Social Sciences & Humanities Open 6 (2022) 100292

Contents lists available at ScienceDirect

Social Sciences & Humanities Open

journal homepage: www.sciencedirect.com/journal/social-sciences-and-humanities-open

Regular Article

Emotions’ influence on customers’ e-banking satisfaction evaluation in

e-service failure and e-service recovery circumstances
Brighton Nyagadza a, *, Gideon Mazuruse b, Asphat Muposhi c, Tinashe Chuchu d,
Tendai Makoni e, Brain Kusotera f
a
Department of Marketing, Marondera University of Agricultural Sciences and Technology (MUAST), Zimbabwe
b
Teaching and Learning Institute (TLI), Marondera University of Agricultural Sciences and Technology (MUAST), Zimbabwe
c
Department of Information & Marketing Sciences, Midlands State University (MSU), Zimbabwe
d
Division of Marketing, School of Business Sciences, University of Witwatersrand, Johannesburg, South Africa
e
Department of Mathematics and Computer Science, School of Natural Sciences, Great Zimbabwe University (GZU), Masvingo, Zimbabwe
f
Department of Statistics and Mathematics, Bindura University of Science Education (BUSE), Bindura, Zimbabwe

A R T I C L E I N F O A B S T R A C T

JEL classification: Purpose: The growth in electronic banking platforms resulted in the unintended effect of minimising the
M3 customer-service provider physical interactions that were instrumental in managing negative emotions
Keywords: emanating from service failure. This study investigates the influence of emotions related to e-service failure and
Customer satisfaction e-service recovery strategies on customer satisfaction.
Emotions Design/methodology/approach: Cross sectional data was collected from 433 e-banking customers using a struc­
e-banking
tured, respondent administered questionnaire. Respondents were randomly intercepted as they moved out of
e-service failure and e-service recovery
Zimbabwe
personal service platforms of four selected banks. Hypotheses were tested using structural equation modelling.
Findings: Emotions were found to have a direct positive influence on customers’ e-banking satisfaction evaluation
during e-service failure and e-service recovery.
Research limitations/implications: Research was conducted in Marondera, an agro-based emerging town
geographically located in Mashonaland East province of Zimbabwe. The generalisation of the study findings in
Zimbabwe may be enhanced by extending the study to other cities.
Practical implications: During the formulation of e-service marketing strategies, bank managers are encouraged to
consider the influence of emotions on customers’ e-banking satisfaction in order to enhance the effectiveness of e-
service recovery initiatives.
Originality/value: The findings of this study contribute to efforts towards effective management of emotions
related to e-service failure and evaluation of e-service recovery strategies, a study area which remains under
researched in Zimbabwe’s banking sector.

1. Introduction and background independence, there were only 9 banks that were operating (Makanyeza
& Chikazhe, 2017; Nyamutowa & Masunda, 2013). Although the
Customer satisfaction and e-service recovery are related concepts banking industry grew during the post-independence era, more banks
anchored on emotions exerted when one experiences service encounters had to close from 2006 to 2008 due to liquidity and operational chal­
via electronic or physical platforms (Chiou, Chao, & Hsieh, 2020). lenges (RBZ, 2021). In an attempt to improve operating efficiencies and
Customer satisfaction and e-service recovery on e-banking platforms are competitiveness, consistent with global trends, most banks in Zimbabwe
critical in maintaining customer loyalty (Chou, 2015; Smith & Mpin­ are adopting electronic banking (e-banking) platforms (Deng et al.,
ganjira, 2015). Online platforms offer a competitive edge to banks to 2010; Nyagadza, Pashapa, Chare, Mazuruse, & Hove, 2022). Such
survive in competitive circumstances (Budianto, 2019). The Zimbab­ platforms include artificial intelligence enabled digital customer service
wean banking history shows that in 1980, when the country gained its chatbots. In 2018, Steward Bank introduced ‘Batsi’ which is connected

* Corresponding author.
E-mail addresses: brightonnyagadza@gmail.com (B. Nyagadza), gmazuruse@gmail.com (G. Mazuruse), muposhia@msu.ac.zw (A. Muposhi), tinashe.chuchu@
wits.ac.za (T. Chuchu), tpmakoni@gmail.com (T. Makoni), kusoterabk@gmail.com (B. Kusotera).

https://doi.org/10.1016/j.ssaho.2022.100292
Received 28 September 2021; Received in revised form 2 May 2022; Accepted 11 May 2022
Available online 31 May 2022
2590-2911/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

to Facebook, Square Mobile App, and e-banking service platform. FBC to this, Equity theory by Adams (1963) proposes that in an exchange
bank followed suit same year by introducing ‘Chido’. ZB bank has its relationship, if an individual perceives that he/she is being treated fairly
own ‘Kesto’, BancABC ‘Ally’ the superhero, and NMB ‘Sera’ (Nyagadza, and there is fair distribution of resources, the individual will be satisfied
Muposhi, et al., 2022). By early 2021, First Capital Bank announced the (Tan, 2014). The SET (Homans, 1961) and Equity Theory (Kahneman &
introduction of ‘Alisa’, the WhatsApp e-banking chatbot. Almost 80% of Tversky, 1979; Adams, 1963) provide the theoretical basis for the
the Zimbabwean banks’ chatbots use animated characters, which mimic concept of perceived justice, which is central to Justice theory. Related
regular conversations that are more or the same as real-life interactions. to this study, the fair-mindedness of the supplier’s service recovery can
Notwithstanding these technological developments, the banking sector be viewed as perceived justice (Kuo & Wu, 2012). Perceived justice has
in Zimbabwe continues to experience a fair share of service failure three dimensions namely distributive, procedural and interactional
(Nyagadza, Pashapa, et al., 2022; RBZ, 2021). justice. Perceived justice is commonly applied in service studies to un­
As a result of the opportunities offered by the Fourth Industrial derstand customers’ reactions to service recovery strategies (Matikiti
Revolution (Sarawathi, 2016) along with the emergency of pandemics et al., 2018; Wirtz & McColl-Kennedy, 2010). Perceived justice is the­
such as the COVID-19 virus (Rukasha, Nyagadza, Pashapa, & Muposhi, orised as the main factor influencing the way customers evaluate service
2021), the adoption of customer e-service platforms by banks is imper­ recovery efforts (Kuo & Wu, 2012; Lovelock & Wirtz, 2011; Matikiti
ative. However, there is a growing realisation that, if not properly et al., 2018; Tan, 2014).
managed, service failure can have negative impact on the success of
e-service platforms (Adams, 1965; Matikiti, Roberts-Lombard, & Mpin­ 2.2. Expectations’ disconfirmation and its impact on customer
ganjira, 2018; Stratemeyer, Geringer, & Canton, 2014). Operating satisfaction
challenges related to e-banking services are unavoidable due to variation
in customer expectations and service satisfaction evaluation criteria Customer satisfaction is a consequence of the comparison between
(Patterson, Cowley, & Prasongsukarn, 2006; Stratemeyer et al., 2014). the expected e-service and service delivered or brand performance and
Therefore, banks’ marketing managers are expected to understand customer expectations (Rita, Oliveira, & Farisa, 2019; Smith, Bolton, &
cognitive and affective factors that influence customer satisfaction. This Wagner, 1999). Expectancy disconfirmation model is a common theory
is critical given that e-service delivery platforms play a key role in that is normally applied when forecasting or predicting customer satis­
influencing customer satisfaction perceptions (Hoffman & Kelley, 2000). faction (Szymanski & Henard, 2001; Varela-Neira et al., 2008). This
Moreover, customers’ perceptions and evaluation of e-service recovery model has proven in previous studies to be effective in explaining
methods used by banks significantly influence patronage intentions and post-purchase evaluation of customer behaviours (Erevelles & Leavitt,
electronic word of mouth (Varela-Neira, Vázquez-Casielles, & Iglesia­ 1992; Varela-Neira et al., 2008; Yi, 1990). Under the conditions of
s-Argüelles, 2008). e-banking service failure and e-banking service recovery, recovery of the
It is important to note that little research efforts have been directed customers’ expectations is their beliefs that they hold at a certain
towards understanding e-banking service failures and e-banking service acceptable repair level (Nyagadza, 2019), which means recovery is
recovery strategies. Most studies on service failure and recovery have another service performance (Lewis & Spyrakopoulos, 2001; Zeithaml,
been conducted in the travel and hospitality industry (Chiou et al., 2020; Berry, & Parasuraman, 1993).
De Souza & Desai, 2013; Koc, 2019). In an attempt to bridge this In normal circumstances, e-banking service recovery disconfirmation
research gap, this study investigates cognitive and affective antecedents has some implications on the customer satisfaction levels with the ser­
of e-banking customer satisfaction in the context of e-banking service vice encounter. Customers may experience proper satisfaction if the
failure and e-banking service recovery circumstances. The specific ob­ bank’s feedback on e-banking is in tandem with their expectations
jectives of this study are: i) to understand the extent to which customer (Menon & Dube, 2000; Varela-Neira et al., 2008), when expectations are
expectations, perceived justice in e-banking service recovery and emo­ exceeded by the bank’s performance of e-service recovery. Conversely,
tions influence customer satisfaction, and ii) to examine whether emo­ e-banking service recovery response can negatively disconfirm expec­
tions affect customer e-banking satisfaction through cognition tations which may reduce customer satisfaction. In line with this, the
antecedents. This study complements the research done by Varela-Neira following hypothesis is formed:
et al. (2008), in Asturias, Spain. This study intends to contribute to
managerial practice by providing insights on effective formulation and H1. Positive expectations disconfirmation in e-banking service failure
implementation of e-service delivery and service recovery strategies. and e-banking service recovery contexts have a positive impact on
This study also contributes to theory by extending the application of the customer satisfaction.
Justice Theory to explain the complexities associated with e-banking
service recovery process. 2.3. Perceived justice in e-service recovery and its impact on customer
The remainder of this paper is structured as follows: Theory and satisfaction
literature review, hypotheses and the conceptual model development
are discussed in the first section. This is followed by a section on The Social Exchange theory has incubated the idea behind Perceived
research methodology, then analysis of results, and finally, the conclu­ Justice (Adams, 1963; Blau, 1964). The theory suggests that, according
sions, research implications, limitations and future research directions to Adams (1963), people evaluate the balance of a transaction by the
are presented. comparison of the ratio of outcomes to investments to the ratio for others
in relationship of the exchange. Multiple conflict situations’ reactions by
2. Literature review, hypotheses and conceptual framework individual customers (Gilliland, 1993; Lind & Tyler, 1988) have been
used to explain the justice concept. Customer evaluative judgements’
2.1. Justice theory formation on a bank’s responses to e-service failure have indicated that
perceived justice is an integral concept in prior research inquiries
The Justice theory was derived from the Social Exchange theory (Mattila, 2001; Varela-Neira et al., 2008). In reality, perceived justice
(Homans, 1961) and the Equity theory (Adams, 1963). The general theory is defined in three dimensional ways, which include Distributive
maxim in exchange states that if the cost is higher than the gains, justice, Procedural justice, and Interactive justice (Smith et al., 1999). All
something must be done to balance the two so that a level of fairness is these elements have some effects to customer satisfaction.
reached (Kuo & Wu, 2012; Matikiti et al., 2018). This is in line with the Distributive justice relates to customers’ evaluation of the satisfaction
tenets of the Social Exchange Theory (SET) which suggests that the of the e-service encounter, thus whatever the bank offers the customer to
benefits of a service should outweighs the cost (Homans, 1961). Further recover from the e-banking service failure and resolve the customers’

