INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Towards Fair Pricing and an Optimized Customer Experience: A  
Study of the Drivers of Premiums and Insured Satisfaction  
Amel Ghachem1*, Karima Ghzaiel2, Jawhar Ben Brahem3  
1Higher Institute of Management of Sousse, Tunisia  
2Faculty of Economic Sciences and Management of Sfax, LRM Laboratory, Tunisia  
3International School of Management and Marketing, Sousse, Tunisia  
*Corresponding Author  
DOI: https://dx.doi.org/10.47772/IJRISS.2025.91100604  
Received: 10 December 2025; Accepted: 18 December 2025; Published: 26 December 2025  
ABSTRACT  
This paper aims to identify ways to improve insurance premium pricing and claims management to optimize  
customer experience and increase customer satisfaction. It examines the interaction between variables with  
two objectives in mind: firstly, to assess the impact of certain aspects of claims management and insurance  
premiums on customer satisfaction and, secondly, to analyze the role of the cost and severity of claims in  
insurance premium pricing. Based on a sample of 90 SECA Assurances customers from the 2018–2024 period,  
and after confirming the endogeneity of the insurance premium variable, we opted for an instrumental variable  
estimation using the two-stage least squares (2SLS) method.  
The results show that policyholder satisfaction depends largely on the human and informational dimensions of  
the service, which reinforce the technical and financial aspects. In addition, insurance premiums are mainly  
determined by the severity of claims, while the cost of claims does not provide a statistically significant  
explanation for the variation in premiums.  
The originality of this paper stems from its integrated approach, considering pricing policies and claims  
management as interrelated constructs shaping customer satisfaction, thereby offering both theoretical and  
managerial contributions to life insurance service.  
Keywords: Insurance, Premium pricing, Claims management, Customer satisfaction  
Ethical Considerations: Before starting this work, we made sure that all ethical requirements were respected.  
All participants were informed about the purpose of the research and agreed to take part voluntarily.  
Conflict of Interest: The authors state that they have no conflicts of interest that could have influenced  
research or its results.  
Data Availability Statement: For confidentiality reasons, the data cannot be shared publicly. However, they  
are available from the corresponding author upon reasonable request.  
INTRODUCTION  
In the insurance sector, customer satisfaction is a major strategic issue. Faced with growing competition and an  
increasingly demanding and volatile customer base, insurers seek to acquire, but above all to retain, their  
policyholders over the long term by focusing on the quality of communication, the reliability of information,  
the transparency of procedures, etc. (Hussen, 2015; Rijanto & Suryani, 2016; Chege, 2021). In this context, it  
should be noted that customer relations are based on a delicate balance between the customer's experience and  
the promise of support made to the customer at the time of subscription, which only truly materializes at a  
specific moment: when a claim is made. This balance is mainly expressed through two dimensions: the  
Page 7753  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
insurance premium and claims management. The premium is the price of protection, and its perceived fairness  
plays a crucial role in the trust placed in the insurer (Xia et al, 2004; Bolton and Lemon, 2022). As for claims  
management, it reflects the moment of truth when the promise of protection is transformed into a concrete  
experience. It is at this critical level that the insurer demonstrates its credibility and commitment to the  
customer. The literature shows that speed of processing, clarity of communication, and personalized support  
are key determinants of satisfaction (Wirtz & Zeithaml, 2018). Conversely, poor claims management can lead  
to dissatisfaction and even customer loss. In reality, a premium that is considered high may be accepted if it is  
offset by efficient and empathetic service when settling a claim. Conversely, a competitive premium can lose  
all appeal if the customer experience is marked by delays, a lack of transparency, or a lack of consideration.  
Thus, customer satisfaction in insurance does not depend solely on the price level, but on an overall equation  
that combines perceived fairness in pricing and the quality of claims management. This raises a key question:  
“What is the combined effect of pricing and claims management on policyholder satisfaction, and what are the  
levers for improvement that can optimize the customer experience?”.  
This issue takes on particular significance in the Tunisian context, which is characterized by a rapidly changing  
insurance market, intensifying competition, and growing demands from policyholders for transparency and  
quality of service. In this context, SECA is a relevant case study, as it illustrates the current challenges facing  
Tunisian insurers: finding the right balance between fair pricing and efficient claims management in order to  
sustainably improve policyholder satisfaction. Therefore, analyzing the impact of pricing policy, especially that  
related to risk and claims management, on policyholder satisfaction is necessary in order to identify the main  
determinants and propose areas for improvement.  
To address this issue, our study will be divided into four main sections: in the first section, we will present a  
review of the literature in order to assess the research hypotheses. Next, we will outline the methodology  
before presenting the results obtained, as well as the managerial implications, limitations, and future avenues  
for research.  
LITERATURE REVIEW  
Following on from the literature review, it appears necessary to develop a conceptual framework that  
formalizes the main constructs used in this research and explains the theoretical relationships that may link  
them.  
1. Satisfaction  
Customer satisfaction plays a decisive role in the insurance sector, as it reflects the extent to which customers'  
expectations are met by the services they receive. According to Oliver (1997), satisfaction can be defined as  
“the consumer's judgment of a consumption experience, resulting from a comparison between their initial  
expectations and the perceived performance of the service.” Applied to the insurance industry, this concept  
reflects the insurer's ability to effectively meet the needs for protection, transparency, and prompt claims  
processing. A satisfied policyholder is more likely to remain loyal, renew their contracts, and share a positive  
image of the company, thereby contributing to its competitiveness and sustainability (Anderson & Srinivasan,  
2003). Today's market is increasingly demanding and marked by the digitization of financial services.  
Improving policyholder satisfaction is therefore becoming a strategic lever for building trust, reducing  
cancellation rates, and ensuring sustainable growth (Ben Omrane & Bellouma, 2021).  
