Investigating Factors Influencing 

Pro-Environmental Behaviour 

Among Students at the University of 

the Witwatersrand 

Nyashadzashe Chipoka 

Student Number: 776853  

 

 

A research report submitted to the Faculty of Commerce, Law and 

Management, University of the Witwatersrand, in partial fulfilment of the 

requirements for the degree of Master of Management in Energy 

Leadership 

Johannesburg, 2024 

Protocol number: WBS/EL776853/812 

 



 
ii 

ABSTRACT  

This quantitative study investigated the factors influencing pro-environmental 

behaviour (PEB) among students at the University of the Witwatersrand (Wits), 

guided by the theory of planned behaviour (TPB). This research explored the 

roles of environmental knowledge, social influence, sense of control, and 

demographic characteristics in shaping PEB among a diverse student population 

in South Africa. The study revealed that social influence, particularly within the 

university context, emerged as a powerful predictor of PEB, with the university 

environment exerting a stronger influence than family. Results indicated a 

significant positive correlation between environmental knowledge and PEB yet 

highlighted a gap between awareness and proactive information-seeking. 

Perceived behavioural control (PBC) also significantly and positively correlated 

with PEB, although variability across different actions suggested barriers unique 

to each situation. Notably, while gender and education level did not significantly 

predict PEB, age and field of study did, emphasising the need to consider the 

specific institutional and cultural context when examining the impact of any 

demographics on PEB. The findings from the study suggest that while the TPB 

provides a useful framework for understanding PEB, incorporating insights from 

other behavioural theories and considering demographic moderators may 

enhance its explanatory power. Recommendations for Wits include integrating 

pro-environmental practices into educational programmes, campus culture, 

infrastructure, academic disciplines, and providing financial incentives and 

promoting a sense of collective responsibility. 

  



 
iii 

KEY WORDS  

Environmental awareness, Environmental knowledge, Perceived 

Behavioural Control, Pro-Environmental Behaviour, South Africa, 

Subjective Norms, Theory of Planned Behaviour, the University of the 

Witwatersrand, University students  



 
iv 

DECLARATION 

I, Nyashadzashe Chipoka, declare that this research report is my own work 

except as indicated in the references and acknowledgements. It is submitted in 

partial fulfilment of the requirements for the degree of Master of Management in 

Energy Leadership at the University of the Witwatersrand, Johannesburg. It has 

not been submitted before for any degree or examination in this or any other 

university. 

 

Name:  Nyashadzashe Chipoka   Signature: 

 

Signed at Langebaan, West Coast, Cape Town 

 

On the 30th day of December 2024 

 



 
v 

DEDICATION 

This work is lovingly dedicated to my wife, Patience, my daughters, Samantha, 

Tessa and Tanya, and my sons, Sasha and Troy, whose unwavering support and 

encouragement have been my greatest inspiration.  



 
vi 

ACKNOWLEDGEMENTS 

I would like to extend my heartfelt gratitude to all those who played a role in 

bringing this study to fruition. 

A special thank you goes to my supervisor, Dr Nomusa Mazonde, whose 

expertise, guidance, and patience were indispensable throughout this journey. I 

am profoundly grateful for her mentorship and unwavering support. 

I also extend my sincere appreciation to the exceptional lecturers and 

administrative staff at the Wits Business School for their invaluable contributions. 

Lastly, I am deeply thankful for the constant encouragement and support from my 

partner, family, and friends. Their belief in me has been an enduring source of 

strength and motivation. 



 
vii 

TABLE OF CONTENTS 

ABSTRACT ....................................................................................................... II 

DECLARATION ................................................................................................ IV 

DEDICATION ..................................................................................................... V 

ACKNOWLEDGEMENTS ................................................................................. VI 

TABLE OF CONTENTS .................................................................................. VII 

LIST OF TABLES ............................................................................................. XI 

LIST OF FIGURES .......................................................................................... XII 

LIST OF ACRONYMS AND ABBREVIATIONS ............................................. XIII 

CHAPTER 1. INTRODUCTION .......................................................................... 1 

1.1 PURPOSE OF THE STUDY ........................................................................ 1 

1.2 CONTEXT OF THE STUDY ......................................................................... 1 

1.3 RESEARCH PROBLEM ............................................................................. 2 

1.4 RESEARCH QUESTIONS .......................................................................... 3 

1.5 SIGNIFICANCE OF THE STUDY .................................................................. 3 

1.6 DEFINITION OF TERMS ............................................................................ 5 

1.7 ASSUMPTIONS ........................................................................................ 6 

1.8 SUMMARY OF THE CHAPTER .................................................................... 6 

CHAPTER 2. LITERATURE REVIEW ................................................................ 7 

2.1 THEORETICAL FRAMEWORK..................................................................... 7 

2.1.1 Theory of Planned Behaviour ........................................................... 8 

2.1.2 Defining Pro-Environmental Behaviour .......................................... 11 

2.2 FACTORS INFLUENCING PRO-ENVIRONMENTAL BEHAVIOUR ..................... 12 

2.2.1 Impact of Environmental Knowledge .............................................. 13 

2.2.2 Impact of Subjective Norms ........................................................... 14 

2.2.3 Impact of Perceived Behavioural Control ....................................... 15 

2.2.4 Impact of Demographic Variables .................................................. 16 

2.3 CONCEPTUAL FRAMEWORK ................................................................... 18 

2.4 SUMMARY OF THE CHAPTER .................................................................. 19 



 
viii 

CHAPTER 3. RESEARCH METHODOLOGY .................................................. 20 

3.1 RESEARCH APPROACH ......................................................................... 20 

3.2 RESEARCH DESIGN .............................................................................. 21 

3.3 DATA COLLECTION METHODS ................................................................ 21 

3.4 POPULATION AND SAMPLE .................................................................... 22 

3.4.1 Population ...................................................................................... 22 

3.4.2 Sample and Sampling Method ....................................................... 22 

3.5 THE RESEARCH INSTRUMENT ................................................................ 24 

3.6 PROCEDURE FOR DATA COLLECTION ..................................................... 24 

3.7 DATA ANALYSIS AND INTERPRETATION ................................................... 25 

3.8 VALIDITY AND RELIABILITY ..................................................................... 26 

3.8.1 External Validity (Generalisability) .................................................. 26 

3.8.2 Internal Validity and Reliability ....................................................... 27 

3.8.3 Objectivity ....................................................................................... 27 

3.9 ETHICAL CONSIDERATIONS ................................................................... 28 

3.10 SUMMARY OF THE CHAPTER .................................................................. 28 

CHAPTER 4. PRESENTATION OF RESULTS ................................................ 30 

4.1 INTRODUCTION ..................................................................................... 30 

4.2 DEMOGRAPHICS INFORMATION OF THE RESPONDENTS ............................ 30 

4.2.1 Age ................................................................................................. 30 

4.2.2 Gender ........................................................................................... 31 

4.2.3 Level of Education .......................................................................... 32 

4.2.4 Field of Study ................................................................................. 33 

4.3 ENVIRONMENTAL ISSUES AND BARRIERS ................................................ 34 

4.4 RELIABILITY STATISTICS ........................................................................ 36 

4.5 RESULTS PERTAINING TO PRO-ENVIRONMENTAL BEHAVIOUR .................. 36 

4.6 RESULTS PERTAINING TO HYPOTHESIS 1................................................ 39 

4.6.1 Environmental Knowledge .............................................................. 39 

4.7 RESULTS PERTAINING TO HYPOTHESIS 2................................................ 40 

4.7.1 Subjective Norms ........................................................................... 40 

4.8 RESULTS PERTAINING TO HYPOTHESIS 3................................................ 42 

4.8.1 Perceived Behavioural Control ....................................................... 42 

4.9 RESULTS PERTAINING TO HYPOTHESIS 4................................................ 43 



 
ix 

4.9.1 Demographics (Age, Gender, Level of Education, Field of Study) . 43 

4.10 INFLUENCE OF INDEPENDENT VARIABLES ON PRO-ENVIRONMENTAL 

BEHAVIOUR .................................................................................................... 46 

4.11 SUMMARY OF THE CHAPTER .................................................................. 48 

CHAPTER 5. DISCUSSION OF THE RESULTS .............................................. 51 

5.1 INTRODUCTION ..................................................................................... 51 

5.2 DISCUSSION PERTAINING TO HYPOTHESIS 1 ........................................... 51 

5.2.1 Environmental Knowledge .............................................................. 51 

5.2.2 Influence of Environmental Knowledge on Pro-Environmental 

Behaviour .................................................................................................. 52 

5.3 DISCUSSION PERTAINING TO HYPOTHESIS 2 ........................................... 53 

5.3.1 Subjective Norms ........................................................................... 53 

5.3.2 Influence of Subjective Norms and Pro-Environmental Behaviour . 53 

5.4 DISCUSSION PERTAINING TO HYPOTHESIS 3 ........................................... 54 

5.4.1 Perceived Behavioural Control ....................................................... 54 

5.4.2 Influence of Perceived Behavioural Control on Pro-Environmental 

Behaviour .................................................................................................. 55 

5.5 DISCUSSION PERTAINING TO HYPOTHESIS 4 ........................................... 56 

5.5.1 Demographics (Age, Gender, Level of Education, Field of Study) . 56 

5.5.2 Influence of Demographic Characteristics on Pro-Environmental 

Behaviour .................................................................................................. 57 

5.6 SUMMARY OF THE CHAPTER .................................................................. 59 

CHAPTER 6. CONCLUSIONS & RECOMMENDATIONS ............................... 60 

6.1 INTRODUCTION ..................................................................................... 60 

6.2 CONCLUSIONS REGARDING RESEARCH QUESTION 1 ............................... 60 

6.3 CONCLUSIONS REGARDING RESEARCH QUESTION 2 ............................... 60 

6.4 CONCLUSIONS REGARDING RESEARCH QUESTION 3 ............................... 61 

6.5 CONCLUSIONS REGARDING RESEARCH QUESTION 4 ............................... 61 

6.6 IMPLICATIONS OF THE STUDY FOR CURRENT THEORY .............................. 65 

6.7 DELIMITATIONS AND LIMITATIONS OF THE STUDY ..................................... 66 

6.8 RECOMMENDATIONS ............................................................................. 67 

6.9 SUGGESTIONS FOR FURTHER RESEARCH ............................................... 68 



 
x 

REFERENCES ................................................................................................. 69 

APPENDIX A PARTICIPANT INFORMATION SHEET ................................... 78 

APPENDIX B RESPONDENT CONSENT FORM ........................................... 80 

APPENDIX C RESEARCH QUESTIONNAIRE SAMPLE ............................... 81 

APPENDIX D ETHICS CERTIFICATE ............................................................ 88 

 



 
xi 

 LIST OF TABLES 

Table 1: Consistency table: research questions and hypotheses..................... 19 

