EFFECT OF SOCIAL MEDIA INFLUENCER ATTRIBUTES ON CONSUMER 

BEHAVIOUR: A CASE STUDY OF COSMETIC MULTINATIONAL FIRMS IN 

NAIROBI COUNTY 

 

 

 

BY 

MUHUNI JACKLINE KERU 

MCOM/ 072369 

 

 

 

 

 

THESIS SUBMITTED TO THE STRATHMORE BUSINESS SCHOOL IN 

PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF 

DEGREE OF MASTER OF COMMERCE OF STRATHMORE UNIVERSITY 

 

MAY, 2024



 

i 

 

DECLARATION 

I declare that this work has not been previously submitted and approved for the award of a degree 

by this or any other University. To the best of my knowledge and belief, the dissertation contains 

no material previously published or written by another person except where due reference is made 

in the dissertation itself.  

Name:  Jackline Keru 

Signature: …J.K.M…………                                                   Date…30th May 2024…………                                        

 Approval  

The research thesis has been submitted for examination with my approval as a university 

supervisor.  

 

Dr. Nancy Njiraini 

Strathmore University 

Signature: …………………….                                                  Date………………….  

  



 

ii 

 

ABSTRACT 

The changing environment of influencer marketing presents both opportunities and difficulties for 

multinational cosmetic companies in Nairobi, Kenya. A deep awareness of the qualities that lead 

to fruitful collaborations is necessary for the identification and engagement of suitable social media 

influencers. However, there is little research on the specific attributes of social media influencers 

and their effect on consumer behavior. This study sought to fill this gap by reviewing how various 

social media influencer attributes; perceived expertise, perceived trustworthiness and perceived 

loyalty have influenced consumer behaviour in cosmetic multinational firms in Nairobi County. 

The study was anchored on social influence theory and source credibility theory. The study was 

based on descriptive research design and positivism research philosophy. The unit of observation 

was 75 multinational cosmetic companies operating in Nairobi, Kenya. The unit of analysis was 

comprised of 5 consumers from each of the 75 selected multinational cosmetic companies, 

resulting in a total of 375 individual consumers. Therefore, a sample size of 375 was selected 

through systematic sampling. The study relied on primary data collected using questionnaires. 

Data was analyzed using descriptive and inferential statistics. Findings are presented in figures 

and tables. Results of the study depicted that there was a positive statistically significant effect of 

perceived expertise of social media influencers on consumer behaviour. Secondly, the study 

depicted that there was a positive statistically significant effect of perceived trustworthiness of 

social media influencers on consumer behaviour. Thirdly, there was a positive statistically 

significant effect of perceived loyalty of social media influencers on consumer behaviour. From 

the findings it can be concluded that social media influencers should enhance their expertise on 

the products that they serve as ambassadors to leverage their influential role on consumer 

preference and choices. Cosmetic products distributors in Kenya should evaluate their social media 

influencer collaborations and strictly engage those who demonstrate authenticity, reliability and 

capacity not only to connect but also influence consumer decision making. Consumers tend to 

develop affinity towards products whose social media influencers depicts loyalty on cosmetics 

they endorse that would nurture a culture of loyalty linked customer engagement. It was 

recommended that there is a need for development of training programs for employees and social 

media influencers, transparent dissemination of product information and development of social 

media influencer selection criteria.  

 

 

 

 

 

 



 

iii 

 

TABLE OF CONTENTS 

 

DECLARATION............................................................................................................................ i 
ABSTRACT ................................................................................................................................... ii 
TABLE OF CONTENTS ............................................................................................................ iii 
LIST OF TABLES ....................................................................................................................... vi 
LIST OF FIGURES .................................................................................................................... vii 
ABBREVIATIONS AND ACRYNOMS .................................................................................. viii 
OPERATIONAL DEFINITION OF KEY TERMS ................................................................. ix 
ACKNOWLEDGEMENTS ......................................................................................................... x 
DEDICATION.............................................................................................................................. xi 
CHAPTER ONE ........................................................................................................................... 1 
INTRODUCTION......................................................................................................................... 1 

1.1 Background of the Study ...................................................................................................... 1 
1.1.1 Social Media Influencers ............................................................................................... 2 
1.1.2 Attributes of Social Media Influencers .......................................................................... 4 
1.1.3 Consumer Behaviour...................................................................................................... 6 
1.1.4 Cosmetics Industry in Kenya ......................................................................................... 8 

1.2 Statement of the Problem ...................................................................................................... 9 
1.3 Objectives of the Study ....................................................................................................... 11 

1.3.1 General Objective of the Study .................................................................................... 11 
1.3.2 Specific Objectives of the Study .................................................................................. 11 

1.4 Research Questions ............................................................................................................. 11 
1.5 Scope of the Study .............................................................................................................. 11 
1.6 Significance of the Study .................................................................................................... 12 
1.7 Chapter Summary ............................................................................................................... 13 

CHAPTER TWO ........................................................................................................................ 14 
LITERATURE REVIEW .......................................................................................................... 14 

2.1 Introduction ......................................................................................................................... 14 
2.2 Theoretical Review ............................................................................................................. 14 

2.2.1 Social Influence Theory ............................................................................................... 14 
2.2.2 Source Credibility Theory ............................................................................................ 16 

2.3 Empirical Literature Review ............................................................................................... 17 
2.3.1 Perceived Expertise and Consumer Behaviour ............................................................ 17 
2.3.2 Perceived Trustworthiness and Consumer Behaviour ................................................. 19 
2.3.3 Perceived Loyalty and Consumer Behaviour............................................................... 21 

2.4 Research Gaps ..................................................................................................................... 23 
2.5 Conceptual Framework ....................................................................................................... 24 
2.6 Chapter Summary ............................................................................................................... 26 

CHAPTER THREE .................................................................................................................... 27 
RESEARCH METHODOLOGY .............................................................................................. 27 

3.1 Introduction ......................................................................................................................... 27 
3.2 Research Philosophy ........................................................................................................... 27 
3.3 Research Design ................................................................................................................. 28 
3.4 Target Population ................................................................................................................ 28 



 

iv 

 

3.5 Sample Size and Sampling Technique ................................................................................ 29 
3.6 Data Collection Instruments ............................................................................................... 30 
3.7 Data Collection Procedures ................................................................................................ 30 
3.8 Research Quality ................................................................................................................. 31 

3.8.1 Reliability of Instrument .............................................................................................. 31 
3.8.2 Validity of Instrument .................................................................................................. 32 

3.9 Data Analysis and Presentation .......................................................................................... 32 
3.10 Diagnostic Tests ................................................................................................................ 33 

3.10.1 Multicollinearity Test ................................................................................................. 33 
3.10.2 Heteroskedasticity Test .............................................................................................. 33 
3.10.3 Normality Test ........................................................................................................... 33 

3.11 Diagnostics Tests Results ................................................................................................. 34 
3.11.1 Multicollinearity Test ................................................................................................. 34 
3.11.2 Heteroskedasticity Test .............................................................................................. 34 
3.11.3 Normality Test ........................................................................................................... 34 

3.12 Ethical Considerations ...................................................................................................... 35 
3.13 Chapter Summary ............................................................................................................. 36 

CHAPTER FOUR ....................................................................................................................... 37 
FINDINGS AND INTERPRETATIONS .................................................................................. 37 

4.1 Introduction ......................................................................................................................... 37 
4.2 Response Rate ..................................................................................................................... 37 
4.3 Demographic Information ................................................................................................... 37 
4.4 Descriptive Statistics on Perceived Expertise ..................................................................... 38 
4.5 Descriptive Statistics on Perceived Trustworthiness .......................................................... 40 
4.6 Descriptive Statistics on Perceived Loyalty ....................................................................... 41 
4.7 Descriptive Statistics on Consumer Behaviour .................................................................. 43 
4.8 Spearman’s Rank Correlation ............................................................................................. 44 

4.8.1 Consumer Behavior and Perceived Expertise: ............................................................. 45 
4.8.2 Consumer Behavior and Perceived Trustworthiness ................................................... 45 
4.8.3 Consumer Behavior and Perceived Loyalty ................................................................ 46 
4.8.4 Summary and Implications .......................................................................................... 46 

4.9 Regression Analysis ............................................................................................................ 46 
4.9.1 Model Summary ........................................................................................................... 46 
4.9.2 ANOVA ....................................................................................................................... 47 
4.9.3 Regression Coefficients ............................................................................................... 47 
4.9.4 Implications .................................................................................................................. 48 

4.10 Chapter Summary ............................................................................................................. 48 
CHAPTER FIVE ........................................................................................................................ 49 
SUMMARY, DISCUSSION, CONCLUSION AND RECOMMENDATIONS .................... 49 

5.1 Introduction ......................................................................................................................... 49 
5.2 Summary of Findings .......................................................................................................... 49 
5.3 Discussions ......................................................................................................................... 50 

5.3.1 Perceived Expertise and Consumer Behaviour ............................................................ 50 
5.3.2 Perceived Trustworthiness and Consumer Behaviour ................................................. 50 
5.3.3 Perceived Loyalty and Consumer Behaviour............................................................... 51 

5.4 Conclusion .......................................................................................................................... 51 



 

v 

 

5.4 Recommendations ............................................................................................................... 52 
5.5 Suggestions for Further Studies .......................................................................................... 55 
5.6 Limitations of the Study ..................................................................................................... 55 
5.7 Chapter Summary ............................................................................................................... 56 

REFERENCES ............................................................................................................................ 57 
APPENDICES ............................................................................................................................. 60 

Appendix I: Introduction Letter ................................................................................................ 60 
Appendix II: Institution Review Board Approval .................................................................... 61 
Appendix III: NACOSTI Research Licence ............................................................................. 62 
Appendix IV: Participant Consent Form .................................................................................. 63 
Appendix V: Questionnaire ...................................................................................................... 67 
Appendix VII: List of Multinational Cosmetic Firms .............................................................. 71 
Appendix VIII: Dissemination Plan ......................................................................................... 73 