2. Claims management  
The claims management process is a central element in the relationship between the insurer and the insured, as  
it is the moment when the insurance company fulfills its promise of protection. It encompasses all stages from  
the reporting of the claim to the final settlement of the compensation (Cummins & Tennyson, 1996). The  
literature emphasizes that the quality of this process is multidimensional and based on several essential criteria,  
such as speed and efficiency of processing, transparency and clarity of procedures, fairness and accuracy of  
compensation, and the quality of communication and follow-up provided to policyholders (Kasilingam &  
Ramasamy, 2009). Some authors also highlight the importance of the relational dimension, where  
attentiveness, empathy, and responsiveness of claims managers directly influence policyholder satisfaction and  
Page 7754  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
loyalty (Harrington & Niehaus, 2004). Thus, claim management cannot be reduced to a simple technical or  
administrative operation, but must be understood as a comprehensive and multidimensional process that plays  
a decisive role in building a lasting relationship of trust with insured parties. In this study, claims management  
will be considered as a construct comprising several key dimensions, namely: clarity of information,  
processing time, and perceived empathy. This approach allows claims management to be understood not as an  
isolated operation, but as a comprehensive and integrated process that plays a decisive role in building a lasting  
relationship of trust with policyholders.  
The relationship between these dimensions and policyholder satisfaction will be presented below:  
Clarity of information  
According to the theory of informational transparency, customer satisfaction depends on their understanding of  
the processes, rights, and responsibilities associated with the insurance contract in the event of a claim. This  
approach is linked to the foundations of interactional justice theory, which emphasizes the importance of  
quality communication and the perception of fairness in service (Colquitt et al., 2001). In this context, Hussen  
(2015) conducted a study in Ethiopia showing that the reliability and clarity of the information communicated  
positively influences the satisfaction and loyalty of policyholders. This research highlights that policyholders  
attach great importance to transparent communication, particularly with regard to coverage and claims  
processes. In a similar vein, Rijanto and Suryani (2016) explored the impact of information transparency on  
customer cancellation behavior in the auto insurance sector. Their study shows that customers who receive  
clear and accessible information are less likely to cancel their contracts, reflecting a higher level of trust in  
their insurance. Such transparency therefore plays a decisive role in customer retention. Furthermore, Chege  
(2021), in a study of the Kenyan insurance market, emphasized that the clarity of information provided during  
customer interactions is a key factor in satisfaction. Such reliability not only improves customers' perception of  
their insurer but also contributes to strengthening their loyalty.  
H1: Clarity of information has a positive effect on policyholder satisfaction.  
Processing time  
The time taken to process a claim is generally perceived as an indicator of the insurer's commitment to its  
customers and its organizational competence. So, based on the theory of expectation disconfirmation (Oliver,  
1980), delayed claim processing causes frustration and dissatisfaction among customers, while fast claim  
processing, which is often expected by the insured, increases their satisfaction. This approach is complemented  
by the theory of procedural justice (Thibaut & Walker, 1975; Tyler, 1984; Greenberg, 1990), which highlights  
the importance of the perception of a fair and efficient process. In this context, Smith & Bolton (1998)  
demonstrated that in the US auto insurance sector, a rapid response to claims significantly increases customer  
satisfaction, which reduces their propensity to switch insurers.  
This pioneering study highlighted that perceived speed of service was a major predictor of loyalty. Davidow  
(2003) explored this topic further in a study focused on the health insurance sector, indicating that short  
turnaround times for claims processing strengthen customer trust and commitment to the insurer, thereby  
improving long-term relationships. More recently, contemporary studies have confirmed and supplemented  
these findings, Kim et al. (2019) found that the speed of claims processing in South Korea reduces stress for  
insured customers and improves their overall satisfaction. Similarly, both Ahmed & Ghoneim (2020) and Zhou  
et al. (2021) validated the positive impact of speed in the health insurance sector in Egypt and the home  
insurance sector in China. Wang et al. (2018) showed that in the auto insurance sector in China, perceived  
speed in claims processing significantly improves customer satisfaction and loyalty. Khan & Ali (2020) also  
observed that the perception of long delays in health insurance claims resolution in Pakistan leads to a  
significant decline in satisfaction and increases the risk of policy cancellation. However, Singh and Sharma  
(2019) pointed out that communicating regularly about the status of the case reduces negative perceptions of  
delays and improves satisfaction, even when processing times are long. These studies support the idea that  
proactive time management and transparency play a key role in the relationship between perceived delays and  
customer satisfaction.  
Page 7755  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
In Tunisia, a study by Nasri et al (2023) analyzed road accidents based on auto insurance claims data,  
examining in particular the determinants of policyholder satisfaction. Their investigation focused on a sample  
of claims reports collected from several local companies, with a focus on the relationship between driver  
behavior, level of coverage, and policyholder satisfaction. The results show that policyholder behavior directly  
influences the claims-to-premiums ratio, a key factor in assessing service quality. In addition, the perceived  
level of coverage (liquidity of settlement, time taken to process the claim) determines the level of customer  
satisfaction. This study provides a relevant empirical framework in Tunisia regarding the impact of claim  
settlement times and procedures on customer satisfaction.  
H2: A perceived long processing time negatively affects policyholder satisfaction.  
Perceived empathy  
According to the SERVQUAL model developed by Parasuraman et al (1988), which identifies empathy as one  
of the key dimensions of service quality, referring to the ability of staff to demonstrate attention, listening, and  
understanding toward customer needs, respectful and attentive treatment promotes satisfaction by  
strengthening customer trust in the service provider. In this context, Bahadur et al (2018) show that employee  
empathy has an indirect positive effect on customer satisfaction through customer emotional engagement and  
perception of service quality. Chege et al. (2019) showed that in the insurance sector in Kenya, the ability of  
employees to understand and respond to customers' emotional needs significantly improves their satisfaction.  
This conclusion is reinforced by the work of Karavadi and Reddy (2025), who identified empathy as a key  
factor in satisfaction in the life insurance industry in India, highlighting the importance of human interactions  
in the service process.  