2
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

complaints (Ding & Lii, 2016; Homburg & Furst, 2005). It is focused on 1999). However, only a few studies have managed to consider e-service
the tangible things that customers receive during service recovery, failure and e-service recovery in connection with emotions (Menon &
which include rewards in the form of money, refunds, future purchase Dube, 2000; Zeelenberg & Pieters, 2004). The lack of prior research
discounts, coupons, and exchange service (Kuo & Wu, 2012; Matikiti related to emotions on e-service encounters has prompted many studies
et al., 2018). In another sense, it shows how the customers feel and their recently (Smith, 2006) including the current one. Emotions are formed
evaluation on whether they have received proper treatment regarding from affect, which is a sub-branch of mental processes. The affective
the redress of e-service recovery (Mattila & Cranage, 2005; Maxham & psychological element includes moods and emotions (Westbrook &
Netemeyer, 2002). Other recent research (Chang, Lai, & Hsu, 2012; Ding Oliver, 1991). The current study only focuses on emotions from the
& Lii, 2016; Kuo & Wu, 2012; Matikiti et al., 2018) has depicted that negative side. When customers experience negative effects, they are
satisfaction can be predicted by distributive justice. It is expected that exposed to lower satisfaction than those that have little to no emotions
distributive justice influences customers’ satisfaction. Therefore, we (Oliver, 1997). This is due to the fact that customer satisfaction in this
suggest the following hypothesis: context has two psychological components, which are cognitive and
affective. Thus, it can be hypothesised:
H2. Distributive justice in e-banking service recovery has a positive
impact on customer satisfaction. H5. Degree of negative emotions experienced in e-banking service
Procedural justice explains the perception of customers on the rules failure circumstances has a negative impact on customer satisfaction.
and guidelines applied by the bank in dealing with the complaints of the E-banking service failure creates some negative bias which results in
stakeholders (Blodgett, Hill, & Tax, 1997; Nikbin, Ismail, Marimuthu, & customer negative emotions during the process of the evaluation of the
Armesh, 2012). The main idea here is of probing the bank on whether it latter. This is deemed to be perceived justice and disconfirmation of
has followed adequate procedures in dealing with the complaint raised. expectations. Negative emotions after e-banking service failure causes
Procedural justice influences satisfaction levels (Ding & Lii, 2016; Lopes customers to have a negative evaluation (Kuo & Wu, 2012; Matikiti
& Silva, 2015) after e-banking service recovery. It may include taking et al., 2018), than when they experience no emotions at all (Taylor,
quick action to rectify service failure, fair customer treatment and the 1994). Therefore, precisely, the customers or people in general who are
elimination of problems (Nikbin et al., 2012; Lopes & Silva, 2015). in negative state tend to see things negatively (Clark & Isen, 1985), are
Therefore, we suggest the following hypothesis: very pessimistic and their behaviours reflect this negativity in their ex­
pectations (Dickinson-Delaporte, Beverland, & Lindgreen, 2010; He,
H3. Procedural justice in e-banking service recovery has a positive
Cai, & Gao, 2016). Thus, it can be hypothesised:
impact on customer satisfaction.
Interactional justice accounts for the extent to which customers are H6. Degree of negative emotions experienced in e-banking service
handled carefully in their contact with the banks or organisation’s em­ failure circumstances has a negative impact on expectations disconfir­
ployees taking responsibility of e-banking service recovery via electronic mation in e-banking service recovery.
means (Matikiti et al., 2018; Orsingher, Valentini, & Angelis, 2010). The In both developing and developed economies e-banking services are
way the employees behave or treat customers during the e-service re­ fast becoming more demanding to be a competitive prerequisite
covery process (Varela-Neira et al., 2008), and behaviour exhibited in (Mathew, Jose, Rejikumar, & Chacko, 2020). An increase in the devel­
handling complaints (Gountas, Gountas, & Mavondo, 2013; Homburg & opment of techno savvy driven approaches in banks has led to creation
Furst, 2005) clearly defines interactional justice. The judgement formed of a vast number of systems failures resulting in expectancy disconfir­
by customers in line with e-service recovery process does have an impact mation (Komunda & Osarenkhoe, 2012), leading to significant decline
on customers’ satisfaction (Nikbin et al., 2012). Some research has in customer satisfaction, and e-banking service perceptions (Piha &
indicated the positive effect of various justice dimensions on satisfaction Avlonitis, 2015). It is focused on the tangible things that customers
with complaint handling or e-banking service recovery (Homburg & receive during service recovery, which include rewards in the form of
Furst, 2005; Lopes & Silva, 2015; Maxham & Netemeyer, 2002), and on money, refunds, future purchase discounts, coupons, and exchange
general customer satisfaction (Myers & Majluf, 1984; Kuo & Wu, 2012). service (Kuo & Wu, 2012; Matikiti et al., 2018). The most frequent
Therefore, we suggest the following hypothesis: e-banking service errors are linked to the self-service technologies and
mobile devices (Menshikova, Romolini, Sabbatelli, & De Marco, 2017),
H4. Interactional justice in e-banking service recovery has a positive
and they cause serious negative effects to customers’ emotional state
impact on customer satisfaction.
(Mathew et al., 2020), further leading to defection and dissatisfaction
Many studies (for example, Kau & Loh, 2006; Patterson et al., 2006;
(Joireman, Gr_egoire, Devezer, & Tripp, 2013). At times the e-banking
Smith et al., 1999) have given proof that distributive justice has a higher
service provider might not be having the room to control the e-banking
effect than interactional justice on customers’ satisfaction with e-service
service failures, hence the need for recovery of the lost customer satis­
recovery. Distributive Justice is claimed to have more impact on the
faction (Jose & Mathew, 2016). Therefore, we suggest the following
satisfaction of customers (Clemmer, 1993), while Interactional Justice is
hypothesis:
deemed to be more impactful than Distributive Justice (Matikiti et al.,
2018; Orsingher et al., 2010; Seiders & Berry, 1998). Due to the fact that H7. Degree of negative emotions experienced in e-banking service
there is some lack of convergence between the conclusions in the prior failure circumstances has a negative impact on distributive justice with
research studies, this has prompted the researchers’ interests in regard to e-banking service recovery strategies.
continuing to explore the relative influence of perceived justice elements The paradox of e-banking service recovery is closely linked to pro­
on customer satisfaction. Thus, it is proposed that: cedural justice. This is so because normally there is a paradox (Wu &
Huang, 2015; Gupta et al., 2017), where customers who have some
e-banking service failure encounters are supplemented by an ultimately
2.4. The effect of emotions and their direct and indirect impact on
good e-banking service, they end up being elevated in terms satisfaction
customer satisfaction
(Hoffman et al., 2016; Rejikumar, 2015), compared to the way they
would have not experienced failures or errors (Sousa & Voss, 2009). Due
Emotions can be defined as those responses that customers have after
to this, procedural justice influences satisfaction levels (Ding & Lii,
an evaluation made in a specific situation related to e-banking service
2016; Lopes & Silva, 2015) after e-banking service recovery. It may
and an experience of consumption (Varela-Neira et al., 2008). For cus­
include taking quick action to rectify service failure, fair customer
tomers and other stakeholders to be able to evaluate the level of satis­
treatment and the elimination of problems (Nikbin et al., 2012; Lopes &
faction in the e-service offered via e-banking platforms (Nyagadza,
Silva, 2015). In line with this, digital efficacy perceptions are likely to
Kadembo, & Makasi, 2021cortant role (Bagozzi, Gopinath, & Nyer,

3
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

cause rationality emotion among the customers to trust and believe measured using item scales adapted from literature specifically related
online platforms and to remain committed to such technologies (Yeoh, to e-banking services. An example of the items is Negative Emotions
Woolford, Eshghi, & Butaney, 2014). Therefore, we suggest the (NE), shown in Appendix 1, with three items that can be found in
following hypothesis: Liljander and Strandvik (1997), Yu and Dean (2001), and Varela-Neira
et al. (2008). Disconfirmation of Expectations (DE) (in Appendix 2 with
H8. Degree of negative emotions experienced in e-banking service
three items) (Hess, Ganesan, & Klein, 2003; Smith et al., 1999).
failure circumstances has a negative impact on procedural justice with
Distributive Justice in Appendix 3 with three items can be found in
regard to e-banking service recovery strategies.
Homburg and Furst (2005) and Maxham and Netemeyer (2003). Pro­
Probability of customers’ need for the e-banking service recovery is
cedural Justice (PJ) (Appendix 4 with three items, were utilised in
very high when failure ocurrs during the process of engagement and
Homburg and Furst (2005), Maxham and Netemeyer (2003) and Smith
interaction. This prompts an urge for e-banking service providers to
et al. (1999). Interactional Justice (IJ) (Appendix 5 with three items)
precisely understand the process of recovery, in a bid to win customers’
(Homburg & Furst, 2005; Maxham & Netemeyer, 2003; Smith et al.,
hearts (Nguyen, McColl-Kennedy, & Dagger, 2012). The judgement
1999; Varela-Neira et al., 2008) and Customer Satisfaction (CS) (Ap­
formed by customers in line with e-service recovery process does have an
pendix 6 with five items) (Makanyeza & Chikazhe, 2017; Maxham &
impact on customers’ satisfaction (Nikbin et al., 2012). Some research
Netemeyer, 2003; Oliver, 1997; Varela-Neira et al., 2008; Yu & Dean,
has indicated the positive effect of various justice dimensions on satis­
2001), were subjected to examination via Confirmatory Factor Analysis
faction with complaint handling or e-banking service recovery (Hom­
(CFA). Further to this, Cronbach’s alpha coefficient, the Average Vari­
burg & Furst, 2005; Lopes & Silva, 2015; Maxham & Netemeyer, 2002),
ance Extracted (AVE), Composite Reliability (CR) coefficient, means,
and on general customer satisfaction (Hooper et al., 2008; Kuo & Wu,
standard deviations and the correlation between the set of constructs
2012). If e-banking service recovery is not to the level of customers’
have been examined.
expectations during the interactions, the propensity of switching the
service provider is very high (Chang, Chen, & Lan, 2013). Therefore, we
suggest the following hypothesis: 3.2. Sampling and data collection