 

 

  



 

vi 

 

LIST OF TABLES 

Table 2.1 Summary of literature gaps 

Table 2.2 Operationalization of variables 

Table 3.1 Reliability Analysis 

Table 3.2 Multicollinearity TestError! Bookmark not defined.Table 3.3 Heteroskedasticity Test

  

Table 4.1 Response Rate  

Table 4.2 Background Information38Table 4.3 Descriptive Statistics on Perceived Expertise

39Table 4.4 Descriptive Statistics on Perceived Trustworthiness

40Table 4.5 Descriptive Statistics on Perceived Loyalty

 42Table 4.6 Descriptive Statistics on Consumer Behaviour  

Table 4.7 Spearman’s Rank Correlation44 Table 4.8 Model Summary

47Table 4.9 ANOVA

 47Table 4.10 Regression Coefficients  

  



 

vii 

 

LIST OF FIGURES 

Figure 2.1 Conceptual Framework 25Figure 4.1 Normality Test 35 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ABBREVIATIONS AND ACRYNOMS 

NACOSTI National Commission for Science Technology and Innovation 



 

viii 

 

PWC Price Waterhouse Coopers  

SMIs Social Media Influencers  

SPSS Statistical Packages for Social Scientists  

 

  



 

ix 

 

OPERATIONAL DEFINITION OF KEY TERMS 

Consumer behaviour  Lee and Young (2020) emphasizes that marketers' trust in influencers is 

influenced by their trustworthiness, subject matter knowledge, and 

authenticity, which in turn affects their intentions to partner with them. 

In this study consumer behaviour will be defined as perception of value, 

purchase intention, confidence and satisfaction.  

Perceived Expertise  Refers to capacity of social media influencer to show authority on 

product information, been informative and knowledgeable (Chan, 2022) 

Perceived Loyalty 

 

Perceived Trustworthiness 

Refers to capacity to social media to have social proof, endorsement and 

attractiveness towards customers (Khasbulloh & Gede, 2022) 

Lou & Yuan (2019), the level of trustworthiness is determined by how 

much the audience believes the influencer to be sincere, honest, or true. 

In this study it will be defined as social proof, endorsement and 

attractiveness. 

 

 

  



 

x 

 

ACKNOWLEDGEMENTS 

I thank God Almighty for granting me strength to achieve this milestone. I thank my family for 

the support given to complete this thesis and for their encouragement through prayers, emotional 

support and understanding throughout this journey. I thank my supervisor Dr. Nancy Njiraini for 

challenging me and offering her guidance during the research journey and providing unaverred 

support. To my friends and colleagues that provided guidance and support along the way, thank 

you. 

 

 

 

 

 

 

 

 

 

 

 

 



 

xi 

 

DEDICATION 

I’m dedicating this thesis to my parents, husband and our lovely son Kieran King’ori.



 

1 

 

CHAPTER ONE 

INTRODUCTION 

1.1 Background of the Study 

Social media influencers (SMI), by virtue of their public presence and personal brand, are 

susceptible to controversies and scandals, which can have profound negative impacts on the brands 

they endorse (Maharaj, 2021). Influencer controversies can significantly alter consumer behavior. 

Consumer behaviour is defined as the study of individuals, groups, or organizations and the 

processes they use to select, secure, use, and dispose of products, services, experiences, or ideas 

to satisfy needs and the impacts that these processes have on the consumer and society (Hoyer & 

Pieters, 2016) . Erdogan (2019) states that when an influencer's behavior or actions lead to public 

outrage, the fallout can extend to associated brands, resulting in a loss of consumer trust, decreased 

sales, and long-term damage to brand reputation.  

Over the years, social media has grown quickly and significantly, and billions of people use it 

every day for a wide range of activities around the world. According to (Shah & Qureshi, 2019), 

about one-third of the world's population is online and regularly uses social media. This has 

changed consumer behavior. Social media has fundamentally transformed global marketing 

practices, enabling companies to engage with consumers across the world in real-time (Gaenssle 

& Budzinski, 2021). Social media influencers (SMIs) have become pivotal in these interactions, 

wielding considerable influence over consumer behaviors and brand perceptions. With the advent 

of platforms like Instagram, YouTube, and TikTok, influencers have emerged as key players in 

the marketing landscape (Berger & Iyengar, 2013). These influencers, who range from celebrities 

to niche content creators, leverage their large followings to promote products and services, thereby 

influencing consumer choices on a global scale (Khamia & Welling, 2017). Their global reach is 

enhanced by the internet's borderless nature, allowing influencers to connect with diverse 

audiences across different regions and cultures (De Veirman, 2017). 

Social media influencers have transformed marketing practices across various industries globally 

(Ahmad, N. et al. 2015). For instance, the cosmetics industry has witnessed a significant shift due 

to the influence of beauty bloggers and makeup artists on platforms like Instagram and YouTube. 



 

2 

 

Influencers like Huda Kattan, founder of Huda Beauty has amassed millions of followers, making 

her recommendations and reviews highly influential (Freberg & McGaughey, 2011). Huda Kattan 

started as a beauty blogger and successfully launched her own cosmetics line, which quickly 

became a global brand largely due to her social media influence and direct engagement with her 

followers (Freberg & McGaughey, 2011). 

In the context of Nairobi County, Kenya, social media influencers are playing an increasingly 

important role in shaping consumer behavior, particularly in the cosmetics industry. Cosmetic 

multinational firms in Nairobi leverage influencer marketing to navigate the competitive landscape 

and connect with consumers on a personal level ( (NJuguna & Njeru, 2020). Influencers such as 

Joanna Kinuthia and Patricia Kihoro have become prominent figures in the Kenyan beauty scene. 

Joanna Kinuthia, for instance, is known for her beauty tutorials and product reviews (Kinuthia, 

2021). Her influence has led to collaborations with both local and international cosmetic brands, 

thereby impacting consumer purchase decisions and brand perceptions in Nairobi (Kinuthia, 

2021). This study investigated the effect of social media influencer attributes on consumer 

behaviour in cosmetic multinational firms in Nairobi County. 

1.1.1 Social Media Influencers  

Social Media Influencer (SMI) are defined as individuals with a sizable social media following 

who act as trustworthy authority and trend-setters in a particular business (De Veirman, 2017). In 

the quickly changing digital environment, we are witnessing the new symbols of these generations 

developing quickly. (Ruiz-Gomez, 2021) refers to these heroes as social media influencers due to 

their enormous social media following. According to (Mahmud, 2018) many academics view 

influencers as "special individuals" who possess the ability to create noteworthy content and a 

solid reputation in a certain sector. Customers' faith in influencers may stem from their expertise 

in their specific field. According to (Zhou & Jin, 2023)it has been demonstrated that using brand 

influencers increases view-through rates by 193%, that 35% of users learn about items and 

companies from influencers, and that 65% of users engage with the material when influencers post 

about them. 

Research on social media influencers (SMIs) has evolved significantly over the past decade, with 

scholars exploring various conceptualizations of influencers' roles and impacts across different 



 

3 

 

channels. (Solomon, 2017) states that despite the similarity of the platforms used such as 

Instagram, YouTube, TikTok, and Facebook the ways in which influencers are perceived, utilized, 

and studied can vary greatly. (Schiffman & Kanuk, 2010) opine that social media influencers 

(SMI) create and share content through social media platforms to drive engagement and increase 

awareness. (Solomon, 2017) opined that despite social media influencers having a significant 

influence on consumer behaviour the core benefits had not been exploited. (Simiyu, 2021) 

determined that SMI significantly influences consumer behaviour.  

Studies have explored influencers around parasocial relationships (PSRs), which describe the one-

sided relationships that followers develop with influencers. (Horton, 2016) originally introduced 

the concept of PSRs to describe interactions between television personalities and viewers. The 

influence of these personalities can be attributed to their perceived authenticity and relatability, 

which traditional celebrity endorsements often lack (Kim, 2011). This concept has been extended 

to social media, where followers feel a sense of intimacy and friendship with influencers, despite 

the lack of reciprocal interaction (Chung, 2017). Influencers often share content related to lifestyle, 

beauty, fashion, and other consumer interests, making them valuable partners for cosmetic brands 

(Abidin, 2016). 

Other studies have categorized influencers into different archetypes based on their content and 

interaction styles. (De Veirman, 2017) identified different types of influencers, such as megastars, 

who have a large following but less personal interaction, and micro-influencers, who have smaller 

followings but higher engagement rates. Campbell & Farrel (2020) discovered one of the most 

often used classifications which included four distinct groups of influencers: macro, micro, nano, 

and mega influencers. Still, there is a lot of confusion around the considerations that consumers 

make when making a purchasing decision based on interacting with social media influencers’ 

contents. 

More recently, studies have started to focus on negative aspects such as influencer fatigue and the 

authenticity crisis. Influencer fatigue refers to the diminishing effectiveness of influencer 

marketing due to oversaturation and excessive commercial content, which can lead to consumer 

distrust and disengagement (Schouten & Verspaget, 2020). Authenticity is increasingly seen as a 

crucial factor, with research suggesting that consumers are more influenced by content that appears 



 

4 

 

genuine and relatable, rather than overly polished or promotional (Audrezet & Moulard, 2020). 

Negative traits such as inauthenticity, over-commercialization, and controversies can harm an 

influencer's credibility. For instance, scandals involving influencers can significantly damage their 

and the associated brand's reputation (Hwang & Zhang, 2018). In Kenya, the Shaffie Weru scandal 

exemplifies how negative publicity can impact public perception and trust (Wainaina, 2021). 

