67      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

How to cite this article:
Alfi, C. F. (2023). A meta-analysis of the relationship between religiosity and saving 
behaviour. International Journal of Banking and Finance, 18(1), 67-94. https://doi. 
org/10.32890/ ijbf2023.18.1.4

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Coky Fauzi Alfi
State Polytechnic of Sriwijaya, Indonesia

coky.fauzi.alfi@polsri.ac.id

Received: 2/2/2022      Revised: 2/3/2022      Accepted: 30/3/2022     Published: 5/1/2023

ABSTRACT 

The purpose of this study was to synthesize the findings of previous 
studies on the relationship between religiosity and saving behaviour 
by using a meta-analysis approach. It also sought to determine the 
strength of the relationship, besides its direction. Eleven studies which 
met the five criteria and four techniques used in the study were used as 
samples for the meta-analytic analysis. The size of the effect in each 
study was then determined by Pearson’s product-moment correlations 
(r). To estimate the average distribution of relationship true effects, the 
Fisher r-to-z transformation and random-effects methods were used.  
The empirical evidence showed that there was a positive correlation 
between religiosity and saving behaviour. However, according to 
Guilford’s convention, the true effect size (r = 0.303) would mean 
that religiosity had a weak correlation with saving behaviour. It is 
recommended that authorities and financial institutions use the 
findings of this study to develop plans focused on advocating and 
facilitating saving behaviour among religious people.

https://e-journal.uum.edu.my/index.php/ijbf

INTERNATIONAL JOURNAL 
OF BANKING AND FINANCE



68        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Keywords: Meta-analysis, religiosity, saving behaviour, Fisher r-to-z 
transformation, random-effects method.
 
JEL Classification: Z120, G410.

INTRODUCTION

It appears that the global community is becoming more religious. 
Based on a Pew Research Center (2015) survey, all major religions 
are estimated to show a rise in the number of followers by 2050. The 
survey found that 84 percent of the world population was religiously 
affiliated in 2010, with projections predicting that this share would 
rise to 87 percent by 2050. These findings and projections appear to 
contradict the views of several influential scholars, such as Karl Marx, 
Emile Durkheim, and Max Weber, who predicted that religion would be 
less important in various socioeconomic activities as industrialization 
progressed, economic markets expanded, and science, technology, 
and education advanced rapidly (Basedau et al., 2018). 

Furthermore, the findings of various studies in economics (e.g. Azzi 
& Ehrenberg, 1975; Iannaccone, 1998; Iyer, 2016), sociology (e.g. 
Geertz, 1973; Inglehart, 2018; Lenski, 1961), and psychology (e.g. 
Allport & Ross, 1967; Berry et al., 2002; Pargament, 1999) have 
acknowledged the importance of religion in human society. For 
example, it plays an important role in, energising people to work for 
social change, promoting mental well-being, or acting as a social 
control agent. 

The investigation into the relationship between religion and economic 
growth has received considerable attention, ever since Max Weber 
(1905) recognized the significance of religious affiliations in economic 
performance. He argued that the values in Protestant teachings would 
shape their adherents’ work ethic, resulting in professionalism and 
efficiency in economic activities. More than a century after Weber’s 
thesis, a large body of literature has noticed a link between religion 
and macroeconomic prosperity. For instance, it has been discovered 
that religious beliefs, particularly beliefs in hell and heaven, have a 
positive effect on economic attitudes, leading to higher incomes and 
Gross Domestic Product (GDP), and Christianity is the religion with 



    69      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

the greatest impact on economic growth, with Protestants being more 
capitalists than other Christians (Barro & McCleary, 2003; Filipova, 
2012; Guiso et al., 2003; Hayward & Kemmelmeier, 2011).

Moreover, religion has long been associated with the teachings of 
thriftiness and customary living. The research findings however, show 
that there are differences in which religions adhere to the most frugal 
and conventional ways of life. According to Keister (2003), Guiso et 
al. (2003), and Renneboog and Spaenjers (2012), Catholics appeared 
to value frugality and convenient living more than Protestants, 
whereas Arruñada (2010) and Filipova (2012) discovered the opposite. 
Although many studies found a correlation between religious belief 
and thriftiness, their findings are less convincing when used to explain 
a link between religious belief and saving decisions. This is due to the 
distinction between thriftiness and saving decisions. Thriftiness is the 
trait to try and reduce spending, whereas saving decisions are initiated 
by residual income. Therefore, research into how religion influences 
economic behaviour and financial decisions at the microeconomic 
level seems to remain limited (see Klaubert, 2010; Yayeh, 2014). In 
terms of empirical assessments that link individual saving attitudes 
to religious preferences or practices, it needs to be explored further. 
This investigation should be beneficial because it could help us to 
solve pressing issues in the national economy, for instance, pressing 
concerns such as wealth inequality (Bilen, 2016; Keister, 2003) and 
consumerism (Tjahjono, 2014), or even the issues of conserving 
energy and natural resources (Singh et al., 2021).

Since the investigation of religiosity on saving behaviour is an 
emerging research area, the present study is interested in knowing 
the ‘true’ effect size of the relationship between these variables. 
Therefore, this study has performed a meta-analysis to gain a more 
objective, robust, and less biased understanding of the relationship 
between the variables by investigating the distribution of effect sizes. 
A meta-analysis is an approach for aggregating effect-size indices from 
multiple studies (Borenstein et al., 2011). It contributes to answering 
the question of whether the observed variations in effect sizes across 
studies are due to a single population effect size (Law, 1995). 

To date, there has been no other study examining religiosity and saving 
behaviour across samples, methodologies, and time. This study has 
utilized meta-analysis as a quantitative tool to synthesize the findings 



70        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

of previous studies, and to determine the strength of the relationship 
between religiosity and saving behaviour, as well as the direction of 
that relationship. As a result, the strength of the correlation between 
previous findings and their direction, whether positive or negative, 
has been held to the same standards, as long as they meet the inclusion 
criteria set for the study sample. One of the inclusion criteria, for 
example, was that the study sample would include religiosity, religious 
belief, or religious faith as an independent variable. In addition, the 
other objective is to contribute to the growth of the literature in this 
area of study. This study can provide a retrospective summary of the 
existing literature and provide further empirical evidence of the true 
effect of religiosity on saving behaviour. It could help shape new 
research by describing what was already known and synthesizing the 
new body of evidence.

After reviewing previous studies and establishing the inclusion 
criteria, eleven journal articles were identified as the study sample; all 
together these sources provided a total of 1,063 participants coming 
from various locations. More specifically, Yayeh (2014) collected 
samples in Ethiopia, while Ababio and Mawutor (2015) did so in 
Ghana. Satsios and Hadjidakis (2017) gathered data in Greece. In 
Indonesia, questionnaires were administered by Murdayanti et al. 
(2020); Prastiwi (2021); Priyo Nugroho et al. (2017); Wijaya et al. 
(2019). Meanwhile, data was collected in Malaysia by Abdullah and 
Abd. Majid (2001); Ismail et al. (2018); Kassim et al. (2019); Mei Teh 
et al. (2019). As a result, this meta-analytic study was able to generate 
numerous plot functions, such as the forest plot, standardized residual 
plot, and Cook’s Distance plot, as well as measurements, such as the 
random-effect model, heterogeneity statistics, outliers, and influential 
case diagnostics.

LITERATURE REVIEW

The relationship between religiosity and saving behaviour is typically 
measured using one of two methods: methods which are either 
economically or psychologically oriented. In the economic approach, 
the goal is to create forecasts about behaviour as accurately as 
possible. However, this approach often neglects to explore the true 
underlying causes of why individuals behave the way they do (Nyhus, 



    71      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

2017). Researchers who employ this approach rely on secondary data 
surveys, such as the World Values Survey (WVS), the International 
Social Survey Program (ISSP), the Gallup Millennium Survey, the 
Panel Study of Income Dynamics (PSID), or the Konda Araştrma ve 
Danşmanlk, to determine an individual’s religious behaviour. From 
these data sources, researchers discovered that the ‘average’ person’s 
religiosity could be measured in the following five ways, namely 
participation in religious services, belief in heaven and hell, belief in 
the afterlife, faith in God, and self-identification as a religious person 
(Barro & McCleary, 2003). These religious aspects are then examined 
in relation to the adherents’ amount of income or consumption using 
various econometric methodologies so as to understand the significance 
of religiosity in saving behaviour. Klaubert (2010), for example, used 
the PSID to investigate the link between individual saving decisions 
and religiosity, as measured by church attendance, in the United 
States. Using the Konda data survey, Davutyan and Öztürkkal (2016) 
investigated the effect of religious affiliation on financial behaviour 
in Turkey. They discovered however, weak evidence that religious 
people have distinct preferences for saving decisions. This was due to 
there being no difference between religious and non-religious people 
when it comes to saving decisions. Meanwhile, Guiso et al. (2003) 
discovered a link between religious intensity and thriftiness in a cross-
national study using the WVS sample statistics. 