H9. Degree of negative emotions experienced in e-banking service The research study applied a cross-sectional survey of 433 e-banking
failure circumstances has a negative impact on interactional justice with customers conducted in Marondera district (Image 1), Mashonaland East
regard to e-banking service recovery strategies. province of Zimbabwe. The researchers divided the population of po­
Based on the theoretical and literature review and posited hypoth­ tential e-banking respondents into more relevant and significant strata
eses, the conceptual model supporting this study is illustrated in Fig. 1: (Nunnally, 1967; Muposhi, Nyagadza, & Mafini, 2021) based on subsets
where a random sample was drawn from each of the strata (Saunders,
Lewis, & Thornhill, 2009) such as the customers’ profiles (low, middle
3. Research methodology
and high income earning capacities) as well as the geographical loca­
tions to which they belong to. Stratified random sampling technique was
The sample, design of the questionnaire and measures, as well as
applied due to its accuracy and easy-to-use advantages (Bagozzi & Yi,
data collection methods applied in the research are explained in this
1988; Saunders et al., 2009). In order to determine adequacy of sample
section. The research follows a quantitative approach, with a deductive
size, Krejcie & Morgan, 1970 formular was applied, necessary to
logic following positivism.
construct a confidence interval (generally +5%) (Chan & Idris, 2017a,
2017b). A total of 433 valid responses were considered for analysis,
3.1. Design of questionnaire and measures translating to a response rate of 87%. Pilot study was conducted on the
respondents using stratified random sampling from the selected banks.
All variables were measured using validated scales adapted from These respondents represented the recommended 5% of the research
previous studies using a 7-point Likert scale. Study constructs were study sample. Participation was voluntary and the objectives of the

Fig. 1. Hypothesised conceptual research study model.

Key: NE: Negative Emotions; DI: Disconfirmation; DJ: Distributive Justice; PJ: Procedural Justice; IJ: Interactional Justice, CS: Customer Satisfaction.
Source: Researchers’ conception (2021)

4
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

study were explained to the participants in the research study, before ethical considerations by maintaining integrity and professionalism
respondents completed the questionnaire. The researchers collected the about the morals of academic research.
data from January 2021 to April 2021. Stretching of the data collection
period was as a result of Covid-19 restrictions, which delayed the whole 4. Data analysis and results
process. The research study applied an online web-based cross-sectional
survey with the aid of 20 fieldworkers to supplement the process. A Testing of the developed hypotheses was done after data were
cross-sectional approach was applied to collect data through the use of assessed for convergent and discriminant validity.
questionnaires whereby data was collected from the sample at once.
Banking customers who had experienced in the use of e-banking were 4.1. Description of respondents
involved in the collection of data.
Sample profile is presented in Table 1. As depicted in the table, fe­
3.3. Non-response bias test males were more than males. The majority of the respondents (34.6
percent) were aged between 20 and 39 years. Most of the respondents
Armstrong and Overton’s technique (1977) was used to check for (43.9 percent) had already earned at least a Bachelor degree. The ma­
non-response bias test. The process involved the use of t-tests to compare jority of the respondents (34.1 percent) were earning less than USD$1,
the means of each of the items of the succeeding responses against the 500 per month.
rest of the responses. There were no larger differences in the means. This
suggests that non-response bias was not a threat to the research study.
4.2. Measurement model
3.4. Data analysis method justification
Model Fit Indices (MFI), Composite Reliability (CR), Standardised
Factor Loadings (SFL), Individual Item Reliabilities (IIR), Critical Ratios
Both descriptive and inferential statistics were used in analysing
(CRs) and Average Variance Extracted (AVE), have been used to assess
quantitative data from the questionnaire. Structural Equation Modelling
convergent validity (Reisinger & Mavondo, 2007). A discriminant val­
(SEM) was used to test the posited hypotheses. Descriptive statistical
idity and reliability test has to be done when fitting a Structural Equa­
analysis was achieved through the functional application of charts, ta­
tion Model (SEM).
bles, graphs and diagrams, and this fed into inferential statistics (Nya­
For reliability checks, the Cronbach’s alpha test (CA) that tests in­
gadza et al., 2020a; Attallah, 2015). These included frequencies, mean,
ternal consistency, Composite Reliability (CR) and Average Extracted
and standard deviation. Software packages used for data visualisation
Variance (AVE) were calculated and the results are summarised in
were WarpPLS and SPSS, version 7 and version 25, respectively.
Table 2. The Variance Inflation Factors (VIFs) values which examine the
Exploratory Factor Analysis (EFA) was used to identify the underlying
presence of multicollinearity among the variables are also presented.
relationships between the variables measured (Gerald, 2018). Chan and
The considered variables are Disconfirmation (DI), Distributive Justice
Idris (2017a, 2017b) advise researchers to carry out an Exploratory
(DJ), Procedural Justice (PJ), Interactional Justice (IJ), Negative Emo­
Factor Analysis (EFA) at the beginning of data analysis as part of scale
tions (NE) and Customer Satisfaction (CS).
validation. Keller and Kros (2011) postulate that Exploratory Factor
Both the structure and cross-loadings used to test convergent validity
Analysis (EFA) is used to measure the dimensionality of a survey, to
are summarised in Table 3. The figures in bold represent high satura­
recognize precarious and non-critical items (Attallah, 2015), to decrease
tions and the non-bold are with lower loads. Meaningful p-values for
the quantity of items and to re-examine the content of the factor. Effendi,
these results are 95%, and this indicates the presents of convergent and
Matore, Khairani, and Adnan (2019) consent that Exploratory Factor
discriminant validity (Shapiro & Nieman-Gonder, 2006; Bagozzi & Yi,
Analysis (EFA) help researchers who do not know how many factors
which explain the interrelationship among a set of items (Maat, Zakaria,
Nordin, & Meerah, 2011). Exploratory Factor Analysis (EFA) was per­ Table 1
formed so as to refine and decrease the number of related variables to a Sample profile demographic characteristics.
more relevant (Keller & Kros, 2011), and manageable number prior to Characteristics Frequency %
using them for further analysis (Alexander et al., 2016; Schoefer, 2008). Gender
To assess adequacy of the measurement model, the researchers applied Male 205 47.3
Confirmatory Factor Analysis (CFA) (Worthington, Russell-Bennett, & Female 228 52.7
Total 433 100
Härtel, 2010). The researchers also utilised Principal Component Anal­
Age
ysis (PCA) to consider the total variance in the data (Muposhi et al., <20 years 117 27.0
2021), and establishing minimum number of factors that will account 20–29 years 150 34.6
for the maximum variance (Da Costa Carvalho, 2015). In addition, the 30–39 years 105 24.2
Bartlett’s test of sphericity was applied to examine the hypothesis that 40–49 years 45 10.4
>50 years 16 3.7
the variables were uncorrelated (Alalwan, Dwivedi, Rana, & Algharabat, Total 433 100
2018). It was used to see whether there were some relationships be­ Education
tween variables, which is necessary for factor analysis to be appropriate High School Level 39 9.0
(Field, Miles, & Field, 2012). Certificate Level 53 12.2
Diploma Level 50 11.5
Bachelor’s Degree Level 190 43.9
3.5. Ethical considerations Master’s Degree Level 91 21.0
Doctoral Degree Level 10 2.3
Ethical considerations related to participating e-banking customers’ Total 433 100
privacy, informed consent, freedom of response, professionalism, Monthly Income (US$)
139 32.1
integrity, accuracy and values of research have been adhered to by the
<500
500–999 148 34.1
researchers, in line with the provisions made by the Marketing Research 1000–1499 79 18.2
Society (MRS) (2022). Due to this, the researchers were obliged to 1500–1999 49 11.3
observe the practices that take note of the values, and integrity of >2000 18 4.2
Total 433 100
research by not making manipulations to ethical issues (Schumacker &
Lomax, 2010; Muposhi et al., 2021). They made sure that they upheld Source: Field Data (2021)

5
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Table 2 4.3. Structural model