Furthermore, in a study of life insurance companies in Nepal, Upadhyay and Adhikari (2021) confirmed that  
empathy positively influences perceptions of service quality and, consequently, customer satisfaction and  
loyalty. Bahadur (2020) highlights that employee empathy strengthens trust in employees, which in turn  
improves satisfaction, itself a mediator of loyalty. Wieseke et al (2012) conceptualize empathy as a  
multidimensional construct (customer and employee empathy). They show that empathy increases customer  
satisfaction and that empathetic customers can mitigate the impact of an unsatisfactory experience on their  
loyalty. These results converge to show that empathy is an essential lever for strengthening the relationship  
between insurers and customers, particularly in this sector where trust and understanding of individual needs  
are paramount.  
H3: Perceived empathy in claims management increases customer satisfaction.  
3. The insurance premium  
The insurance premium, which is the amount paid by the insured to get coverage, is also crucial to overall  
satisfaction. Indeed, according to consumer economic theory (Adam Smith, 1776), price is a determining  
factor in the purchase decision and in the evaluation of the perceived value of the service (Zeithaml, 1988). A  
premium that is considered too high can generate a feeling of overload and reduce satisfaction, while a  
premium that is perceived as fair or competitive reinforces the feeling of value for money and loyalty (Bolton  
& Lemon, 1999).  
In this context, numerous studies have highlighted the impact of premiums, and more specifically perceived  
premiums, which reflect how customers evaluate the cost and fairness of insurance premiums, on policyholder  
satisfaction. Lee and Kim (2018) showed that in the auto insurance sector, a premium perceived as fair or  
competitive increases customer satisfaction, even when the actual amount is high. Similarly, Zhang and Zhao  
(2020) observed that the perceived premium influences the relationship between service quality and  
satisfaction in health insurance, emphasizing that customers constantly compare the perceived price to the  
value they receive.  
However, Singh and Kaur (2021) demonstrated that transparency and communication about the composition of  
premiums reduce negative perceptions of cost, thereby increasing satisfaction and trust. These studies  
emphasize the importance of integrating perceived premiums into marketing and customer relationship  
Page 7756  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
management strategies in the insurance sector. In Tunisia, a study by Ben Amor et al. (2018) indicates that the  
perception of premium fairness significantly influences policyholder satisfaction and their propensity to renew  
their contract. Therefore, based on these studies, it can be said that the insurance premium, beyond its  
economic role, is a key lever in customer relationship management and overall satisfaction.  
H4: A perceived expensive insurance premium reduces customer satisfaction and loyalty.  
However, it should also be noted that the insurance premium must enable the insurer to build up the technical  
provisions necessary to meet its commitments, while generating a profit margin. Its calculation aims to  
determine a fair balance between the cost of risk for the insurer and the acceptable price for the insured. Thus,  
to set the amount of the insurance premium, the insurer considers several components that reflect both the  
technical aspects related to risk assessment and its own commercial aspects1.  
Understanding these different elements provides a better understanding of how insurance contracts are priced.  
The pure premium, also known as the risk premium, forms the technical basis for calculating the premium. It  
corresponds to the statistical estimate of the average cost of claims that the insurer will have to cover,  
distinguishing between normal claims and serious claims. It is calculated using complex actuarial methods that  
consider the frequency and average cost of claims for a given risk profile. In addition to this pure premium, a  
safety loading is added to take into account the volatility of technical results and to protect against unfavorable  
fluctuations in claims from one year to the next, as well as commercial components that include all costs  
incurred by the insurer to manage contracts and claims, the profit margin that the insurer wishes to generate on  
each contract, and applicable taxes.  
This calculation process combines actuarial rigor and commercial considerations to offer premiums that are  
both fair and competitive, while taking into account factors that may influence the premium, such as the type  
of coverage (e.g., all risks insurance), the guarantees included, the perceived level of risk (e.g., for auto  
insurance: vehicle power, use (private/professional), annual mileage, claims history), and the insured's profile  
(age, gender, occupation, geographic location, history, etc.).  
In this study, we will focus on the technical aspect of premium setting related to risk assessment by examining  
the effect of the average cost of claims and their severity.  
The average cost of claims  
According to actuarial risk theory, Dionne and Harrington (1992) assert that the premium is proportional to the  
expected loss, considering both the frequency and severity of claims. In addition, actuarial pricing theory deals  
with determining a fair and equitable premium for insurance contracts. Its objective is to set a premium that  
best reflects the insured risk while ensuring the solvency of the insurance company. It is in this context that  
Cummins and Weiss (2014) demonstrated that an increase in average claim costs leads to a proportional  
increase in premiums, particularly in the auto and home insurance sectors.  
Furthermore, Boucher et al. (2020) show that changes in claims costs, particularly due to factors such as  
inflation or the complexity of repairs, are prompting insurers to regularly adjust premiums to maintain their  
profitability.  
In Tunisia, while empirical studies remain scarce, sectoral analyses conducted by the Tunisian Federation of  
Insurance Companies confirm that premium trends are closely correlated with claims-related expenses,  
particularly in the motor insurance segment. These findings suggest that claims’ costs are a key explanatory  
variable in setting premiums, in line with the economic fundamentals of insurance pricing.  
H5: A high claims cost leads to higher premiums and generates dissatisfaction among insured clients.  
The severity of the disaster  
The claim severity, which corresponds to the extent of losses related to an insured event, is also considered a  
1 Insurance premium calculation: methods and examples.  
Page 7757  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
determining factor in setting insurance premiums. Cummins et al (1988) show the impact of serious claims  
(requiring hospitalization) on changes in auto and health insurance premiums. Dionne and Eeckhoudt (1985)  
showed that high claim severity leads to a significant increase in premiums, as it increases the insurer's  
financial exposure. In addition, Cohen and Einav (2007) studied data from the automotive sector and  
demonstrated that premiums are correlated with the severity of previous claims, reflecting the need to adjust  
rates based on potential costs. Dionne et al (2006) showed that accurate modeling of claim severity, combined  
with claim frequency, improves pricing and risk management in insurance companies.  