On the other hand, the psychological viewpoint begins from a different 
place. This approach frequently concentrates on psychological factors, 
and examines individual differences rather than average human 
behaviour. Therefore, various explanatory variables and methods have 
been employed in the analysis of the relationship between religiosity 
and saving behaviour, which makes it different from the economic 
approach. The psychological viewpoint often employs primary data 
sources and applies behavioural theories, for example the theory of 
planned behavior (Ajzen, 1991) or social learning theory (Bandura, 
1977). The theory of planned behavior identifies specific factors, 
namely intentions and perceived behavioral control, that can be 
utilised to estimate and describe human behavior in various contexts. 
Intentions are motivational variables that demonstrate how far 
individuals are willing to go and how much effort they intend to put in, 
whereas perceived behavioral control refers to the perception of how 
easy or difficult it is to control an interest (Ajzen, 1991). Meanwhile, 



72        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

attitudes toward behavior, subjective norms, and perceived behavioral 
control can all have an impact on intentions. Furthermore, social 
learning theory is often used as a base theory to describe the role of 
financial literacy in saving behavior. The theory hypothesizes that the 
cognitive abilities of individuals, i.e., knowledge and skills, impact 
on changing their behaviors. This cognitive ability can be learned by 
seeing, imitating, practicing, and processing information from the 
behaviour of others and its environments, including families, friends, 
neighbours, the workplace, or the media. 

The psychological viewpoint also employs religiosity measurement 
scales, such as the orthodoxy measurement (Glock, 1962) or the 
religious orientations (Allport & Ross, 1967). The orthodoxy 
measurement uses the following five scales: belief, practice, 
knowledge, experience, and consequences, and these would inform 
the preferred faith. Belief is an ideological dimension that a religious 
person will adhere to. Prayer, fasting, involvement in special 
sacraments, worship, and other ritualistic activities are included in the 
practice. Knowledge refers to the understanding of the fundamental 
tenets of a religious person’s faith and its sacred scriptures. Experience 
gives a religious emotional experience, and consequences are all of 
the religious prescriptions for what a religious person should do. 

Meanwhile, the measurement of religious orientation uses the 
following two dimensions: intrinsic and extrinsic, and they would 
inform us of the primary motive for life in religion. Those who are 
intrinsically motivated find that their primary motive in religion is 
to live according to their religious convictions and prescriptions. 
However, extrinsically oriented people may find religion useful in a 
variety of ways, including stability and reassurance, social connection 
and diversionary tactics, status, and self-justification. Few researchers 
have adopted this approach in their studies. For example, Priyo 
Nugroho et al. (2017) expanded the theory of planned behaviour by 
including two new variables: religiosity and self-efficacy. They then 
employed Allport and Ross’s (1967) scale for measuring religiosity 
to investigate the saving behaviour of Islamic bank customers. In the 
meantime, Kassim et al. (2019) who used the social learning theoretical 
framework and the religiosity measurement scale which had its root 
in Glock’s (1962) work, discovered that whereas religiosity had no 
effect on saving behaviour, financial literacy did.



    73      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

METHODOLOGY

Criteria and Search Procedure

The samples for the present meta-analytic study were selected because 
they had discussed the influence of religiosity on saving behaviour 
directly. To be included in the meta-analytic sample, the studies must 
fulfil five criteria. They are as follows:

• The studies used religiosity, religious belief, or religious faith 
as an independent variable.

• The studies used saving behaviour or saving habits, saving 
money, or saving decisions as a dependent variable.

• The studies used a quantitative research approach.
• The studies used primary data at a micro analytical level.
• The studies presented the Pearson’s r effect size clearly or 

could be processed using another statistical method.

Studies would be excluded if they had found a relationship between 
religiosity and saving behaviour indirectly. 

Finding studies from various journals, such as journals on economics, 
business, management, finance, marketing, religion, culture, and 
social science, that fit the inclusion and exclusion criteria for a meta-
analysis study was challenging. For example, to avoid the possibility 
that this might turn out to be a time-consuming process, an effective 
search strategy was used from start to finish. The present study has 
implemented four techniques to conduct a wide-ranging literature 
search. They were as follows: (1) deciding search terms and keywords, 
(2) searching for specific phrases, (3) using truncated and wildcard 
searches as well as Boolean logic, and (4) using citation searching. 
Firstly, these terms and keywords were applied in the search process: 
religiosity, religious belief, religious faith, saving behaviour, saving 
habits, saving money, and saving decisions. Secondly, quotation 
marks were used for words which appear next to each other, e.g., 
“religious belief,” “religious faith,” “saving behaviour,” “saving 
habits,” “saving money,” “saving decisions.” Thirdly, the search 
used combined truncation and wildcard searches with Boolean logic, 
e.g., “religio*” AND “saving behavio?r”. Fourthly, articles that were 
cited in other publications were also included in the search. These 
techniques were then employed to search for studies in the various 
research search engines and databases, such as Semantic Scholar, 



74        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Google Scholar, Research Gate, EBSCOhost, ProQuest, JSTOR, and 
Wiley Online Library. 

All potentially relevant titles and abstracts were then saved and 
managed systematically for the next stage, which was screening. 
The first stage of the screening was to import all the references into 
a reference management software package and de-duplicate them. In 
this case, a software called EndNote was used. The next stage was to 
read and identify all the saved articles of study, and filter them out if 
they were irrelevant. The stages of screening resulted in 11 journal 
articles identified as the relevant data selected for the meta-analysis.

Data Extraction

Once screening has been done and all relevant articles selected for 
the study have been identified, the next step is the data extraction (see 
Teshome et al., 2018; Zuckerman et al., 2013). In this process, the key 
aspects that will be used for the statistical meta-analysis will have to 
be extracted from the articles. Some key aspects of the articles are set, 
namely the authors’ name and year of publication, sampling methods, 
measurement techniques, variables identification (independent 
and dependent), methods of statistical analysis, and a summary of 
the results (see Table 1). The characteristics of each article that met 
the criteria for inclusion were also highlighted. For example, the 
eleven studies used various themes related to religiosity and saving 
behaviour as independent and dependent variables, respectively. 
These are described in the column on variables. In another column, 
such as the measurement technique, it is stated that all studies applied 
a questionnaire survey to ensure that primary data was used. The most 
useful information, however, is in the results column. It discusses 
the significance of the relationship between religiosity and saving 
behaviour, as well as various goodness of fit tests, e.g., chi-square, 
odds ratio, or t-statistic, that can be used to compute the effect size r.

Effect Size Computation

Following data extraction, the next task was to determine the size of 
the effect in each study and ensure that this effect size was expressed 
in the same way. The effect sizes are used to describe the strength of 
the relationship between the variables. There are two common types 
of effect size: the r type and the d type. The two most commonly 



    75      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

used of the r type are Pearson’s product-moment correlations (r) and 
Fisher’s r-to-z transformation (Zr), whereas the three most commonly 
used d type are Cohen’s d, Hedges’s g, and Glass’s D (Rosenthal, 
1995). In this study, Pearson’s r was the preferred effect size. In this 
regard, it involved calculating the r value for each study carrying out 
the meta-analysis. There was no need to do anything if a study had 
used the r value. However, because some studies had no effect size 
value and only provided various fit test indicators (e.g., chi-square, 
odds ratio, t-test statistic), a conversion to Pearson’s r was performed 
via an online calculator at www.psychometrica.de/effect_size.html 
(Lenhard & Lenhard, 2016). Meanwhile, if the authors did not provide 
the indicators and could not compute a conversion to the r value, they 
were contacted via email to gain the relevant information. A reminder 
was sent if they did not reply.