1.1.2 Attributes of Social Media Influencers 

Social media influencer attributes are the distinct characteristics or qualities that define their 

effectiveness and impact on social media platforms (Dessart et al. 2015). These attributes shape 

how influencers are perceived by their audience, influencing followers' opinions, attitudes, and 

behaviors (Vrontis & Christofi, 2021). Studies have examined the role of influencer attributes such 

as credibility, attractiveness, and relatability in shaping consumer attitudes and purchasing 

decisions (Jin & Phua, 2014). However, there is a tendency to generalize findings across different 

demographic and cultural groups within the country, neglecting the diverse consumer segments 

that exist. For instance, the influence of social media personalities may vary significantly between 

urban and rural populations or among different ethnic groups, which has not been thoroughly 

explored  (De Veirman, 2017) 

Previous studies have often conflated these attributes or generalized findings across diverse 

demographic and cultural groups (De Veirman, 2017). According to (Lou & Yuan, 2019) 

credibility encompasses the trustworthiness and perceived honesty of the influencer, which can 

significantly affect consumer attitudes towards promoted products. According to (Uzunoglu & 

Kip, 2014) credibility is mostly pertaining to the fact that the influencers have used the product 

themselves, and it has been demonstrated that this favorably influences followers' purchase 

intentions. Djafarova and Rushworth (2017) explored credibility but used a broad measure that 

included both trustworthiness and expertise without differentiating between them, potentially 

conflating their distinct impacts.  

Research indicates that higher credibility leads to greater persuasive power and influence over 

followers (Hovland et al. 2020). (Lou & Yuan, 2019) both found a direct correlation between a 

communication source's persuasiveness and level of influence. (Lou & Yuan, 2019) studied the 

effects of social media influencer marketing but did not distinctly define perceived expertise, 



 

5 

 

instead lumping it with overall influencer credibility. This broad approach may obscure the 

specific impact of expertise on consumer behavior. Freberg et al. (2011) examined public 

perceptions of social media influencers but did not isolate perceived trustworthiness as a separate 

construct, treating it as part of overall influencer appeal. This general approach can dilute the 

specific role of trustworthiness in influencing consumer behavior. 

Attractiveness is defined as a set of attributes that cause an individual to be favorably perceived 

by others. These attributes can include physical features such as facial symmetry, body shape, and 

grooming, as well as non-physical characteristics such as confidence, kindness, and intelligence 

(Langlois et al. 2000). The social and physical appeal of the influencers was characterized by 

(Delbaere et al. 2021) as attractiveness. Most prior studies have only been based solely on physical 

appearance. Lee & Watkins (2016) demonstrated that physical beauty encourages a certain 

behavior that the influencer has advocated to have significance in shaping brand views and to 

affect followers' purchase intentions. According to Sakib et al. (2020) influencers' ability to evoke 

strong feelings and create a sense of community had a positive effect on consumer’s purchase 

behaviour.  

Relatabilty measures the level of engagement between the influencer and their audience. High 

engagement indicates a strong connection and active communication, which can enhance the 

influencer's impact on their followers (Dessart & Thomas, 2015). Metrics such as follower count, 

post reach, likes, shares, and comments are used to evaluate an influencer's audience size and 

engagement rates (Kemp, 2021). Jin and Phua (2014) looked at social identification with 

celebrities but did not explicitly measure perceived loyalty to influencers, which could have 

provided deeper insights into consumer allegiance and repeated engagement with endorsed brands. 

The current study examined the effect of social media influencer attributes on consumer behaviour 

using the attributes: perceived expertise, perceived trustworthiness and perceived loyalty. Key 

attributes enhancing influencer effectiveness include perceived expertise, perceived 

trustworthiness, and perceived loyalty (Djafarova & Rushworth, 2017). This is as also ascribed in 

(Jepkoech, 2020) study. These measures were also used by Mwangi, J. (2018). The focus on these 

attributes is justified by their significant roles in shaping consumer behavior (Djafarova & 

Rushworth, 2017).  



 

6 

 

Perceived Expertise refers to the influencer's knowledge and proficiency in a specific domain, 

enhancing consumer confidence in product recommendations (Hovland et al. 2020). Consumers 

are more likely to trust and act on advice from influencers considered experts in cosmetics. Evans 

et al. (2017) found that influencers' expertise significantly impacts consumer attitudes and 

behaviors. Perceived trustworthiness involves the influencer's perceived honesty and integrity, 

which can significantly affect consumer attitudes towards promoted products (Xiao et al. 2018). 

Trustworthiness is enhanced when influencers use the products themselves and share genuine 

experiences (Uzunoglu & Kip, 2014). Djafarova and Rushworth (2017) emphasize that higher 

credibility leads to greater persuasive power and influence over followers. 

Perceived loyalty refers to the degree to which consumers feel a connection and commitment to 

the influencer, fostering a sense of loyalty and repeated engagement (Djafarova & Rushworth, 

2017). This connection can enhance the influencer's impact by making their endorsements more 

personally relevant to the consumer (Horton & Wohl, 2016). High engagement indicates a strong 

connection, enhancing the influencer's impact on their followers (Dessart & Thomas, 2015). This 

study aims to provide a clearer understanding of the specific impacts of perceived expertise, 

perceived trustworthiness, and perceived loyalty on consumer behavior within the context of 

cosmetic multinational firms in Nairobi County. 

1.1.3 Consumer Behaviour 

Consumer behavior is an intricate process encompassing the decision-making activities and actions 

involved in the acquisition, use, and disposal of goods and services. It includes searching for 

information, evaluating alternatives, and ultimately deciding whether to purchase or reject a 

product (Valaskova & Kovacova, 2015). Kotler and Keller (2016) highlight that consumer 

behavior also encompasses the experiential and emotional responses of consumers during and after 

the consumption process. For this study, Njeru's (2019) definition of consumer behavior is 

adopted, focusing on the various activities, decision-making processes, and social influences that 

drive individuals and households to purchase products in Nairobi County. 

The study of consumer behavior, including purchase intention, perceived value, and satisfaction, 

is informed by established theories such as the Theory of Planned Behavior (Ajzen, 1991) and the 

Consumer Loyalty Framework (Oliver, 1999). These frameworks provide a basis for 



 

7 

 

understanding how attitudes, intentions, and actual behaviors are influenced by external factors 

like marketing communications. To systematically study consumer behavior, the following 

constructs will be measured in this study: purchase intention, perceived value, and satisfaction. 

Purchase intention refers to the likelihood that a consumer will buy a product based on SMI 

recommendations. This will be assessed using likert scale items adapted from prior research 

(Chetioui & Lebdaoui, 2020). Studies show that influencer endorsements can lead to increased 

brand awareness, positive brand perceptions, and higher purchase intentions (Chetioui & 

Lebdaoui, 2020). In Nairobi County, where digital adoption is growing, the impact of influencers 

on consumer behavior is becoming increasingly significant, particularly among younger 

demographics who are active social media users (Kamau, 2020). 

Perceived value refers to consumers' evaluation of the benefits and costs of a product as influenced 

by an SMI. This will be measured using scales developed by Sweeney and Soutar (2001). When 

elucidating consumer behavior, it is paramount to consider unique individual traits such as culture, 

lifestyle, buying habits, and the surrounding environment (Zak & Maria, 2019). Waqas et al. 

(2021) argue that these attributes contribute to the creation of a context, thus producing unique 

experiences while consuming purchased products. Consumers often engage in repetitive purchases 

of similar products not solely for optimal satisfaction but to reduce cognitive effort in responding 

to their needs (Zak & Maria, 2019). Cognitive behavior can be described as the dynamic interaction 

of affection and cognition, behavior, and the surrounding environment (Khasbulloh & Gede, 

2022). Given that consumption is behavioral, there is a need to explore the emotional and 

behavioral aspects attributed to an individual upon product consumption (Frah & Siew, 2021). 

Satisfaction refers to the post-purchase evaluation of a product endorsed by an SMI. The 

elucidation of consumer behavior should incorporate an examination of consumer purchasing 

decisions and their determinants. Mishra and Ashfaq (2023) argue that consumers behave in 

certain ways when consuming products, often purchasing those that provide maximum 

satisfaction. Thus, purchase decisions are based on satisfaction derived from financial investment 

and pleasant product service (Atika et al., 2019). Repurchase decisions are associated with 

products that elevate consumer satisfaction levels (Ahmed et al., 2023). Consumers are guided by 

consumption and experimentation in response to their needs. 



 

8 

 

Adopting Njeru's (2019) definition ensures that the unique cultural, economic, and social factors 

influencing Kenyan consumers are considered, making the study's findings more relevant and 

applicable to Nairobi County's market dynamics. This definition includes purchase intention, 

perceived value and satisfaction, offering a holistic view of consumer interactions with products. 

This comprehensive scope is essential for understanding the full impact of social media influencers 

(SMIs) on consumer behavior. 

1.1.4 Cosmetics Industry in Kenya  

The cosmetics industry is a vibrant and rapidly expanding sector. In Kenya, the cosmetics market 

is robust and growing, driven by rising disposable incomes and increasing awareness of beauty 

(KIPPRA, 2021). According to a report by (Euromonitor, 2023), the Kenyan cosmetics market 

was valued at approximately USD 200 million in 2022, with an annual growth rate of around 7%. 

This growth is expected to continue, driven by a youthful population and increasing demand for 

beauty products. The cosmetics industry in Kenya can be broadly categorized into skincare, 

haircare, makeup, and fragrances (Kenya Cosmetics Industry Report, 2022). Skincare and haircare 

products dominate the market, accounting for over 60% of total sales. Makeup and fragrances are 

also growing segments, particularly among urban consumers (Kenya Cosmetics Industry Report, 

2022). 