Descriptive statistics for construct variables.
NE DI DJ PJ IJ CS Variance Inflation Factor (VIF), Coefficient of determination (R2),
Effect size (f2) and the Predictive Relevance of the model (Q2) and
CR 0.838 0.868 0.871 0.905 0.849 0.904
CA 0.710 0.772 0.774 0.790 0.733 0.856 Overall Assessment of the model were used to evaluate the structural
AVEs 0.634 0.687 0.694 0.826 0.653 0.703 model for the current study. Partial Least Squares (PLS) approach was
VIFs 2.279 3.131 3.948 3.465 3.987 4.273 used to calculate the relationships between the variables in the struc­
KEY: CR = Composite reliability, CA = Composite reliability, AVEs = Average tural model. WarpPLS 7.0 software was employed in calculating the
Variance Extracted, VIFs = Variance Inflation Factors. model’s path coefficients and the p-values. The reason behind this was to
Source: Field Data (2021) check the impact of each relationship of the variables and whether the
data fits to the proposed model. Table 5 presents the results.
With reference to the results in Table 5, DI had a negative effect on
Table 3 CS (β = − 0.163, p < 0.001). This is due to the fact that e-banking service
Structure and cross loadings. recovery response can negatively disconfirm expectations which may
NE DI DJ PJ IJ CS reduce customer satisfaction (Smith & Mpinganjira, 2015; Varela-Neira
NE1 0.785 0.351 0.403 0.326 0.432 0.477 et al., 2008). DJ had a positive effect on CS (β = 0.253, p < 0.001).
NE2 0.747 0.611 0.589 0.624 0.593 0.672 Customer evaluative judgements’ formation on a bank’s responses to
NE3 0.853 0.555 0.589 0.481 0.602 0.559 e-service failure have indicated that perceived justice is an integral
DI1 0.679 0.808 0.594 0.516 0.679 0.63 concept, in prior research inquiries (Homburg & Furst, 2005). PJ had a
DI2 0.495 0.824 0.588 0.581 0.628 0.628
DI3 0.408 0.853 0.662 0.548 0.669 0.540
positive effect on CS (β = 0.419, p < 0.001) and IJ had a positive effect
DJ1 0.505 0.690 0.703 0.607 0.636 0.609 on CS (β = 0.195, p < 0.001) (Blodgett et al., 1997). It can be further
DJ2 0.554 0.673 0.915 0.699 0.700 0.687 noted that NE had a positive effect on CS (β = 0.197, p < 0.001.). For
DJ3 0.595 0.514 0.867 0.659 0.615 0.684 customers and other stakeholders to be able to evaluate the level of
PJ2 0.511 0.608 0.719 0.909 0.646 0.765
satisfaction in the e-service offered via e-banking platforms (Nyagadza
PJ3 0.57 0.595 0.711 0.909 0.688 0.688
IJ1 0.603 0.661 0.751 0.739 0.861 0.735 et al., 2021c important role (Bagozzi et al., 1999). NE had a positive
IJ2 0.533 0.539 0.493 0.427 0.748 0.484 effect on DI (β = 0.754, p < 0.001) (He et al., 2016), NE had a positive
IJ3 0.514 0.721 0.628 0.595 0.813 0.637 effect on DJ (β = 0.753, p < 0.001), NE had a positive effect on PJ (β =
CS2 0.607 0.675 0.530 0.527 0.627 0.725 0.706, p < 0.001) (Kau & Loh, 2006) and NE had a positive effect on IJ
CS3 0.620 0.578 0.738 0.760 0.697 0.904
CS4 0.579 0.639 0.764 0.780 0.706 0.890
(β = 0.758, p < 0.001) (Patterson et al., 2006). Customers or people in
CS5 0.589 0.546 0.601 0.589 0.554 0.823 general, who are in a negative state tend to see things negatively (Clark
& Isen, 1985; Matikiti et al., 2018), are very pessimistic and their be­
Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ),
haviours reflect this negativity in their expectations (Dick­
Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS).
More details on these are found in Appendix 1.
inson-Delaporte et al., 2010; He et al., 2016). Fig. 2 is a summary of the
Source: Field Data (2021) fitted model with coefficients.

2012).
4.4. Mediation analysis
In line with the results presented in Table 3, all the cross-loadings
exceed 0.6, implying that convergent validity on the measurement
Mediation is an extension of simple linear regression in that it adds
constructs exist (Fornell & Larcker, 1981). Discriminant validity of the
one or more variables to the regression equation (Cude et al., 2006).
constructs was also examined through inter-constructs correlations.
Simply defined, mediating variables are mechanisms through which R
The inter-constructs correlations, together with the square root of
[Independent variable] influences S [Dependent variable] (Hayes,
AVEs, are shown in Table 4. Average Variance Extracted (AVE) was
2013). In mediation analysis, researchers assume that the Independent
compared with squared inter-construct correlations in a bid to assess
variable (R) affects the Mediator (T), which in turn, affects the Depen­
discriminant validity.
dent variable (S). Mediation analysis was done using Sobel’s test in this
It can be noted from Table 4 results, that discriminant validity exists
study. The Sobel’s test uses the product of coefficients. In this study,
because the square root of the AVE values (diagonal elements) for the
emotional attachment mediates the relationship between every
latent variables exceeded the corresponding correlation coefficient
construct variable to corporate brand perception. The results are pre­
values of other latent variables (Fornell & Larcker, 1981; Segars, 1997).
sented in Table 6.

Table 5
Structural Equation Modelling (SEM) results.
Table 4
Hypothesis Relationship Coefficient p-values Decision
Inter-constructs Correlations among latent variables with square root of AVEs.
H1 DI → CS − 0.163 <0.001 Supported and significant
NE DI DJ PJ IJ CS
H2 DJ →CS 0.253 <0.001 Supported and significant
NE 0.796 H3 PJ → CS 0.419 <0.001 Supported and significant
DI 0.634 0.829 H4 IJ → CS 0.195 <0.001 Supported and significant
DJ 0.662 0.742 0.833 H5 NE → CS 0.197 <0.001 Supported and significant
PJ 0.595 0.662 0.786 0.909 H6 NE → DI 0.754 <0.001 Supported and significant
IJ 0.681 0.795 0.778 0.734 0.808 H7 NE → DJ 0.753 <0.001 Supported and significant
CS 0.711 0.722 0.792 0.8 0.771 0.838 H8 NE → PJ 0.706 <0.001 Supported and significant
AVE 0.634 0.687 0.694 0.826 0.653 0.702 H9 NE → IJ 0.758 <0.001 Supported and significant

NB: Square root of AVEs is in bold. P-value level of significance is less than 0.001.
Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ), Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ),
Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS). Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS).
More details on these are found in Appendix 1. More details on these are found in Appendix 1.
Source: Field Data (2021) Source: Field Data (2021)

6
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Fig. 2. Summary of the fitted model with coefficients.

Source: Field Data (2021)

Image 1. Marondera District in Zimbabwean Geographical Map depicting the study area.
Source: Google Maps (2021)

Table 6
Mediating effect analysis via Sobel test.
Hypothesis Path Std Beta Std Error T Statistics P Values Decision Bootstrapping confidence interval

95% CI LL 95% CI UL

H10 NE→ DI→ CS − 0.12 0.124 2.293 <0.001 Supported 0.084 0.132
H11 NE →DJ → CS 0.19 0.04 2.873 <0.001 Supported 0.106 0.251
H12 NE→ PJ→ CS 0.30 0.021 2.306 <0.001 Supported 0.163 0.621
H13 NE → IJ→ CS 0.13 0.082 8.644 <0.001 Supported 0.093 0.247

Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ), Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS). More details on
these are found in Appendix 1.
Source: Field Data (2021)

7
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

In reference to Table 6, the path NE → DI → CS, there is the product Table 8

of 0.75 and 0.16 which are beta values for NE→DI and DI → CS Collinearity.
respectively that is 0.75 × 0.16 results in 0.12. Further to this, the Exogenous Variables VIF Tolerance
relationship between negative emotions and customer satisfaction is
NE→DI 2.33 0.43
significantly mediated by disconfirmation (β = 0.12, p < 0.001), the DI→CS 3.85 0.26
relationship between negative emotions and customer satisfaction is NE →DJ 2.33 0.43
significantly mediated by distributive justice (β = 0.19, p < 0.001) and DJ→CS 3.85 0.26
the relationship between negative emotions and customer satisfaction is NE→PJ 2 0.5
PJ→CS 3.85 0.26
significantly mediated by interactional injustice (β = 0.13, p < 0.001). NE→IJ 2.33 0.43
Mediation effect may result in some of the relationship between the IJ→CS 3.85 0.26
variables being statistically insignificant whilst some remain significant.
Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ),
From the analysis, it shows in both a direct and indirect relationship that
Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS).
there was no change in terms of the significance of the constructs. The More details on these are found in Appendix 1.
confidence interval also confirms the results since the interval of the Source: Field Data (2021)
beta value excludes zero.
It is important to note the magnitude of the indirect effects, the
4.5.2. Coefficient of determination (R2)
variance accounted for (VAF). Hair, Hult, Ringle, and Sarstedt (2017)
Schumacher, Erol, and Sihn (2016) define R2 value as the percentage
suggested that VAF >80% indicates full mediation, 20% ≤ VAF ≤ 80%
of variance in the variable that is accounted for by association in the
partial mediation and VAF <20% indicates no mediation. The formula
independent variable groups. R2 values of 0.75, 0.5 and 0.25 can be
for VAF helps to know how the independent variables contribute in
considered substantial, moderate, and weak respectively (Hair et al.,
explaining the variation in the dependant variable through the medi­
2011). Very high values of R2 may result in the model overfitting the
ator. VAF formula was used:
data and may result in a spurious relationship provided the R2 value is
Indirect ​ effects greater than the Durbin Watson. In the current study, disconfirmation
VAF =
Total ​ effects has an R2 value of 0.57 which is explained by negative emotions. The
predictor has a direct effect towards disconfirmation. The distributive
From the results displayed in Table 7, it can be concluded that the
justice has an R2 value of 0.57, contributed by negative emotions.
disconfirmation, distributive justice, procedural justice and interac­
Moreover, the procedural justice has an R2 value of 0.50. The interac­
tional justice played the role of mediating between negative emotions
tional justice has an R2 value of 0.57 which is explained by negative
and customer satisfaction.
emotions. The customer satisfaction has an R2 value of 0.74 which is
explained by negative emotions, interactional justice, procedural jus­
4.5. Evaluation of the structural model
tice, distributive justice, and disconfirmation. Overall, the developed
model has a moderate explaining power.
After the recognition that the measurement model was valid and
reliable, the next step was to measure the structural model outcomes.
4.5.3. The effect size (F2)
This included examining by the Variance Inflation Factor (VIF), Coeffi­
Effect size (f2) (shown in Table 9) is a measurement that tells the
cient of determination (R2), Effect size (f2) and the Predictive Relevance
impact of change in the R2 value when a specified exogenous construct is
of the model (Q2).
ignored in the model. (Hair et al., 2011). An effect size f2 ≤ 0.30, 0.3 < f2
≤ 0.50 and f2 >0.50 is thought to represent a weak, moderate, and
4.5.1. Collinearity of structural model
strong effect respectively (Bliwise, 2006).
Multicollinearity is one of the assumptions of structural equation
Effect size is calculated using the following equation:
modelling. Before judging the structural interactions, collinearity must
be studied to make sure it does not bias the results. The Variance R2
Inflation Factor (VIF) and tolerance is often used to evaluate collinearity Effect size =
1 − R2
of the predictors. Hair, Sarstedt, Ringle, and Mena (2011) noted that VIF
values of 5 or above indicate critical collinearity issues among the var­ where, R2 is the coefficient of determination.
iables whilst Field (2012) noted that tolerance values (>0.10) are The effect size calculated from the model shows a strong effect as
desirable. However, collinearity issues can also occur at lower VIF depicted by Bliwise (2006).
values of 3 (Becker, Ringle, Sarstedt, & Völckner, 2015). Ideally, the VIF
values should be close to 3 and lower. From the model results, VIFs 4.5.4. The predictive relevance (Q2)
values of the most variables (Table 8) were below the rules of thumb of 5 In addition to R2 as a predictive criterion, Hair et al. (2017)
and tolerance values were ranging from 0.26 to 0.5 which moderately
approves the multicollinearity assumption.
Table 9
Effect size.
Exogenous Latent Variables Effect Size Total Effect
Table 7 NE→DI 1.33 Strong
Variance Accounted displayed. DI→CS 2.85 Strong
NE →DJ 1.33 Strong
Path VAR Ranking
DJ→CS 2.85 Strong
NE→ DI→ CS 0.375 Partial mediation NE→PJ 1 Strong
NE →DJ → CS 0.487 Partial mediation PJ→CS 2.85 Strong
NE → PJ→ CS 0.6 Partial mediation NE→IJ 1.33 Strong
NE → IJ→ CS 0.394 Partial mediation IJ→CS 2.85 Strong

Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ), Key: Disconfirmation (DI), Distributive Justice (DJ), Procedural Justice (PJ),
Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS). Interactional Justice (IJ), Negative Emotions (NE) and Customer Satisfaction (CS).
More details on these are found in Appendix 1. More details on these are found in Appendix 1.
Source: Field Data (2021) Source: Field Data (2021)

8
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

recommend that researchers examine Q2 to assess the predictive rele­ Table 11

vance of the structural model. Chin (1998) mentions that the predictive Goodness of Fit index calculation.
relevance of constructs must be positive and with values greater than Construct AVE R2
zero; so also, Hair et al. (2011). The size of the Q2 effect allows the
NE 0.634
evaluation of how an exogenous construct contributes to an endogenous DI 0.687 0.57
latent construct Q2 as a measure of predictive relevance, which can be DJ 0.694 0.57
small (0.02), medium (0.15) or large (0.35). The Q2 values for this study PJ 0.826 0.50
model was 0.532 which was higher than the threshold limit and supports IJ 0.653 0.57
CS 0.702 0.74
that the path model’s predictive relevance was adequate for the AVE 0.699
endogenous construct. AVE × R2 0.412
The Standardised Root Mean Square Residual (SMRMR) is an index GoF 0.642
of the average of standardised residuals between the observed and the Source: Field Data (2021)
hypothesised covariance matrices (Chen, 2007). SRMR is a measure of
the estimated model fit. When SRMR ≤ 0.08, then the study model has a
(NE) had a positive effect on Customer Satisfaction (CS). For customers
good fit (Hu, 1998), with a lower SRMR being a better fit. According to
and other stakeholders to be able to evaluate the level of satisfaction in
Table 10 results, the SRMR value for the fitted model is 0.063 which is
the e-service offered via e-banking platforms (Nyagadza et al.,
less than the threshold value of 0.08, suggesting that the model can be
2021cortant role (Bagozzi et al., 1999). Negative Emotions (NE) had a
accepted. Furthermore, the NFI value for the model is 0.804 which is
positive effect on Disconfirmation (DI) (He et al., 2016), Negative
slightly above the recommended threshold value of 0.9. These results
Emotions (NE) had a positive effect on Distributional Justice (DJ),
suggest that the fitted model is a good model, whereas the Chi-Square
Negative Emotions (NE) had a positive effect on Procedural Justice (PJ)
was equal to 3482.826.
(Kau & Loh, 2006) and Negative Emotions (NE) had a positive effect on
Interactional Justice (IJ) (Patterson et al., 2006). Customers or people in
4.6. Overall assessment general, who are in negative state tend to see things negatively (Clark &
Isen, 1985), are very pessimistic and their behaviours reflect this
Goodness of Fit (GoF) is defined as the geometric mean of both negativity in their expectations (Dickinson-Delaporte et al., 2010; He
average variances extracted (AVE) and the average of R2 of all endog­ et al., 2016). Thus, emotions affect customer e-banking satisfaction
enous variables (Akter, D’Ambra, & Ray, 2011). PLS results can be evaluations directly (Nyagadza, 2022; Varela-Neira et al., 2008). The
assessed globally for the overall mode and locally for the measurement better the customers’ evaluation, the more they are satisfied and become
model and the structural model (Henseler, 2017). The formula for loyal to their banks. As such, the way complaints are solved has an
calculating GoF was adopted from Akter et al. (2011) as follows: important signal in constructing future long-term relationships between
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ the banks and their customers.
GoF = AVE × R2
The criteria of GoF (in Table 11) to decide whether GoF values are 6. Conclusion
not fit, small, medium, or large to be considered as globally valid PLS
model are given by Akter et al. (2011) as GoF less than 0.1 (not fit), GoF Major lesson learnt in the current study points to the fact that general
between 0.1 and 0.25 (small), GoF between 0.25 and 0.36 (medium) and errors committed during the e-banking service are unavoidable due to
GoF greater than 0.36 (large). Therefore, the GoF value for this study is complications in the e-delivery process of the promise as a consequence
0.642 which is above 0.36 as indicated (Akter et al., 2011). This proves of alternating customer expectations, simultaneous production and
that the developed model is large in explaining emotions’ influence on consumption of the e-banking service. Therefore, banks’ leaders are
customers’ e-banking satisfaction evaluation in e-service failure and encouraged to understand customers’ cognitive and affective psycho­
e-service recovery circumstances. logical needs in order to counter competition efficiently. The importance
of the current study is that it closes gaps and adds value to the existing e-
5. Discussion banking, services marketing and justice theory literature and body of
knowledge. Customers may experience proper satisfaction if the bank’s
The results depicted that Disconfirmation (DI) had a negative effect feedback on e-banking is in tandem with their expectations. The
on Customer Satisfaction (CS). This is due to the fact that e-banking judgement formed by customers in line with e-service recovery process
service recovery response can negatively disconfirm expectations which have a direct impact on customers’ satisfaction. As the world drives
may reduce customer satisfaction (Varela-Neira et al., 2008). Distribu­ towards digitalisation of work life, as a result of the dynamics of the
tional Justice (DJ) had a positive effect on Customer Satisfaction (CS). Fourth Industrial Revolution (4IR) and the rise of the dangerous pan­
Customer evaluative judgements’ formation on a bank’s responses to demics, banks are pushing for the agenda of adopting e-service via e-
e-service failure have indicated that perceived justice is an integral banking to minimise direct contact with customers, despite the costs in
concept, in prior research (Homburg & Furst, 2005). Procedural Justice e-banking service recovery. Theoretical, practical, and future research
(PJ) had a positive effect on Customer Satisfaction (CS) and Interac­ implications as well as limitations of the study are discussed in the
tional Justice (IJ) had a positive effect on Customer Satisfaction (CS) following section.
(Blodgett et al., 1997). It can be further noted that Negative Emotions
7. Theoretical implications
Table 10
The study contributes to theory as it supports the use of Justice
Goodness of Fit (GoF) results.
Theory in explaining the complexity of the e-banking service recovery
Estimated Model
process. Since the Justice Theory has been used in service recovery
SRMR 0.063 studies to understand customer satisfaction, this study takes a step
d_ULS 11.483 further in also investigating customer satisfaction in e-banking service,
d_ G 15.905
together with the negative effect initially on customers. The dimensions
Chi-Square 3482.826
NFI 0.804 of Justice Theory (distributive, procedural and interactional justice)
have not been widely used in the e-banking service failure and e-banking
Source: Field Data (2021)

9
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

service recovery. There is a paucity of theoretical and empirical research 9. Study limitations
studies in Zimbabwe on emotions’ influence on customers’ e-banking
satisfaction evaluation in e-service failure and e-service recovery. This The study has limitations which may affect the generalisability of the
current study was carried out in order to fill this gap of knowledge. Prior results, since they can only be applied to the population studied. Com­
studies have made inquiries on general service recovery on customer plementary research studies can be done in other parts of the world to be
satisfaction (for example, Blut, 2016; Budianto, 2019; Chiou et al., 2020; able to come up with cross-cultural comparisons, as well as methodo­
Firend & Abadi, 2014). Due to the support of all the research study logical validation.
hypotheses, this proposes the direct link between the emotions’ influ­
ence on customers’ e-banking satisfaction evaluation in e-service failure 10. Future research implications
and e-service recovery circumstances. Thus, this study extends the cur­
rent body of knowledge on justice theory by depicting that emotions In future, longitudinal research studies can be made in order to check
positively affect customers’ e-service failure and recovery evaluations different variations of economic situations in other relevant studies.
(Matikiti et al., 2018), no matter the changes that may take place in any Future research studies can include evaluating other relevant theoretical
banking environment. The research shows that disconfirmation of ex­ frameworks in e-service failure and recovery than Expectations Dis­
pectations and perceived justice in e-banking service recovery (cognitive confirmation theory. Despite the limitations of the current study, the
antecedents) and negative emotions experienced, to a greater extent, are results have contributed to the better understanding of e-service failure
as a result of e-banking service failure (affective antecedents), and that and recovery in e-banking. The results hopefully may influence further
there is a link between cognitive and affective antecedents (Kuo & Wu, future research study inquiries.
2012). The model developed in the current study depicts this.
Funding
8. Practical implications
This research did not receive any specific grant from funding
From a practical perspective, the study contributes by incubating agencies in the public, commercial, or not-for-profit sectors.
insights to banks in developing and emerging economies to have an
improved understanding of how Justice Theory can strengthen customer CRediT authorship contribution statement
satisfaction through effective e-banking service delivery during the re­
covery process. Further to this, banks are urged to offer the proper e- Brighton Nyagadza: Conceptualization, Funding acquisition,
banking customer services to timeously resolve any e-service failure, in a Investigation, Methodology, Project administration, Supervision,
bid to build satisfaction and loyalty. To improve procedural justice, Writing – original draft, Writing – review & editing. Gideon Mazuruse:
practices, senior bank managers and customer service employees need to Data curation, Formal analysis, Investigation, Methodology, Software,
focus their efforts on superior-quality e-banking service delivery. Visualization. Asphat Muposhi: Conceptualization, Writing – review &
Appropriate interactional justice for e-service recovery may call for editing. Tinashe Chuchu: Conceptualization, Writing – review & edit­
making online apologies, empathising, attentiveness, courteous ing. Tendai Makoni: Data curation, Formal analysis, Methodology,
appearance, and being respectful to the affected customers. Where Software, Validation, Visualization. Brain Kusotera: Data curation,
possible, offering compensation as a form of distributive justice, is Formal analysis, Methodology, Software, Validation, Visualization.
essential. Further to this, attending to queries on time when they are
needed will reduce disloyalty and general negative emotional brand
attachment. Investing in e-banking quality is absolutely necessary by Declaration of interest statement
targeting the e-service elements as proposed by Lovelock and Wirtz
(2011). Improper and ineffective e-banking service handling results in The researchers declare that the research was conducted in the
further e-banking service failures, worsening customer dissatisfaction, absence of any commercial or financial relationships that could be
customer switching, negative word-of-mouth, reducing trust and double construed as a potential conflict of interest.
deviance. Loyalty e-programmes are also necessary in shaping customers
satisfaction in e-banking service. On the other hand, higher probability Acknowledgement
of e-banking service failure causes an increase in customers’ risk
perception. Managers are also encouraged to provide for the requisite Gratitude is given to the customers of the sampled banks listed on the
training and development on dealing with e-banking service failures for Zimbabwe Stock Exchange (ZSE), who gave the researchers author­
a sustainable e-service recovery process. isation to carry out the surveys.