Method of Analysis

The analysis is carried out using the Fisher r-to-z transformed 
correlation coefficient as the outcome measure. The Fisher’s r-to-z 
transformation is commonly used because samples from a meta-
analysis contain a variety of effect sizes. It is also to achieve normality 
in the effect sizes (Cheung et al., 2012). There are three steps in 
implementing this method (Borenstein et al., 2011; Field & Gillett, 
2010). To begin, use Fisher’s r-to-z transformation to convert the 
effect size in each study into a standard normal metric. The Fisher’s 
r-to-z transformation formula is given as                                where   
     is the effect size in each study. After that, for each study, a weighted  

average of       scores are computed by the formula                               

where    is the number of studies and     is the sample size. Finally, it 

should be converted back to     using the formula  

In addition, the random-effects statistical model was applied to 
estimate the average distribution of true effects. The random-effects 
method was chosen because the effect size was extracted from a 
series of studies conducted by various authors in various populations 
at various times. The present analysis also reported on the estimate 
of the    index, the H2 index, the I2 index, and the Q-test (Cochran, 
1954) with a p-value as the heterogeneity statistics outcome. The 
Q-test was used to assess the null hypothesis that all effect sizes 

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

three steps to implementing this method (Borenstein et al., 2011; Field & Gillett, 2010). To begin, use 
Fisher's r-to-z transformation to convert the effect size in each study into a standard normal metric. The 
Fisher's r-to-z transformation formula is given by 𝑧𝑧𝑟𝑟𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒 (

1+𝑟𝑟𝑖𝑖
1−𝑟𝑟𝑖𝑖

), where 𝑟𝑟𝑖𝑖 is the effect size in each 

study. After that, for each study, a weighted average of 𝑧𝑧𝑟𝑟 scores are computed by �̅�𝑧𝑟𝑟𝑖𝑖 =
∑ 𝑛𝑛𝑖𝑖𝑧𝑧𝑟𝑟𝑖𝑖
𝑘𝑘
𝑖𝑖=1
∑ 𝑛𝑛𝑖𝑖𝑘𝑘𝑖𝑖=1

, where 𝑘𝑘 
is the number of studies and 𝑛𝑛𝑖𝑖 is the sample size. Finally, it should be converted back to 𝑟𝑟𝑖𝑖 using the 
formula 𝑟𝑟𝑖𝑖 =

𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖−1
𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖+1

.  

In addition, the random-effects statistical model is applied to estimate the average distribution of true 
effects. The random-effects method was chosen because the effect size was extracted from a series of 
studies conducted by various authors in various populations at various times. The analysis also reports the 
estimate of the 2 index, the H2 index, the I2 index, and the Q-test (Cochran, 1954) with a p-value as the 
heterogeneity statistics outcome. The Q-test is used to assess the null hypothesis that all effect sizes from 
all studies are homogenous (Chen & Peace, 2021). If the p-value is less than  (the typical significance 
level is 0.05), the null hypothesis should be rejected, indicating that the effect sizes from all studies are not 
homogenous. Meanwhile, 2, H2, and I2 are used to determine the strength of the distribution of true effect 
sizes. The 2 index is estimated using the Hedges’ estimator (Hedges & Olkin, 1985) to measure the variance 
of the true effect sizes, and the index should be greater than zero. The H2 index is quantified using Higgins 
and Thompson’s (2002) formula to inform the relative extent of heterogeneity in comparison to all studies, 
and the index should be greater than 1. The I2 index is also calculated using Higgins and Thompson's (2002) 
formula to determine the percentage of observed heterogeneity versus real heterogeneity. As a rule of 
thumb, the I2 index could be considered as having low heterogeneity (I2 = 25%), moderate heterogeneity 
(I2 = 50%), and high heterogeneity (I2 = 75%). To display the conclusions of meta-analyses, forest plots are 
generated. Forest plots provide information about each study’s effect size and confidence interval, as well 
as the average distribution of true effects. 

The analysis also examines whether studies may be outliers and/or influential in the random-effect model. 
They can have a significant impact on the value of the estimated random-effect model coefficients, i.e., the 
intercept. If they had remained in the analysis, they could have changed the entire outcome. The 
standardized residuals are used to detect outliers, while the Cook's distances (Cook, 1977) and DFFITS 
(Difference in Fits) are applied to diagnose the influential studies. Studies are considered as potential 
outliers if they have a standardized residual larger than 3 or smaller than -3 (rstudent > ± 3), while they are 
considered to be influential if the Cook's distance value is more than 1 (cook.D > 1) and DFFITS is larger 
than 2 (dffits > 2) (Gerbing, 2014). 

The meta-analysis was carried out with the help of open-source statistical software Jamovi version 1.6.23 
(The Jamovi Project, 2021). The MAJOR meta-analysis module library was used to compute r-to-z 
transformations, as well as to generate a random-effect model, heterogeneity statistics, a forest plot, and 
outlier and influential case diagnostics. 

RESULTS 

Table 1 shows the results of data extraction processing. The eleven studies were published between 2001 
and 2021, with Abdullah and Abd. Majid (2001) is the longest and Prastiwi (2021) is the most recent. The 
studies used two types of sampling methods: probability sampling and non-probability sampling. Ababio 
and Mawutor (2015); Kassim et al. (2019); Murdayanti et al. (2020); and Yayeh (2014) applied the 
probability sampling method, whereas Ismail et al. (2018); Mei Teh et al. (2019); Priyo Nugroho et al. 

three steps to implementing this method (Borenstein et al., 2011; Field & Gillett, 2010). To begin, use 
Fisher's r-to-z transformation to convert the effect size in each study into a standard normal metric. The 
Fisher's r-to-z transformation formula is given by 𝑧𝑧𝑟𝑟𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒 (

1+𝑟𝑟𝑖𝑖
1−𝑟𝑟𝑖𝑖

), where 𝑟𝑟𝑖𝑖 is the effect size in each 

study. After that, for each study, a weighted average of 𝑧𝑧𝑟𝑟 scores are computed by �̅�𝑧𝑟𝑟𝑖𝑖 =
∑ 𝑛𝑛𝑖𝑖𝑧𝑧𝑟𝑟𝑖𝑖
𝑘𝑘
𝑖𝑖=1
∑ 𝑛𝑛𝑖𝑖𝑘𝑘𝑖𝑖=1

, where 𝑘𝑘 
is the number of studies and 𝑛𝑛𝑖𝑖 is the sample size. Finally, it should be converted back to 𝑟𝑟𝑖𝑖 using the 
formula 𝑟𝑟𝑖𝑖 =

𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖−1
𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖+1

.  

In addition, the random-effects statistical model is applied to estimate the average distribution of true 
effects. The random-effects method was chosen because the effect size was extracted from a series of 
studies conducted by various authors in various populations at various times. The analysis also reports the 
estimate of the 2 index, the H2 index, the I2 index, and the Q-test (Cochran, 1954) with a p-value as the 
heterogeneity statistics outcome. The Q-test is used to assess the null hypothesis that all effect sizes from 
all studies are homogenous (Chen & Peace, 2021). If the p-value is less than  (the typical significance 
level is 0.05), the null hypothesis should be rejected, indicating that the effect sizes from all studies are not 
homogenous. Meanwhile, 2, H2, and I2 are used to determine the strength of the distribution of true effect 
sizes. The 2 index is estimated using the Hedges’ estimator (Hedges & Olkin, 1985) to measure the variance 
of the true effect sizes, and the index should be greater than zero. The H2 index is quantified using Higgins 
and Thompson’s (2002) formula to inform the relative extent of heterogeneity in comparison to all studies, 
and the index should be greater than 1. The I2 index is also calculated using Higgins and Thompson's (2002) 
formula to determine the percentage of observed heterogeneity versus real heterogeneity. As a rule of 
thumb, the I2 index could be considered as having low heterogeneity (I2 = 25%), moderate heterogeneity 
(I2 = 50%), and high heterogeneity (I2 = 75%). To display the conclusions of meta-analyses, forest plots are 
generated. Forest plots provide information about each study’s effect size and confidence interval, as well 
as the average distribution of true effects. 

The analysis also examines whether studies may be outliers and/or influential in the random-effect model. 
They can have a significant impact on the value of the estimated random-effect model coefficients, i.e., the 
intercept. If they had remained in the analysis, they could have changed the entire outcome. The 
standardized residuals are used to detect outliers, while the Cook's distances (Cook, 1977) and DFFITS 
(Difference in Fits) are applied to diagnose the influential studies. Studies are considered as potential 
outliers if they have a standardized residual larger than 3 or smaller than -3 (rstudent > ± 3), while they are 
considered to be influential if the Cook's distance value is more than 1 (cook.D > 1) and DFFITS is larger 
than 2 (dffits > 2) (Gerbing, 2014). 

The meta-analysis was carried out with the help of open-source statistical software Jamovi version 1.6.23 
(The Jamovi Project, 2021). The MAJOR meta-analysis module library was used to compute r-to-z 
transformations, as well as to generate a random-effect model, heterogeneity statistics, a forest plot, and 
outlier and influential case diagnostics. 

