessment of the model      154 

Timeframes for change over time     156 

Empirical measurement of the model      158 

Scalogram utilisation to examine stage progression properties 

of the model   163 

Generalisation of the model in the current study    165 

Limitations of the current study       167 



 xiii 

Future Directions         169 

Conclusion          170 

References           173 

Appendices           181 

Appendix A: Teacher Information Sheet 

Appendix B: Teacher Consent Form 

Appendix C: Child Participant Information Sheet 

Appendix D: Child Participant Consent 

Appendix E: Parent Participant Information Sheet 

Appendix F: Parent Participant Consent 

Appendix G: Teacher Confidentiality Agreement 

Appendix H: Karate Authorisation 

Appendix I: Session Plans 

Appendix J: Individual fitted OLS trajectories 

Appendix K: OLS Regressions 

Appendix L: Scalograms 



 xiv 

List of Tables 

 

Table 1: Hellison’s Teaching Personal and Social Responsibility levels as Goals. 

Table 2: Summarised research conducted on the Teaching Personal and Social 

Responsibility (2003) model. 

Table 3: Demographic characteristics of participants. 

Table 4: Mapping the lesson plan to the goals of the Hellison (2003) model. 

Table 5: Measures used to assess the constructs of the responsibility goals. 

Table 6: Testing protocol. 

Table 7: Timelines of the current study: 

Intervention and test administration timeline. 
Session plan timeline. 

Table 8: Means and Standard Deviations (SD) for all child self-reporting measures 

(including subscales) for control and intervention groups. 

Table 9: Means and Standard Deviations (SD) for the Positive Behavior Scale 

(teachers(T) and parents(P) by subscale) for control and intervention 

groups. 

Table 10: Reliability analysis of all measures (including subscales) calculated with 

Cronbach’s alpha. 

Table 11: Control correlations for the Positive Behavior Scale between teachers and 

parents (Initial Estimates and Rates of Change). 

 



 xv 

Table 12: Control correlations between child self-reporting measures 

(Initial Estimates and Rates of Change). 

Table 13: Intervention correlations for the Positive Behavior Scale between parents 

and teachers (Initial Estimates and Rates of Change). 

Table 14: Intervention correlations between child self-reporting measures 

(Initial Estimates and Rates of Change). 

Table 15: Multivariate tests using Pillai’s Trace. 

Table 16: Estimated marginal means by Group. 

Table 17: Estimated marginal means by Time. 

Table 18: Mean change difference between Time points. 

Table 19: Coefficients of Reproducibility for all measures – control (n=49) and 

intervention groups (n=36). 

Table 20: Between groups comparison, rates of progress at all time points, 

goals 1 - 4. 

Table 21: Between groups comparison, rates of progress at all time points, 

goals 1 - 5. 

Table 22: Control results from exploratory OLS regression models for the PBS, 

SMI-C9, and S-ES measures. 

Table 23: Control results from exploratory OLS regression models for the SEQ-C 

subscales. 

Table 24: Control results from exploratory OLS regression models for the PBS(T) 

subscales. 



 xvi 

Table 25: Control results from exploratory OLS regression models for the PBS(P) 

subscales. 

Table 26: Intervention results from exploratory OLS regression models for PBS, 

SMI-C9, and S-ES measures. 

Table 27: Intervention results from exploratory OLS regression models for the 

SEQ-C subscales. 

Table 28: Intervention results from exploratory OLS regression models for the 

PBS(T) subscales. 

Table 29: Intervention results from exploratory OLS regression models for the 

PBS(P) subscales. 



 xvii 

List of Figures 
 

Figure 1: A model of ‘How’ Hellison’s model should work. 

Figure 2: A schemata of independently developing personal and social responsibility 

skills of varying difficulty. 

Figure 3: A schemata of concurrently developing personal and social responsibility 

skills of equal difficulty. 

Figure 4: An exemplar of the lesson plan for Respect. 

Figure 5: Control OLS trajectories for the S-ES measure. 

Figure 6: Intervention OLS trajectories for the S-ES measure. 

Figure 7: Control OLS trajectories for the SMI-C9 measure. 

Figure 8: Intervention OLS trajectories for the SMI-C9 measure. 

Figure 9: Control OLS trajectories for the SEQ-C (A subscale) measure. 

Figure 10: Intervention OLS trajectories for the SEQ-C (A subscale) measure. 