Eling and Luhnen (2010) discussed the ways in which the severity of claims (including hospitalization) is  
integrated into premium calculations and risk management. Barseghyan et al (2013) explain how the severity  
of claims impacts premiums via the choice behavior of policyholders. Outreville (2018) emphasizes that the  
frequency and severity of claims directly influence pricing, taking into account the specific risks of each  
customer segment.  
To this end, the following hypothesis is formulated:  
H6: The severity of the claim has a positive influence on high insurance premiums and generates  
dissatisfaction among insured parties.  
In light of this literature review, customer satisfaction in the insurance sector can be influenced by several  
management practices, in particular the quality of the information provided, the time taken to process claims,  
managers' empathy, and the amount of the insurance premium. This in turn may depend on the characteristics  
of the claim related to risk, such as the cost of the claim, its severity, etc.  
In this perspective, our study aims to go beyond traditional approaches. The objective is to identify levers for  
improving insurance premium pricing and claims management, with a view to optimizing the customer  
experience and enhancing policyholder satisfaction by examining the interactions between variables with a  
twofold objective: on the one hand, to assess the impact of certain aspects of claims management and  
insurance premiums on customer satisfaction, and secondly, to analyze the role of the cost and severity of  
claims in insurance premium pricing.  
RESEARCH METHODOLOGY  
The data is collected from customers whose car insurance policies are managed by SECA Assurances and Risk  
Management, a Tunisian brokerage firm that plays a key role in claims management on behalf of its customers,  
and interacts directly with several partner insurance companies such as GAT Assurance, Assurances  
MAGHREBIA, C.O.M.A.R., etc. Although SECA is not strictly speaking an insurance company, its status as  
an intermediary gives it a broad and in-depth understanding of management practices and the perceptions of  
policyholders.  
The initial sample included 122 SECA clients who reported auto claims during the period from 2018 to 2024.  
Excluding cases currently being processed and unusable respondents, we arrived at a final sample of 90 valid  
observations.  
As part of this study, we used an instrumental variable (IV) model to analyze, on the one hand, the  
determinants of policyholder satisfaction and, on the other hand, the factors influencing insurance premiums.  
This model allows us to correct for the potential endogeneity bias between insurance premiums and customer  
satisfaction. Indeed, premiums can influence satisfaction, while themselves being determined by risk-related  
characteristics of claims, which makes the ordinary least squares (OLS) estimator biased. We therefore use a  
two-stage estimation (2SLS) method, which allows us to obtain unbiased and economically interpretable  
coefficients. In the first stage, the insurance premium variable “PRIME” is explained by instruments  
considered exogenous, namely the cost of the claim ‘CS’ and the severity of the claim “GS.” In the second  
stage, the customer satisfaction variable “SATISF” is regressed by the insurance premium ‘PRIME’ as an  
instrumental variable and by the dimensions of claims management, namely the clarity of information “CI,”  
processing time “DT,” and empathy “EMP.” This econometric framework provides a more rigorous  
Page 7758  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
interpretation of the causal effect of the explanatory variables and constitutes a solid basis for analyzing the  
results.  
SATISFi = α₀ + α₁·PRIMEi + α₂·DTi + α₃·EMPi + α₄·CIi + εi₁  
PRIMEi = β₀ + β₁·CSi + β₂·GSi + εi₂  
(1)  
(2)  
SATISFi: Overall customer satisfaction is measured by questions inviting the insured to rate their overall level  
of satisfaction with their insurer on a Likert scale. The items used aim to capture the overall assessment of the  
experience with the insurer, the degree of conformity between initial expectations and the services received,  
the intention to renew or remain loyal, and the propensity to recommend the company to other customers. The  
overall score is obtained by calculating the average of the responses to the various items (Fornell et al, 1996;  
Taylor, 2002; Yusuf et al, 2018).  
PRIMEi: The measurement of the insurance premium was based on a subjective approach, relying on the  
perception of the insured regarding the premium, in particular their assessment of its high or costly nature. The  
data is collected using a questionnaire that includes four items designed to capture this perception, namely the  
assessment of whether the premium is expensive or excessive, the fairness of the amount paid in relation to the  
coverage offered, the perceived value for money, and satisfaction with the cost of insurance. These items,  
formulated in terms of a premium perceived as expensive, were rated on a Likert scale. A composite score for  
perceived premium cost was constructed by calculating the average of the responses to the various items, after  
recoding the items where necessary. This approach is consistent with the academic literature, which  
emphasizes that the perception of a costly premium is a key determinant of customer satisfaction (Pooser and  
Browne, 2023; Sukmawan and Zulganef, 2023).  
DTi: Claim processing time: Using a questionnaire, claim processing time was measured across five items  
focused on the perception of slow processing. These items captured different aspects of this perception,  
including the general impression of slow processing, exceeding the initially expected time frame, perceived  
delays at different stages of the process, late communication on the progress of the claim, and the total duration  
considered excessive. Each item was evaluated on a Likert scale to construct a composite score reflecting  
satisfaction with the speed of processing.  
EMPi : Empathy: Respondents rated their perception of the claims manager's empathetic attitude using four  
items based on active listening skills, understanding emotions, considering the customer's point of view, and  
responding kindly to their concerns. All items were rated on a Likert scale, and the overall empathy score was  
calculated as the average responses. This commonly used approach captures the subjective and relational  
dimensions of interaction while providing a reliable measure of perceived empathy in financial and insurance  
services.  
CIi : The clarity of information was measured using items relating to the impact of communication on  
customer satisfaction in the service sector. This approach assesses the insurer's ability to provide simple,  
understandable explanations of procedures, coverage, and conditions related to the contract and claims  
processing. The respondents were asked to rate, on a Likert scale, their level of agreement with statements  
about the clarity of information, the explanation of technical terms, and the availability of all the information  
needed to make an informed decision. This approach makes it possible to assess policyholders' perceptions of  
the quality of communication, a key element of client satisfaction.  
CSi : The average cost of claims (compensation + related expenses) / number of claims, measured objectively  
based on internal data, expressed as the average amount in dinars compensated per claim.  
GSi : Severity of the incident measured using a dichotomous variable coded 0 if the incidents are deemed non-  
serious and 1 for at least one serious incident. An injury is considered serious if it requires hospitalization.  