RESULTS 

Table 1 shows the results of data extraction processing. The eleven studies were published between 2001 
and 2021, with Abdullah and Abd. Majid (2001) is the longest and Prastiwi (2021) is the most recent. The 
studies used two types of sampling methods: probability sampling and non-probability sampling. Ababio 
and Mawutor (2015); Kassim et al. (2019); Murdayanti et al. (2020); and Yayeh (2014) applied the 
probability sampling method, whereas Ismail et al. (2018); Mei Teh et al. (2019); Priyo Nugroho et al. 

three steps to implementing this method (Borenstein et al., 2011; Field & Gillett, 2010). To begin, use 
Fisher's r-to-z transformation to convert the effect size in each study into a standard normal metric. The 
Fisher's r-to-z transformation formula is given by 𝑧𝑧𝑟𝑟𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒 (

1+𝑟𝑟𝑖𝑖
1−𝑟𝑟𝑖𝑖

), where 𝑟𝑟𝑖𝑖 is the effect size in each 

study. After that, for each study, a weighted average of 𝑧𝑧𝑟𝑟 scores are computed by �̅�𝑧𝑟𝑟𝑖𝑖 =
∑ 𝑛𝑛𝑖𝑖𝑧𝑧𝑟𝑟𝑖𝑖
𝑘𝑘
𝑖𝑖=1
∑ 𝑛𝑛𝑖𝑖𝑘𝑘𝑖𝑖=1

, where 𝑘𝑘 
is the number of studies and 𝑛𝑛𝑖𝑖 is the sample size. Finally, it should be converted back to 𝑟𝑟𝑖𝑖 using the 
formula 𝑟𝑟𝑖𝑖 =

𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖−1
𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖+1

.  

In addition, the random-effects statistical model is applied to estimate the average distribution of true 
effects. The random-effects method was chosen because the effect size was extracted from a series of 
studies conducted by various authors in various populations at various times. The analysis also reports the 
estimate of the 2 index, the H2 index, the I2 index, and the Q-test (Cochran, 1954) with a p-value as the 
heterogeneity statistics outcome. The Q-test is used to assess the null hypothesis that all effect sizes from 
all studies are homogenous (Chen & Peace, 2021). If the p-value is less than  (the typical significance 
level is 0.05), the null hypothesis should be rejected, indicating that the effect sizes from all studies are not 
homogenous. Meanwhile, 2, H2, and I2 are used to determine the strength of the distribution of true effect 
sizes. The 2 index is estimated using the Hedges’ estimator (Hedges & Olkin, 1985) to measure the variance 
of the true effect sizes, and the index should be greater than zero. The H2 index is quantified using Higgins 
and Thompson’s (2002) formula to inform the relative extent of heterogeneity in comparison to all studies, 
and the index should be greater than 1. The I2 index is also calculated using Higgins and Thompson's (2002) 
formula to determine the percentage of observed heterogeneity versus real heterogeneity. As a rule of 
thumb, the I2 index could be considered as having low heterogeneity (I2 = 25%), moderate heterogeneity 
(I2 = 50%), and high heterogeneity (I2 = 75%). To display the conclusions of meta-analyses, forest plots are 
generated. Forest plots provide information about each study’s effect size and confidence interval, as well 
as the average distribution of true effects. 

The analysis also examines whether studies may be outliers and/or influential in the random-effect model. 
They can have a significant impact on the value of the estimated random-effect model coefficients, i.e., the 
intercept. If they had remained in the analysis, they could have changed the entire outcome. The 
standardized residuals are used to detect outliers, while the Cook's distances (Cook, 1977) and DFFITS 
(Difference in Fits) are applied to diagnose the influential studies. Studies are considered as potential 
outliers if they have a standardized residual larger than 3 or smaller than -3 (rstudent > ± 3), while they are 
considered to be influential if the Cook's distance value is more than 1 (cook.D > 1) and DFFITS is larger 
than 2 (dffits > 2) (Gerbing, 2014). 

The meta-analysis was carried out with the help of open-source statistical software Jamovi version 1.6.23 
(The Jamovi Project, 2021). The MAJOR meta-analysis module library was used to compute r-to-z 
transformations, as well as to generate a random-effect model, heterogeneity statistics, a forest plot, and 
outlier and influential case diagnostics. 

RESULTS 

Table 1 shows the results of data extraction processing. The eleven studies were published between 2001 
and 2021, with Abdullah and Abd. Majid (2001) is the longest and Prastiwi (2021) is the most recent. The 
studies used two types of sampling methods: probability sampling and non-probability sampling. Ababio 
and Mawutor (2015); Kassim et al. (2019); Murdayanti et al. (2020); and Yayeh (2014) applied the 
probability sampling method, whereas Ismail et al. (2018); Mei Teh et al. (2019); Priyo Nugroho et al. 



76        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

from all studies were homogenous (Chen & Peace, 2021). If the 
p-value was less than    (the typical significance level is 0.05), the null 
hypothesis should be rejected, indicating that the effect sizes from all 
studies were not homogenous. Meanwhile,    , H2, and I2 were used 
to determine the strength of the distribution of true effect sizes. The  
     index was estimated using the Hedges’ estimator (Hedges & Olkin, 
1985) to measure the variance of the true effect sizes, and the index 
should be greater than zero. The H2 index was quantified using the 
Higgins and Thompson’s (2002) formula to inform the relative extent 
of heterogeneity in comparison to all studies, and the index should 
be greater than 1. The I2 index was also calculated using the Higgins 
and Thompson’s (2002) formula to determine the percentage of 
observed heterogeneity versus real heterogeneity. As a rule of thumb, 
the I2 index could be considered as having low heterogeneity (I2 = 
25%), moderate heterogeneity (I2 = 50%), and high heterogeneity (I2 
= 75%). To display the conclusions of the meta-analyses, forest plots 
were generated. Forest plots provided information about each study’s 
effect size and confidence interval, as well as the average distribution 
of true effects.

The analysis also examined whether studies may be outliers and/or 
influential in the random-effect model. They can have a significant 
impact on the value of the estimated random-effect model coefficients, 
i.e., the intercept. If they had remained in the analysis, they could have 
changed the entire outcome. The standardized residuals are used to 
detect outliers, while the Cook’s distances (Cook, 1977) and DFFITS 
(Difference in Fits) are applied to diagnose the influential studies. 
Studies are considered as potential outliers if they have a standardized 
residual larger than 3 or smaller than -3 (rstudent > ± 3), while they 
are considered to be influential if the Cook’s distance value is more 
than 1 (cook.D > 1) and DFFITS is larger than 2 (dffits > 2) (Gerbing, 
2014).

The meta-analysis was carried out with the help of an open-source 
statistical software the Jamovi version 1.6.23 (The Jamovi Project, 
2021). The MAJOR meta-analysis module library was used to compute 
r-to-z transformations, as well as to generate a random-effect model, 
heterogeneity statistics, a forest plot, and outlier and influential case 
diagnostics.

three steps to implementing this method (Borenstein et al., 2011; Field & Gillett, 2010). To begin, use 
Fisher's r-to-z transformation to convert the effect size in each study into a standard normal metric. The 
Fisher's r-to-z transformation formula is given by 𝑧𝑧𝑟𝑟𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒 (

1+𝑟𝑟𝑖𝑖
1−𝑟𝑟𝑖𝑖

), where 𝑟𝑟𝑖𝑖 is the effect size in each 

study. After that, for each study, a weighted average of 𝑧𝑧𝑟𝑟 scores are computed by �̅�𝑧𝑟𝑟𝑖𝑖 =
∑ 𝑛𝑛𝑖𝑖𝑧𝑧𝑟𝑟𝑖𝑖
𝑘𝑘
𝑖𝑖=1
∑ 𝑛𝑛𝑖𝑖𝑘𝑘𝑖𝑖=1

, where 𝑘𝑘 
is the number of studies and 𝑛𝑛𝑖𝑖 is the sample size. Finally, it should be converted back to 𝑟𝑟𝑖𝑖 using the 
formula 𝑟𝑟𝑖𝑖 =

𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖−1
𝑒𝑒2�̅�𝑧𝑟𝑟𝑖𝑖+1

.  