Table 2: Profile of respondents ........................................................................ 24 

Table 3: Consistency table: research questions, hypotheses, data collection and 

data analysis ............................................................................................... 29 

Table 4: Age ..................................................................................................... 31 

Table 5: Gender ............................................................................................... 31 

Table 6: Level of education .............................................................................. 32 

Table 7: Field of study ...................................................................................... 33 

Table 8: Cronbach's alpha test ......................................................................... 36 

Table 9: Likert type scale ................................................................................. 37 

Table 10: Measuring pro-environmental behaviour .......................................... 37 

Table 11: Group variable for pro-environmental behaviour .............................. 38 

Table 12: Measuring knowledge of environmental issues ................................ 39 

Table 13: Group variable for Knowledge .......................................................... 39 

Table 14: Measuring Subjective Norms ........................................................... 40 

Table 15: Group variable for subjective norms ................................................. 41 

Table 16: Measuring perceived behavioural control ......................................... 42 

Table 17: Group variable for PBC .................................................................... 43 

Table 18: Measuring Gender............................................................................ 44 

Table 19: Measuring Age ................................................................................. 44 

Table 20: Measuring level of education ............................................................ 44 

Table 21: Measuring field of study ................................................................... 45 

Table 22: Model summary ................................................................................ 46 

Table 23: Analysis of variance (ANOVA) ......................................................... 47 

Table 24: Coefficients ...................................................................................... 47 

Table 25: Coefficients for demographics .......................................................... 48 

Table 26: Hypotheses testing ........................................................................... 48 

Table 27: Comparison of literature review and findings ................................... 49 

Table 28: Consistency table: research questions, conclusions and contribution 

to knowledge ............................................................................................... 63 

 



 
xii 

LIST OF FIGURES 

Figure 1: Literature review structure ................................................................... 7 

Figure 2: Pro-environmental behaviours studied through the lens of the TPB 

since 1995 .................................................................................................... 8 

Figure 3: Theory of planned behaviour .............................................................. 9 

Figure 4: Defining pro-environmental behaviour .............................................. 12 

Figure 5: Conceptual framework ...................................................................... 18 

Figure 6: Methodologies in previous related studies ........................................ 21 

Figure 7: Age .................................................................................................... 31 

Figure 8: Gender .............................................................................................. 32 

Figure 9: Level of education ............................................................................. 33 

Figure 10: Field of study ................................................................................... 34 

Figure 11: Environmental concerns .................................................................. 35 

Figure 12: Barriers to pro-environmental behaviour ......................................... 36 

Figure 13: Distribution for the group variable for pro-environmental behaviour 38 

Figure 14: Distribution for the group variable for knowledge ............................ 40 

Figure 15: Distribution for the group variable for subjective norms .................. 41 

Figure 16: Distribution of the group variable for PBC ....................................... 43 

 

  



 
xiii 

LIST OF ACRONYMS AND ABBREVIATIONS 

PBC  - Perceived behavioural control 

PEB  - Pro-environmental behaviour 

RES  - Renewable energy sources  

SN  - Subjective norms 

SSPS  - Statistical Package for Social Sciences 

SDGs  - Sustainable Development Goals  

TPB  - Theory of planned behaviour 

Wits  - the University of the Witwatersrand



 
1 

CHAPTER 1. INTRODUCTION 

1.1 Purpose of the Study 

This quantitative study investigated the factors that influence pro-environmental 

behaviour (PEB) of students at the University of the Witwatersrand (Wits) on and 

off campus.  

1.2 Context of the Study 

Targets in goal 13 of the United Nations Sustainable Development Goals (SDGs) 

involve "enhancing education, awareness, and capacity on climate change 

mitigation, adaptation, impact reduction, and early warning" (United Nations, 

2015, Target 13.3). Engaging the public in this area can promote PEB and 

increase public awareness of climate change (Smith et al., 2021).  

Public awareness would result in individual PEB including responsibly interacting 

with nature, adapting to climate change impacts, purchasing sustainable products 

(such as local food and eco-friendly cleaning supplies), recycling household 

waste, conserving water and energy, changing travel habits (e.g., switching from 

driving to walking or cycling), and purchasing an electric vehicle (Mbokane & 

Modley, 2024).  

Within South Africa, universities are uniquely positioned to cultivate 

environmentally responsible citizens who can drive these behavioural changes. 

With approximately two million students enrolled annually across the country's 

universities, representing a significant portion of South Africa's population of 

approximately 62 million (Statistics South Africa, 2024), these institutions have 

the potential to shape the attitudes and behaviours of a substantial segment of 

society. Wits, with its approximately 40 000 students, serves as a microcosm of 

South Africa's diverse racial, cultural, and socioeconomic landscape. This diverse 

setting provided a valuable opportunity to explore how different demographic 

groups engage with environmental issues and to identify effective strategies for 

promoting PEB. 



 
2 

Previous studies in this area have largely been anchored in the theory of planned 

behaviour (TPB) by Ajzen (1991) which serves as a foundational framework for 

understanding how attitudes, subjective norms, and perceived behavioural 

control influence students' intentions and behaviours towards the environment 

(Rahmawati et al., 2024). Existing studies situated within various cultural and 

educational contexts, including Indonesia, Pakistan, Portugal, South Africa, the 

UAE, and the United States, highlight the global relevance of fostering PEB in 

higher education institutions (Anwar et al., 2024; Kotyza et al., 2024; Rahmawati 

et al., 2024). These studies collectively highlight the significant role that 

universities can play in shaping environmental attitudes and behaviours through 

curriculum design, awareness programmes, and policy implementation. This 

study builds upon this existing global body of research, providing context-specific 

insights into the drivers of PEB within a major South African university, ultimately 

contributing to the development of more effective strategies for promoting 

sustainability within higher education and beyond. 

1.3 Research Problem 

Despite the recognised importance of PEB, there is a gap in understanding the 

specific factors that drive these behaviours among university students across 

different cultural contexts in South Africa.  

Many available studies on the topic of sustainability in South Africa discuss the 

broader context of energy transition but do not appear to focus on the 

perspectives of university students regarding sustainability (Akinbami et al., 2021; 

Nkoana, 2018; Nkosi & Dikgang, 2018).  

Consequently, there is a notable absence of wide empirical research conducted 

within South African universities (including Wits) that examines the impact of 

knowledge levels concerning environmental issues and the corresponding 

attitudes, personal, social norms, behavioural control and demographic 

characteristics among individual students in promoting subsequent PEB. This 

study contributes to the topic by examining these elements' impact on 



 
3 

environmental perspectives and sustainable actions among students in tertiary 

education in South Africa.  

1.4 Research Questions 

Research questions for the study were: 

1. What is the relationship between environmental knowledge and pro-

environmental behaviour among students at Wits? 

2. What is the relationship between social influence and pro-environmental 

behaviour among students at Wits? 

3. What is the relationship between perceived behavioural control and pro-

environmental behaviour among students at Wits? 

4. What is the relationship between demographic variables, such as age, 

gender, and education level and pro-environmental behaviour among 

students at Wits? 

1.5 Significance of the Study 

The significance of this research lies in its potential to inform policy and 

educational practices aimed at enhancing PEB among university students and to 

contribute to the academic body of knowledge on this topic. By understanding the 

factors that drive PEB, universities can develop targeted interventions to foster 

environmental consciousness and sustainable practices. The findings can guide 

the integration of sustainability into educational curricula, thereby equipping 

students with the knowledge and motivation needed to address environmental 

challenges (Torroba Diaz et al., 2023).  

Universities have the potential to act as incubators for sustainable practices and 

innovations. By integrating sustainability into curricula and campus operations, 

universities can create environments that encourage students to adopt and 

innovate PEB. This can lead to the development of new technologies and 

practices that benefit both the environment and the economy. Moreover, this 

research contributes to the broader discussion on sustainability in higher 



 
4 

education by highlighting the importance of cultural and contextual factors in 

shaping environmental behaviours (Anwar et al., 2024; Phang & Ilham, 2023).  

In terms of academic contribution, analysis of existing academic literature on PEB 

at universities indicate that contextually most studies have been conducted in the 

North America, Europe, China and a few other Asian countries. Much of this 

research has focused on PEB in general populations, and there is limited 

understanding of how university students, specifically at Wits, engage in these 

behaviours.  

Previous studies on energy use and recycling at Rhodes University (Mtutu & 

Thondhlana, 2016; Thondhlana & Hlatshwayo, 2018) provide valuable insights 

into these specific behaviours but overlook the potential influence of demographic 

factors like gender, age, and year of study. This omission limits understanding of 

how individual characteristics may shape students' engagement in sustainable 

practices. Evert et al. (2022) investigated environmental attitudes among 

undergraduates at North-West University, but their focus remained on attitudes 

rather than actual behaviours. That study was more exploratory and descriptive, 

focusing on measuring environmental attitudes without explicitly grounding the 

research in a specific theoretical framework. Moreover, the study also did not fully 

explore the influence of demographic variables on environmental attitudes. 

This study will enable the development of targeted interventions that can 

effectively promote PEB among students and ultimately among the future 

leaders, consumers and decision makers. This research also adds baseline data 

pertinent to South African scholars and may serve as a reference point for 

comparative analyses in future research on this topic.  



 
5 

1.6 Definition of Terms 

Definitions used in this research report are presented below. 

Attitude: An attitude can be defined as a concise assessment of a particular 

thought object. Essentially, anything that an individual distinguishes or considers 

can constitute an attitude object (Mtutu & Thondhlana, 2016).  

Environmental awareness: Understanding and consciousness of the impact on 

the environment as a result of human behaviours (Aikowe & Mazancova, 2023; 

Karasmanaki & Tsantopoulos, 2019; Torres et al., 2023). 

