Can Social Media-Driven Trust and Customer positive Emotions with a Car Dealership Increase Brand Love?

1. Introduction

Brand love is a reflection of the emotional responses of satisfied consumers. This ‘love-like’ attachment stems from a deep satisfaction with the brand. Researchers have argued that positive emotions play a strong role in brand love [5]. Positive emotions are associated with consumers’ willingness to invest time, effort and resources into the brands they love [1]. In the highly competitive automobile business environment, every car dealer tries hard to make their customers happy. There is evidence in the previous literature that the customer´s positive emotions with service facilities is the key to increasing his/her brand love [6].

Although brand love is acknowledged as being an important construct in consumer–brand relationships [1], only limited research has taken place into the combined influence of the cognitive and affective drivers of brand love, and its consequences for these relationships. Offering automobile customers high-quality services, based on their needs and expectations, is increasingly seen as a condition for developing successful marketing strategies that satisfy and retain customers. The novelty of the present study lies in its exploration of the links between the dimensions of perceived service quality, brand love and its effects on behavioural intentions.

2. Literature Review and Hypotheses Development

The cognition-affection-conation (C-A-C) framework [10] is based on the sequential linkages of cognition-affection and conation in human decision-making processes: cognitive factors lead to affective outcomes that consequently impact on conation, which ultimately stimulates actual behaviours. Cognition relates to a person’s thoughts, beliefs and values regarding an object; affect relates to a person’s feelings or emotions felt towards the object; and conation represents the development of the individual´s behavioural intentions and actual behaviours towards that object. The C-A-C framework illustrates how consumers translate their experiences, learning and perceptions into behaviours through affect, which is precisely in line with our objective of understanding the behaviours of car brand loving clients. Based on literature review, we propose that:

Hypothesis 1 (H1).

Positive emotions evoked by the car purchasing experience at the dealership increases brand love for the car brand.

Hypothesis 2 (

H2). Trust in the car dealership influences car brand love:

Hypothesis 3 (H3).

The perceived service quality provided by the car dealership influences the customer´s trust in the car dealership.

Hypothesis 4 (H4).

The perceived service quality provided by the car dealership evokes customer positive emotions.

3. Methodology

The empirical study took place March-October 2023, in collaboration with a leading automobile dealership. The quantitative study was based on data obtained from an online questionnaire.

We operationalised the study constructs using multi-item scales measured mostly by five-point Likert scales, ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5).

Car brand love was measured using a 10-item scale adapted from [2]. The respondents´ self-reported trust was measured using a 3-item scale adapted from [36]. Perceived service quality (PSQ) was measured as a formative second-order construct. We used three dimensions to capture PSQ, that is “personnel”, “price” and “operational service performance”.

4. Results

4.1. Psychometric Properties of the Measurement Model

We established the theoretical structure of the model using the “repeated indicator approach” [39] and partial least squares (PLS). We opted for PLS-SEM as the estimation method due to the complexity of the model, which involves numerous constructs (including a second-order formative construct), indicators and model relationships [40].

The parameter estimation was carried out using Smart-PLS 4.0 [41], and we performed bootstrapping with 5,000 samples to determine the significance of the parameters (see Table 2). The composite reliability of the constructs exceeded the recommended threshold of 0.60 [42]. In addition, the average variance extracted (AVE) for all constructs surpassed the 0.50 threshold [43]. As an indicator of convergent validity, the confirmatory factor analysis (CFA) demonstrated that all items were significantly related (p < 0.01) to their respective factors [44].

Table 1 displays also the mean factor loadings for all the dimensions, each of which exceeds the criteria for convergent validity. Finally, the results show that the model has no collinearity problems in the reflective-formative second-order construct of perceived service quality.

Table 1. Reliability and convergent validity of the measurement instrument.

First order factors	Item	Std loadings	t value (bootstrapping)	rho_A	CR	AVE
PSQ Personnel (PER)	PER1	.93	130.40**	.97	.98	.88
	PER2	.94	158.42**
	PER3	.94	173.62**
	PER4	.95	167.98**
	PER5	.93	153.46**
PSQ Price (PRI)	PRI1	.95	171.37**	.97	.98	.88
	PRI2	.93	98.73**
	PRI3	.95	132.73**
	PRI4	.94	145.13**
	PRI5	.93	139.50**
PSQ Operational service performance (OSP)	OSP1	.90	80.12**	.97	.98	.86
	OSP2	.94	120.20**
	OSP3	.93	99.78**
	OSP4	.95	189.26**
	OSP5	.95	184.06**
	OSP6	.89	64.49**
	OSP7	.93	99.65**
Car dealership trust (TRU)	TRU1	.96	203.12**	.95	.97	.90
	TRU2	.95	168.17**
	TRU3	.95	165.75**
Car brand love (LOV)	LOV1	.92	129.47**	.98	.98	.85
	LOV2	.91	69.69**
	LOV3	.93	140.14**
	LOV4	.88	54.4**
	LOV5	.94	152.08**
	LOV6	.92	100.07**
	LOV7	.93	138.51**
	LOV8	.93	151.48**
	LOV9	.93	135.56**
	LOV10	.92	94.97**
Second-order factor				VIF
Perceived Service Quality (PSQ)	Personnel	.48	20.732**	2.24
	Price	.31	12.484**	2.58
	Oper. service performance	.32	10.567**	3.56

N/A: Single item latent variable (Not applicable). VIF: Variance Inflation Factor. rho_A=Dijkstra-Henseler rho; CR=Composite reliability; AVE=Average variance extracted. **p<0.0

Table 1. Reliability and convergent validity of the measurement instrument.