The cosmetics sector has recently emerged as Kenya's new investment hotspot, drawing significant 

interest from multinational firms through multimillion shilling acquisitions. Kenya is home to 226 

foreign multinational corporations (Mwaniki, 2020), with Nairobi, as a commercial hub, hosting 

several multinational cosmetic firms that actively engage influencers to promote their products. 

According to data from the Kenya Chamber of Commerce (2023), 75 of these 226 multinational 

corporations are in the cosmetics sector. This market is characterized by the presence of both 

international and local brands competing on product quality, pricing, and brand loyalty. Major 

international players include L'Oréal, Unilever, and Procter & Gamble, while notable local brands 

comprise SuzieBeauty, Haco Tiger Brands, and Ashleys Kenya Limited. 

The marketing strategies of these companies have increasingly leaned towards digital platforms, 

with companies like Estée Lauder reporting that 75% of their marketing budget is now allocated 

to digital marketing, heavily focusing on influencers (Cohen, 2018). However, this strategy comes 



 

9 

 

with its own set of challenges. Mwangi (2021) argues that brands face several obstacles when 

utilizing influencer marketing in the Kenyan cosmetic sector. Consumers are becoming 

increasingly savvy and can often discern when an endorsement is genuine or purely transactional. 

Additionally, Gikandi (2021) highlights that indiscriminate promotion by influencers can erode 

trust among their followers, which can extend to the brands they endorse, negatively impacting 

both brand reputation and sales. 

Brands must ensure that the influencer’s audience aligns with their target market and that the 

influencer’s values match the brand’s image in order to navigate these challenges (Gikandi, 2021). 

Moreover, the market is flooded with influencers, making it difficult to identify those who 

genuinely have a substantial impact and those who might have inflated follower counts through 

unethical means (Njenga, 2021). Therefore, while the cosmetics sector in Kenya holds significant 

potential, leveraging influencer marketing effectively requires a strategic and discerning approach. 

1.2 Statement of the Problem 

The vast and diverse landscape of social media platforms makes it difficult to identify influencers 

with the right demographics, interests, and engagement levels for a particular product or service 

(Smith, 2020). Some marketing campaigns have experienced negative outcomes that may be 

linked to the choice of social media influencers. Factors such as lack of authenticity, over-

promotion, and controversies surrounding influencers have led to brand damage and loss of 

consumer trust (Johnson et al., 2019). Moreover, little is known about the characteristics that make 

an influencer successful for a particular marketing campaign. Furthermore, there is a homogeneity 

of the social media influencer attributes on consumer behaviour in different sectors. 

Existing studies have employed various methodologies, including experimental designs and 

mixed-method approaches. For example, Chung, Lee, and Kim (2023) adopted an experimental 

research design, which may not fully capture the complexities of real-world social media 

interactions. Additionally, Gong and Holiday (2023) used a combination of primary data (via 

questionnaires) and secondary data (through document content analysis), which highlights the 

need for a more integrated approach. Contextually, studies have drawn respondents from 

continents whose macro-economic aspects are not common thus they had documented mixed 

results. For instance, in Western markets, social media influencer individuality and authenticity 



 

10 

 

are highly valued, whereas in many Asian markets, collectivism and social proof may play a more 

substantial role in consumer decision-making (Hofstede, 2001). There is variability in how studies 

have conceptualized and operationalized social media influence. Many have not adequately 

defined key attributes such as perceived expertise, perceived trustworthiness, and perceived 

loyalty. For instance, (Lou & Yuan, 2019) studied the effects of social media influencer marketing 

but did not distinctly define perceived expertise, instead lumping it with overall influencer 

credibility. 

This study adopts a comprehensive methodological approach by employing systematic sampling 

to ensure a representative sample of consumers. It focuses on primary data collection through 

structured questionnaires tailored to capture nuanced consumer behaviors and attitudes towards 

social media influencers in the cosmetics industry. This approach allows for a more in-depth 

understanding of the specific context of Nairobi County, enhancing the robustness and relevance 

of the findings (Creswell, 2014). By focusing on Nairobi County, this study provides empirical 

insights specific to an African urban setting, which is underrepresented in existing literature. It 

examines how local consumers respond to social media influencers within the cosmetics industry, 

offering valuable empirical data that reflects the unique socio-economic and cultural environment 

of Kenya (Pew Research Center, 2019).  

This study specifically contextualizes the impact of social media influencers within Nairobi 

County, considering local economic conditions, internet penetration rates, and cultural factors. By 

doing so, it provides contextually relevant insights that can inform more effective influencer 

marketing strategies tailored to the Kenyan market (Mbogo, 2020). This study clearly defines and 

operationalizes these attributes, drawing on established theoretical frameworks such as Social 

Influence Theory and Source Credibility Theory. By focusing on perceived expertise, 

trustworthiness, and loyalty, it provides a structured and theoretically grounded approach to 

understanding influencer effectiveness (Hovland & Weiss, 1951).  



 

11 

 

1.3 Objectives of the Study  

1.3.1 General Objective of the Study  

The general objective of the study was to examine the influence of social media influencers on 

consumer behaviour in cosmetic multinational firms in Nairobi Kenya.  

1.3.2 Specific Objectives of the Study  

The specific objectives of the study were:  

i) To examine the effect of perceived expertise of social media influencers on consumer 

behavior in cosmetic multinational firms in Nairobi Kenya.  

ii) To examine the effect of perceived trustworthiness of social media influencers on 

consumer behavior in cosmetic multinational firms in Nairobi Kenya.  

iii) To examine the effect of perceived loyalty of social media influencers on consumer 

behavior in cosmetic multinational firms in Nairobi Kenya. 

1.4 Research Questions  

The study was responding to the following research questions:  

(i) What is the effect of perceived expertise of social media influencers on consumer behavior 

in cosmetic multinational firms in Nairobi Kenya?  

(ii) What is the effect of perceived trustworthiness of social media influencers on consumer 

behavior in cosmetic multinational firms in Nairobi Kenya?  

(iii)What is the effect of perceived loyalty of social media influencers on consumer behavior 

in cosmetic multinational firms in Nairobi Kenya? 

 

1.5 Scope of the Study  

This study focused specifically on the influence of social media influencer attributes on the 

consumer behaviour in cosmetic multinational firms in Nairobi, Kenya. The study sourced primary 



 

12 

 

data amongst customers of respective cosmetic multinationals. Quantitative data approach was 

adopted, and questionnaires administered among the consumers of MNCs in the Cosmetic 

industry. The study was executed in Kenya, Nairobi County in April 2024. Even though there are 

several social media influencers attributes, the current study was limited to perceived expertise, 

perceived trustworthiness and perceived loyalty. The study adopted multiple regression analysis, 

correlation, exploratory factor analysis and descriptive statistics to analyze the data. 

1.6 Significance of the Study  

Accomplishment of the current empirical enquiry is advantageous to different stakeholders. Policy 

makers are guided on influencer selection criterion that would guide in investment of resources. 

This would be possible through understanding on how social influencer attributes resonate with 

consumers hence, companies would make informed decisions on their choice of social media 

influencers.  

The study will benefit management of cosmetic multinationals and they may rely on the findings 

to develop strategies that would enhance achievement of organization goals and objectives. 

Incorporation of social media influencers in marketing of cosmetic products would guide in 

development of customized marketing strategies in the respective target market. Further, since the 

business operation is dynamic there is a need for incorporation of non-traditional marketing 

strategies.   

An understanding of consumer behaviour would enable management in multinationals to develop 

strategies on integration of social media influencer in elucidation of purchasing decisions. An 

examination of the influence of influencer attributes would aid in identification of mechanisms for 

influencing consumer decisions in a digital market space. Further, the management would benefit 

in understanding the most optimal marketing strategy that would be customized to target 

audiences, hence increasing the effectiveness of firm marketing efforts.  

To theorists and empirical researchers, the study provides empirical and theoretical methods that 

can be applied in a wide variety of marketing. It serves as the basis of benchmarking future 

empirical studies to either refute or support existing findings and conclusions to be drawn on social 



 

13 

 

media influencers and consumer behaviours of marketing professionals in cosmetic multinational 

companies in Kenya.  

1.7 Chapter Summary  

The foregoing chapter has presented the background of the study, statement of the problem, 

objectives of the study, scope of the study and significance of the study. In the subsequent chapter 

theoretical and empirical literature review will be presented.  

 

 

 

 

 

 

 

 

 

 

 



 

14 

 

CHAPTER TWO 

LITERATURE REVIEW 

2.1 Introduction  

This chapter provides an extensive overview of existing literature related to the impact of social 

media influencer attributes on the consumer behaviour. The focus of this review is on the context 

of cosmetic multinational firms in Nairobi, Kenya. The chapter begins by discussing the theoretical 

foundations of the study. It then delves into the attributes of social media influencers that influence 

consumer behaviour. Finally, the literature review addresses the unique context of cosmetic 

multinational firms operating in Nairobi, Kenya, and their specific interactions with social media 

influencers. 

2.2 Theoretical Review  

The study was grounded on the Social Influence Theory and Source Credibility Theory to 

understand the impact of social media influencer attributes on consumer behavior in the context of 

cosmetic multinational firms in Nairobi County. These theories offer a comprehensive framework 

for analyzing how influencer characteristics such as credibility, attractiveness, relatability, and 

expertise influence consumer decisions and brand perceptions.  

2.2.1 Social Influence Theory 

Social Influence Theory explains how individuals change their attitudes, beliefs, and behaviors 

based on the social context and the influence of others (Kelman, 1958). This theory encompasses 

three processes: compliance, identification, and internalization, which are directly influenced by 

the perceived attributes of social media influencers (Kelman, 1958). 