Environmental knowledge: Environmental knowledge refers to an individual’s 

understanding of the fragile nature of the environment and the importance of 

preserving it for future generations. This includes the understanding, 

perspectives, abilities, and engagement in actions that contribute to safeguarding 

and enhancing the environment (Janmaimool & Chontanawat, 2021). 

Personal norm: refers to an individual's internalised sense of moral obligation to 

engage in PEB. Research consistently shows that personal norms are the 

strongest predictor of actual PEB, including support for renewable energy 

(Helferich et al., 2023). 

Pro-environmental behaviour: Actions that are intentional, conscious, and 

beneficial to the environment, reflecting individuals' attitudes, values, and beliefs 

towards ecological issues (Farrow et al., 2017; Rahmawati et al., 2024; Zehui, 

2023). This involves for example, adaptive measures such as buying sustainable 

products (like local food and eco-friendly cleaning supplies), conserving water 

and energy, altering travel habits (e.g., switching from driving to walking or 

cycling), to purchasing an electric vehicle. 

Socioeconomic characteristics: Socioeconomic characteristics refer to the 

attributes of a population, such as people’ age, their gender, ethnicity, marital 

status, level of income, education, and profession (Mtutu & Thondhlana, 2016). 

Social Norms: This refers to the perceived behavioural standards within a social 

group or society. They can exert a powerful influence on individual behaviour, 



 
6 

even if they are not explicitly stated (Chan et al., 2022; Dwyer et al., 2015). 

Studies show that both injunctive norms (perceptions of what is socially 

approved) and descriptive norms (the behaviour exhibited by others) can impact 

pro-environmental actions (Karakislak et al., 2023). 

1.7 Assumptions 

The research made the following assumptions: 

a. Respondents would reflect their typical experiences and real-life 

situations. 

b. The sample size would be representative of the population. 

1.8 Summary of the Chapter 

This chapter introduced the research topic, explained the problem statement, and 

defined the research questions relevant to the study area. Furthermore, the 

significancy of the study was also been articulated. 



 
7 

CHAPTER 2. LITERATURE REVIEW 

This chapter lays the theoretical background on the topic of PEB by identifying 

and evaluating both academic and technical literature. The literature review also 

provides the building blocks for a theoretical framework from which hypotheses 

were developed for testing (Sekaran & Bougie, 2016). 

The chapter structure as illustrated in Figure 1 begins by reviewing the theories 

of PEB to establish a detailed definition of the term. This is followed by the 

identification and analysis of potential antecedents of PEB from within the 

relevant literature, leading to the development of hypotheses.  

 

Figure 1: Literature review structure  
Source: Developed by the researcher (2024) 

 

The review incorporated both academic and technical industry literature on the 

theories of PEB. 

2.1 Theoretical Framework  

This study defines PEB through the lens of established research in psychology 

and sociology, fields that have long sought to understand the complex interplay 

of factors that drive environmentally conscious actions. Understanding PEB 

requires examining the underlying values, beliefs, and norms that shape 

individual choices (Stern, 2000; Whitley et al., 2018). These internal drivers 

contribute to the broader relationship between attitudes and behaviours, a 

connection explored in Ajzen's TPB (Ajzen, 1991, 2011; Ajzen et al., 2018). In 



 
8 

recent years, a growing body of research has applied various frameworks, such 

as value-belief-norm, norm activation model, protection motivation theory, 

specifically to understand how young adults and students engage with renewable 

energy and sustainability (Arya & Kumar, 2023; Bask et al., 2020; Helferich et al., 

2023; Janmaimool & Chontanawat, 2021; Phang & Ilham, 2023; Yuriev et al., 

2020). Among these frameworks, TPB has emerged as a particularly powerful 

tool for understanding environmentally conscious actions, and it serves as the 

theoretical foundation for this investigation (Ajzen, 1991; Kotyza et al., 2024; 

Phang & Ilham, 2023; Yuriev et al., 2020). This study builds upon this tradition of 

research, seeking to further explain the factors that encourage or hinder PEB 

among young adults in the context of students at Wits, a major university in South 

Africa. 

 

Figure 2: Pro-environmental behaviours studied through the lens of the TPB since 1995 
Source: Yuriev et al. (2020) 

 

The use of TPB on studies on PEB has grown over the years as shown with the 

solid line on Figure 2 above. As the world’ attention towards the human 

environmental behaviour increases, TPB’ application to future environmental 

science studies will increase.  

2.1.1 Theory of Planned Behaviour 

Within the framework of this research, the TPB (Ajzen, 1991, 2011; Ajzen et al., 

2018) posits that the intentions of university students to engage in 

environmentally friendly behaviour are significantly influenced by three key 



 
9 

predictors, as illustrated in Figure 3, namely: attitudes towards environmental 

issues, subjective norms, and perceived behavioural control. According to the 

TPB, these factors collectively shape an individual’s behavioural intentions, which 

are strong predictors of actual behaviour (Ajzen, 1991, 2011; Ajzen et al., 2018). 

For instance, if a student might believe recycling is important (positive attitude), 

think their friends also recycle (subjective norms), and knows it's easy to do 

(perceived control), they are more likely to recycle (intention and behaviour).  

 

Figure 3: Theory of planned behaviour  
Source: Ajzen (1991), Yuriev et al. (2020) 

 

The following is a breakdown of these three predictors of intention of PEB namely 

behavioural attitude, subjective norms and perceived behavioural control.  

First predictor of intentions is behavioural attitude. In the context of TPB, 

attitudes refer to an individual's positive or negative evaluations of performing a 

specific behaviour (Ajzen, 1991, 2011; Ajzen et al., 2018). These evaluations can 

significantly influence intentions to act. This means that if a person has a positive 

attitude towards a behaviour, they are more likely to form an intention to engage 

in that behaviour (Sussman & Gifford, 2019). For example, if a person believes 

that recycling benefits the environment and they place a high value on 

environmental conservation, they are likely to develop a positive attitude towards 

recycling, which increases their intention to recycle. 

The second predictor of intentions is subjective norms. The second determinant 

of intentions is subjective norms. These norms encompass the perceived social 



 
10 

pressure to either engage in or refrain from a specific action (Ajzen, 1991, 2011; 

Ajzen et al., 2018). This construct is shaped by two main components: 

• Normative Beliefs: These represent convictions regarding the approval 

or disapproval of significant others (e.g., family, friends, colleagues) 

towards a particular behaviour. For instance, if an individual believes that 

their peers regard recycling as important, this belief becomes an integral 

aspect of their normative beliefs (Ajzen, 1991, 2011; Ajzen et al., 2018).  

• Motivation to Comply: This denotes the extent to which an individual is 

motivated to align their behaviour to the expectations they believe are held 

by significant others. For example, if an individual places considerable 

importance on their friends' opinions, they are more inclined to adhere to 

the perceived expectations concerning recycling. Collectively, these 

components constitute the subjective norm, which impacts an individual's 

intention to undertake a specific behaviour (Ajzen, 1991, 2011; Ajzen et 

al., 2018). 

The final predictor of intentions and an additional contribution to the TPB is 

perceived behavioural control. This concept relates to an individual's 

assessment of the ease or difficulty associated with executing a specific action 

(Ajzen, 1991, 2011; Ajzen et al., 2018). This assessment is shaped by two 

factors: control beliefs and perceived power. 

• Control beliefs: Refers to an individual’s perception of the factors that 

may either support or hinder their ability to carry out a specific behaviour. 

For example, if an individual believes that there are easily accessible 

recycling facilities in their community, this belief can enhance their control 

over recycling behaviour. Conversely, a lack of facilities may hinder their 

ability to recycle (Ajzen, 1991, 2011; Ajzen et al., 2018).  

• Perceived power: This refers to the perceived strength of these facilitating 

or impeding factors. Continuing with the recycling example, the individual 

may perceive that their recycling efforts significantly contribute to 

environmental conservation. This perception of power can strengthen their 

intention to engage in recycling behaviour, as they feel their actions have 

a meaningful impact (Ajzen, 1991; Phang & Ilham, 2023). 



 
11 

In summary, people are more likely to engage in an activity and have stronger 

intentions when they see the activity as enjoyable and beneficial, receive support 

and encouragement from others, observe members of their social group 

engaging in the behaviour, and ultimately believe they can complete the task. 

2.1.2 Defining Pro-Environmental Behaviour 

This study focused on PEB as its dependent variable. PEB encompasses a wide 

spectrum of actions, practices, and activities that impact the environment, either 

positively by minimising harm or negatively by contributing to its degradation 

(Walton & Jones, 2022). The concept includes actions undertaken by individuals 

or groups across different contexts and timeframes, spanning past behaviours, 

current practices, and even intentions for future action (Walton & Jones, 2022). 

The frequency of these behaviours can vary considerably, from everyday habits 

like utilising public transport to less frequent efforts such as purchasing energy-

efficient appliances, or singular actions like planting a tree. Furthermore, PEB 

differs in the level of effort involved and the scale of their impact, affecting 

individuals, communities, or the broader environment. This inherent complexity 

underscores the challenges in both measuring and understanding the behaviours 

that contribute to environmental sustainability. 

This study adopts a definition of PEB informed by the TPB, which emphasises 

the crucial roles of attitude, subjective norms, and perceived behavioural control 

in shaping behavioural intentions and subsequent actions (Ajzen, 1991; Phang & 

Ilham, 2023; Sussman & Gifford, 2019; Yuriev et al., 2020). Drawing upon these 

theoretical underpinnings, Figure 4 synthesises the diverse elements of PEB into 

a concise definition tailored for this research. This definition recognises that PEB 

is not merely a collection of isolated actions, but rather a reflection of underlying 

psychological factors as outlined by TPB.  



 
12 

 

Figure 4: Defining pro-environmental behaviour  
Source: Developed by the researcher (2024) 

 

Therefore, PEB in the context of this study, and based on the Figure 4 above, 

can be understood as actions taken by Wits students that are intentional, 

conscious, and beneficial to the environment, like recycling, cycling or walking, 

conserving water and energy, reflecting the students’ attitudes, values, and 

beliefs towards ecological issues. 