Measurement Instrument Appendices

Appendix 1. Items to measure Negative EmotionsThink about the way you felt when the problem aroused. Indicate the degree to which you have
experienced the following emotions. (Likert: 1 = not at all, 7 = very much)

1 2 3 4 5 6 7

EM1 Angry
EM2 Humiliated
EM3 Disappointed
Source: Liljander and Strandvik (1997), Dean (2001), and Varela-Neira et al. (2008).

10
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Appendix 2. Items to measure Disconfirmation of ExpectationsIndicate your agreement level with the following statements (Likert: 1 = much worse, 7
= much better)

Code Construct 1 2 3 4 5 6 7

DE1 Bank’s response in this occasion has been ———— than expected.
DE2 I expected something ———— than the e-banking service I have obtained from the bank in this particular occasion.
DE3 When I compare this experience with the bank with my expectations, I believe the e-banking service experience has been ————.
Source: Hess et al. (2003), Smith and Mpinganjira (2018), Smith et al. (1999).

Appendix 3. Items to measure Distributive Justice Indicate your agreement level with the following statements (Likert: 1 = total disagreement, 7 = total
agreement)

Code Construct 1 2 3 4 5 6 7

DJ1 Given the inconvenience caused by the problem and the time lost, the response (e-banking service) I received from the bank has been
correct.
DJ2 The bank has been quite fair when solving the problem.
DJ3 The outcome I received from the bank in response to the problem in the e-banking service performance has been adequate.

Source: Homburg and Furst (2005), Matikiti et al. (2018) and Maxham and Netemeyer (2003).

Appendix 4. Items to measure Procedural Justice Indicate your agreement level with the following statements (Likert: 1 = total disagreement, 7 = total
agreement)

Code Construct 1 2 3 4 5 6 7

PJ1 The bank has given me the opportunity to explain my point of view of the problem.
PJ2 The bank has fair policies and practices to handle the problem.
PJ3 The bank has shown adequate flexibility in dealing with the problem.
Source: Homburg and Furst (2005), Maxham and Netemeyer (2003) Smith and Mpinganjira (2018) and Smith et al. (1999).

Appendix 5. Items to measure Interactional Justice Indicate your agreement level with the following statements (Likert: 1 = total disagreement, 7 = total
agreement)

Code Construct 1 2 3 4 5 6 7

IJ1 In response to the problem, the bank personnel via e-banking service platform has treated me with courtesy.
IJ2 Bank employees’ communication and care via e-banking service when solving the problem has been appropriate.
IJ3 Bank employees’ have been honest and ethical via e-banking service when solving the problem.
Source: Homburg and Furst (2005), Matikiti et al. (2018), Maxham and Netemeyer (2003), Smith et al. (1999), and Varela-Neira et al. (2008).

Appendix 6. Items to measure Customer Satisfaction Indicate your agreement level with the following statements (Likert: 1 = strongly disagree, 7
= strongly agree)

Code Construct 1 2 3 4 5 6 7

CS1 I am satisfied with e-banking services provided this bank.

CS2 I am satisfied with my decision of choosing this bank.
CS3 I am satisfied with my e-banking service relationship through time with this bank.
CS4 I am satisfied by my bank’s adequate e-banking services.
CS5 I am satisfied by my bank’s promise fulfilment.
Source: Makanyeza and Chikazhe (2017), Yu and Dean (2001), Maxham and Netemeyer (2003), Smith and Mpinganjira (2018), Oliver (1997), and Varela-Neira et al.
(2008).

References Alexander, J. R., Houghton, D. C., Twohig, M. P., Franklin, M. E., Saunders, Neal-
Barnett, A. M., et al. (2016). Factor analysis of the Milwaukee inventory for subtypes
of trichotillomania-adult version. Journal of obsessive-compulsive and related disorders,
Bliwise, N, G (2006). Web-based tutorials for teaching introductory statistics. Journal of
11, 31–38.
Educational Computing Research, 33(3), 309–325.
Armstrong, J. S., & Overton, T. S. (1977). Estimating non-response bias in mail surveys.
Adams, J. S. (1963). Toward an understanding of inequity. Journal of Abnormal and Social
Journal of Marketing Research, 14(3), 396–402.
Psychology, 67, 422–436.
Attallah, N. F. (2015). Evaluation of perceived service quality provided by tourism
Adams, J. S. (1965). Inequity in social exchange. In L. Berkovitz (Ed.), 2. Advances in
establishments in Egypt. Tourism and Hospitality Research, 15(3), 149–160.
experimental social psychology (pp. 267–299). New York: Academic Press.
Bagozzi, R. P., Gopinath, M., & Nyer, P. U. (1999). The role of emotions in marketing.
Akter, S., D’Ambra, J., & Ray, P. (2011). Trustworthiness in mHealth information
Journal of the Academy of Marketing Science, 27, 184–206.
services: An assessment of a hierarchical model with mediating and moderating
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal
effects using partial least squares (PLS). Journal of the American Society for
of the Academy of Marketing Science, 16(1), 74–94.
Information Science and Technology, 62(1), 100–116.
Bagozzi, R. P., & Yi, Y. (2012). Specification, evaluation, and interpretation of structural
Alalwan, A. A., Dwivedi, Y. K., Rana, N. P., & Algharabat, R. (2018). Examining factors
equation models. Journal of the Academy of Marketing Science, 40, 8–34.
influencing jordanian customers’ intentions and adoption of internet banking:
Becker, J.-M., Ringle, C. M., Sarstedt, M., & Völckner, F. (2015). How collinearity affects
Extending UTAUT2 with risk. Journal of Retailing and Consumer Services, 40,
mixture regression results. Marketing Letters, 643–659.
125–138.
Blau, P. M. (1964). Exchange and power in social life. New York: Wiley.