Compliance occurs when individuals conform to the expectations of others to gain social approval 

or avoid social disapproval (Kelman, 1958). In the context of social media, followers might 

purchase products endorsed by influencers to align with perceived social norms or trends. For 

instance, Evans (2017) states that when followers perceive influencers as experts in cosmetics, 

they are more likely to internalize their recommendations, believing them to be knowledgeable 



 

15 

 

and credible sources. This internalization process leads consumers to adopt the influencer’s 

suggestions as their own beliefs and preferences. An influencer with a background in dermatology 

or extensive experience in makeup artistry can shape consumer behavior by providing expert 

advice on skincare routines or makeup application techniques, thereby influencing purchasing 

decisions (Lou & Yuan, 2019). 

Identification involves adopting behaviors and attitudes to establish or maintain a relationship with 

a person or group (Kelman, 1958). Followers often identify with influencers who they perceive as 

role models, leading to emulation of the influencer’s consumption behaviors. Trustworthiness 

enhances the likelihood of compliance and identification (Djafarova & Rushworth, 2017). 

Consumers are more likely to comply with recommendations from influencers they perceive as 

honest and reliable. Additionally, when consumers identify with a trustworthy influencer, they are 

more likely to adopt behaviors and attitudes that align with the influencer's (Lou & Yuan, 2019). 

An influencer who transparently shares both positive and negative aspects of cosmetic products 

builds trust, encouraging followers to follow their advice and fostering a deeper identification with 

their brand (Djafarova & Rushworth, 2017). 

Internalization happens when individuals accept influence because the content of the influence 

aligns with their value system (Kelman, 1958). When influencers are perceived as credible and 

knowledgeable, their endorsements can lead to genuine changes in consumer attitudes and 

behaviors (Lou & Yuan, 2019). Loyalty towards an influencer is often a result of strong 

identification. When followers feel a sense of loyalty, they are more likely to emulate the 

influencer’s behavior and purchase the products they endorse.  

This loyalty can lead to sustained consumer engagement and repeated purchases. Influencers who 

consistently engage with their audience and show genuine appreciation for their support cultivate 

a loyal following, which in turn influences consumer behavior and brand loyalty (Evans et al., 

2017). Social Influence Theory helps explain how influencers shape consumer behavior through 

compliance, identification, and internalization processes. This theory is critical in understanding 

the psychological mechanisms behind consumer responses to influencer endorsements. 



 

16 

 

2.2.2 Source Credibility Theory 

Source Credibility Theory asserts that the persuasiveness of a message is significantly influenced 

by the perceived credibility of the source (Hovland & Weiss, 1951). This theory is particularly 

pertinent to understanding how social media influencers impact consumer behavior through 

attributes such as expertise, trustworthiness, and attractiveness (Lou & Yuan, 2019). 

Expertise refers to the perceived knowledge, skill, or experience of the source in a particular 

domain (Hovland & Weiss, 1951). Influencers with recognized expertise in cosmetics, such as 

those with professional training in dermatology or extensive experience in makeup artistry, are 

seen as more credible and persuasive sources of information (Lou & Yuan, 2019). Expertise is a 

core component of source credibility. An influencer’s knowledge and skill in cosmetics enhance 

their credibility, making their endorsements more persuasive (De Veirman, 2017).  

Trustworthiness denotes the perceived honesty, integrity, and reliability of the source (Hovland, 

Janis, & Kelley, 1953). Influencers who are perceived as trustworthy can effectively shape 

consumer attitudes and behaviors. Influencers who transparently disclose sponsored content and 

provide balanced reviews, including potential drawbacks of products, are perceived as more 

trustworthy, leading to a stronger influence on consumer behavior (Djafarova & Rushworth, 2017). 

Honest and ethical behavior from influencers leads to higher trust and more significant influence 

on consumer behavior (Freberg et al., 2011). Influencers who are transparent about sponsored 

content and provide honest reviews are more likely to build trust with their audience, leading to 

higher consumer loyalty and engagement (Djafarova & Rushworth, 2017). 

Attractiveness includes both physical attractiveness and personal charm or likability (Hovland, 

Janis, & Kelley, 1953). Influencers who are attractive in these ways are often more persuasive. 

Influencers who are perceived as likable and relatable can capture and retain audience attention, 

enhancing their persuasive power (De Veirman et al. 2017). While not a primary focus, 

attractiveness enhances the persuasive power of an influencer. Attractive influencers build strong 

relationships with their audience, increasing their influence (Jin & Phua, 2014). 

Source Credibility Theory emphasizes the importance of an influencer's perceived credibility—

comprising expertise, trustworthiness, and attractiveness—in enhancing the effectiveness of 



 

17 

 

marketing messages (Lou & Yuan, 2019). Together, these theories provide a comprehensive 

framework for analyzing the impact of social media influencers on consumer behavior in the 

cosmetics market. 

2.3 Empirical Literature Review  

The conceptual literature defines Social Media Influencers (SMIs) as individuals with significant 

followings on social platforms who act as trusted authorities and trend-setters in specific industries 

(De Veirman, 2017). Empirical studies align with these definitions by demonstrating the 

significant influence SMIs have on consumer behavior. 

2.3.1 Perceived Expertise and Consumer Behaviour   

Serman and Sims (2020) explored how consumers' purchasing patterns are influenced by social 

media influencers. The study utilized cross sectional research design and gathered primary data 

through administration of 202 questionnaires. Univariate, exploratory factors and structural 

equation modeling examined the information. The study's findings showed that social 

attractiveness, sponsorship, perceived utility, trustworthiness, past experience, and subjective 

norms significantly influenced consumers' purchasing behavior. Further, social attractiveness 

displayed a statistically significant mediating effect on the attitude and intentions of consumers. 

The study has methodological and contextual gaps. Methodologically the study may have 

considered blending qualitative with quantitative data and there was need for documentation of 

diagnostic tests before modeling. Contextually, the study may have considered drawing data 

amongst management since they better understand the essence of social media influencers in 

relation to their brands.  

Chan (2022) examined social media influencers and the impact they have on consumer purchasing 

habits in the United Kingdom. The study asserted that brands collaborate with and exploit social 

media influencers to attain their target audiences, maximize revenue, enhance brand recognition, 

and enhance brand image. The study specifically aim,ed at identifying personality and content 

features which were elicited by SMIs, elucidate measures that can be deployed to increase 

consumer loyalty and investigate how demographic factors—such as gender, income, and 

education—affect the impact of SMIs. The study adopted descriptive research design and obtained 



 

18 

 

information by means of administration of questionnaires. The data were examined using multiple 

regression and univariate analysis. The study's findings showed that content features (closeness 

and interactivity) and personality features (perceived attractiveness and perceived credibility) had 

a statistically significant effect on customer loyalty and choice imitation. Further, there was a 

statistically significant mediating effect of customer loyalty on the effect of content features and 

personality features on choice imitation.  

Chopra and Gupta (2020) elucidated the impact of social media on consumer behaviour. 

Descriptive research design was used in the study to assess how customers obtain and analyze 

information before making a purchase as well as the role that social media plays at various phases 

of the decision-making process. Primary data was collected through administration of 

questionnaires. According to the study's findings, the majority of respondents preferred online 

shopping and spent an average of two hours each day on the internet. At least 60% of respondents 

followed brands in the social media and 61% perceived social media to have more superior 

information as compared to word of mouth. Moreover, 47% asserted that their brand vision was 

influenced by social media and 46% believed that their decision was influenced by social media 

through reviews and comments on given products. In addition, 88% believed that brand viability 

and attractiveness was anchored on social media feedback.  

Kumar and Vimal (2018) explored the impact of social media on consumer buying behaviour using 

technology acceptance model. The study hypothesized that purchase intention was influenced by 

perceived usefulness, perceived value and perceived risk. Descriptive research design was used in 

the study, and 265 postgraduate students were given questionnaires to complete in order to gather 

primary data. Univariate and multiple regression analysis analyzed the data. It was found that 

perceived usefulness, perceived value and perceived risk has a statistically significant effect on 

purchase intention whenever customers purchase intentions were influenced by social media. 

Through the study it was easy to stimulate organization products demand through use of social 

media.  

Hazar et al. (2022) examined the effect of social media attributes on the intention to buy. By using 

a descriptive research methodology, 439 questionnaires were sent out to participants in the study 

to gather primary data. Descriptive statistics and partial least squares models analyzed the data. 



 

19 

 

The study's findings showed that factors such as information quality, expertise, attractiveness, 

meaning transfer, and reliability significantly influenced consumers' inclinations to make 

purchases. It was concluded that company’s decision-makers should develop measures aimed at 

enhancing their sales and rely on social media influencers' insight since they can be relied on for 

customer engagement and influencing decision making.  

De Veirman (2017) and Ruiz-Gomez (2021) highlight the large followings and authority of 

influencers. Chan (2022) empirically confirms that perceived credibility, including influencer 

expertise and trustworthiness, affects customer loyalty and choice imitation. Mahmud (2018) notes 

the trust consumers place in influencers due to their expertise. This is reflected in Ahmed et al. 

(2023), who shows that source quality and argument credibility positively impact purchase 

intentions. 

2.3.2 Perceived Trustworthiness and Consumer Behaviour   

Mishra and Ashfaq (2023) evaluated the influencers impact on consumer buying behavior and 

purchasing patterns in India. Descriptive research design was utilized in the study and relied on 

primary data that was gathered using 358 questionnaires. Data was analyzed using descriptive and 

bivariate statistics. Results of the study depicted that the majority of social media users aged 

between 18 to 30 years, 49.2% spent at least three hours per day on social media, majority of those 

who visited social media they were seeking for entertainment or food. 51.7% reported that 

Instagram gives them to follow influencers. Concerning the type of influencers, 26% preferred 

product reviews, 17.6% preferred sponsored posts and 30% had preference for live streams. 

Majority 46% reported that they do not regret having undertaken purchases through social media 

and it was agreed that the majority of social media influencers had a significant impact in 

introduction of new products in the market.  

Zak and Maria (2019) examined how social media influencers affect consumers' decision-making. 