According to Baker et al. (1974), this criterion allows serious injuries to be objectively distinguished from  
minor injuries. It indicates a sufficiently serious physical injury requiring intensive medical care.  
εi₁, εi₂: Random error terms: As part of this research, the Likert scale used to measure respondents' perceptions  
and attitudes is a numerical scale ranging from 1 to 10, where 1 means “strongly disagree” and 10 means  
Page 7759  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
“strongly agree”. The choice of this scale is justified by the desire to capture greater granularity in the  
responses, allowing for a finer distinction between different levels of satisfaction, perception, or attitude.  
According to Coelho and Esteves (2007), the 10-point scale improves statistical sensitivity without  
compromising psychometric reliability. Moreover, in cultural contexts where individuals in Tunisia are familiar  
with 10-point ratings, such as in education systems, this scale is more intuitive and easier for respondents to  
interpret. Finally, the 10-point scale facilitates statistical processing, particularly in the context of multivariate  
analyses, by providing a more varied distribution of data and reducing the effects of capping or excessive  
grouping of responses.  
FINDINGS AND DISCUSSION  
Using level data, initial examination of descriptive statistics and preliminary results revealed an asymmetric  
distribution of the claim cost (CC) variable, accompanied by marked heteroscedasticity, calling into question  
the validity of ordinary least squares (OLS) estimates. In order to stabilize the variance, minimize the effect of  
extreme values, and improve the linearity of the model, we opted for a logarithmic transformation of CS. The  
econometric tests performed on the basis of this transformation confirmed the validity of the instrumental  
variable estimation. The Durbin–Wu–Hausman test of endogeneity of the premium (PRIME) produced a  
significant chi-square statistic (p= 0.0062 < 0.05), indicating that the PRIME variable is indeed endogenous  
and therefore requires estimation using instrumental variables. Subsequently, the relevance of the instruments  
was tested through first-stage regression. The severity of loss variable (GS) was found to be significantly and  
positively correlated with the endogenous PRIME variable, with a statistically significant coefficient (coeff =  
2.46; p = 0.025< 0.05), while the disaster cost variable lCS is weakly significant (coeff =0.623: p= 0.079< 0.1).  
In addition, the Sargan over-identification test (Chi² (1) = 0.034; p = 0.8532) shows that the instruments are  
valid in terms of exclusion and are not correlated with error, but they are not powerful enough. Indeed, the F  
statistic for the excluded instruments is less than 10 (F = 2.77), which attests that the lCS and GS instruments  
remain too weak to guarantee a solid identification of the causal effect of the strength of the instruments used.  
The Anderson identification test (Chi² = 5.57; p = 0.0617) also supports the limited identification.  
Furthermore, the multicollinearity test shows that there are no major multicollinearity issues. However, when  
testing the normality of the residuals, the skewness and kurtosis results reject the normality hypothesis (p-value  
= 0.000 < 0.05). On the other hand, the Breusch-Pagan/Cook-Weisberg heteroscedasticity test Cook-Weisberg  
test reveals no evidence of heteroscedasticity in the errors of the first-stage model (model explaining PRIME)  
(p-value=0.8610 > 0.05), indicating that the errors of the first-stage model comply with the assumption of  
constant variance. Nevertheless, there is heteroscedasticity in the residuals in the second-stage model  
explaining satisfaction (p-value=0.000 < 0.05).  
The correction of normality and heteroscedasticity leads us to apply robust errors on Stata to confirm the  
stability and overall reliability of the estimates. Even if the tests reveal a relative instrumental weakness, the  
use of robust errors makes statistical inference more reliable.  
Page 7760  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
Table 1 : The estimation  
Estimates efficient for homoskedasticity only  
Statistics robust to heteroskedasticity  
Number of obs =  
F( 4, 85) = 2241.62  
Prob > F = 0.0000  
Centered R2 = 0.9951  
Uncentered R2 = 0.9996  
Root MSE .1038  
90  
Total (centered) SS  
Total (uncentered) SS =  
=
199.225  
2316.25  
= .9704940517  
Residual SS  
SATISF  
=
Robust  
Coef. Std. Err.  
z
P>|z|  
[95% Conf. Interval]  
PRIME  
DT  
CI  
EMP  
_cons  
.2525636 .0174998  
.2611802  
.2511515 .0040143  
.2575272 .0048069  
-.1532523 .1145897  
14.43 0.000  
50.86 0.000  
62.56 0.000  
53.57 0.000  
-1.34 0.181  
.2182646  
.2511158  
.2432837  
.2481058  
-.3778439  
.2868626  
.2712447  
.2590193  
.2669485  
.0713394  
.005135  
Underidentification test (Kleibergen-Paap rk LM statistic):  
Chi-sq(2) P-val =  
5.898  
0.0524  
Weak identification test (Cragg-Donald Wald F statistic):  
(Kleibergen-Paap rk Wald F statistic):  
Stock-Yogo weak ID test critical values: 10% maximal IV size  
15% maximal IV size  
2.771  
3.090  
19.93  
11.59  
8.75  
20% maximal IV size  
25% maximal IV size  
7.25  
Source: Stock-Yogo (2005). Reproduced by permission.  
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.  
Hansen J statistic (overidentification test of all instruments):  
Chi-sq(1) P-val =  
0.028  
0.8660  
Instrumented:  
Included instruments: DT CI EMP  
Excluded instruments: lCS GS  
PRIME  
As shown in the table above, the estimation by instrumental variables via the two-stage least squares method  
(2SLS) indicates that the satisfaction model is highly significant overall, with high explanatory power (R² ≈  
0.995), high predictive accuracy (Root MSE ≈ 0.10) and F (4, 85) = 2241.62 (Prob > F = 0.0000). Therefore,  
99.5% of the variation in customer satisfaction (SATISF) is significantly explained by the explanatory  
variables (PRIME, DT, EMP, CI). Indeed, clarity of information has a positive and significant effect on  
customer satisfaction (Coef. = 0.2511515; P-value = 0.000). This suggests that customers who receive clearer  
and more understandable information are generally more satisfied, which is consistent with the findings of  
Hussen (2015) and Chege (2021). It therefore confirms that better communication and well-explained  
information contribute to a positive customer experience. In addition, customer-perceived empathy  
significantly increases customer satisfaction, with a slightly higher coefficient (Coeff = 0.2575272; P-value =  
0.000). Insureds are therefore more sensitive to empathy. This result confirms the findings of previous studies  
by Chege et al. (2019), Karavadi and Reddy (2025), Upadhyay and Adhikari (2021), and Bahadur (2020),  
which show that empathy is a key lever for strengthening the relationship between insurers and customers.  