First order factors	Item	Std loadings	t value (bootstrapping)	rho_A	CR	AVE
PSQ Personnel (PER)	PER1	.93	130.40**	.97	.98	.88
	PER2	.94	158.42**
	PER3	.94	173.62**
	PER4	.95	167.98**
	PER5	.93	153.46**
PSQ Price (PRI)	PRI1	.95	171.37**	.97	.98	.88
	PRI2	.93	98.73**
	PRI3	.95	132.73**
	PRI4	.94	145.13**
	PRI5	.93	139.50**
PSQ Operational service performance (OSP)	OSP1	.90	80.12**	.97	.98	.86
	OSP2	.94	120.20**
	OSP3	.93	99.78**
	OSP4	.95	189.26**
	OSP5	.95	184.06**
	OSP6	.89	64.49**
	OSP7	.93	99.65**
Car dealership trust (TRU)	TRU1	.96	203.12**	.95	.97	.90
	TRU2	.95	168.17**
	TRU3	.95	165.75**
Car brand love (LOV)	LOV1	.92	129.47**	.98	.98	.85
	LOV2	.91	69.69**
	LOV3	.93	140.14**
	LOV4	.88	54.4**
	LOV5	.94	152.08**
	LOV6	.92	100.07**
	LOV7	.93	138.51**
	LOV8	.93	151.48**
	LOV9	.93	135.56**
	LOV10	.92	94.97**
Second-order factor				VIF
Perceived Service Quality (PSQ)	Personnel	.48	20.732**	2.24
	Price	.31	12.484**	2.58
	Oper. service performance	.32	10.567**	3.56

N/A: Single item latent variable (Not applicable). VIF: Variance Inflation Factor. rho_A=Dijkstra-Henseler rho; CR=Composite reliability; AVE=Average variance extracted. **p<0.0

The measurement model’s ability to discriminate between constructs was verified using the [43], the square roots of the AVEs exceeding the inter-construct correlations. In addition, the heterotrait-monotrait ratio values support the conclusion that the measurement model possesses discriminant validity.

Table 2 presents the outcomes of the hypotheses testing and displays the standardised coefficients for each structural relationship and the corresponding significance levels of the t statistics.

Table 2. Hypotheses testing.

Hypotheses	Std. beta	t value (bootstrapping)	f2	95% CI
				2.50%	97.50%
H1 Positive emotions -> Car brand love	.42*	6.79*	.10	.29	.54
H2 Car dealership trust -> Car brand love	.28*	4.40*	.05	.16	.41
H3 Perceived service quality -> Car dealership trust	.82*	40.94*	2.02	.78	.86
H4 Perceived service quality -> Positive emotions	.81*	37.09*	1.85	.76	.850

Table 2. Hypotheses testing.

Hypotheses	Std. beta	t value (bootstrapping)	f2	95% CI
				2.50%	97.50%
H1 Positive emotions -> Car brand love	.42*	6.79*	.10	.29	.54
H2 Car dealership trust -> Car brand love	.28*	4.40*	.05	.16	.41
H3 Perceived service quality -> Car dealership trust	.82*	40.94*	2.02	.78	.86
H4 Perceived service quality -> Positive emotions	.81*	37.09*	1.85	.76	.850

The results show that all the structural paths achieved bootstrap t values over 2.326, at a significance level of 1%, that is, 99% confidence. The estimation of the model confirms that PSQ is a powerful predictor of trust in the car dealership (β=.82) and the emotions experienced by the customer (β=.82), and that these two emotional responses towards the car dealership translate into greater car brand love for the manufacturer (emotions: β=.42, H1 accepted; trust: β=.81). This satisfactory estimation was accompanied by R² and Q² values that met the minimum established criteria; thus, the structural model shows predictive relevance

5. Conclusions

This research contributes to the existing literature on brand love and emotions and has implications for the automobile industry. The recent literature on emotions explains their importance in enhancing brand love. Understanding how perceived service quality generates brand love will help automobile retailers gain competitive advantage and respond to consumers’ demands.

The study has some limitations that open promising avenues for future research. First, the cross-sectional nature of the study can provide only a snapshot of the hypothesised paths; a longitudinal design could be used to identify time-based facets and provide more rigorous empirical support for the hypotheses.

Second, we are keenly aware that focusing on a single company and product type (new cars) limits the generalisability of the results (hence, it would be worthwhile to test the model with a sample of second-hand cars). Future studies might validate the results by examining other companies and product categories. For example, it is conceivable that perceived service quality could be a more important issue for service than for product providers, and for hedonic than for utilitarian offers.

Third, the present research is limited by the constructs examined. Future studies might delve into other brand-level effects and interactions, such as the role of brand uniqueness and brand desire, to shed light on the potential boundary conditions.