In addition, the random-effects statistical model is applied to estimate the average distribution of true 
effects. The random-effects method was chosen because the effect size was extracted from a series of 
studies conducted by various authors in various populations at various times. The analysis also reports the 
estimate of the 2 index, the H2 index, the I2 index, and the Q-test (Cochran, 1954) with a p-value as the 
heterogeneity statistics outcome. The Q-test is used to assess the null hypothesis that all effect sizes from 
all studies are homogenous (Chen & Peace, 2021). If the p-value is less than  (the typical significance 
level is 0.05), the null hypothesis should be rejected, indicating that the effect sizes from all studies are not 
homogenous. Meanwhile, 2, H2, and I2 are used to determine the strength of the distribution of true effect 
sizes. The 2 index is estimated using the Hedges’ estimator (Hedges & Olkin, 1985) to measure the variance 
of the true effect sizes, and the index should be greater than zero. The H2 index is quantified using Higgins 
and Thompson’s (2002) formula to inform the relative extent of heterogeneity in comparison to all studies, 
and the index should be greater than 1. The I2 index is also calculated using Higgins and Thompson's (2002) 
formula to determine the percentage of observed heterogeneity versus real heterogeneity. As a rule of 
thumb, the I2 index could be considered as having low heterogeneity (I2 = 25%), moderate heterogeneity 
(I2 = 50%), and high heterogeneity (I2 = 75%). To display the conclusions of meta-analyses, forest plots are 
generated. Forest plots provide information about each study’s effect size and confidence interval, as well 
as the average distribution of true effects. 

The analysis also examines whether studies may be outliers and/or influential in the random-effect model. 
They can have a significant impact on the value of the estimated random-effect model coefficients, i.e., the 
intercept. If they had remained in the analysis, they could have changed the entire outcome. The 
standardized residuals are used to detect outliers, while the Cook's distances (Cook, 1977) and DFFITS 
(Difference in Fits) are applied to diagnose the influential studies. Studies are considered as potential 
outliers if they have a standardized residual larger than 3 or smaller than -3 (rstudent > ± 3), while they are 
considered to be influential if the Cook's distance value is more than 1 (cook.D > 1) and DFFITS is larger 
than 2 (dffits > 2) (Gerbing, 2014). 

The meta-analysis was carried out with the help of open-source statistical software Jamovi version 1.6.23 
(The Jamovi Project, 2021). The MAJOR meta-analysis module library was used to compute r-to-z 
transformations, as well as to generate a random-effect model, heterogeneity statistics, a forest plot, and 
outlier and influential case diagnostics. 

RESULTS 

Table 1 shows the results of data extraction processing. The eleven studies were published between 2001 
and 2021, with Abdullah and Abd. Majid (2001) is the longest and Prastiwi (2021) is the most recent. The 
studies used two types of sampling methods: probability sampling and non-probability sampling. Ababio 
and Mawutor (2015); Kassim et al. (2019); Murdayanti et al. (2020); and Yayeh (2014) applied the 
probability sampling method, whereas Ismail et al. (2018); Mei Teh et al. (2019); Priyo Nugroho et al. 

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2

A META-ANALYSIS OF THE RELATIONSHIP BETWEEN 
RELIGIOSITY AND SAVING BEHAVIOUR

Method of Analysis

𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 = 0.5𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑒𝑒𝑒𝑒 �
1+𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
1−𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖

�,

𝑧𝑧𝑧𝑧�̅�𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖𝑧𝑧𝑧𝑧𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

∑ 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖
𝑘𝑘𝑘𝑘
𝑖𝑖𝑖𝑖=1

,

𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖 =
𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖−1

𝑒𝑒𝑒𝑒2𝑧𝑧𝑧𝑧�𝑟𝑟𝑟𝑟𝑖𝑖𝑖𝑖+1
.

τ2



    77      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

Ta
bl

e 
1 

O
ve

rv
ie

w
 o

f S
tu

di
es

 In
cl

ud
ed

 in
 M

et
a-

A
na

ly
si

s

A
ut

ho
rs

 (y
ea

r)
Sa

m
pl

in
g 

M
et

ho
d

M
ea

su
re

m
en

t 
Te

ch
ni

qu
e

V
ar

ia
bl

es
St

at
is

tic
al

 
A

na
ly

si
s

R
es

ul
t

   
  I

nd
ep

en
de

nt
 

D
ep

en
de

nt
A

bd
ul

la
h 

an
d 

A
bd

. M
aj

id
 

(2
00

1)

N
ot

 m
en

tio
ne

d
Q

ue
st

io
nn

ai
re

 
su

rv
ey

R
el

ig
io

si
ty

 in
de

x 
an

d 
in

co
m

e
Sa

vi
ng

M
ul

tip
le

 li
ne

ar
 

re
gr

es
si

on
“…

 th
er

e 
ex

is
t a

 c
on

cl
us

iv
e 

re
la

tio
ns

hi
p 

be
tw

ee
n 

sa
vi

ng
 

an
d 

R
el

ig
io

si
ty

 In
de

x 
…

” 
(t

-s
ta

tis
tic

s 
= 

1.
99

3,
 p

 <
 0

.0
5)

 
(p

. 7
5)

.
Y

ay
eh

 (2
01

4)
M

ul
tis

ta
ge

 
cl

us
te

r s
am

pl
in

g 
an

d 
pr

ob
ab

ili
ty

 
pr

op
or

tio
na

l t
o 

si
ze

 
(P

PS
) s

am
pl

in
g

Q
ue

st
io

nn
ai

re
 

su
rv

ey
re

lig
io

n 
af

fil
ia

tio
n,

 re
lig

io
us

 
at

te
nd

an
ce

, r
el

ig
io

n 
id

en
tit

y,
 

ho
us

eh
ol

d 
ne

t i
nc

om
e 

pe
r 

m
on

th
, g

en
de

r, 
ho

us
eh

ol
d 

ac
ce

pt
in

g 
in

te
re

st
 p

ay
m

en
t, 

le
ve

l o
f e

du
ca

tio
n,

 fa
m

ily
 

si
ze

, a
ge

, m
ar

ita
l s

ta
tu

s,
 

w
ea

lth
, a

nd
 k

no
w

le
dg

e 
ab

ou
t 

sa
vi

ng
 in

te
re

st
 p

ay
m

en
t.

sa
vi

ng
L

in
ea

r a
nd

 p
ro

bi
t 

re
gr

es
si

on
“…

 th
e 

m
or

e 
of

te
n 

pe
op

le
 

go
in

g 
to

 c
hu

rc
h/

m
os

qu
e,

 i.
e.

 
th

e 
m

or
e 

re
lig

io
us

 th
ey

 a
re

, 
th

e 
lo

w
er

 is
 th

ei
r 

pr
op

en
si

ty
 

to
 s

av
e 

m
on

ey
.”

 (W
al

d 
ch

i-
sq

ua
re

 o
f 6

2.
58

 w
ith

 p
-v

al
ue

 
of

 0
.0

00
)

A
ba

bi
o 

an
d 

M
aw

ut
or

 (2
01

5)
Si

m
pl

e 
ra

nd
om

 
sa

m
pl

in
g 

an
d 

co
nv

en
ie

nc
e 

sa
m

pl
in

g

Q
ue

st
io

nn
ai

re
 

su
rv

ey
re

lig
io

si
ty

, u
nc

er
ta

in
ty

, 
liq

ui
di

ty
 c

on
st

ra
in

t, 
st

ag
e 

in
 

lif
e,

 a
nd

 in
te

rg
en

er
at

io
na

l 
ef

fe
ct

 in
co

m
e

sa
vi

ng
L

og
it 

m
od

el
C

hu
rc

h 
at

te
nd

an
ce

 v
er

y 
si

gn
ifi

ca
nt

ly
 e

xp
la

in
s 

th
at

 
re

lig
io

si
ty

 e
ff

ec
ts

 s
av

in
g 

be
ha

vi
or

 (o
dd

s 
ra

tio
 =

 0
.8

22
, p

 
< 

0.
05

) (
p.

 5
5)

.

(c
on

tin
ue

d)



78        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

A
ut

ho
rs

 (y
ea

r)
Sa

m
pl

in
g 

M
et

ho
d

M
ea

su
re

m
en

t 
Te

ch
ni

qu
e

V
ar

ia
bl

es
St

at
is

tic
al

 
A

na
ly

si
s

R
es

ul
t

   
  I

nd
ep

en
de

nt
 

D
ep

en
de

nt
Pr

iy
o 

N
ug

ro
ho

 
et

 a
l. 