The study applied cross sectional research design and relied on primary data that was sourced 

through administration of questionnaires. Through descriptive statistical analysis results of the 

study indicated that personalities from the sports were the highly regarded influencers, 55% 

perceived some social media influencers to better than others, 30% purchased products since they 



 

20 

 

were promoted by social media influencers and 45% alluded that social media influencer has the 

highest influence on their decision making. The study may have considered carrying out inferential 

statistics such as correlation or regression so as to examine the causality between social media 

influencers and consumers decision making.  

Rifki and Arif (2023) established the impact of social media influencers on consumer behaviours 

and intentions in Indonesia. The study used quantitative research methods, administering 180 

questionnaires to gather data. The data were evaluated using multiple regression analysis and 

descriptive statistics. Results of the study indicate that there was a statistically significant influence 

of social media attraction, credibility, product sustainability and transfer of meaning on consumer 

attitudes and purchase intentions. Since the study applied multiple regression analysis there was a 

need for the study to report on diagnostic tests prior to modeling. This might have reduced the 

likelihood of making skewed judgments.  

Ahmed et al. (2023) studied the impact of social media influencers on consumers' wellbeing and 

purchase intention. The study applied experimental research design and issued 190 questionnaires 

among experimental and control groups. Structural equation modeling was applied to carry out 

multi group analysis. Results of the study depicted that source quality, argument credibility and 

kindliness had a positive statistically notable outcome on consumer intentions. Further, social 

media influencers had a positive statistically significant effect on purchase intentions. The study 

is methodological since the current study will apply multiple regression analysis to examine the 

influence of social media influencers on consumer behaviour.  

Frah and Siew (2021) explored social media influencers and their impact on consumers behaviour 

in Malaysia. The study adopted descriptive research design and sourced primary data through 

administration of 200 questionnaires among respondents who were selected through snowball 

sampling. To analyze the data, descriptive and inferential statistics were applied. Results of the 

study depicts that the majority 35% moderately agreed that they follow social media influencers 

for opinion. Majority reported that the social media influencers influenced their purchase decision. 

Further, there was a statistically significant influence of electronic words while making purchase 

decisions. Positive and negative social media comments had a significant effect on consumer 

decision making.  



 

21 

 

Mahmud (2018) and Zhou & Jin (2023) describe influencers as "special individuals" with 

noteworthy content and solid reputations. Empirical findings by Hazar et al. (2022) support this, 

showing that attributes like attractiveness and reliability significantly influence purchase 

intentions. Delbaere et al. (2021) and Lee & Watkins (2016) highlight the role of physical 

attractiveness. Serman and Sims (2020) find that social attractiveness mediates consumer attitudes 

and intentions, reinforcing attractiveness as a significant influencer attribute. Sakib et al. (2020) 

note the positive effect of influencers' ability to evoke strong feelings, which Frah and Siew (2021) 

support by showing that both positive and negative comments influence consumer decision-

making. 

2.3.3 Perceived Loyalty and Consumer Behaviour   

The impact of social media influencers on brand image, self-concept, and purchase intention was 

investigated by Atika et al. (2019). The study used descriptive methods and used 219 

questionnaires to gather primary data. Data analysis was done using structural equation modeling 

and descriptive modeling. The study's findings showed that while there was a positive significant 

relationship between social media influencer and brand image, there was an inverse significant 

influence of both self-concept and social media influencer on purchase intention. The study may 

have considered documenting exploratory factor analysis prior to structural equation modelling.  

Zehra and Julian (2022) studied the social credibility theory through social media posting during 

COVID 19. Using a cross-sectional research methodology, 449 questionnaires were distributed to 

gather primary data for the study. For data analysis, structural equation modeling and descriptive 

statistics were used. The study's findings showed that credibility was positively impacted by 

reputation and trustworthiness, but expertise and promotional incentives had no discernible impact 

on credibility. Additionally, the study showed that the relationship between loyalty and credibility 

was mediated by the sharing of unverified information. Since the study was carried out in the 

service industry unlike multinationals that operate in different sectors there is a need for 

consideration of drawing data from alternative sectors so as to respond to existing contextual gaps.  

Habibi and Mevdi (2022) assessed the mediating role of electronic word-of-mouth marketing in 

the Iranian Jabama corporation in order to determine the impact of perceived value on customer 



 

22 

 

value. The study used a descriptive research approach and distributed 200 questionnaires to gather 

primary data. Descriptive and structural equation modeling analyzed the data. Results of the study 

depicted that perceived word of mouth has a positive statistically significant effect on customer 

value. Moreover, the impact of perceived value on customer value was positively mediated by 

electronic word-of-mouth marketing.  

The impact of perceived risk and perceived value on customer satisfaction and loyalty among 

Bukalap users was investigated by Khasbulloh and Gede (2022). 108 questionnaires were 

distributed in order to collect primary data for the descriptive study design. Structural equation 

modeling analyzed the data. Results of the study depicted that perceived risk has an inverse effect 

while perceived value has positive statistically significant effect on customer loyalty and customer 

satisfaction. There is a population gap because the study never took social media influencers into 

account.  

Anchal (2023) investigated how social media affects the preferences and behavior of consumers. 

The study used a descriptive research methodology and used questionnaires to collect primary 

data. Descriptive and inferential statistics analyzed the data. 88% reported that social media 

increased brand exposure amongst those aged between 18 to 24 years. Social media enhanced 

social proof since 70% of respondents agreed that they trusted internet review and evaluations 

from social media comments. Instagram availed an opportunity for evaluation of likes and 

comments of respective brands. The dominant social media platform was reported to be you-tube 

(73%) followed by face-book that accounted for 69% and Instagram at 40%. Further, 48% of 

respondents reported that they perceived companies that responded to enquiries in the social media 

as trustworthy and 79% argued that their purchasing decision was dependent on user generated 

contents. 80% reported that they can make purchase decisions through personalized experiences. 

The study may have considered incorporating inferential statistics so as to minimize the likelihood 

of making skewed judgments.  

 

 



 

23 

 

2.4 Research Gaps  

Table 2.1 Summary of Literature Gaps 

Author(s) Title Findings Research Gaps 

Zhou & Jin (2023) 

"The Efficacy of Social 

Media Influencers: Metrics 

and Measurements" 

Using brand influencers 

increases view-through rates 

by 193%, and 35% of users 

learn about products from 

influencers. 

The study conducted a need for longitudinal studies 

on the sustained impact of influencers on consumer 

behavior and brand loyalty. The current study 

examines the specific impacts of perceived expertise, 

trustworthiness, and loyalty of influencers on 

consumer behavior, aiming to provide a more detailed 

understanding of these attributes in the context of the 

cosmetics industry. 

Simiyu (2021) 

"Influence of Social Media 

on Consumer Behavior" 

SMIs significantly influence 

consumer behavior. 

The study provided a detailed analysis of the 

underlying psychological mechanisms of influencer 

impact on consumer decision-making. 

The current study focuses on the psychological 

constructs such as perceived expertise, 

trustworthiness, and loyalty, and their effects on 

consumer purchase intention, perceived value, and 

satisfaction. 

Schouten & 

Verspaget (2020) 

"Under the Influence: The 

Downsides of Influencer 

Marketing" 

Influencer fatigue and 

authenticity crisis are 

emerging issues. 

The study investigated long-term impacts of 

influencer fatigue and strategies to mitigate negative 

perceptions. 

The current study includes authenticity indirectly by 

focusing on perceived trustworthiness and loyalty, 

which contribute to an influencer's overall 

authenticity in the eyes of consumers. 

Audrezet et al. 

(2020) 

"Authenticity 

under Threat: When Social 

Media Influencers Need to 

Consumers are more 

influenced by genuine and 

relatable content. 

The study emphasized the need for empirical studies 

on how influencers can maintain authenticity while 

engaging in commercial activities. The current study 

distinctly examines perceived expertise, 



 

24 

 

Author(s) Title Findings Research Gaps 

Go Beyond Self-

presentation" 

trustworthiness, and loyalty as separate attributes to 

understand their individual and combined effects on 

consumer behavior. 

Djafarova & 

Rushworth (2017) 

"Exploring the Credibility 

of Social Media Influencers 

and Its Impact on 

Consumers’ Purchase 

Intentions" 

Credibility impacts followers' 

purchase intentions, but often 

conflates trustworthiness and 

expertise. 

The study only gives a differentiation between the 

impacts of trustworthiness and expertise on consumer 

behavior. The current study distinctly examines 

perceived expertise, trustworthiness, and loyalty as 

separate attributes to understand their individual and 

combined effects on consumer behavior. 

Lee & Watkins 

(2016) 

"YouTube Vlogs as a 

Source of Information on 

Beauty and Lifestyle 

Products" 

Physical beauty of influencers 

encourages consumer 

behavior towards endorsed 

brands. 

The study investigated how influencers' non-physical 

attributes affect consumer behavior in the beauty and 

lifestyle sectors. The current study includes both 

physical and non-physical attributes (expertise, 

trustworthiness, loyalty) to provide a comprehensive 

view of how these characteristics influence consumer 

behavior in the cosmetics industry. 

 

2.5 Conceptual Framework  

This diagram illustrates the relationship between the variables being studied. The current study 

postulates that social media influencer attributes influence consumer behaviour. Social media 

influencer attributes are conceptualized as perceived expertise, perceived trustworthiness, 

perceived loyalty and unverified information sharing. Consumer behaviour was operationalized as 

perception of value, purchase intention and satisfaction. Perceived expertise was operationalized 

as authoritative, informative and knowledgeable. Perceived trustworthiness was measured as 

social proof, endorsement and attractiveness. Perceived loyalty will be measured as multiple 

products like, customer churn and customer lifetime value. The hypothesized relationship is as 

shown in Figure 2.1.  