2.2 Factors Influencing Pro-Environmental Behaviour 

Building upon the established definition of PEB as the dependent variable, this 

study investigates the key antecedents that influence such behaviour among 

university students. These antecedents are multifaceted, from individual 

characteristics, educational experiences, social influences, and contextual 

factors, all of which contribute to a student's tendency to engage in 

environmentally friendly actions. The impact of demographic variables such as 

gender and age, coupled with educational background, has been found to be 

significant (Sailesh & Reddy, 2024). For instance, a student's academic 

discipline, their exposure to environmental issues within the curriculum, or the 

norms and values prevalent within their social circles can all play a role in shaping 

their environmental attitudes and behaviours. This study particularly examined 

the roles of environmental knowledge, subjective norms, perceived behavioural 

control, and included demographics in predicting PEB among university students. 



 
13 

2.2.1 Impact of Environmental Knowledge 

A substantial body of research globally, highlights the influence of environmental 

awareness and knowledge on PEB. The prevailing view suggests that individuals 

with greater awareness of environmental issues and a deeper understanding of 

sustainable solutions are more likely to engage in pro-environmental actions 

(Liobikienė & Poškus, 2019; Paço & Lavrador, 2017; Simiyu et al., 2022). This 

connection is often rooted in underlying values and beliefs about the importance 

of nature and sustainability. In other studies, Ulkhaq and George Joseph (2024) 

found a positive correlation between higher levels of environmental concern 

among students in Swedish universities and more favourable attitudes towards 

eco-friendly actions. This could point to the university students in Sweden 

participation in eco-friendly endeavours. 

However, the relationship between environmental awareness and knowledge and 

PEB can be nuanced and context-dependent. Emerging research suggests that 

educational context plays a crucial role in predicting PEB. For example, a study 

by Karasmanaki and Tsantopoulos (2019) in Greece found that while future 

scientists generally possessed high levels of environmental awareness and 

positive attitudes towards renewable energy sources (RES), there were 

significant differences based on their year of study. Specifically, fourth and fifth 

year students demonstrated a greater understanding of RES technologies and a 

stronger interest in pursuing careers in the RES sector compared to their younger 

peers. This suggests that exposure to specialised coursework and practical 

experiences within their field of study may play a crucial role in shaping students' 

knowledge and career aspirations related to renewable energy. Further 

highlighting the contextual nature of this relationship, a study at the University of 

Johannesburg by Mbokane and Modley (2024) found that students' purchasing 

decisions were primarily driven by price and quality rather than environmental 

impact. Specifically, the study revealed that only 27% of students consistently 

prioritised eco-labels when making purchasing decisions, with 63% admitting that 

price remained their primary concern. This suggests a significant gap between 

students' environmental values and their consumer choices, highlighting the 

challenges in translating awareness into sustainable action. These findings 



 
14 

underscore the need to consider financial constraints and promote affordability of 

eco-friendly products alongside enhancing green initiatives within universities. 

Other studies also suggest that environmental knowledge and awareness alone 

may not be sufficient to drive PEB among university students (Carducci et al., 

2021). This perspective is supported by research indicating that while knowledge 

may be associated with certain aspects of PEB, other factors such as social 

influence and personal norms play a more significant role in shaping actual 

behavioural outcomes (Janmaimool, 2019; Torres et al., 2023; Zeng et al., 2023). 

These findings highlight the complex interplay of factors that influence PEB and 

suggest that interventions solely focused on increasing knowledge may have 

limited effectiveness without addressing social and personal dimensions. This 

aligns with the TPB's emphasis on the importance of attitudes, subjective norms, 

and perceived behavioural control alongside knowledge in predicting behaviour. 

H1: Higher levels of environmental knowledge significantly influence pro-

environmental behaviour among university students. 

2.2.2 Impact of Subjective Norms 

Social influence, encompassing the attitudes and behaviours of peers, family, 

and broader society, plays a significant role in shaping PEB. A foundational 

principle in this area is that individuals are more likely to adopt environmentally 

friendly behaviours when these actions are perceived as the norm within their 

social circles (Cialdini et al., 1990; Gould et al., 2023; Steg & Vlek, 2009). This 

suggests that observing and conforming to the behaviours of others is a powerful 

motivator for engaging in PEB. 

Studies by Farrow et al. (2017) highlight the connection between social norms 

and PEB, demonstrating how these norms influence individuals' intentions and 

actions through established psychological pathways, such as those outlined in 

the TPB (behavioural attitudes, subjective norms, and perceived behavioural 

control). Essentially, when individuals believe that significant others approve of 

and engage in PEB (subjective norms), they are more likely to develop positive 

attitudes towards these behaviours and feel capable of performing them. This 



 
15 

perspective is further reinforced by research demonstrating the impact of social 

influence on students' environmental knowledge and subsequent PEB 

(Chakraborty et al., 2017; Simiyu et al., 2022). These findings underscore the 

importance of fostering a social environment where PEB is valued and practised. 

The influence of social factors extends to the realm of education. Longitudinal 

studies, such as the study by Collado et al. (2022), demonstrate that structured 

educational interventions can have lasting effects on students' environmental 

knowledge, personal norms, and PEB. This suggests that educational 

programmes have the potential to not only impart knowledge but also to shape 

social norms and instil enduring pro-environmental values among university 

students. Moreover, research by Aikowe and Mazancova (2023) emphasises the 

crucial role of environmental education in raising awareness and promoting PEB, 

particularly in developing country contexts like Nigeria, a situation likely 

applicable to South Africa as well. These studies highlight the need to consider 

contextual factors and the long-term impact of interventions when promoting 

sustainable practices among university students. They also reinforce the idea that 

educational institutions can serve as important agents of social change by 

fostering environments that encourage and support PEB. 

H2: Greater social influence is positively correlated with university 

students' intentions to engage in pro-environmental behaviour. 

2.2.3 Impact of Perceived Behavioural Control 

Perceived behavioural control is an individual's belief in their ability to perform a 

specific behaviour and plays a crucial role in shaping PEB. Alongside attitudes 

and subjective norms, PBC, as a key component of the TPB, has a positive 

impact on behavioural intentions, which in turn influence environmentally 

responsible actions (Rahmawati et al., 2024). Empirical evidence underscores 

the significance of PBC, with a multi-country study demonstrating that it accounts 

for a substantial portion of the variance in pro-environmental intentions among 

students (Kotyza et al., 2024). This suggests that interventions aimed at 

bolstering students' confidence in their ability to engage in PEB, such as providing 



 
16 

resources, skill-building opportunities, or highlighting successful examples, can 

be effective in promoting sustainable behaviours. 

The relationship between PBC and PEB is further nuanced by factors such as 

environmental awareness and a sense of green self-efficacy. Research suggests 

that environmental awareness and a strong belief in one's ability to make a 

difference can strengthen the positive impact of PBC on PEB (Anwar et al., 2024). 

In essence, students who are not only confident in their ability to act but also 

knowledgeable about environmental issues and committed to environmental 

protection are more likely to translate that confidence into tangible eco-friendly 

actions. This highlights the importance of fostering both the perceived ability and 

the motivation to engage in PEB. 

While social norms can influence behaviour, their effectiveness in promoting PEB 

appears to be limited in some contexts. This suggests that interventions focused 

solely on highlighting social norms may not be as effective as those that prioritise 

enhancing PBC and environmental concern. Therefore, universities seeking to 

foster sustainable behaviours should focus on strategies that empower students 

by increasing their PBC through providing them with the skills, resources, and 

confidence to act while simultaneously cultivating a deeper sense of 

environmental awareness and commitment. By focusing on these key drivers, 

universities can create an environment that supports and encourages sustained 

pro-environmental engagement. 

H3: Increased perceived behavioural control is positively related to 

university students' ability to perform pro-environmental actions. 

2.2.4 Impact of Demographic Variables 

Research focusing on demographic variables, such as gender and age, as well 

as objective and subjective socioeconomic status, reveals their complex interplay 

in shaping individual engagement with environmental issues. 

Age appears to play a role in shaping PEB, with younger individuals often 

displaying greater engagement in pro-environmental actions compared to older 



 
17 

individuals (Vrselja et al., 2024). This generational difference could be attributed 

to various factors, including evolving social norms, increased exposure to 

environmental education, or a greater sense of urgency regarding environmental 

issues among younger generations. This finding, combined with the work of 

Vrelselja et al. (2024) supporting the influence of age, suggests that targeted 

approaches based on both age and gender could be effective in fostering PEB. 

Gender emerges as a particularly influential factor, with numerous studies 

highlighting differences in how men and women perceive and approach 

environmental issues. Research suggests a consistent trend of women often 

exhibiting more favourable environmental attitudes, and more engagement in 

PEB, than men (Fathonah & Nastiti, 2024; Martins et al., 2020; Raman, 2016; 

Strydom et al., 2020; Vrselja et al., 2024). For instance, Raman's (2016) content 

analysis revealed that female students displayed a more positive outlook and 

conduct towards the environment compared to their male counterparts. Similarly, 

Martins et al. (2020) found that while women exhibited slightly lower levels of 

energy literacy, they demonstrated more positive attitudes and behaviours 

regarding energy consumption. These findings suggest that gender targeted 

interventions may be valuable in promoting PEB. 

Socioeconomic status (encompassing both objective factors such as income, 

education level, and field of study), as well as subjective perceptions of social 

standing, also seems to exert some influence on PEB. Generally, higher 

socioeconomic status seems to be associated with greater environmental 

concern and action (Vrselja et al., 2024). This correlation may be linked to 

increased access to resources, information, and educational opportunities that 

facilitate environmentally friendly choices. For example, Dalu et al. (2023) found 

that staff members at a South African university, who likely have a higher 

socioeconomic status than students, possessed greater knowledge about plastic 

pollution. Furthermore, the influence of academic background is highlighted by 

Karasmanaki and Tsantopoulos (2019), who found that students from different 

disciplines in Greece exhibited varying levels of environmental awareness, 

impacting their attitudes towards sustainable actions. These findings highlight the 

importance of considering socioeconomic disparities when designing 



 
18 

environmental education and intervention programmes. They also reinforce the 

need for targeted approaches that address the specific needs and contexts of 

different socioeconomic groups. 

H4: Demographic variables, such as age, gender, and education level, 

significantly influence the likelihood of university students participating in 

pro-environmental activities. 