11
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Blodgett, J. G., Hill, D. J., & Tax, S. S. (1997). The effects of distributive, procedural, and He, A. Z., Cai, L., & Gao, J. (2016). A study about the influence of brand media content
interactional justice on post complaint behaviour. Journal of Retailing, 73(2), marketing to consumer brand attitudes. Chinese Journal of Management, 13(10),
185–210. 1534–1545. In Chinese.
Blut, M. (2016). E-service quality: Development of a hierarchical model. Journal of Henseler, J. (2017). Bridging design and behavioral research with variance-based
Retailing, 92(4), 500–517. structural equation modeling. Journal of Advertising, 46(1), 178–192.
Budianto, A. (2019). Customer loyalty: Quality of service. Journal of Management Review, Hess, R. L., Ganesan, S., & Klein, N. M. (2003). Service failure and recovery: The impact
3(1), 299–305. of relationship factor on customer satisfaction. Journal of the Academy of Marketing
Chang, C.-S., Chen, S.-Y., & Lan, Y.-T. (2013). Service quality, trust, and patient Science, 31(2), 127–145.
satisfaction in interpersonal-based medical service encounters. BMC Health Services Hoffman, K. D., & Kelley, S. W. (2000). Perceived justice needs and recovery evaluation:
Research, 13(1), 22. A contingency approach. European Journal of Marketing, 34(3/4), 418–432.
Chang, H. H., Lai, M.-K., & Hsu, C.-H. (2012). Recovery of online service: Perceived Hoffman, K. D., Kelley, S. W., & Rotalsky, H. M. (2016). Retrospective: Tracking service
justice and transaction frequency. Computers in Human Behavior, 8(6), 2199–2208. failures and employee recovery efforts. Journal of Services Marketing, 30(1), 7–10.
Chan, L. L., & Idris, N. (2017a). Validity and reliability of the instrument using Homans, G. C. (1961). Social behaviour its elementary forms. New York, NY: Harcourt,
exploratory factor analysis and Cronbach’s alpha. International Journal of Academic Brace, and World.
Research in Business and Social Sciences, 7(10), 400–410. Homburg, C., & Furst, A. (2005). How organizational complaint handling drives
Chan, L. L., & Idris, N. (2017b). Validity and reliability of the instrument using customer loyalty: An analysis of the mechanistic and the organic approach. Journal
exploratory factor analysis and Cronbach’s alpha. International Journal of Academic of Marketing, 69, 95–114.
Research in Business and Social Sciences, 7(10), 400–410. Hooper, D., Coughlan, J., & Mullen, M. (2008). Structural equation modelling:
Chin, W. W. (1998). The partial least squares approach to structural equation modeling. Guidelines for determining model fit. Electronic Journal of Business Research Methods,
In G. A. Marcoulides (Ed.), Modern methods for business research (pp. 295–358). 6(1), 53–60.
Mahwah, NJ: Lawrence Erlbaum Associates. Hu, L, & Bentler, P. M.. (1998). Fit indices in covariance structure modeling: Sensitivity
Chiou, M.-R., Chao, S.-L., & Hsieh, H.-Y. (2020). The moderating role of service recovery to underparameterized model misspecification. Psychological Methods, 3(4),
on customer loyalty in the context of cruise passengers. Maritime Policy & 424–453.
Management, 1–18. Joireman, J., Gr_egoire, Y., Devezer, B., & Tripp, T. M. (2013). When do customers offer
Chou, P.-F. (2015). An analysis of the relationship between service failure, service firms a “second chance” following a double deviation? The impact of inferred firm
recovery and loyalty for low cost carrier travellers. Journal of Air Transport motives on customer revenge and reconciliation. Journal of Retailing, 89(3),
Management, 17, 119–125. 315–337.
Clark, M., & Isen, A. (1985). Toward understanding the relationship between feeling Jose, A., & Mathew, S. (2016). Initial negativity-A moderator in customer satisfaction
status and social behavior. In A. Hastorf, & A. Isen (Eds.), Cognitive social psychology during the service recovery process in the life insurance context. Rajagiri Management
(pp. 73–108). New York: Elsevier/North-Holland. Journal, 10(1), 15–36.
Clemmer, E. C. (1993). An investigation into the relationships of justice and customer Kahneman, D., & Tversky, A. (1979). On the interpretation of intuitive probability: A
satisfaction with services. In R. Cropanzano (Ed.), Justice in the workplace: reply to Jonathan Cohen. Cognition, 7(4), 409–411.
Approaching fairness in human resources management (pp. 193–207). Hillsdale, NJ: Kau, A., & Loh, E. W. (2006). The effects of service recovery on consumer satisfaction: A
Erlbaum. comparison between complainants and non-complainants. Journal of Services
Cude, B., Lawrence, F., Lyons, A., Metzger, K., LeJeune, E., Marks, L., et al. (2006). Marketing, 20(2), 101–111.
College students and financial literacy: What they know and what we need to learn. Keller, C. M., & Kros, J. F. (2011). An innovative Excel application to improve exam
Proceedings of the Eastern Family Economics and Resource Management Association, 102 reliability in marketing courses. Marketing Education Review, 21(1), 21–28.
(9), 106–109. Koc, E. (2019). Service failures and recovery in hospitality and tourism: A review of
Da Costa Carvalho, P. D. (2015). An integrated conceptual model of destination branding literature and recommendations for future research. Journal of Hospitality Marketing
– Touristmind. Journal of Tourism Management Research, 2(2), 24–40. & Management, 28(5), 513–537.
De Souza, M. F., & Desai, P. H. (2013). Model of justice in service recovery efforts in Komunda, M., & Osarenkhoe, A. (2012). Remedy or cure for service failure: Effects of
airline expectations of service. American Journal of Economics, 3(6), 330–335. service recovery on customer satisfaction and loyalty. Business Process Management
Deng, Z., Lu, Y., Weib, K. K., & Zhang, J. (2010). Understanding customer satisfaction Journal, 18(1), 82–103.
and loyalty: An empirical study of mobile instant messages in China. International Krejcie, R. V., & Morgan, D. W. (1970). Determining sample size for research activities.
Journal of Information Management, 30(1), 289–300. Educational and Psychological Measurement, 30, 607–610.
Dickinson-Delaporte, S., Beverland, M., & Lindgreen, A. (2010). Building corporate Kuo, Y.-F., & Wu, C.-M. (2012). Satisfaction and post-purchase intentions with service
reputation with stakeholders: Exploring the role of message ambiguity for social recovery of online shopping websites: Perspectives on perceived justice and
marketers. European Journal of Marketing, 44(11/12), 1856–1874. emotions. International Journal of Information Management, 32, 127–138.
Ding, M., & Lii, Y. (2016). Handling online service recovery: Effects of perceived justice Lewis, B. R., & Spyrakopoulos, S. (2001). Service failures and recovery in retail banking:
on online games. Telematics and Informatics, 33, 881–895. The customers’ perspective. International Journal of Bank Marketing, 19(1), 37–47.
Effendi, M., Matore, E. M., Khairani, A. Z., & Adnan, R. (2019). Exploratory factor Liljander, V., & Strandvik, T. (1997). Emotions in service satisfaction. International
Analysis (EFA) for adversity quotient (AQ) instrument among Youth. Journal of Journal of Service Industry Management, 8(2), 148–169.
Critical Reviews, 6(6), 234–242. Lind, E. A., & Tyler, T. R. (1988). The social psychology of procedural justice. New York:
Erevelles, S., & Leavitt, C. (1992). A comparison of current models of consumer Plenum Press.
satisfaction/dissatisfaction. Journal of Consumer Satisfaction, Dissatisfaction and Lopes, E. L., & Silva, M. A. (2015). The effect of justice in the history of loyalty: A study in
Complaining Behavior, 5, 104–114. failure recovery in the retail context. Journal of Retailing and Consumer Services, 24,
Field, A., Miles, J., and Field, Z. (2012) Field, A (2012). Discovering statistics using R. 110–120.
London: Sage. Lovelock, C., & Wirtz, J. (2011). Services marketing: People, technology, strategy (7th ed.).
Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R. London: Sage. Upper Saddle River, NJ: Pearson Education/Prentice Hall.
Firend, A. R., & Abadi, M. F. (2014). Impact of service quality, trust and perceived value Maat, S., Zakaria, E., Nordin, N., & Meerah, T. (2011). Confirmatory factor analysis of the
on customer loyalty in Malaysia services industries. Procedia-Social and Behavioural mathematics teachers’ teaching practices instrument. World Applied Sciences Journal,
Sciences, 164, 298–304. 12(11), 2092–2096.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with Makanyeza, C., & Chikazhe, L. (2017). Mediators of the relationship between service
unobservable variables and measurement error. Journal of Marketing Research, 18(1), quality and customer loyalty. Evidence from the banking sector in Zimbabwe.
39–50. International Journal of Bank Marketing, 35(3), 540–557. https://doi.org/10.1108/
Gerald, B. (2018). A brief review of independent, dependent and one sample t-test. IJBM-11-2016-0164
International Journal of Applied Mathematics and Theoretical Physics, 4(2), 50. Marketing Research Society (MRS). (2022). Marketing research ethics. Retrieved from
Gilliland, S. W. (1993). The perceived fairness of selection systems: An organizational https://www.mrs.org.uk/.
justice perspective. Academy of Management Review, 18(4), 694–734. Mathew, S., Jose, A., Rejikumar, G., & Chacko, D. P. (2020). Examining the relationship
Gountas, S., Gountas, J., & Mavondo, F. (2013). Exploring the associations between between e-service recovery quality and e-service recovery satisfaction moderated by
standards for service delivery (organisational culture), co-worker support, self- perceived justice in the banking context. Benchmarking: An International Journal, 27
efficacy, job satisfaction and customer orientation in the real estate industry. (6), 1951–1980.
Australian Journal of Management, 39(2), 107–126. Matikiti, R., Roberts-Lombard, M., & Mpinganjira, M. (2018). The influence of perceived
Gupta, S., Yun, H., Xu, H., & Kim, H.-W. (2017). An exploratory study on mobile banking justice on recovery satisfaction in the airline industry. Journal of African Business, 19
adoption in Indian metropolitan and urban areas: A scenario-based experiment. (4), 512–530.
Information Technology for Development, 23(1), 127–152. Mattila, A. S. (2001). The effectiveness of service recovery in a multi-industry setting.
Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A primer on partial least Journal of Services Marketing, 15(7), 583–596.
squares structural equation modeling (PLS-SEM) (2nd ed.). Thousand Oaks, CA: Sage Mattila, A. S., & Cranage, D. (2005). The impact of choice on fairness in the context of
Publications Inc. service recovery. Journal of Services Marketing, 19(5), 271–279.
Hair, J. F., Sarstedt, M., Ringle, C. M., & Mena, J. A. (2011). An assessment of the use of Maxham, J. G., & Netemeyer, R. G. (2002). Modelling customer perceptions of complaint
partial least squares structural equation modeling in marketing research. Journal of handling over time: The effects of perceived justice on satisfaction and intent.
the Academy of Marketing Science, 40(3), 414–433. Journal of Retailing, 78, 239–252.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional ProcessAnalysis: Maxham, J. G., & Netemeyer, R. G. (2003). Firms reap what they sow: The effects of
A regression-based approach. New York, NY: The Guilford Press. shared values and perceived organizational justice on customers’ evaluations of
complaint handling. Journal of Marketing, 67, 46–62.