 

25 

 

     Independent Variables                                                                 Dependent Variable 

 

 

 

 

 

 

 

 

 

 

Figure 2.1 Conceptual Framework 

 

The aim of the study was to determine the relationship between social media influencer attributes 

and consumer behaviour in cosmetic multinational firms in Nairobi County. The study variables 

were operationalized as shown below. 

      

Perceived Expertise  

 

• Authoritative  

 

• Informative  

• Knowledgeable 

           Perceived Trustworthiness 

 

• Social Proof  

 

• Endorsement 

 

• Attractive 

Consumer Behaviour  

• Perception of value  

• Purchase intention             

• Satisfaction  

Perceived Loyalty  

• Multiple products like  

 

• Customer churn 

 

• Customer lifetime value  



 

26 

 

 

 

Table 2.2 Operationalization of variables 

2.6 Chapter Summary  

The foregoing chapter presents the empirical and theoretical review. The study was anchored on 

social influence and source credibility theory. The study conceptualized that social media 

influencer attributes affects consumer behaviour.  

Variable Type Indicators Measurement Source 

Perceived Expertise Independent 

variable 

• Knowledge 

• Skill level 

• Qualifications 

• Experience 

5-point Likert 

Quantitative 

analysis 

Lou & Yuan (2019); 

Hovland et al. 

(1953) 

Perceived 

Trustworthiness 

Independent 

variable 

• Honesty 

• Reliability 

• Integrity 

 

5-point Likert 

Quantitative 

analysis 

Djafarova & 

Rushworth (2017); 

Hovland et al. 

(1953) 

Perceived Loyalty Independent 

variable 

• Repeated interactions 

• Sustained interest 

• Consistent support 

5-point Likert 

Quantitative 

analysis 

Evans et al. (2017); 

Oliver (1999) 

Consumer 

behaviour 

Depenedent 

Variable 

• Perception of value 

• Purchase Intention 

• Satisfaction 

5-point Likert 

Quantitative 

analysis 

Kotler & Keller 

(2016); Schiffman & 

Kanuk (2000) 

 



 

27 

 

CHAPTER THREE 

RESEARCH METHODOLOGY 

3.1 Introduction  

A thorough examination of research design, study area, population, sample, data collection 

techniques and procedures, data validity and reliability, data analysis and presentation, and ethical 

considerations are presented in this chapter.  

3.2 Research Philosophy  

Research philosophy is deemed to be a framework stipulating how data can be gathered and 

processed in response to research questions (Saunders, Lewis & Thornhill, 2014). It can be 

perceived to be the source of the methodological foundation that avails minimize variations and 

hastens understanding of empirical issues under examination (Sekeran & Bougie, 2013). 

According to Creswell (2014), there are four different research philosophies, namely positivism, 

constructivism, critical realism and pragmatism. Positivism research philosophy alludes that 

reality will always be stable, observable and be easily described in an objective point of view 

(Saunders et al., 2014).  

It is mostly applicable when issues under examination can be observed repetitively and notable 

differences avails information of elements under examination (Sekaran & Bougie, 2013). Though, 

applicable is social sciences it is not void of short comings since it relies on past predictions, but 

some may have had unmeasurable attributes. Interpretivism argues that to fully comprehend the 

research under interest then there is need for subjective interpretation. In this situation, the study 

ought to be carried out in nature settings where research implementers have no capacity to 

influence attributes under examination (Bryman & Bell, 2015). The current study adopted 

positivism research philosophy. Positivism was applied since the researcher aimed at collecting 

quantitative data to respond to research questions.   



 

28 

 

3.3 Research Design  

The step-by-step guideline stipulating how a research study is undertaken to attain study objectives 

is known as research design (Sekaran & Bougie, 2013; Saunders et al., 2014). Research design’s 

choice is pegged on research problem, resources endowment, study motivation and research ethics 

compliance (Sekaran & Bougie, 2013). This study adopted descriptive research design. The basic 

tenet of descriptive research is description of research phenomena in aspects such as what, where 

and how (Sekaran & Bougie, 2013). Descriptive research examined the events that lead to 

inferences and relationships between variables. Hence, it was vital in this study to identify the 

relationship between the research variables. Further, the design allowed for the examination of the 

study variables in its natural state with minimal interference. 

3.4 Target Population  

The study population refers to all constituents of any clearly described group of people, events, or 

objects who are the focus of an investigation (Patten & Newhart, 2017). The focus of this research 

was 75 multinational cosmetic companies within Nairobi who are registered with the Kenya 

Chamber of Commerce (KNCCI, 2023). The unit of observation for this study consisted of the 75 

multinational cosmetic companies operating in Nairobi (See Appendix VII). These companies 

have been selected based on their significant market presence and influence within the Kenyan 

cosmetic industry. The focus on multinational companies is motivated by their extensive reach and 

diverse consumer base, which provide a robust framework for understanding consumer behavior 

and preferences in the cosmetic sector. While the population of interest is the multinational 

cosmetic firms, data collection from customers allowed the researcher to gain insights into 

consumer behavior, perceptions, and preferences.  

Customers are the end-users of cosmetic products, and their opinions and experiences are valuable 

for understanding market trends and making strategic decisions (Lou & Yuan, 2019). Therefore, 

the unit of analysis in this study was comprised of 5 consumers from each of the 75 selected 

multinational cosmetic companies, resulting in a total of 375 individual consumers. While it may 

not have been feasible to survey all customers of each firm due to time and resource constraints, 

selecting 5 customers from each firm provided a representative sample that reflects the overall 

customer base. This sample size strikes a balance between obtaining sufficient data for analysis 



 

29 

 

and managing practical limitations. These consumers represent a diverse demographic cross-

section, including variations in age, gender, socio-economic status. The selection of these 

consumers aims to capture a broad perspective on the consumption patterns, brand loyalty, and 

satisfaction levels associated with multinational cosmetic products. 

3.5 Sample Size and Sampling Technique 

A sampling design is a process of developing a plan in which elements of interest are selected 

(Bloomfield & Fisher, 2019). The design outlines the probability of developing a reliable sample 

that is related to a population (Leavy, 2017). The sampling design consists of the sampling 

technique, the sample frame, and the sample size. The purpose of the sampling technique is to 

outline a process that was used to collect data that enables the researcher to make conclusions 

about the research population (Ghauri, Grønhaug, & Strange, 2020). The study was based on 

descriptive research design and positivism research philosophy. In this study the target population 

were cosmetic multinational firms operating in Nairobi. Systematic sampling will be used to 

select the consumers from each company. This technique ensures that the sample is spread evenly 

across the entire population, reducing the likelihood of bias. Therefore, each member of the 

population has a known chance of participating in the study (Creswell & Clark, 2017).  

 

According to Leavy (2017), when choosing a sample frame, one must consider the composition 

of the sample, the representativeness of the sample, and the size of the sample. The sample frame 

for this study was limited to 5 consumers of each of the 75 cosmetic multinational companies 

operating in Nairobi County. Selecting 5 customers from each firm was logistically feasible and 

manageable within the scope of the study. It allowed the researcher to collect a substantial amount 

of data while ensuring data quality and thoroughness in the analysis process. The sample size 

should be based on a general representation of a sample population (Bloomfield & Fisher, 2019). 

It is vital to select a sample size that provides accurate results and identifies differences. The 

selected sample size should give adequate power to the results of the study (Ghauri, Grønhaug, 

& Strange, 2020). The sample size for the research was computed as below: 

Sample size per company = 5 consumers Number of companies = 75 

Total sample size = Sample size per company × Number of companies  



 

30 

 

= 5 consumers/company × 75 companies = 375 consumers 

So, the sample size for this study is 375 consumers. 

3.6 Data Collection Instruments  

According to Saunders et al. (2014), instruments are the tools that was used to collect data and the 

methods by which they were created. The strength of questionnaires in social science studies is 

based on its degrees to sustain confidentiality, save time and is rarely faced with distribution 

logistical challenges because one can administer them online (Sekaran & Bougie, 2013; Saunders 

et al., 2014). Furthermore, Crewell (2014) supported questionnaires because they portray 

convenience for both researcher and respondent, therefore the researcher has high chances of 

gathering truth information and knowledge. Secondly, it is the best research tool for those 

researchers who are faced by budgetary and time constraints.  

3.7 Data Collection Procedures  

A research procedure outlines the process that was followed to collect information about a research 

study (Bloomfield & Fisher, 2019). The research procedure of this study includes the development 

of a questionnaire that was reviewed and approved for distribution by the supervisor. This tool was 

designed to capture relevant data related to consumer behavior, perceptions, and preferences 

regarding multinational cosmetic products. Further, the study obtained any relevant approval from 

Strathmore University Ethic review committee, and it was used to seek research permit from 

National Commission for Science Technology and Innovation (NACOSTI). Additionally, hiring 

and training a research assistant was carried out. To achieve optimal efficiency and response rate, 

either the drop and pick method or self-administered questionnaires was used.  

The research agent was stationed at selected multinational cosmetic outlets within shopping malls 

across Nairobi County and the CBD. These outlets were chosen based on their popularity and 

representation of the target multinational cosmetic firms. As customers exited these outlets, the 

research agent identified potential respondents who had made purchases or visited the outlets. 

Upon identifying potential respondents, the research agent approached them and administered the 

questionnaire. The respondents were asked to complete the questionnaire on-site, providing their 



 

31 

 

responses to the questions included. The research agent played a crucial role in facilitating the data 

collection process. He aided the respondents, clarifying any questions or concerns they had about 

the questionnaire. Additionally, the agents ensured that respondents understood the purpose of the 

study and the confidentiality of their responses. After the respondents completed the questionnaire, 

they returned the completed documents to the research agent. The research agent then collected 

the completed questionnaires and ensured their safekeeping until the data collection process was 

concluded.  