2.3 Conceptual Framework 

This study tested whether there was a relationship between independent 

variables, socioeconomic characteristics (age, gender, level of education and 

field of study), behavioural attitudes (environmental awareness and knowledge), 

subjective norms (social and personal norms), and perceived behavioural control 

on the dependent variable PEB, among a group of students at Wits. The 

conceptual framework illustrated in Figure 5 outlines the hypothesised 

relationships between the above-mentioned variables.  

 

Figure 5: Conceptual framework  
Source: Adapted from Ajzen (1991), Yuriev et al. (2020) 



 
19 

2.4 Summary of the Chapter  

This chapter started by looking at the various theories governing the concept of 

PEB. Building on this understanding, a definition of PEB was advanced. Potential 

PEB antecedents were then derived from the literature, leading to the 

development of four hypotheses as presented in Table 1. 

Table 1: Consistency table: research questions and hypotheses 

RQ 
# 

Research Question  Hyp 
# 

Hypothesis 

1 What is the relationship between 
environmental knowledge and pro-
environmental behaviour among university 
students at Wits? 

H1 Higher levels of environmental 
knowledge significantly influence 
pro-environmental behaviour 
among university students. 

2 What is the relationship between social 
influence and pro-environmental behaviour 
among university students at Wits? 

H2 Greater social influence is positively 
correlated with university students' 
intentions to engage in pro-
environmental behaviour. 

3 What is the relationship between perceived 
behavioural control and pro-environmental 
behaviour among university students at Wits? 

H3 Increased perceived behavioural 
control is positively related to 
university students' ability to 
perform pro-environmental actions. 

4 What is the relationship between 
demographic variables, such as age, gender, 
and education level and pro-environmental 
behaviour among university students at Wits? 

H4 Demographic variables, such as 
age, gender, and education level, 
significantly influence the likelihood 
of university students participating 
in pro-environmental activities. 

The next chapter defines the methodological approach that was employed to 

explore the hypotheses established from the literature review, which were 

suggested as possible answers to the research questions. 



 
20 

CHAPTER 3. RESEARCH METHODOLOGY 

This chapter outlines the approach that was used to examine the hypotheses 

identified from the literature review that were put forward as possible solutions to 

the research questions. 

3.1 Research Approach 

The research approach for this study involved a quantitative paradigm, focusing 

on numerical analysis and measurements to examine the prevalence and factors 

influencing PEB among university students (Mohajan, 2020).  

Existing literature on PEB at universities indicates that contextually most studies 

had been conducted in the North America, Europe, China and a few other Asian 

countries. Much of this research had focused on PEB in general populations, and 

there was limited understanding of how university students in South Africa 

engage in these behaviours.  

An analysis of 37 existing academic research journal articles on PEB indicated 

that from a methodological point of view, approximately 90% of the studies have 

been quantitative studies as illustrated in Figure 6. While this dominance might 

suggest the need for a balance of qualitative research, it also strongly suggests 

that a quantitative approach is a well-established method for previous related 

studies. It is on that basis that this study adopted a quantitative approach. 



 
21 

  

Figure 6: Methodologies in previous related studies  
Source: Developed by the researcher (2024) 

Additionally, this approach aligns with the need to investigate generalisable 

trends and causal mechanisms in behaviours, such as those between 

behavioural attitudes, subjective norms, perceived behavioural control, 

demographic characteristic and pro-environmental actions, within a specific 

population (Mohajan, 2020).  

3.2 Research Design 

The study used a cross-sectional design as its research framework as it is 

relatively cost effective design. Additionally, cross-sectional designs are used to 

test hypotheses and are especially valuable when the objective is to comprehend 

existing patterns or tendencies at a specific point in time (McCusker & Gunaydin, 

2015). Creswell and Creswell (2017) add that cross-sectional designs are 

beneficial for gathering data rapidly and efficiently and can be employed to 

examine independent variables (environmental knowledge, subjective norms, 

PBC, demographics) and a dependent variable/s (PEB).  

3.3 Data Collection Methods 

The study utilised data collected through a structured online questionnaire using 

the Qualtrics platform. The questionnaire included questions on PEB, 



 
22 

environmental knowledge, subjective norms, PBC and demographic information 

from a sample of students at Wits. Questionnaires are a common data collection 

tool in quantitative research, allowing for the systematic collection of standardised 

responses that can be analysed using statistical methods (Blair & Blair, 2020).  

3.4 Population and Sample 

3.4.1 Population 

In research, population refers to the collection of all the entities that are eligible 

to participate in a study (Creswell & Creswell, 2017). The population for this study 

consisted of 38 792 university students from various academic disciplines and 

different years of study currently studying at Wits. The students were 18 years of 

age or older as this is the legal age in South Africa. 

3.4.2 Sample and Sampling Method 

A simple random sampling technique was employed for this study. A 

comprehensive email list of all registered students at Wits was used, and each 

student was subsequently sent an email containing a link to the online 

questionnaire. This method ensured that every student had an equal probability 

of being included in the sample, minimising potential selection bias. Utilising email 

as the distribution medium facilitated efficient and rapid dissemination of the 

questionnaire to the entire student population at Wits. 

This study required the use of a sample size calculation formula to determine the 

sample size and based on Cochran (1977) and Lee and Landers (2022). The size 

of the population, the desired level of precision, and the desired level of 

confidence in the estimates were all considered by this formula.  

The equation for calculating the sample size for a population proportion was 

based on the following formula: 

𝑵 =
𝒁𝟐𝒑𝒒

𝒆𝟐
 



 
23 

where: 

N is the sample size, Z is the z-score corresponding to the level of confidence 

desired, p is the estimated proportion of the population with the characteristic of 

interest (e.g., the proportion of the population with high levels of environmental 

awareness and knowledge), q is 1 – p, and e is the desired level of precision 

(e.g., 0.05 for a 5% margin of error). 

The sample size is dependent on the size of the population until the population 

reaches approximately 40 000, after which the increase is close to none. 

Therefore, if the estimated population is greater than 40 000, the theoretical 

sample size, as computed for an unknown population, is about equal to the one 

generated by the modified formula. For this study, the sample size was therefore:  

Z = 1.96,  

p = 0.5 (approximate proportion of 50%),  

e = 0.05.  

𝑁 =
1.962 ∗ 0.5 ∗ 0.5

0.052
= 384 𝑟𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠 

To improve the rate of responses, emails were sent out to 38 792 students, out 

of which 337 responses were received, yielding a response rate of approximately 

87%. Among the responses, 267 were fully completed, while 28 had some data 

fields missing and 42 had many data fields missing. To address the missing data, 

mean imputation was employed, which involved replacing the missing values with 

the mean of the available data for each respective variable. This approach was 

chosen to preserve the overall distribution of the data while minimising the impact 

of incomplete responses on the analysis. However, it is acknowledged that mean 

imputation may reduce variability in the data, and sensitivity analyses or 

alternative methods (for example, multiple imputation) could be explored in future 

studies to validate the robustness of the findings. The timing of the survey should 

be carefully planned in future studies to avoid overlapping with students' year-

end exams, as this can contribute to response fatigue, lower participation rates, 

and an increase in incomplete or rushed responses. 



 
24 

Table 2: Profile of respondents 

Description of respondent type Responses 

Undergraduates 

 

Postgraduates 

 

162 

 

172 

 

Other 3 

TOTAL number of respondents 337 

3.5 The Research Instrument 

The study utilised data collected through a structured online questionnaire using 

the Qualtrics platform. The questionnaire included questions on demographic 

information, PEB and environmental attitudes. A five-point Likert scale was 

employed to assess all the survey items, with a range from 1 to 5, with 1 signifying 

strong disagreement, and 5 denoting strong agreement. Some open-ended 

questions were also included for each of the variables. 

The research adhered to recognised best methods for designing questionnaires, 

which involved testing the questionnaire initially to guarantee its reliability and 

validity (Dillman et al., 2014). 

3.6 Procedure for Data Collection 

Data collection is important in conducting academic research. Collection of the 

data in research is normally done through surveys, completion of questionnaires 

or through interviews (Johnson & Christensen, 2019). Data for this study was 

collected through use of an online self-administered questionnaire created using 

Qualtrics. The final questionnaire was distributed via email and university online 

platforms to 38 792 students from which 337 responses were received. The 

collected data was analysed to draw inferences in answering the research 

questions.  



 
25 

3.7 Data Analysis and Interpretation 

The data from the research was analysed using best practices in data 

management and statistical analysis. This process included arranging and 

scrubbing the data in IBM SPSS and Microsoft Excel to prepare it for analysis 

(Blair & Blair, 2020; Creswell & Creswell, 2017; Johnson & Christensen, 2014). 

This step ensured that the data was clean and ready for more complex analyses, 

as highlighted by the importance of data cleaning and preprocessing in 

descriptive analysis (Monsen, 2018). 

Data was analysed using IBM Statistical package for Social Sciences (IBM SPSS 

30.0) to generate percentages, means, standard deviations, and frequencies. 

The following statistical techniques were applied: 

Descriptive statistics - To summarise the data and describe the sample 

characteristics. This involved calculating measures such as mean, median, 

mode, standard deviation, and variance to offer a comprehensive overview of the 

data set. Descriptive statistics helped in understanding the distribution and 

central tendencies of the data, which is crucial for subsequent analyses (Monsen, 

2018; Nag & Ahmad Malik, 2023).  

Correlation analysis - To examine the relationships between socioeconomic 

factors, environmental knowledge, attitudes, and behaviours. This technique 

involved calculating correlation coefficients to determine the strength and 

direction of relationships between variables. Understanding these relationships 

are critical for identifying potential associations that can inform further analysis 

(Nag & Ahmad Malik, 2023).  

Regression analysis - To identify predictors of PEB among university students. 

This involved using regression models to assess the impact of independent 

variables (for example, demographic characteristics, environmental knowledge, 

subjective norms and PBC) on a dependent variable (PEB). Multiple regression 

analysis allows for the evaluation of multiple predictors simultaneously, providing 

insights into which factors significantly influence the behaviour in question (Nag 

& Ahmad Malik, 2023). The application of regression analysis requires careful 



 
26 

consideration of assumptions and model fit to ensure valid and reliable results 

(Teli et al., 2023). The integration of these statistical techniques provide a 

comprehensive approach to data analysis and interpretation.  