In addition, we note that the long processing time perceived by customers has a positive impact on customer  
satisfaction. Every additional day in processing time is associated with an increase in customer satisfaction of  
0.2611802. That is, the longer the perceived processing time, the more satisfied the customer. This result  
dismisses the hypothesis that customers appreciate a company that responds quickly to their requests and  
contradicts the findings of several studies (Smith and Bolton, 1998; Davidow, 2003; Wang et al, 2018; Kim et  
al, 2019; Khan and Ali, 2020; Ahmed and Ghoneim, 2020; Zhou et al, 2021). Though this may seem  
counterintuitive, it could reflect a situation where longer processing times are perceived as a more rigorous or  
complete process, thereby increasing satisfaction.  
On the other hand, the perceived expensive insurance premium positively affects customer satisfaction, with a  
significant coefficient of 0.2525636. This suggests that when insurance premiums increase, customers remain  
satisfied. This result rejects the hypothesis that a premium perceived as expensive leads to dissatisfaction and  
confirms the findings of Lee and Kim (2018), which show that in the auto insurance sector, a premium  
Page 7761  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
perceived as fair or competitive increases customer satisfaction, even when the actual amount is high. We can  
therefore argue that premiums have a positive causal effect on satisfaction, subject to caution due to the  
demonstrated instrumental weakness.  
Using the first-stage model explaining the insurance premium by the cost and severity of the claim  
(Appendices), we admit that the cost of the claim and its severity fail to explain the insurance premium  
strongly. In fact, although the severity of the claim (GS) leads to a significant increase in the amount of  
premium paid by insureds (p<0.05), the negative effect observed for the cost of the claim (lCS) is not  
statistically significant, thus rejecting the hypothesis that the higher the cost, the more the insurer is obliged to  
increase the premium to cover them.  
CONCLUSION  
To study the combined effect of pricing and claims management on policyholder satisfaction and identify areas  
for improvement to optimize customer satisfaction and experience, we conducted this study on a sample of 90  
observations, adopting an instrumental variable model to identify, firstly, the effect of cost and severity of  
claim on insurance premium, and secondly, the effect of premium, clarity of information, empathy and  
processing time on customer satisfaction.  
The findings reveal that a slow turnaround time is perceived by customers as a more rigorous and complete  
process, which in turn increases their satisfaction. This goes against conventional theories. What's more, an  
increase in the insurance premium also seems to lead to an increase in customer satisfaction. Both of these  
results can be explained by the empathy of managers and the clarity of the information they provide, which  
play an essential role in customer satisfaction and confidence, even in the presence of longer processing times  
or higher premiums. Customers highly value the quality of the insurer's relationship and transparency, which  
offsets the impact of perceived slow processing times and high premiums. In other words, when customers  
perceive that the insurer shows empathy and communicates clearly, they are more tolerant of longer delays and  
higher premiums, considering that the service provided justifies these constraints.  
In line with the findings of Singh and Sharma (2019), regular communication on the status of the case reduces  
negative perceptions of delays and improves satisfaction, even when processing times are indeed long. In  
addition, as Singh and Kaur (2021) mention, transparency and communication about the composition of the  
premium reduce negative perceptions of cost, thereby increasing satisfaction and trust.  
Thus, policyholder satisfaction depends largely on the human and informational aspects of the service, which  
justify the technical and financial aspects. It is therefore recommended that advisors and managers be trained  
in empathetic communication and transparency to strengthen relationships of trust and improve the quality of  
information and clarity of explanations regarding pricing through digital media that allow easy access and  
simplified understanding of contracts and claims.  
It is also advisable to digitize and implement performance indicators to measure responsiveness and customer  
satisfaction, thereby reducing file processing times. Finally, the pricing policy must be adjusted to ensure  
perceived fairness by introducing customization mechanisms tailored to risk profiles and loyalty programs.  
According to the results, the severity of the claim (GS) plays a significant role in determining the premium. It  
is therefore recommended to refine pricing by giving greater weight to severity indicators in order to better  
reflect the actual risk and ensure fair pricing. All of these actions contribute to aligning technical performance  
and relationship quality, thus ensuring a better customer experience and sustainable profitability for the  
company. However, the remarkably insignificant influence of the average cost of claims (CS) on insurance  
premiums requires further investigation. This result does not necessarily justify insurers not using the cost of  
past claims to adjust premiums, but rather that the cost of claims does not provide a statistically robust  
explanation for the variation in premiums.  
In addition to the severity of claims, the introduction of other factors such as claims history, claims frequency,  
and the nature of coverage would provide a better understanding of the complexity of premium pricing.  
Furthermore, the sample of 90 SECA ASSURANCE customers who have made automobile claims appears to  
be limited in relation to the total number of customers. It would therefore be appropriate to expand the  
Page 7762  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
database in order to improve the statistical robustness of the results and their representativeness in relation to  
the overall population of policyholders. Extending the study to a larger and more diverse sample, including  
different categories of contracts and several types of claims, would allow for more generalizable conclusions.  
In addition, the integration of longitudinal data, tracking policyholders over several periods, would provide a  
better understanding of rate changes and customer satisfaction over time. Ultimately, these elements could be  
taken into account in other improvement projects in future research.  
REFERENCES  
1. Ahmed, R., & Ghoneim, A. (2020). Impact of claim service quality on customer satisfaction and  
loyalty in health insurance industry. Journal of Business and Management, 22(3), 45-58.  