Figure 11: Control OLS trajectories for the SEQ-C (S subscale) measure. 

Figure 12: Intervention OLS trajectories for the SEQ-C (S subscale) measure. 

Figure 13: Control OLS trajectories for the SEQ-C (E subscale) measure. 

Figure 14: Intervention OLS trajectories for the SEQ-C (E subscale) measure. 

Figure 15: Control OLS trajectories for the PBS measure. 

Figure 16: Intervention OLS trajectories for the PBS measure. 

Figure 17: Standardised OLS mean trajectories by measure - Control group. 

Figure 18: Standardised OLS mean trajectories by measure - Intervention Group. 



 xviii 

Figure 19: Control OLS trajectories for the Positive Behavior Scale (T) 

(subscale SC). 

Figure 20: Intervention OLS trajectories for the Positive Behavior Scale (T) 

(subscale SC). 

Figure 21: Control OLS trajectories for the Positive Behavior Scale (T) 

(subscale A). 

Figure 22: Intervention OLS trajectories for the Positive Behavior Scale (T) 

(subscale A). 

Figure 23: Control OLS trajectories for the Positive Behavior Scale (T) 

(subscale C). 

Figure 24: Intervention OLS trajectories for the Positive Behavior Scale (T) 

(subscale C). 

Figure 25: Control OLS trajectories for the Positive Behavior Scale (P) 

(subscale SC). 

Figure 26: Intervention OLS trajectories for the Positive Behavior Scale (P) 

(subscale SC). 

Figure 27: Control OLS trajectories for the Positive Behavior Scale (P) 

(subscale A). 

Figure 28: Intervention OLS trajectories for the Positive Behavior Scale (P) 

(subscale A). 

Figure 29: Control OLS trajectories for the Positive Behavior Scale (P) 

(subscale C). 

Figure 30: Intervention OLS trajectories for the Positive Behavior Scale (P) 

(subscale C). 



 xix 

Figure 31: Teachers standardised OLS mean trajectories by subscale – 

Control group. 

Figure 32: Teachers standardised OLS mean trajectories by subscale – 

Intervention group. 

Figure 33: Parents standardised OLS mean trajectories by subscale – 

Control group. 

Figure 34: Parents standardised OLS mean trajectories by subscale – 

Intervention group. 

Figure 35: Estimated Marginal Means of self-esteem by group. 

Figure 36: Estimated Marginal Means of self-motivation by group. 

Figure 37: Estimated Marginal Means of self-efficacy by group. 

Figure 38: Estimated Marginal Means of prosocial behaviour by group. 

Figure 39: Group Statistics and Independent Samples Test for the Positive 

Behaviour Scale (PBS) completed by teachers (T). 

Figure 40: Group Statistics and Independent Samples Test for the Positive 

Behaviour Scale (PBS) completed by parents (P). 

Figure 41: Control scalograms: Child measures – Time 1 and Time 2. 

Figure 42: Control scalograms: Child measures – Time 3 and Time 4. 

Figure 43: Control scalograms: Child measures – Time 5 and Time 6. 

Figure 44: Intervention scalograms: Child measures – Time 1 and Time 2. 

Figure 45: Intervention scalograms: Child measures – Time 3 and Time 4. 



 xx 

Figure 46: Intervention scalograms: Child measures – Time 5 and Time 6. 

Figure 47: Control teacher / parent scalograms - Time 1 and Time 2. 

Figure 48: Control teacher / parent scalograms - Time 3 and Time 4. 

Figure 49: Control teacher / parent scalograms - Time 5 and Time 6. 

Figure 50: Intervention teacher / parent scalograms - Time 1 and Time 2. 

Figure 51: Intervention teacher / parent scalograms - Time 3 and Time 4. 

Figure 52: Intervention teacher / parent scalograms - Time 5 and Time 6. 

Figure 53: A representation of a perfect Guttman scale for the current study, 

according to the Hellison (2003) model. 

Figure 54: Rearranging the columns according to Guttman (1950) – Intervention 

group at Time 1. 

Figure 55: Counting errors (rearranging the rows according to Guttman (1950)) – 

Intervention group at Time 1. 

Figure 56: Mean scale scores as a function of time and treatment group – 

Goals 1 - 4 only. 

Figure 57: Quasi-scalogram by treatment group – Goals 1 - 5.