Descriptive statistics offer a baseline understanding of the data, correlation 

analysis reveals potential relationships, and regression analysis identifies key 

predictors, collectively contributing to a nuanced understanding of the factors 

influencing PEB among university students.  

The process of data interpretation involves creating meaning from the analysis, 

which requires clarity regarding the purpose of the study and the questions being 

addressed (Maxwell, 2021). This involves transforming and displaying data to 

reveal patterns and trends, which can be critical in understanding complex 

behaviours and interactions (Maxwell, 2021).  

In conclusion, the application of descriptive statistics, correlation analysis, and 

regression analysis provided a structured approach to understanding the 

relationships and predictors of PEB among university student at Wits. This 

approach, supported by careful data interpretation, ensured that the findings are 

meaningful and actionable, contributing to the broader understanding of 

environmental behaviours in higher educational contexts. 

3.8 Validity and Reliability  

3.8.1 External Validity (Generalisability) 

The concept of generalisability involves determining whether the findings of a 

study can be applied or extended to a broader population beyond the sample that 

was studied. In order to meet the conditions for generalisability, a study needs to 

utilise a sample that accurately represents the population of interest and 

guarantee that the results are not restricted to the particular sample examined 

(McCusker & Gunaydin, 2015; Trochim & Donnelly, 2001). 



 
27 

The study adhered to recognised guidelines for creating questionnaires, such as 

conducting a trial run of the questionnaire to verify its reliability and validity 

(Dillman et al., 2014). 

3.8.2 Internal Validity and Reliability  

The internal validity and reliability of a study are determined by the extent to which 

the outcomes can be attributed to the specific variables under investigation, 

rather than to other irrelevant factors (Cook et al., 2002). In other words, validity 

is concerned with whether the instrument measures what it is supposed to 

measure, and reliability is whether it measures it in a consistent and stable way. 

To achieve internal validity, a study needs to eliminate or manage errors that are 

naturally present in the research tool which could impact the results. It also needs 

to confirm that the observed influences between variables are not the result of 

chance occurrences or other random factors (Cook et al., 2002). 

The reliability of the questionnaire used for this study was tested using 

Cronbach's alpha, the result of which was higher than the minimum acceptable 

value. The overall Cronbach's alpha was calculated yielding a coefficient of 0.881 

and a standardised value of 0.889. This overall coefficient suggests that the 

instrument demonstrated good reliability. A Cronbach's alpha value of 0.7 or 

higher is considered acceptable for internal consistency. 

3.8.3 Objectivity  

The term "objectivity" relates to how impartial and free from personal or subjective 

influences the findings of a study are (Creswell & Creswell, 2017).  

Questions in the research instrument were phrased neutrally to avoid leading 

responses. Respondents' anonymity was guaranteed to encourage honest and 

unbiased responses. Data collection and analysis procedures were standardised 

to minimise researcher bias (Neuman & Robson, 2014). 



 
28 

3.9 Ethical Considerations 

Any research project, including those at Wits, must consider ethical factors. 

These factors are based on the principles and values that govern research 

conduct, ensuring responsible and trustworthy practices. 

The principal ethical concern for this study was around obtaining informed 

consent from participants. To guarantee that all participants were adequately 

informed to give meaningful consent, an introductory letter outlining the research 

was provided at the start of the survey. This introductory page explicitly stated 

that all responses would be kept entirely anonymous, that participants were not 

obligated to disclose their names or identities, and that involvement in the study 

was entirely voluntary, with participants having the right to withdraw from the 

survey at any point. Furthermore, it was clarified that the data gathered from the 

survey would be used solely for research purposes. 

Another key ethical consideration is to safeguard the confidentiality and privacy 

of respondents. Personal information and research data was safeguarded and 

kept private, with measures in place to prevent unauthorised access or use. 

Further, the researcher applied and obtained an ethics clearance certificate from 

Wits before embarking on any data collection. 

3.10 Summary of the Chapter 

This chapter outlined the approach that was used to examine the hypotheses 

identified from the literature review that were put forward as possible solutions to 

the research questions. Table 3 is a consolidation of the research questions, 

hypotheses, data collection detail and data analysis methods for each. 

 

 



 
29 

Table 3: Consistency table: research questions, hypotheses, data collection and data analysis 

RQ 
# 

Research Question  Hyp 
# 

Hypothesis Data collection detail Data analysis 
method 

1 What is the relationship between 
environmental knowledge and pro-
environmental behaviour among 
university students at Wits? 

H1 Higher levels of environmental knowledge 
significantly influence pro-environmental 
behaviour among university students. 

Questionnaires assessing 
environmental knowledge levels and 
self-reported pro-environmental 
behaviours among university 
students. 

Descriptive 
statistics, 
correlation 
analysis, 
regression 
analysis 

2 What is the relationship between social 
influence and pro-environmental 
behaviour among university students at 
Wits? 

H2 Greater social influence is positively 
correlated with university students' 
intentions to engage in pro-environmental 
behaviour. 

Surveys measuring perceived social 
influence (peer pressure, social 
norms) and pro-environmental 
behavioural intentions among 
students 

Descriptive 
statistics, 
correlation 
analysis, 
regression 
analysis 

3 What is the relationship between 
perceived behavioural control and pro-
environmental behaviour among 
university students at Wits? 

H3 Increased perceived behavioural control 
is positively related to university students' 
ability to perform pro-environmental 
actions. 

Questionnaires assessing students 
perceived behavioural control 
(confidence in their ability to engage 
in environmental behaviours) and 
self-reported environmental 
behaviours. 

Descriptive 
statistics, 
correlation 
analysis, 
regression 
analysis 

4 What is the relationship between 
demographic variables, such as age, 
gender, and education level and pro-
environmental behaviour among 
university students at Wits? 

H4 Demographic variables, such as age, 
gender, and education level, significantly 
influence the likelihood of university 
students participating in pro-
environmental activities. 

Collection of demographic data (age, 
gender, education level) alongside 
self-reported engagement in pro-
environmental activities. 

Descriptive 
statistics, 
correlation 
analysis, 
regression 
analysis 



 
30 

CHAPTER 4. PRESENTATION OF RESULTS 

4.1 Introduction 

This chapter interprets the study’ findings in light of the guiding conceptual 

framework. It draws on the conceptual framework outlined in Chapter 2 (Figure 

5), which was rooted in Ajzen’ Theory of Planned Behaviour (TPB). This 

framework hypothesised that environmental knowledge, subjective norms, and 

perceived behavioural control, along with demographic factors (age, gender, level 

of education, and field of study), collectively influence students’ pro-

environmental behaviour. The following sections therefore situate each result 

within this framework and broader literature, examining how the evidence from 

Wits aligns with or diverges from existing studies. The chapter starts with the 

demographic profile of the respondents, followed by insights from the open-

ended questions and then the reliability statistics. Analysis was by means of IBM 

SPSS version 30. 

4.2 Demographics Information of the Respondents 

A demographic profile provides detail about the research respondents offering 

perceptions of the characterises of the respondents in the context of the study 

(Creswell & Creswell, 2017).  

A link to the survey was generated using Qualtrics and sent to all current students 

at Wits. The respondents were explicitly informed that involvement in the study 

was entirely voluntary, and that their responses would be kept anonymous, 

confidential, and used solely for academic purposes. 

Data from 337 respondents were collected and analysed to create a demographic 

profile based on age, gender, educational level, and field of study.  

4.2.1 Age 

The age of the respondents is summarised in Table 4 and Figure 7. 



 
31 

Table 4: Age 

Age Frequency Percent 

18-24 169 50.1 

25-30 40 11.9 

31-35 39 11.6 

36+ 89 26.4 

 

Figure 7: Age 

 

Table 4: Age and Figure 7 show that 50.1% of the 337 respondents were aged 

between 18 and 24 years, 11% between 25 and 35, and 26.4% were 36 years or 

older.  

4.2.2 Gender 

The gender of the respondents is summarised in Table 5 and Figure 8. 

Table 5: Gender 

Gender Frequency Percent 

Male 109 32.3 

Female 219 65 

Non-binary / third gender 7 2.1 

Prefer not to say 2 0.6 



 
32 

 

Figure 8: Gender 

 

As presented in Table 5 and Figure 8, 65% of the 337 respondents were female 

while 32.3% were male, with 2.1% being non-binary. 

4.2.3 Level of Education 

The highest level of education of the respondents is summarised in Table 6 and 

Figure 9. 

Table 6: Level of education 

Level of education Frequency Percent 

Undergraduate 162 47.9 

Postgraduate 172 51.2 

Other 3 0.9 



 
33 

 

Figure 9: Level of education 

 

The bulk of the respondents as per Table 6 and Figure 9 were undergraduates 

and postgraduates with 47.9% and 51.2% respectively, while 0.9% indicated as 

other. 

4.2.4 Field of Study 

The respondents' field of study is summarised in Table 7 and Figure 10. 

Table 7: Field of study 

Field of study Frequency Percent 

Commerce Law and Management 72 21.4 

Engineering and the Built Environment 69 20.5 

Health Sciences 65 19.3 

Humanities 79 23.4 

Science 45 13.4 

Other 7 2.1 



 
34 

 

Figure 10: Field of study 

 

There was an even spread of respondents across all schools at Wits with slightly 

more respondents (23.4%) coming from the Humanities.  

4.3 Environmental Issues and Barriers 

Respondents were asked an open-ended question about what the most pressing 

environmental issue was they were currently concerned about. Figure 11 shows 

a word cloud of their responses. The word cloud visually represents the most 

frequent terms used by respondents of this study when discussing environmental 

issues. "Climate change" and "pollution" are the most prominent, highlighting 

students' awareness of these critical challenges.  

Concerns about "waste" and its "disposal" are also evident, alongside "plastic", 

indicating an understanding of pollution sources. Terms like "water", "energy", 

and "carbon emission" reflect key themes in environmental discourse, while 

"South Africa" suggests a focus on local contexts. The presence of "global 

warming" and "greenhouse" emphasises the connection between human 

activities and planetary health. Overall, the word cloud illustrates a broad 

understanding of environmental issues among the respondents, with an 

emphasis on climate change, pollution, and resource management. 



 
35 

 

Figure 11: Environmental concerns 

 

Figure 12 is a word cloud of responses relating to barriers that respondents have 

faced in trying to engage in pro-environmental actions. "Expensive", "cost", and 

"lack" feature prominently, suggesting financial barriers to sustainable choices. 