(2
01

7)
Pu

rp
os

iv
e 

sa
m

pl
in

g
Q

ue
st

io
nn

ai
re

 
su

rv
ey

Se
lf

-e
ffi

ca
cy

, R
el

ig
io

si
ty

, 
A

tti
tu

de
, a

nd
 S

ub
je

ct
iv

e 
no

rm
In

te
nt

io
n 

an
d 

B
eh

av
io

r
Si

m
ul

ta
ne

ou
s 

E
qu

at
io

n 
M

od
el

in
g

“…
 r

el
ig

io
si

ty
 h

as
 a

 p
os

iti
ve

 
an

d 
si

gn
ifi

ca
nt

 in
flu

en
ce

 o
n 

cu
st

om
er

 b
eh

av
io

r 
us

in
g 

pr
od

uc
ts

 a
nd

 s
er

vi
ce

s 
of

 
Is

la
m

ic
 b

an
ks

” 
(i

.e
. b

an
k 

sa
vi

ng
s 

or
 d

ep
os

its
) (

p.
 4

4)
.

Sa
ts

io
s 

an
d 

H
ad

jid
ak

is
 

(2
01

7)

Sn
ow

ba
ll 

sa
m

pl
in

g
Q

ue
st

io
nn

ai
re

 
su

rv
ey

re
lig

io
si

ty
 a

nd
 s

el
f-

m
as

te
ry

fiv
e 

in
te

nt
io

ns
 to

 
sa

vi
ng

 s
ub

sc
al

es
: 

th
ri

ft
, s

av
in

g 
in

vo
lv

em
en

t, 
sa

vi
ng

 
ha

bi
ts

, s
ha

m
e 

of
 

de
bt

 a
nd

 n
o 

ne
ed

 
to

 s
av

e

Pe
ar

so
n 

co
rr

el
at

io
n

“…
, r

el
ig

io
si

ty
 is

 s
ig

ni
fic

an
tly

 
po

si
tiv

el
y 

co
rr

el
at

ed
 w

ith
 

al
l 5

 in
te

nt
io

n 
su

bs
ca

le
s,

 
...

” 
(p

. 2
0)

. T
he

 s
ub

sc
al

e 
of

 
sa

vi
ng

 h
ab

its
 is

 s
ig

ni
fic

an
tly

 
po

si
tiv

el
y 

co
rr

el
at

ed
 w

ith
 

re
lig

io
si

ty
 (r

(1
00

) =
 0

.3
32

, p
 

< 
0.

01
).

Is
m

ai
l e

t a
l. 

(2
01

8)
Pu

rp
os

iv
e 

sa
m

pl
in

g
Q

ue
st

io
nn

ai
re

 
su

rv
ey

 
Se

rv
ic

e 
qu

al
ity

, r
el

ig
io

us
 

be
lie

f, 
an

d 
kn

ow
le

dg
e.

sa
vi

ng
 b

eh
av

io
r

M
ul

tip
le

 li
ne

ar
 

re
gr

es
si

on
“…

 r
el

ig
io

us
 b

el
ie

f i
s 

si
gn

ifi
ca

nt
ly

 r
el

at
ed

 to
 s

av
in

g 
be

ha
vi

ou
r 

(t
 =

 4
.6

0,
 p

 =
 

0.
00

).”
 (p

. 1
07

6)

K
as

si
m

 e
t a

l. 
(2

01
9)

D
is

pr
op

or
tio

na
te

 
st

ra
tifi

ed
 s

am
pl

in
g

Q
ue

st
io

nn
ai

re
 

su
rv

ey
Fa

m
ily

 b
ac

kg
ro

un
d,

 
R

el
ig

io
si

ty
, A

tti
tu

de
, L

ite
ra

cy
, 

H
ou

se
ho

ld
 In

co
m

e,
 A

ge
, 

L
ev

el
 o

f e
du

ca
tio

n,
 a

nd
 

L
oc

al
ity

.

Sa
vi

ng
 b

eh
av

io
r

M
ul

tip
le

 li
ne

ar
 

re
gr

es
si

on
“…

 th
e 

re
su

lts
 d

em
on

st
ra

te
 

th
at

 r
el

ig
io

si
ty

, …
 a

re
 n

ot
 

si
gn

ifi
ca

nt
ly

 r
el

at
ed

 to
 s

av
in

g 
be

ha
vi

or
.”

 (p
. 2

48
) (

t-
st

at
is

tic
s 

= 
1.

41
8)

(c
on

tin
ue

d)



    79      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

A
ut

ho
rs

 (y
ea

r)
Sa

m
pl

in
g 

M
et

ho
d

M
ea

su
re

m
en

t 
Te

ch
ni

qu
e

V
ar

ia
bl

es
St

at
is

tic
al

 
A

na
ly

si
s

R
es

ul
t

   
  I

nd
ep

en
de

nt
 

D
ep

en
de

nt
M

ei
 T

eh
 e

t a
l. 

(2
01

9)
C

on
ve

ni
en

ce
 

sa
m

pl
in

g
Q

ue
st

io
nn

ai
re

 
su

rv
ey

In
di

vi
du

al
 c

ha
ra

ct
er

is
tic

, 
So

ci
al

is
at

io
n,

 C
og

ni
tiv

e 
ab

ili
ty

, R
el

ig
io

n 
fa

ith
, a

nd
 

Se
lf

-e
ffi

ca
cy

.

Pr
iv

at
e 

sa
vi

ng
L

og
is

tic
 re

gr
es

si
on

“A
s 

fo
r 

re
lig

io
us

 fa
ith

, d
iv

in
e 

gu
id

an
ce

 (o
dd

s 
ra

tio
 =

 6
.5

1)
 

si
gn

ifi
ca

nt
ly

 p
re

di
ct

ed
 a

n 
in

di
vi

du
al

’s 
lik

el
ih

oo
d 

to
 s

av
e 

m
on

ey
.”

 (p
. 1

0)
W

ija
ya

 e
t a

l. 
(2

01
9)

C
on

ve
ni

en
ce

 
sa

m
pl

in
g

Q
ue

st
io

nn
ai

re
 

su
rv

ey
R

el
ig

io
si

ty
 le

ve
l

Sa
vi

ng
 d

ec
is

io
ns

C
hi

-s
qu

ar
e 

te
st

“…
 a

 c
hi

-s
qu

ar
e 

te
st

 b
et

w
ee

n 
re

lig
io

si
ty

 le
ve

l a
nd

 s
av

in
g 

de
ci

si
on

s 
cr

ite
ri

a,
 w

hi
ch

 
sh

ow
ed

 th
er

e 
is

 a
 s

ig
ni

fic
an

t 
di

ffe
re

nc
e 

(p
 <

 0
.0

1)
. M

or
e 

th
an

 6
0 

pe
r 

ce
nt

 o
f t

he
 

re
sp

on
de

nt
s 

de
ci

de
d 

to
 s

av
e 

m
on

ey
 in

 B
M

Ts
 b

ec
au

se
 

of
 th

ei
r 

pr
od

uc
ts

 b
ei

ng
 in

 
ac

co
rd

an
ce

 w
ith

 S
ha

ri
a.

” 
(p

. 
14

75
) (

ch
i-

sq
ua

re
 =

 6
.4

63
67

)
M

ur
da

ya
nt

i e
t 

al
. (

20
20

)
Pr

op
or

tio
na

te
 

st
ra

tifi
ed

 ra
nd

om
 

sa
m

pl
in

g

Q
ue

st
io

nn
ai

re
 

su
rv

ey
Fi

na
nc

ia
l k

no
w

le
dg

e,
 s

el
f-

co
nt

ro
l, 

an
d 

re
lig

io
us

 b
el

ie
fs

.
Sa

vi
ng

 b
eh

av
io

r
Pa

rt
ia

l L
ea

st
 

Sq
ua

re
“…

 r
el

ig
io

us
 b

el
ie

fs
 h

av
e 

a 
si

gn
ifi

ca
nt

 p
os

iti
ve

 e
ffe

ct
 o

n 
sa

vi
ng

s 
be

ha
vi

or
, …

” 
(p

. 8
) 

(t
-s

ta
tis

tic
s 

= 
6.

77
, p

 <
 0

.0
01

)

Pr
as

tiw
i (

20
21

)
N

ot
 m

en
tio

ne
d

Q
ue

st
io

nn
ai

re
 

su
rv

ey
R

el
ig

io
si

ty
, E

nv
ir

on
m

en
t, 

an
d 

R
ep

ut
at

io
n.