Throughout the data collection process, quality control measures were implemented to ensure the 

accuracy and reliability of the gathered data. The research agent monitored the completeness and 

consistency of the responses, addressing any discrepancies or issues that arose during the 

administration of the questionnaire. 

3.8 Research Quality  

The research conducted a pre-test of the study instrument with 10% of the study sample of 39 

consumers of cosmetic products that were not included in the final research. The pilot test was 

vital to examining the reliability tests of the research instrument. Further, through the pre-test, the 

study tests for the validity of the instrument using the content validity and construct validity 

approach. This sample was appropriate since Sekaran and Bougie (2013) averted that a sample of 

ten percent of the actual study sample is appropriate for piloting the research instruments. 

3.8.1 Reliability of Instrument  

Reliability is the extent to which a measure consistently produces similar results over time, in 

running several tests at different instances. This is also viewed as the consistency and stability of 

a measurement over time (Drost, 2011). Test-retest, parallel reliability, alternative forms, inter-

rater, split-half, and internal consistency are among the tests used to determine reliability 

(Trochim, 2006). While test-retest includes giving the same exam twice, parallel reliability tests 

the same construct using two different copies of the assessment.  Internal consistency measures 

the extent to which various assessments that probe the same construct yield comparable results, 

while inter-rater reliability measures the degree of agreement between raters (Lameck, 2013). The 

Cronbach Alpha coefficient, which has a range of 0 to 1, will be used to assess reliability. If it is 



 

32 

 

greater than 0.7, the research instrument is considered reliable. The reliability results in Table 3.1 

depicts that the research tool was reliable with a minimum Cronbach Alpha of 0.772 and maximum 

of 0.812.  

Table 3.1 Reliability Analysis  

Variables Number of items Cronbach’s Alpha Comments 

Perceived expertise 10 0.812 Accepted 

Perceived trustworthiness  8 0.804 Accepted 

Perceived loyalty  7 0.772 Accepted 

Consumer behaviour  7 0.779 Accepted 

3.8.2 Validity of Instrument  

According to Drost (2011), research validity refers to how useful a test is in measuring the things 

it was designed to evaluate, which can include internal, external, construct, or discriminant 

variables. According to Zohrabi (2013), internal validity refers to a research instrument's capacity 

to measure the things it is intended to measure, whereas external validity is the ability of data to 

be generalized. The degree to which scores on one scale correlate with those on other scales 

intended to measure the same construct is known as convergent validity. According to Lameck 

(2013), discriminant validity presupposes that the scale's scores do not correlate with those of other 

measures meant to measure distinct constructs. Validity in this study was assessed by determining 

whether the findings can be applied to all Kenyan county governments (Zohrabi, 2013). Supervisor 

input and remarks were taken into account throughout evaluation.  

3.9 Data Analysis and Presentation  

Data analysis is the process of inspecting information for decision-making to draw a conclusion 

on the research questions (Bloomfield & Fisher, 2019). This study employed a quantitative data 

analysis method. Data collected using questionnaires was counter checked for completeness, 

coded and entered into Microsoft Access. It was cleaned and then exported to SPSS version 24. 

The study used both descriptive and inferential analysis. The descriptive analysis included 

frequency, percentages, a mean, and standard deviation, which analyses the participant's 



 

33 

 

characteristics and dominant patterns generated from the data. The study applied the following 

diagnostic checks, normality tests, collinearity tests and heteroscedasticity tests. The inferential 

analysis examined the use of regression analysis and correlation tests in determining the 

magnitude of the relationship and direction of the association, respectively. The analysed research 

data was presented using graphs, and tables. Multiple regression model used in this study is given 

as; 

Y= α + β1X1 + β2X2 + β3X3 + ε………………….………………...Equation 3.1 

Y= Consumer behaviour; 𝛼 = constant; β1… β3= the slope; 𝑥1=Perceived expertise; 𝑥2= Perceived 

trustworthiness; 𝑥3= Perceived loyalty; and ε = error term.  

3.10 Diagnostic Tests  

3.10.1 Multicollinearity Test  

There is a strong link between independent variables and multicollinearity. This research employed 

Variance Inflation Factor (VIF) values to assess multicollinearity. To test for multicollinearity, 

tolerance limits was also employed. Baltagi (2005) asserts that multicollinearity exists when the 

variance in functions (VIF) exceeds 10 and when the tolerance limit is less than 0.1. 

3.10.2 Heteroskedasticity Test  

Homoscedasticity is the fourth presumption made by the multiple linear regression analysis.  

Heteroscedasticity was tested using the Breusch Pagan Test.  To determine whether there are 

significant differences between the samples, the multiple linear regression test divides the data into 

high and low values.  

3.10.3 Normality Test 

The normalcy test was conducted using the histogram.  The data may not be normally distributed 

if the p value is less than 0.05; otherwise, it is not normal. For the histogram normally distributed 

ought to have a mean of zero and standard deviation of 1. Normally distributed data ought to be 

bell shaped.  



 

34 

 

3.11 Diagnostics Tests Results  

Regression diagnostics tests were carried out prior to regression modelling.  

3.11.1 Multicollinearity Test  

Multicollinearity test was carried out using variance inflation factors and tolerance limits. Results 

in Table 3.2 depicts the highest VIF was 3.64 and the least 1.87 a clear indication there was no 

multicollinearity.  

Table 3.2 Multicollinearity Test  

  Collinearity Statistics 

  Tolerance VIF 

Perceived Expertise 0.34 2.91 

Perceived Trustworthiness 0.28 3.64 

Perceived Loyalty 0.53 1.87 

3.11.2 Heteroskedasticity Test  

Heteroskedasticity was carried out to examine the uniformity of variance among the error term. 

Results in Table 3.3 depicts that there was no enough evidence to warrant rejection of null 

hypothesis since the p value was greater than 0.05. Thus, it can be concluded that there was 

uniformity of variance of the error term.  

Table 3.3 Heteroskedasticity Test  

  F statistics  P value  

Breusch Pagan Test   32.54 0.0000 

 

3.11.3 Normality Test  

Pictorial presentation in Figure 3.1 depicts that the error term was normally distributed with mean 

0 and standard deviation of 1. Thus, there was no need to carry out data transformation prior to 

modelling.  



 

35 

 

 

 

 

Figure 3.1 Normality Test  

 

3.12 Ethical Considerations  

In social science research ethics refers to a code of conduct or expected societal norm of behaviour 

while conducting research (Sekaran & Bougie, 2010). Moreover, Saunders et al., (2014) averred 

that ethical issues cannot be ignored in all research stages. Respondents ought to be allowed to 

withdraw from the study at any given time, their confidentiality must be protected, none of them 

ought to be exposed to stressful situations upon participation in data collection, respondents ought 

to be assured that their data will not be analyzed to contribute to other issues beyond academic 

objectives. Although different stakeholders are expected to benefit from the study, they ought to 

draw their benefits from processed data and not its raw form. In this study approval was sought 

both from Strathmore University School Ethical Review committee and the NACOSTI. These 

research approvals and permits were presented to human resources departments of respective 

cosmetic multinationals.  



 

36 

 

3.13 Chapter Summary  

The foregoing chapter has presented research philosophy, research design, sample size and 

sampling procedures, data collection instruments and data collection procedures and data 

processing and analysis as well ethical considerations.  

 

 

 

 

 

 

 

 

 

 

 

 

 



 

37 

 

CHAPTER FOUR 

FINDINGS AND INTERPRETATIONS 

4.1 Introduction  

In this section data analysis, findings and interpretations will be presented. The study adopted 

descriptive statistics that include frequency, percentages, mean and standard deviation. 

Spearman’s rank correlation and multiple regression analysis. Before regression analysis 

diagnostic tests are presented on multicollinearity, heteroskedasticity and normality.  

4.2 Response Rate  

The response rate for the current study was 81% whereby out of 375 questionnaires administered 

303 were correctly filled and returned. The response rate was attributed to adoption of drop and 

pick and self-administration of questionnaires with aid of research assistant upon customers as 

their exited from different multinationals outlets dealing with cosmetics.  

Table 4.1 Response Rate  

Questionnaires Frequency Percentage 

Issued and completely filled  

 

303 81 

Not returned  72 19 

Total 375 100 

 

4.3 Demographic Information  

Study findings on the background of respondents was summarized using frequencies and 

percentages as tabulated in Table 4.2. It was found that 69.3% of the respondents were female 

while 30.7% were male a clear indication that cosmetic products were more attractive to females. 

On the age distribution 11.9% aged 30 years and below, 29% aged between 31 to 34 years and 

48.5% aged between 35 to 40 years a clear indication that most of those who aged below 40 years 



 

38 

 

had preference for multinational cosmetic products. Concerning the level of education attained 

40.9% had diploma qualifications, 25.4% had bachelor’s degree and 28.7% had certificate. On the 

years of experience with the products 46.2% had used the products for at most 10 years, followed 

by 28.1% who had used them for a period of 1 to 5 years and 19.5% had used for less than 1 year. 

This depicts that there were those who had used cosmetic products for a longer period while others 

were new entrants.  

Table 4.2 Demographic Information  

    N % 

Gender Male 93 30.7% 

  Female 210 69.3% 

Age Below 30 years 36 11.9% 

  31 - 34 years 88 29.0% 

  35 – 40 years 147 48.5% 

  41 – 44 years 30 9.9% 

  45 – 50 years 2 0.7% 

Highest level of education Certificate 87 28.7% 

  Diploma 124 40.9% 

  Bachelor’s Degree 77 25.4% 

  Post graduate 15 5.0% 

Years of experience Less than 1 year 59 19.5% 

  1-5 years 85 28.1% 

  6-10 years 140 46.2% 

  Over 10 years 19 6.3% 

 

4.4 Descriptive Statistics on Perceived Expertise  

The first objective of the study examined the influence of perceived expertise on consumer 

beha