"Time" is also a significant concern, highlighting the perceived effort involved in 

eco-friendly actions. 

"Transport", "distance", and "access" indicate challenges related to infrastructure 

and availability of sustainable options. This is further emphasised by "car" and 

"public transport". "Sustainable", "environment", "eco-friendly", and "products" 

show a clear focus on environmental consciousness and consumer choices. The 

word cloud reveals that respondents were aware of and interested in sustainable 

living but face barriers in terms of cost, time, and accessibility. 



 
36 

 

Figure 12: Barriers to pro-environmental behaviour 

4.4 Reliability Statistics 

The internal consistency of the survey instrument used in this study was 

assessed using Cronbach's alpha. Typically, a Cronbach's alpha coefficient of 

0.7 or higher is deemed acceptable, while coefficients exceeding 0.8 are 

regarded as good, and those of 0.9 or greater as excellent. 

The overall Cronbach's alpha was calculated and is presented in Table 8, yielding 

a coefficient of 0.881 and a standardised value of 0.889. This overall coefficient 

suggests that the instrument demonstrates good reliability. 

Table 8: Cronbach's alpha test 

Cronbach's Alpha Cronbach's Alpha Based on 
Standardised Items 

N of Items 

0.881 0.889 23 

4.5 Results Pertaining to Pro-Environmental Behaviour  

The study used Likert type scale questions to measure the dependent variable 

PEB, and independent variables environmental knowledge, subjective norms, 

perceived behavioural control (Table 9). Descriptive analysis was also done for 

demographic characteristics. 



 
37 

Table 9: Likert type scale 

Very Easy Easy Neutral Difficult Very Difficult 

Always Often Sometimes Rarely Never 

Strongly agree Agree Neutral Disagree Strongly disagree 

5 4 3 2 1 

The next sections present the results of the survey. 

This segment outlines the descriptive statistics for questions measuring 

respondents' PEB. The means and standard deviation of the responses are 

presented in the Table 10. 

Table 10: Measuring pro-environmental behaviour 

Variable Mean Standard 
Deviation 

1 In the last month, how often have you engaged 
in recycling? 

2.76 1.150 

2 In the last month, how often have you engaged 
in conserving energy (e.g., turning off 
electronics when not in use)? 

3.86 0.951 

3 In the last month, how often have you engaged 
in reducing water consumption? 

3.62 0.996 

4 In the last month, how often have you engaged 
in using public transportation or cycling instead 
of driving? 

2.66 1.423 

5 In the last month, how often have you engaged 
in purchasing eco-friendly products? 

2.59 0.917 

6 I intend to engage in pro-environmental 
behaviour in the future. 

4.12 0.773 

7 Pro-environmental behaviour is important to me 
personally. 

4.03 0.769 

8 How likely are you to participate in university-
organised environmental activities (e.g., tree 
planting, clean-up drives)? 

3.56 1.175 

The outcome in Table 10 shows that the means of the sub-variables range from 

2.59 to 4.12. The mean scores for reducing water consumption and purchasing 

eco-friendly products fall between the extremes, suggesting moderate 

engagement in these behaviours. The intention to engage in PEB and its 

perceived importance are consistently rated high, which could indicate a positive 

trend towards greater environmental awareness and action. 



 
38 

To understand the variable measuring PEB, the mean of sub-variables is shown 

under the group variable in Table 11. 

Table 11: Group variable for pro-environmental behaviour 

Group variable Minimum Maximum Mean Standard Deviation 

Pro-environmental 
behaviour 

1 5 3.399 0.595 

The result in Table 11 shows that the mean of the group variable for measuring 

environmental knowledge is 3.399. The mean value of 3.399 suggests that, on 

average, respondents exhibit PEB slightly above the midpoint of the scale. A 

standard deviation of 0.595 indicates a moderate level of variability around the 

mean.  

Figure 13 shows the distribution of the group variable for measuring PEB among 

the respondents. 

 

Figure 13: Distribution for the group variable for pro-environmental behaviour 



 
39 

4.6 Results Pertaining to Hypothesis 1 

4.6.1 Environmental Knowledge 

This section outlines the descriptive statistics for questions measuring knowledge 

of environmental issues among the respondents. 

Table 12: Measuring knowledge of environmental issues 

Variable Mean 
Standard 
Deviation 

1 How would you rate your general knowledge of 
environmental issues? 

3.6 0.738 

2 
How often do you seek information on 
environmental issues (e.g., climate change, 
pollution, biodiversity)? 

2.99 0.973 

3 
I am aware of the impact my daily activities 
(e.g., energy usage, waste disposal) have on 
the environment. 

3.91 0.801 

4 I know what actions I can take to reduce my 
negative impact on the environment. 

3.73 0.855 

The outcome in Table 12 shows that the means of the variables range from 2.99 

to 3.91. The data suggests that respondents rate themselves relatively high on 

awareness and knowledge about environmental issues and the impact of their 

actions. They also have a good understanding of the actions they can take to 

reduce their negative impact. However, they are less likely to actively seek out 

information, as indicated by the lower mean and higher variability for that item. 

The group variable for measuring knowledge and awareness of environmental 

issues is shown in Table 13. 

Table 13: Group variable for Knowledge 

Group variable Minimum Maximum Mean Standard Deviation 

Knowledge 1 5 3.558 0.653 

The result in Table 13 shows that the mean of the group variable for measuring 

environmental knowledge is 3.558. This suggests that respondents generally 

have a moderate to high level of environmental knowledge. The standard 

deviation of 0.653 indicates a moderate level of variability around the mean.  



 
40 

Figure 14 shows the distribution of the group variable for measuring 

environmental knowledge and awareness among the respondents. 

 

Figure 14: Distribution for the group variable for knowledge 

4.7 Results Pertaining to Hypothesis 2 

4.7.1 Subjective Norms 

This section outlines the descriptive statistics for sub-variables measuring 

subjective norms of environmental issues among the respondents. 

Table 14: Measuring Subjective Norms 

Variable Mean Standard 
Deviation 

1 My family/friends actively encourage me to take 
pro-environmental actions (e.g., recycling, 
reducing energy use). 

2.78 0.948 

2 I feel influenced by the behaviours of those 
around me (friends, classmates, etc.) in regard 
to environmental actions. 

3.07 0.961 

3 In my social circles, pro-environmental 
behaviour is perceived as important. 

3.13 0.983 

4 I believe the university environment encourages 
pro-environmental actions. 

3.57 0.894 



 
41 

The outcome in Table 14 shows that the means of the variables range from 2.78 

to 3.57. The mean score for family influence is 2.78 with a standard deviation of 

0.948, indicating that family and friends offer moderate encouragement for pro-

environmental actions. The mean score on the university environment is 3.57 with 

a standard deviation of 0.894, suggesting that respondents believe the university 

environment encourages pro-environmental actions. 

The group variable for measuring social influence on PEB is shown in Table 15. 

Table 15: Group variable for subjective norms 

Group variable Minimum Maximum Mean Standard Deviation 

Subjective Norms 1 5 3.138 0.683 

The result in Table 15 shows that the mean of the group variable for measuring 

environmental knowledge is 3.138. The mean value of 3.138 suggests that 

respondents perceive a moderate level of social influence on PEB. The standard 

deviation of 0.683 indicates a moderate level of variability around the mean.  

Figure 15 shows the distribution of the group variable for measuring social 

influence among the respondents. 

 

Figure 15: Distribution for the group variable for subjective norms 



 
42 

4.8 Results Pertaining to Hypothesis 3 

4.8.1 Perceived Behavioural Control 

This section outlines the descriptive statistics for questions measuring PBC 

among the respondents. 

Table 16: Measuring perceived behavioural control 

Variable Mean Standard 
Deviation 

1 I have the resources (time, knowledge, tools) to 
engage in pro-environmental behaviour. 

3.37 0.875 

2 I believe that I have control over my ability to 
reduce my negative environmental impact. 

3.84 0.806 

3 Even if it requires extra effort, I am confident 
that I can engage in pro-environmental actions. 

3.85 0.803 

4 How easy or difficult is it for you to engage in 
recycling? 

3.51 0.985 

5 How easy or difficult is it for you to engage in 
reducing energy consumption (e.g., turning off 
lights)? 

3.97 0.886 

6 How easy or difficult is it for you to engage in 
using eco-friendly transportation (e.g., walking, 
cycling, public transport)? 

3.04 1.224 

7 How easy or difficult is it for you to engage in 
buying sustainable products? 

2.98 0.921 

The outcome in Table 16 shows that the means of the variables range from 2.98 

to 3.97 with varying standard deviations. This indicates that respondents have 

high confidence in their ability to perform certain actions (with mean scores of 

3.97 and 3.51) but lower confidence in others (with mean scores of 3.04 and 

2.98). Overall, the data suggests that respondents have a moderate to high level 

of perceived behavioural control. They believe they have the resources and 

control to engage in PEB and find it moderately easy to do so. However, their 

confidence varies depending on the specific action. 



 
43 

The group variable for measuring PBC on PEB is shown in Table 17. 

Table 17: Group variable for PBC 

Group variable Minimum Maximum Mean Standard Deviation 

PBC 1 5 3.508 0.586 

The result in Table 17 shows that the mean of the group variable for measuring 

PBC is 3.508. This suggests that respondents have a moderate to high level of 

perceived behavioural control over PEB. A standard deviation of 0.586 indicates 

a moderate level of variability around the mean. 

Figure 16 shows the distribution of the group variable for measuring 

environmental knowledge among the respondents. 

 

Figure 16: Distribution of the group variable for PBC 

4.9 Results Pertaining to Hypothesis 4 

4.9.1 Demographics (Age, Gender, Level of Education, Field of 

Study) 

This section outlines the descriptive statistics for questions measuring 

demographic characteristics of the respondents. 



 
44 

Table 18: Measuring Gender 

Gender Mean  Standard 
Deviation 

Male 3.335 0.655 

Female 3.445 0.535 

Non-binary / third gender 2.899 1.127 

Prefer not to say 3.637 0.336 

Analysing the data in Table 18, it can be observed that there are some variations 

in PEB across different genders. Females exhibit a slightly higher mean score 

(3.445) compared to males (3.335). Individuals who identify as non-binary or third 

gender have the lowest mean score (2.899), while those who prefer not to say 

have the highest (3.637). The data suggests a potential trend of slightly higher 

PEB scores among females compared to males. 