 

Sa
vi

ng
 d

ec
is

io
n

M
ul

tip
le

 li
ne

ar
 

re
gr

es
si

on
“…

 R
el

ig
io

si
ty

, .
.. 

ha
ve

 a
 

si
gn

ifi
ca

nt
 p

os
iti

ve
 e

ffe
ct

 o
n 

sa
vi

ng
 d

ec
is

io
ns

.”
 (p

. 2
22

) 
(t

-s
ta

tis
tic

s 
= 

2.
16

1,
 p

 <
 0

.0
5)

(c
on

tin
ue

d)



80        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Ta
bl

e 
2 

O
ve

rv
ie

w
 o

f D
es

cr
ip

tiv
e 

St
at

is
tic

s 
an

d 
E

ffe
ct

 S
iz

e

A
ut

ho
rs

 (y
ea

r)
N

   
 A

ge
 (y

/o
)

G
en

de
r

M
ar

ita
l s

ta
tu

s
R

el
ig

io
n

L
oc

at
io

n 
(C

ou
nt

ry
)

Pe
ar

so
n 

r

A
bd

ul
la

h 
an

d 
A

bd
. M

aj
id

 
(2

00
1)

16
0

78
.1

3%
 1

8 
to

 2
3 

19
.3

8%
 2

4 
to

 2
9 

2.
50

%
 3

0 
to

 3
5

34
.3

8%
 M

al
e 

65
.6

3%
 F

em
al

e
95

.6
3%

 S
in

gl
e 

4.
38

%
 M

ar
ri

ed
M

us
lim

In
te

rn
at

io
na

l I
sl

am
ic

 
U

ni
ve

rs
ity

 M
al

ay
si

a 
(I

IU
M

)-
 

Se
la

ng
or

 (M
al

ay
si

a)

0.
15

76

Y
ay

eh
 (2

01
4)

38
4

42
 (M

ea
n)

N
ot

 m
en

tio
ne

d
74

%
 M

ar
ri

ed
 

16
%

 W
id

ow
ed

67
.5

%
 O

rt
ho

do
x 

C
hr

is
tia

n
30

.8
%

 M
us

lim
 

1.
7%

 P
ro

te
st

an
t

W
es

t A
m

ha
ra

 n
at

io
na

l r
eg

io
na

l 
st

at
e 

(E
th

io
pi

a)
0.

40
37

A
ba

bi
o 

an
d 

M
aw

ut
or

 (2
01

5)
20

0
N

ot
 m

en
tio

ne
d

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

C
hr

is
tia

n
A

cc
ra

 M
et

ro
po

lit
an

 (G
ha

na
)

0.
05

4

Pr
iy

o 
N

ug
ro

ho
 

et
 a

l. 
(2

01
7)

22
0

42
%

 le
ss

 3
1 

45
%

 3
1 

to
 4

0 
13

%
 a

bo
ve

 4
0

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

M
us

lim
Y

og
ya

ka
rt

a 
(I

nd
on

es
ia

)
0.

56
6

Sa
ts

io
s 

an
d 

H
ad

jid
ak

is
 

(2
01

7)

10
0

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

N
ot

 m
en

tio
ne

d
M

us
lim

X
an

th
i, 

R
od

op
i a

nd
 E

vr
os

 - 
T

hr
ac

e 
(G

re
ec

e)
0.

33
2

Is
m

ai
l e

t a
l. 

(2
01

8)
15

0
2.

7%
 le

ss
 2

0 
54

%
 2

0 
to

 3
0 

29
.3

%
 3

1 
to

 4
0 

14
%

 a
bo

ve
 4

0

42
.7

%
 M

al
e 

57
.3

%
 F

em
al

e
N

ot
 m

en
tio

ne
d

61
.3

%
 M

us
lim

  
15

.3
%

 B
ud

dh
a 

12
.7

%
 H

in
du

 
7.

3%
 C

hr
is

tia
n 

3.
3%

 o
th

er
s

(M
al

ay
si

a)
0.

37
56

(c
on

tin
ue

d)



    81      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

A
ut

ho
rs

 (y
ea

r)
N

   
 A

ge
 (y

/o
)

G
en

de
r

M
ar

ita
l s

ta
tu

s
R

el
ig

io
n

L
oc

at
io

n 
(C

ou
nt

ry
)

Pe
ar

so
n 

r

K
as

si
m

 e
t a

l. 
(2

01
9)

53
1

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

51
.2

%
 M

ar
ri

ed
 

48
.8

%
 S

in
gl

e
M

us
lim

Se
la

ng
or

 (M
al

ay
si

a)
0.

06
15

M
ei

 T
eh

 e
t a

l. 
(2

01
9)

22
4

16
 to

 6
0

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

M
us

lim
(M

al
ay

si
a)

0.
45

88

W
ija

ya
 e

t a
l. 

(2
01

9)
52

3
2.

68
%

 1
0 

to
 2

0 
13

%
 2

1 
to

 2
5 

18
.1

6%
 2

6 
to

 3
0 

20
.6

5%
 3

1 
to

 3
5 

21
.0

3%
 3

6 
to

 4
0 

24
.4

7%
 a

bo
ve

 
40

50
.8

6%
 M

al
e 

49
.1

4%
 F

em
al

e
N

ot
 m

en
tio

ne
d

M
us

lim
Su

ra
ka

rt
a 

an
d 

Su
ko

ha
rj

o 
(I

nd
on

es
ia

)
0.

11
12

M
ur

da
ya

nt
i e

t 
al

. (
20

20
)

26
8

13
 to

 2
0

N
ot

 m
en

tio
ne

d
N

ot
 m

en
tio

ne
d

M
us

lim
D

ar
un

na
ja

h 
Is

la
m

ic
 B

oa
rd

in
g 

Sc
ho

ol
 - 

Ja
ka

rt
a 

(I
nd

on
es

ia
)

0.
41

35

Pr
as

tiw
i (

20
21

)
10

0
12

%
 le

ss
 1

8 
27

%
 1

8 
to

 2
5 

36
%

 2
5 

to
 3

0 
25

%
 3

0 
to

 4
0

N
ot

 m
en

tio
ne

d
45

%
 M

al
e 

55
%

 F
em

al
e

M
us

lim
K

SP
PS

 B
M

T
 A

m
an

ah
 U

m
m

ah
 

- S
ur

ab
ay

a 
(I

nd
on

es
ia

)
0.

21
61

   



82        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

RESULTS

Table 1 shows the results of data extraction processing. The 11 studies 
were published between 2001 and 2021, with Abdullah and Abd. 
Majid (2001) being the oldest and Prastiwi (2021) is the most recent. 
The studies used two types of sampling methods: probability sampling 
and non-probability sampling. Ababio and Mawutor (2015); Kassim 
et al. (2019); Murdayanti et al. (2020); and Yayeh (2014) applied the 
probability sampling method, whereas Ismail et al. (2018); Mei Teh et 
al. (2019); Priyo Nugroho et al. (2017); Satsios and Hadjidakis (2017); 
and Wijaya et al. (2019) employed the non-probability sampling 
method. Meanwhile, Abdullah and Abd. Majid (2001); and Prastiwi 
(2021) there was no mention of the method used in their studies. 
To collect primary data, all studies developed a self-administered 
questionnaire. Furthermore, various themes of religiosity and saving 
behaviour, such as religious attendance (Yayeh, 2014), religious belief 
(Ismail et al., 2018; Murdayanti et al., 2020), religion faith (Mei Teh 
et al., 2019), saving habits (Satsios & Hadjidakis, 2017), and saving 
decisions (Prastiwi, 2021; Wijaya et al., 2019), were used to as the 
independent and dependent variables. 

Various statistical analyses were also applied, namely Pearson 
correlation (Satsios & Hadjidakis, 2017), chi-square test (Wijaya 
et al., 2019), multiple linear regression (Abdullah & Abd. Majid, 
2001; Ismail et al., 2018; Kassim et al., 2019; Prastiwi, 2021), 
logit regression (Ababio & Mawutor, 2015; Mei Teh et al., 2019), 
probit regression (Yayeh, 2014), partial least square (Murdayanti 
et al., 2020), and simultaneous equation modelling (Priyo Nugroho 
et al., 2017). On the other hand, various fit test indicators, such as 
chi-square (Wijaya et al., 2019; Yayeh, 2014), odds ratio (Ababio & 
Mawutor, 2015; Mei Teh et al., 2019), and t-statistic (Abdullah & 
Abd. Majid, 2001; Ismail et al., 2018; Kassim et al., 2019; Prastiwi, 
2021), have been used to assess the significance of the relationship 
between religiosity and saving behaviour. These indicators had to be 
converted into Pearson’s r before they are used in a meta-analysis. 
Meanwhile, Priyo Nugroho et al. (2017) have provided another fit test 
indicator, namely the critical ratio (8.395) (personal communication).