2. Anderson, R. E., & Srinivasan, S. S. (2003). E-satisfaction and e-loyalty: A contingency framework.  
Psychology & Marketing, 20(2), 123–138.  
3. Bahadur, W. (2020). Effect of Employee Empathy on Service Loyalty Through the Development of  
Trust in and Satisfaction With Service Employee During Service Interactions. International Journal  
of Customer Relationship Marketing and Management, 11(1), 31–49.  
4. Bahadur, W., Aziz, S., & Zulfiqar, S. (2018). Effect of employee empathy on customer satisfaction  
and loyalty during employee–customer interactions: The mediating role of customer affective  
commitment and perceived service quality. Cogent Business & Management, 5(1), 1–21.  
5. Baker, S. P., O’Neill, B., Haddon, W., & Long, W. B. (1974). The Injury Severity Score: A method  
for describing patients with multiple injuries and evaluating emergency care. Journal of Trauma,  
14(3), 187– 196.  
6. Barseghyan, L., Molinari, F., O'Donoghue, T., & Teitelbaum, J. C. (2013). The nature of risk  
preferences: Evidence from insurance choices. American Economic Review, 103(6), 2499–2529.  
7. Ben Amor, H., Triki, A., & Ben Abdallah, H. (2018). Impact of service quality on customer  
satisfaction in the insurance sector: Evidence from Tunisia. International Journal of Business and  
Management, 13(4), 45-56.  
8. Ben Omrane, W., & Bellouma, M. (2021). Determinants of customer satisfaction in the insurance  
sector: Evidence from Tunisia. International Journal of Financial Services Management, 11(3), 205–  
222.  
9. Bolton, R. N., & Lemon, K. N. (1999). A dynamic model of customers’ usage of services: Usage as  
an antecedent and consequence of satisfaction. Journal of Marketing Research, 36(2), 171–186  
10. Bolton, R. N., & Lemon, K. N. (2022). Customer retention and satisfaction in service industries: A  
comprehensive review. Journal of Service Research, 25(1), 5–23.  
11. Boucher, J. P., Dionne, G., & Gagne, R. (2020). The impact of claim costs on insurance premiums:  
Evidence from automobile insurance. Journal of Risk and Insurance, 87(4), 933-957  
12. Bowers, N. L., Gerber, H. U., Hickman, J. C., Jones, D. A., & Nesbitt, C. J. (1997). Actuarial  
Mathematics (2nd ed.). Society of Actuaries  
13. Chege, C. N., Wanjau, K. & Nkirina, S. (2019). Relationship between empathy dimension and  
customer satisfaction in the insurance industry in Kenya. International Journal of Research in  
Business and Social Science, 8(6), 357–366.  
14. Chege, C. N., Wanjau, K. L., & Nkirina, S. (2019). Relationship between empathy dimension and  
customer satisfaction in the insurance industry in Kenya. International Journal of Research in  
Business and Social Science, 8(5), 1-9.  
15. Cohen, A., & Einav, L. (2007). The effects of risk on prices and demand: Evidence from automobile  
insurance. The Journal of Political Economy, 115(1), 1-43.  
16. Cummins, J. D., & Tennyson, S. (1996). Moral hazard in insurance claiming: Evidence from  
automobile insurance. Journal of Risk and Uncertainty, 12(1), 29–50.  
17. Cummins, J. D., & Van Derhei, J. L. (1988). An empirical investigation of moral hazard in  
insurance. Journal of Risk and Insurance, 55(2), 315–336.  
18. Cummins, J. D., & Weiss, M. A. (2014). Systemic risk and the U.S. insurance sector. Journal of Risk  
and Insurance, 81(3), 489–528.  
19. Davidow, M. (2003). Organizational responses to customer complaints: What works and what  
doesn’t. Journal of Service Research, 5(3), 225-250.  
Page 7763  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
20. Dionne, G., & Eeckhoudt, L. (1985). The effect of risk aversion on the demand for insurance: A  
note. Journal of Risk and Insurance, 52(2), 323-331.  
21. Dionne, G., Gagné, R., & Vanasse, C. (2006). Insurance data and risk classification. The Geneva  
Papers on Risk and Insurance - Issues and Practice, 31(2), 219-237.  
22. Eling, M., & Luhnen, M. (2010). What risk measures are used in practice? Journal of Risk Finance,  
11(3), 273–287.  
23. Harrington, S. E., & Niehaus, G. (2004). Risk Management and Insurance. McGraw-Hill/Irwin  
24. Hege, C. N. (2021). Examining the influence of service reliability on customer satisfaction in the  
insurance industry in Kenya. International Journal of Research in Business and Social Science,  
10(1), 259-265.  
25. Hussen, S. O. (2015). Service quality, customers' satisfaction and loyalty: A study on insurance  
companies in Adama, Ethiopia. European Journal of Business and Management, 7(4), 269-277.  
26. Karavadi, S., & Reddy, M. J. (2025). The role of impact of assurance, empathy, responsiveness on  
customer satisfaction in the life insurance industry. Journal of Informatics Education and Research,  
5(1), 683-688.  
27. Kasilingam, R., & Ramasamy, M. (2009). Policyholders’ satisfaction in the insurance sector: An  
empirical study. Journal of Management and Science, 1(1), 1–10.  
28. Khan, M. A., & Ali, R. (2020). Effects of claim settlement delay on customer satisfaction in the  
health insurance sector of Pakistan. Journal of Insurance Issues, 43(1), 12-28.  
29. Kim, S., Lee, H., & Park, J. (2019). Effect of claim handling speed on customer satisfaction in  
automobile insurance: Evidence from Korea. Journal of Insurance Studies, 37(2), 112-130.  
30. Klugman, S. A., Panjer, H. H., & Willmot, G. E. (2012). Loss Models: From Data to Decisions (4th  
ed.). Wiley.  