Table 19: Measuring Age 

Age Group Mean  Standard 
Deviation 

18-24 3.472 0.500 

25-30 3.494 0.527 

31-35 3.183 0.740 

36+ 3.314 0.688 

Analysing the data in Table 19, it can be observed that there are variations in 

PEB across different age groups. The 25-30 age group shows the highest mean 

score (3.494), closely followed by the 18-24 age group (3.472). Interestingly, the 

31-35 age group shows the lowest mean score (3.183). This suggests that PEB 

might not follow a linear trend with age in the context of students at Wits. 

Table 20: Measuring level of education 

Variables Mean Standard 
Deviation 

Undergraduate 3.442 0.558 

Postgraduate 3.355 0.602 

Other 2.875 1.635 



 
45 

The outcome in Table 20 shows that there is a slight variation in PEB across 

different education levels. Undergraduate students exhibit the highest mean PEB 

score (3.442). Postgraduate students have a slightly lower mean score (3.355), 

and the "other" category shows the lowest score (2.875). The data suggests a 

potential trend of higher PEB scores among graduate and undergraduate 

students compared to postgraduate students and those in the "other" category. 

Table 21: Measuring field of study 

Field of Study Mean Standard Deviation 

Commerce, Law, and Management 3.376 0.622 

Engineering and the Built Environment 3.358 0.547 

Health Sciences 3.306 0.541 

Humanities 3.400 0.634 

Science 3.633 0.479 

Other 3.399 1.142 

According to the results in Table 21, there is some variation in PEB across 

different fields of study. Respondents in Science exhibit the highest mean score 

(3.633), followed by those in Humanities (3.400) and Commerce, Law, and 

Management (3.376). Respondents in Engineering and the Built Environment 

and Health Sciences have slightly lower mean scores (3.358 and 3.306, 

respectively). The "Other" category shows a mean score close to the overall 

average (3.399). 

The data suggests a potential trend of higher PEB scores among respondents in 

Science and Humanities. 

The demographic variables age and field of study have statistically significant 

relationships with PEB, while gender and level of study do not. The overall 

predictive power of the models is relatively low, suggesting that these 

demographic variables alone may not be strong predictors of PEB. 



 
46 

4.10 Influence of Independent Variables on Pro-Environmental 

Behaviour 

The relationship between PEB (the dependent variable) and knowledge, 

subjective norms, and perceived behavioural control (the independent variables) 

was analysed using multiple linear regression. This was estimated by the 

following model: 

Pro_Environmental_Behaviour=β1Knowledge+β2Subjective_Norms+β3

Perceived_Behavioural_Control+β3Demographics+ϵ 

Where: 

β1, β2, β3 and β4 are correlation coefficients, and  

ϵ = Error term 

Table 22: Model summary 

Model R R Square Adjusted R 
Square 

Std. Error of the Estimate 

1 0.756 0.571 0.566 0.392 

Table 22 presents the model summary, revealing a moderate to strong 

association between the independent and dependent variables. The R-squared 

value of 0.756 signifies a robust positive correlation between the predictor 

variable(s) and the outcome variable. Furthermore, the coefficient of 

determination (0.571) suggests that roughly 57.1% of the variance in the 

dependent variable can be explained by the independent variable(s) included in 

the model. This suggests that while the model captures important factors 

influencing PEB, there are likely other variables not included in this study that 

also play a role in shaping environmental behaviours among students. 

  



 
47 

Table 23: Analysis of variance (ANOVA) 

  Sum of 
Squares 

df Mean Square F Sig. 

Regression 67.545 3 22.515 145.539 <.001 

Residual 51.515 333 0.155 

  

Total 119.061 336 

   

The results in the ANOVA output in Table 23 show that the multiple regression 

model is statistically significant as indicated by the low p-value of < 0.001, and a 

high F-value (145.539) meaning that the model is effective at predicting the 

outcome variable. 

The coefficients in Table 24 provide detailed information about the individual 

predictors in the regression model. It shows the strength and direction of the 

relationship between each predictor and the outcome variable. 

Table 24: Coefficients 

  Unstandardised 
Coefficients 

Standardised 
Coefficients 

  

B Std. 
Error 

Beta t Sig. 

Constant 0.482 0.482 0.167 2.897 0.004 

Demographics (Age, 
Gender, Level of education, 
Field of education) 

-0.06 0.037 -0.06 -1.643 0.101 

Knowledge 0.282 0.04 0.309 7.108 <.001 

Subjective Norms 0.189 0.037 0.217 5.138 <.001 

Perceived Behavioural 
Control 

0.414 0.044 0.408 9.449 <.001 

The results in the coefficients output in Table 24 shows that three independent 

variables – knowledge, subjective norms and perceived behavioural control – 

have a positive relationship with the dependent variable PEB. Both 

unstandardised and standardised coefficients are positive. These results are 

statistically significant as indicated by the low p-value of < 0.001. The 

independent group variable demographics have a negative relationship with the 

dependent variable PEB with a p-value > 0.001.  

 

 



 
48 

Table 25: Coefficients for demographics 
 

Unstandardised 
Coefficients 

Standardised 
Coefficients 

 

B Std. 
Error 

Beta t Sig. 

Constant 3.420 0.154 

 

22.170 <.001 

Age -0.061 0.035 -0.132 -1.757 0.080 

Gender 0.004 0.062 0.004 0.066 0.948 

Level of Education 0.006 0.044 0.010 0.128 0.898 

Field of Study 0.031 0.023 0.075 1.347 0.179 

However as shown in Table 25, the demographic variables age and field of study 

have statistically significant relationships with PEB, while gender and level of 

study do not. The overall predictive power of the models is relatively low, 

suggesting that these demographic variables alone may not be strong predictors 

of PEB. 

Table 26 summarises the hypotheses testing results. 

Table 26: Hypotheses testing 

Hypotheses P-value Results 

H1 < 0.001 Statistically significant and supported 

H2 < 0.001 Statistically significant and supported 

H3 < 0.001 Statistically significant and supported 

H4 0.101 Age and Field of Study are statistically significant 

Gender and level of Study are not statistically significant 

4.11 Summary of the Chapter 

The chapter presented the results and outcomes of this study. The findings 

indicate a favourable relationship between knowledge and awareness of 

environmental issues and PEB among respondents. This result is statistically 

significant at a 5% level (p-value < 0.001) and addresses the first research 

question to investigate if there is a relationship between knowledge and 

awareness and PEB. There is also a positive correlation between social influence 

and PEB. This result is statistically significant at a 5% level (p-value < 0.001) and 

addresses the second research question to investigate if there is a relationship 

between subjective norms and PEB. In addition, there is a positive relationship 



 
49 

between perceived behavioural control and PEB. This result is statistically 

significant at a 5% level (p-value < 0.001) and addresses the third research 

question to investigate if there is a relationship between PBC and PEB. Finally, 

the demographic variables age and field of study have statistically significant 

relationships with PEB, while gender and level of study do not. This data suggests 

a potential link between age and PEB, with older individuals showing slightly 

lower scores. The overall predictive power of the models is relatively low, 

suggesting that these demographic variables alone may not be strong predictors 

of PEB. 

Comparison of literature review and own findings Table 27 is a consistency table 

to compare the answers obtained to the research questions from the study with 

the answers obtained from the literature review which gave rise to the study's 

hypotheses. 

Table 27: Comparison of literature review and findings 

RQ # Research Question  Hyp # Hypothesis Findings from own study 

1 What is the relationship 
between environmental 
knowledge and pro-
environmental behaviour 
among university students 
at Wits? 

1 Higher levels of 
environmental knowledge 
significantly influence pro-
environmental behaviour 
among university students. 

Respondents generally 
have a high level of 
environmental knowledge, 
with an average score of 
3.558 out of 5. However, 
there is still room for 
improvement, as some 
respondents reported 
having difficulty 
understanding specific 
environmental concepts or 
knowing how to take action 
to reduce their 
environmental impact. 

2 What is the relationship 
between social influence 
and pro-environmental 
behaviour among 
university students at 
Wits? 

2 Greater social influence is 
positively correlated with 
university students' 
intentions to engage in pro-
environmental behaviour. 

The average score for 
social influence was 3.138 
out of 5, suggesting that 
social influence plays a 
moderate role in shaping 
pro-environmental 
behaviour among the 
respondents. Family and 
friends were found to be 
the most influential social 
group, followed by social 
circles and the university 
environment. 

3 What is the relationship 
between perceived 
behavioural control and 

3 Increased perceived 
behavioural control is 
positively related to 

Respondents had a 
moderate to high level of 
perceived behavioural 



 
50 

RQ # Research Question  Hyp # Hypothesis Findings from own study 

pro-environmental 
behaviour among 
university students at 
Wits? 

university students' ability 
to perform pro-
environmental actions. 

control over pro-
environmental behaviour, 
with an average score of 
3.508 out of 5. This means 
that respondents generally 
believe they can engage in 
pro-environmental 
behaviour, even if it 
requires extra effort. 
However, some 
respondents reported 
facing barriers to pro-
environmental behaviour, 
such as the cost of eco-
friendly products and the 
inconvenience of using 
public transportation. 

4 What is the relationship 
between demographic 
variables, such as age, 
gender, and education 
level and pro-
environmental behaviour 
among university students 
at Wits? 

4 Demographic variables, 
such as age, gender, and 
education level, 
significantly influence the 
likelihood of university 
students participating in 
pro-environmental 
activities. 

The regression analysis 
showed that demographic 
variables age and field of 
study have statistically 
significant relationships 
with pro-environmental 
behaviour, while gender 
and level of study do not. 
The overall predictive 
power of the models was 
relatively low, suggesting 
that these demographic 
variables alone may not be 
strong predictors of pro-
environmental behaviour. 

 



 
51 

CHAPTER 5. DISCUSSION OF THE RESULTS  

5.1 Introduction  

This chapter interprets the study’ findings in light of the guiding conceptual 

framework. It draws on the conceptual framework outlined in Chapter 2 (Figure 

5), which was rooted in Ajzen’ Theory of Planned Behaviour (TPB). This 

framework hypothesised that environmental knowledge, subjective norms, and 

perceived be