12
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Menon, K., & Dube, L. (2000). Ensuring greater satisfaction by engineering salesperson Smith, A., & Mpinganjira, M. (2015). The role of perceived justice in service recovery on
response to customer emotions. Journal of Retailing, 76(3), 285–307. banking customers’ satisfaction and behavioral intentions: A case of South Africa.
Menshikova, M., Romolini, A., Sabbatelli, I., & De Marco, M. (2017). The role of digital Banks and Bank Systems, 10(2), 35–43.
tools and platforms for training programmes developed by the organisations of the Sousa, R., & Voss, C. A. (2009). The effects of service failures and recovery on customer
banking sector. May. In International conference on exploring services science (pp. loyalty in eservices: An empirical investigation. International Journal of Operations &
309–322). Cham: Springer. Production Management, 29(8), 834–864.
Muposhi, A., Nyagadza, B., & Mafini, C. (2021). Fashion designers’ attitude-behaviour Stratemeyer, A. W., Geringer, S. D., & Canton, A. C. (2014). An exploratory investigation
inconsistencies towards a sustainable business model: A neutralization theory of the effects of service failure and recovery efforts on customer satisfaction.
perspective. Journal of Fashion Marketing and Management, 1–20. American Journal of Management, 14(3), 20–28.
Myers, S., & Majluf, N. (1984). Corporate financing and investment decisions when firms Szymanski, D. M., & Henard, D. H. (2001). Customer satisfaction: A meta-analysis of the
have information that investors do not have. Journal of Financial Economics, 13, empirical evidence. Journal of the Academy of Marketing Science, 29(1), 16–35.
187–221. Tan, T. A. G. (2014). Integrating justice theory and service quality concept to evaluate
Nguyen, D. T., McColl-Kennedy, J. R., & Dagger, T. S. (2012). Matching service recovery customer recovery satisfaction. IAMURE International Journal of Business and
solutions to customer recovery preferences. European Journal of Marketing, 46(9), Management, 8(1), 24–36.
1171–1194. Taylor, S. (1994). Waiting for service: The relationship between delays and evaluations
Nikbin, D., Ismail, I., Marimuthu, M., & Armesh, H. (2012). Perceived justice in service of service. Journal of Marketing, 58, 56–69.
recovery and switching intention: Evidence from Malaysian mobile Varela-Neira, C., Vázquez-Casielles, R., & Iglesias-Argüelles, V. (2008). The influence of
telecommunication industry. Management Research Review, 35(3/4), 309–325. emotions on customer’s cognitive evaluations and satisfaction in a service failure
Nunnally, J. C. (1967). Psychometric theory. New York: McGraw Hill Book Company. and recovery context. Service Industries Journal, 28(4), 497–512.
Nyagadza, B. (2019). Responding to change and customer value improvement: Pragmatic Westbrook, R. A., & Oliver, R. L. (1991). The dimensionality of consumption emotion
advice to banks. The Marketing Review, 19(3–4), 235–252. patterns and consumer satisfaction. Journal of Consumer Research, 18, 84–91.
Nyagadza, B. (2022). Digital disruption and global businesses’ viability under COVID-19 Wirtz, J., & McColl-Kennedy, J. R. (2010). Opportunistic customer claiming during
(SARS-CoV-2) pandemic. Journal of Digital & Media Policy (JDMP), 1–27. service recovery. Journal of the Academy of Marketing Science, 38, 564–675.
Nyagadza, B., Kadembo, E. M., & Makasi, A. (2020a). Exploring internal stakeholders’ Worthington, S., Russell-Bennett, R., & Härtel, C. (2010). A tri-dimensional approach for
emotional attachment & corporate brand perceptions through corporate storytelling auditing brand loyalty. Journal of Brand Management, 17, 243–253.
for branding. Cogent Business & Management, 7(1), 1–22. Wu, I. L., & Huang, C. Y. (2015). Analysing complaint intentions in online shopping: The
Nyagadza, B., Kadembo, E. M., & Makasi, A. (2021c). When corporate brands tell stories: antecedents of justice and technology use and the mediator of customer satisfaction.
A signalling theory perspective. Cogent Psychology, 8(1), 1–30. Behaviour & Information Technology, 34(1), 69–80.
Nyagadza, B., Muposhi, A., Mazuruse, G., Makoni, T., Chuchu, T., Maziriri, E. T., et al. Yeoh, P. L., Woolford, S. W., Eshghi, A., & Butaney, G. (2014). Customer response to
(2022). Prognosticating chatbots’ anthropomorphic usage intention as an e-banking service recovery in online shopping. Journal of Services Research, 14(2), 33–56.
customer service gateway: Cogitations from Zimbabwe. PSU Research Review (PRR), Yu, Y., & Dean, A. (2001). The contribution of emotional satisfaction to consumer
1–15. loyalty. International Journal of Service Industry Management, 12(3), 234–250.
Nyagadza, B., Pashapa, R., Chare, A., Mazuruse, G., & Hove, P. K. (2022). Digital Zeelenberg, M., & Pieters, R. (2004). Beyond valence in customer dissatisfaction: A
technologies, Fourth Industrial Revolution (4IR) & Global Value Chains (GVCs) review and new findings on behavioural responses to regret and disappointment in
nexus with emerging economies’ future industrial innovation dynamics. Cogent failed services. Journal of Business Research, 57, 445–455.
Economics & Finance, 9(1), 1–23. Zeithaml, V., Berry, L., & Parasuraman, A. (1993). The nature and determinants of
Nyamutowa, C., & Masunda, S. (2013). An analysis of credit risk management practices customer expectations of service. Journal of the Academy of Marketing Science, 21,
in commercial banking institutions in Zimbabwe. International Journal of Economic 1–12.
Research, 4(1), 31–46.
Oliver, R. L. (1997). Satisfaction. A behavioural perspective on the consumer. Boston, MA:
Brighton Nyagadza is a full time lecturer and A/Chairperson department of Marketing
Irwin/McGraw-Hill.
(Digital Marketing) at Marondera University of Agricultural Sciences & Technology
Orsingher, C., Valentini, S., & Angelis, M. (2010). A meta-analysis of satisfaction with
(MUAST), Zimbabwe, full member of the Marketers Association of Zimbabwe (MAZ), an
complaint handling in services recovery. Journal of the Academy of Marketing Science,
Associate of The Chartered Institute of Marketing (ACIM), United Kingdom and Power
38, 169–186.
Member of the Digital Marketing Institute (DMI), Dublin, Ireland. He has published several
Patterson, P. G., Cowley, E., & Prasongsukarn, K. (2006). Service failure recovery: The
book chapters in Routledge books of Taylor & Francis Publishers, New York (USA),
moderating impact of individual-level cultural value orientation on perceptions of
Emerald Insight, United Kingdom (UK), Lexington books of the Rowan & Littlefield Pub­
justice. International Journal of Research in Marketing, 23, 263–277.
lishers, Maryland (USA) and in reputable international journals such as Journal of Digital
Piha, L. P., & Avlonitis, G. J. (2015). Customer defection in retail banking: Attitudinal
Media & Policy (Intellect Publishers, Bristol, UK), Journal of Fashion Marketing & Man­
and behavioural consequences of failed service quality. Journal of Service Theory and
agement (Emerald Insight, UK), Journal of Environmental Media (Intellect, Bristol, UK),
Practice, 25(3), 304–326.
Tourism Critiques: Practice & Theory (Emerald Insight, UK), Journal of Asian & African Studies
RBZ. (2021). List of registered banking institutions. Available at: www.rbz.co.zw/bankin
(SAGE, London, UK), PSU Research Review (PRR) (Emerald Insight, UK), Youth & Society
g-institutions.HTML. (Accessed 10 April 2021).
(SAGE, London, UK), Cogent Business & Management, Cogent Economics & Finance, Cogent
Reisinger, Y., & Mavondo, F. (2007). Structural equation modeling. Journal of Travel &
Psychology, Cogent Social Sciences (Taylor & Francis, England & Wales, UK), The Marketing
Tourism Marketing, 21(4), 41–71.
Review (Westburn Publishers, Scotland), and others. Email: brightonnyagadza@gmail.com
Rejikumar, G. (2015). An empirical study on antecedents of perceived service recovery
quality in e-banking context. Journal of Internet Banking and Commerce, 20(3), 1-24.
Rita, P., Oliveira, T., & Farisa, A. (2019). The impact of e-service quality and customer Gideon Mazuruse is a full-time Mathematics and Statistics lecturer in the Teaching &
satisfaction on customer behavior in online shopping. Heliyon, 5(10), 1-14. Learning Institute of the Marondera University of Agricultural Sciences & Technology
Rukasha, T., Nyagadza, B., Pashapa, R., & Muposhi, A. (2021). Covid-19 impact on (MUAST), Zimbabwe. He holds an MSc in Statistics & Operations Research (NUST), BSc
Zimbabwean agricultural supply chains and markets: A sustainable livelihoods Hons in Statistics & Operations Research (GZU), BSc in Mathematics & Computer Science
perspective. Cogent Social Sciences, 7(1), 1–17. (GZU), Post Graduate Diploma in Education (ZOU). Email: gmazuruse@gmail.com
Sarawathi, S. (2016). Service marketing-customer’s expectation and delight. IOSR
Journal of Business and Management, 2, 41–43.
Asphat Muposhi is a Lecturer in the Department of Information & Marketing Sciences,
Saunders, M., Lewis, P., & Thornhill, A. (2009). Research methods for business students”
Faculty of Business Sciences, Midlands State University (MSU), Gweru, Zimbabwe. He
(5th ed.). Pearson Education Limited.
holds a PhD in Marketing Management from the University of Johannesburg (UJ), South
Schoefer, K. (2008). The role of cognition and affect in the formation of customer
Africa. His research interests are in sustainable development and green supply chain
satisfaction judgements concerning service recovery encounters. Journal of Consumer
management. E-mail: muposhia@msu.ac.zw
Behaviour, 7(3), 210–221.
Schumacher, A., Erol, S., & Sihn, W. (2016). A maturity model for assessing industry 4.0
readiness and maturity of manufacturing enterprises. Procedia CIRP, 52, 161–166. Tinashe Chuchu holds a Doctorate in Marketing from the University of the Witwa­
Schumacker, R. E., & Lomax, R. G. (2010). A beginner’s guide to structural equation tersrand. Currently, he works as a Senior Lecturer in the Marketing Division of the School
modelling. New York, NY: Routledge. of Business Sciences at the University of the Witwatersrand, South Africa. Previously, he
Segars, A. (1997). Assessing the unidimensionality of measurement: A paradigm and worked as a Senior Lecturer in the Department of Marketing Management, University of
illustration within the context of information systems research. Omega International Pretoria, South Africa. He is a consumer behaviour and tourism scholar who has published
Journal of Management Science, 25(1), 107–121. numerous studies in these fields in top journals and has presented at international con­
Seiders, K., & Berry, L. L. (1998). Service fairness: What it is and why it matters. The ferences. Dr Chuchu has published and reviewed for major publishing outlets, which
Academy of Management Executive, 12, 8–20. include Wiley, Taylor & Francis, Elsevier, Emerald Publishing, SAGE and presented at the
Shapiro, T., & Nieman-Gonder, J. (2006). Effect of communication mode in justice-based premier conference for marketing academics (the American Marketing Association Con­
service recovery. Managing Service Quality, 16(2), 124–144. ference) which was held in Chicago, USA in 2019. He is a member of The Academy of
Smith, A. M. (2006). A cross-cultural perspective on the role of emotion in negative Business and Retail Management Conferences, based in the United Kingdom. He was a
service encounters. Service Industries Journal, 26(7), 709–726. recipient of the South African government’s National Research Foundation Doctoral
Smith, A. K., Bolton, R. N., & Wagner, J. (1999). A model of customer satisfaction with Scholarship in 2015. In 2019, Dr Chuchu was awarded the Best Junior Researcher in
service encounters involving failure and recovery. Journal of Marketing Research, 36, Management Sciences at the University of Pretoria, South Africa. He sits on the editorial
356–373. board of the Retail and Marketing Review as well as the African Journal of Business and
Economic Research. Email: tinashe.chuchu@wits.ac.za

13
B. Nyagadza et al. Social Sciences & Humanities Open 6 (2022) 100292

Tendai Makoni is a Mathematics and Statistics Lecturer in the Department of Mathematics Brain Kusotera is a Statistics and Mathematics lecturer and A/Chairperson of the
& Computer Science, School of Natural Sciences, Great Zimbabwe University (GZU), Department of Statistics and Mathematics, Bindura University of Science Education
Masvingo, Zimbabwe. He is a holder of an MSc (Operations Research & Statistics) (NUST) (BUSE), Bindura, Zimbabwe. He is a holder of an MSc (Operations Research & Statistics)
and BSc Hons (Mathematics) (MSU). Email: tpmakoni@gmail.com (NUST) and BSc Hons (Operations Research & Statistics) (NUST). Both degrees are of the
first-class division. Email: kusoterabk@gmail.com

14