31. Lee, H., & Kim, S. (2018). Perceived fairness of insurance premiums and its effect on customer  
satisfaction in the automobile insurance market. Journal of Consumer Policy, 41(3), 365-384.  
https://doi.org/10.1007/s10603-018-9377-6  
32. Nasri, I., Bouzir, A., & Benammou, S. (2023). An analysis of road traffic accidents in Tunisia using  
car insurance claims. Journal of Academic Research in Economics, 15(3), 545–568.  
33. Oliver, R. L. (1977). Effect of expectation and disconfirmation on postexposure product evaluations:  
An alternative interpretation. Journal of Applied Psychology, 62(4), 480–486.  
34. Oliver, R. L. (1997). Satisfaction: A Behavioral Perspective on the Consumer. McGraw-Hill, New  
York.  
35. Outreville, J. F. (2018). Insurance markets in developing countries. The Geneva Papers on Risk and  
Insurance - Issues and Practice, 43(2), 237-253.  
36. Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-item scale for  
measuring consumer perceptions of service quality. Journal of Retailing, 64(1), 12-40.  
37. Pooser, D. M., & Browne, M. J. (2024). The Impact of Regulation on Customer Satisfaction:  
Evidence From the US Auto Insurance Industry. Journal of Insurance Regulation.  
38. Rijanto, S., & Suryani, A. (2016). Information transparency and customer churn: Evidence from the  
car insurance industry. Proceedings of the 2016 International Conference on Information Systems, 1-  
15.  
39. Singh, P., & Sharma, R. (2019). Communication and perceived claim processing time: Effects on  
customer satisfaction in Indian insurance companies. International Journal of Bank Marketing,  
37(6), 1421-1438.  
40. Singh, R., & Kaur, G. (2021). Transparency in premium composition and its impact on customer  
satisfaction in the insurance sector. International Journal of Bank Marketing, 39(6), 1079-1097.  
41. Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations. London : W.  
Strahan and T. Cadell.  
42. Smith, A. K., & Bolton, R. N. (1998). An experimental investigation of customer reactions to service  
failure and recovery encounters: Paradox or peril? Journal of Service Research, 1(1), 65-81.  
43. Sukmawan, R., & Zulganef, Z. (2023). The influence of insurance service reputation, customer  
relationship management, and price attractiveness on insurance service customer customer  
experience: a literature review. International Journal of Business, Economics, and Social  
Development, 4(1), 32-37.  
44. Thibaut, J., & Walker, L. (1975). Procedural justice: A psychological analysis. Erlbaum.  
Page 7764  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025  
45. Tyler, T. R. (1984). Justice in the social system: Procedural justice. In Social Justice (pp. 65–92).  
46. Upadhyay, J. P., & Adhikari, P. R. (2021). Impact of service quality on customer satisfaction and  
firm performance in Nepalese life insurance companies. International Journal of Engineering and  
Advanced Technology, 10(3), 106-110.  
47. Xia, L., Monroe, K. B., & Cox, J. L. (2004). The price is unfair! A conceptual framework of price  
fairness perceptions. Journal of Marketing, 68(4), 1–15.  
48. Wang, J., Li, X., & Zhao, Y. (2018). The impact of perceived claim processing speed on customer  
satisfaction and loyalty: Evidence from the Chinese automobile insurance market. Journal of Service  
Research, 21(4), 455-468.  
49. Wieseke, J., Geigenmüller, A., & Kraus, F. (2012). On the Role of Empathy in Customer–Employee  
Interactions. Journal of Service Research, 15(3), 316–331  
50. Wirtz, J., & Zeithaml, V. A. (2018). Technology-mediated service encounters. In Handbook of  
Service Marketing Research (pp. 345–368). Edward Elgar Publishing.  
51. Zeithaml, V. A. (1988). Consumer perceptions of price, quality, and value: A means-end model and  
synthesis of evidence. Journal of Marketing, 52(3), 2–22.  
52. Zhang, Y., & Zhao, L. (2020). Service quality, perceived premium fairness, and customer  
satisfaction in health insurance: An empirical study. Journal of Health Economics, 69, 102251.  
53. Zhou, Y., Wang, L., & Chen, X. (2021). Influence of claim management efficiency on customer  
retention in property insurance: A study in China. International Journal of Insurance and Risk  
Management, 15(1), 25-40.  
APPENDICES  
. estat endogenous  
Tests of endogeneity  
Ho: variables are exogenous  
Robust score chi2(1)  
= 7.48889 (p = 0.0062)  
= 9.23864 (p = 0.0031)  
Robust regression F(1,87)  
F test of excluded instruments:  
F( 2,  
84) =  
=
2.77  
Prob > F  
0.0684  
Sanderson-Windmeijer multivariate F test of excluded instruments:  
F( 2,  
84) =  
=
2.77  
Prob > F  
0.0684  
. vif  
Variable  
VIF  
1/VIF  
EMP  
DT  
1.07  
1.07  
1.03  
1.02  
0.931668  
0.934337  
0.973923  
0.983882  
PRIME  
CI  
Mean VIF  
1.05  
Page 7765  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025  
. vif  
Variable  
VIF  
1/VIF  
GS  
1.85  
1.85  
0.540698  
0.540698  
lCS  
Mean VIF  
1.85  
. estat hettest  
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity  
Ho: Constant variance  
Variables: fitted values of SATISF  
chi2(1)  
=
27.73  
Prob > chi2 =  
0.0000  
. estat hettest  
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity  
Ho: Constant variance  
Variables: fitted values of PRIME  
chi2(1)  
=
0.03  
Prob > chi2 =  
0.8610  
Skewness/Kurtosis tests for Normality  
joint  
Prob>chi2  
Variable  
resid_iv  
Obs Pr(Skewness) Pr(Kurtosis) adj chi2(2)  
90  
0.0000  
0.0010  
29.16  
0.0000  
. swilk resid_iv  
Shapiro-Wilk W test for normal data  
Variable  
resid_iv  
Obs  
90  
W
V
z
Prob>z  
0.79262  
15.686  
6.071  
0.00000  
Page 7766  
www.rsisinternational.org