Table 2 summarizes the descriptive statistics from each study, 
including the number of observations, age, gender, marital status, 
religion, and location, as well as the estimated effect sizes. As 
previously stated, 11 articles were used as samples for the meta-



    83      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

analytic study, and these gave a total of 1,063 observations. Kassim 
et al. (2019) observed 531 people, making it the largest population 
sample, while Prastiwi (2021); as well as Satsios and Hadjidakis 
(2017) had 100 observations, making it the smallest. Meanwhile, the 
age range varied from 13 to more than 40 years. Furthermore, only 
four studies provided gender information, i.e., Abdullah and Abd. 
Majid (2001); Ismail et al. (2018); Prastiwi (2021); and Wijaya et al. 
(2019), in which most of their respondents were women. Similarly, 
marital status was only provided by three studies, i.e., Abdullah and 
Abd. Majid (2001); Kassim et al. (2019); and Yayeh (2014), with the 
majority of their respondents being married. On the other hand, all 
studies provided religious information, with Muslims representing the 
majority of their respondents. Furthermore, the study locations reveal 
that respondents were from a variety of countries, including Ethiopia, 
Ghana, Greece, Malaysia, and Indonesia. Meanwhile, all studies were 
ready to use the same method to express effect sizes in Pearson’s r. 
The study conducted by Priyo Nugroho et al. (2017) had the largest 
effect size (0.566), while the study performed by Ababio and Mawutor 
(2015) had the smallest one (0.054).

Table 3 

Random-effect Model

Estimate se z P CI Lower Bound CI Upper Bound
Intercept 0.303 0.0587 5.17 < .001 0.188 0.418

Table 4 

Heterogeneity Statistics

I² H² Df Q P

0.181
0.0328  

(SE= 0.0171)
89.08% 9.155 10 108.574 < .001

Table 3 provides the random-effects model, and Table 4 presents 
the heterogeneity statistics. Based on the random-effects model, the 
estimated average Fisher r-to-z transformed correlation coefficient 
was 0.303. (95% CI: 0.188 to 0.418) and was statistically significant 
(z = 5.17, p < 0.001). It showed that statistically there was a positive 
correlation between religiosity and saving behaviour, with a true effect 

Table 3  

Random-effect model 

 Estimate se z P CI Lower Bound CI Upper Bound 

Intercept 0.303. 0.0587 5.17 < .001 0.188 0.418 

 

Table 4  

Heterogeneity statistics 

  ² I² H² Df Q P 

 0.181 
0.0328  

(SE= 0.0171) 
89.08% 9.155 10 108.574 < .001 

Note: Random-effects model (k = 11); ² estimator: Hedges 

 

Table 5  

The source of heterogeneity analysis 

Heterogeneity 
source 

Coefficients T P 
95% Confidence Interval 

Lower Upper 
Publication year -0.131 -0.395 0.703 -0.895 0.634 

Sample size -0.334 -1.007 0.343 -1.098 0.431 
Muslim religion -0.535 -1.79 0.111 -1.22 0.153 

 

 

Figure 1 Forest plot 

Table 3  

Random-effect model 

 Estimate se z P CI Lower Bound CI Upper Bound 

Intercept 0.303. 0.0587 5.17 < .001 0.188 0.418 

 

Table 4  

Heterogeneity statistics 

  ² I² H² Df Q P 

 0.181 
0.0328  

(SE= 0.0171) 
89.08% 9.155 10 108.574 < .001 

Note: Random-effects model (k = 11); ² estimator: Hedges 

 

Table 5  

The source of heterogeneity analysis 

Heterogeneity 
source 

Coefficients T P 
95% Confidence Interval 

Lower Upper 
Publication year -0.131 -0.395 0.703 -0.895 0.634 

Sample size -0.334 -1.007 0.343 -1.098 0.431 
Muslim religion -0.535 -1.79 0.111 -1.22 0.153 

 

 

Figure 1 Forest plot 



84        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

size of 0.303 and a significance value less than 0.001. Meanwhile, 
according to heterogeneity statistics, the true effects appear to be non-
homogenous (Q(10) = 108.574, p < 0.001). By using the Hedges’ 
estimator, the      index (0.0328) agreed with the Q-test result, indicating 
that there was some between-study heterogeneity in the data, whereas     
     = 0.181 indicated that the true effect sizes had an estimated standard 
deviation of SD = 0.181. In addition, the H2 (9.155) and I2 (89.08%) 
indices confirmed that true effect size differences account for more 
than half of the variation in the studies, which meant that the level of 
heterogeneity was high.

Table 5 

The Source of Heterogeneity Analysis

Heterogeneity 
source

Coefficients T P
95% Confidence Interval

Lower Upper
Publication year -0.131 -0.395 0.703 -0.895 0.634

Sample size -0.334 -1.007 0.343 -1.098 0.431
Muslim religion -0.535 -1.79 0.111 -1.22 0.153

Based on a random-effects model, Figure 1 depicts correlation 
coefficients with corresponding 95 percent confidence intervals 
for each study graphically. The estimated r from each study ranged 
from 0.0541 (Ababio & Mawutor, 2015) to 0.6416 (Priyo Nugroho 
et al., 2017). Meanwhile, the weights ranged from 7.99 percent to 
9.92 percent, in Prastiwi (2021), with Satsios and Hadjidakis (2017) 
having the lowest and Kassim et al. (2019) having the highest. On the 
other hand, with Kassim et al. (2019) having the shortest interval and 
Prastiwi (2021) having the longest interval, the 95 percent confidence 
intervals ranged from (-0.02, 0.15) to (0.02, 0.42). Furthermore, the 
majority of confidence intervals were completely positive of zero, 
indicating that the majority of studies had a statistically significant 
positive effect. However, in some studies, i.e., Ababio and Mawutor 
(2015) and Kassim et al. (2019), the confidence intervals were 
not entirely positive of zero, indicating that these studies had a 
statistically insignificant positive effect (0.05 and 0.06). Thus, all 
observed dispersions reflected genuine differences in effect size and 
p-value between studies. However, a meta-analysis method only uses 
the effect size from each study rather than the p-value (Borenstein et 
al., 2011).



    85      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

Figures 2, 3, and 4 provide the outcomes of outlier and influential case 
diagnostics. Figure 2 shows the standardized residual values, which 
are used to identify outliers, while Figures 3 and 4 show the DFFITS 
values and Cook’s distances, which are used to detect influential cases. 
A look at the studentized residuals revealed that none of the studies 
had a value greater than ±3, indicating that there were no outliers in 
the random-effect model. Meanwhile, based on the DFFITS values of 
no more than 2 and Cook’s distances of no more than 1, none of the 
studies could be considered overly influential.

Figure 1 

Forest Plot

Figure 2 

Standardized Residual

Table 3  

Random-effect model 

 Estimate se z P CI Lower Bound CI Upper Bound 

Intercept 0.303. 0.0587 5.17 < .001 0.188 0.418 

 

Table 4  

Heterogeneity statistics 

  ² I² H² Df Q P 

 0.181 
0.0328  

(SE= 0.0171) 
89.08% 9.155 10 108.574 < .001 

Note: Random-effects model (k = 11); ² estimator: Hedges 

 

Table 5  

The source of heterogeneity analysis 

Heterogeneity 
source 

Coefficients T P 
95% Confidence Interval 

Lower Upper 
Publication year -0.131 -0.395 0.703 -0.895 0.634 

Sample size -0.334 -1.007 0.343 -1.098 0.431 
Muslim religion -0.535 -1.79 0.111 -1.22 0.153 

 

 

Figure 1 Forest plot 

14 
 

 
 

Figure 2  
 

Standardized Residual 

 

Figure 3  
 

DFFITS Values 

 

 
 
 
 
 
 
 
 
 
 
 
 
 



86        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Figure 3 

DFFITS Values

Figure 4 

Cook’s Distances

The current study’s goal was to examine the existing empirical 
evidence on the relationship between religiosity and saving behaviour 
through a meta-analysis approach. The findings show that religiosity 
has a positive, but weak correlation with saving behavior with r = 
0.303 (p < 0.001; 95% CI = (0.188; 0.418)). As a rule of thumb, the 
correlation strength is described as negligible (r < 0.2), low (0.2 ≤ r 
< 0.4), moderate (0.4 ≤ r < 0.7), high (0.7 ≤ r < 0.9), or very high (r 
≥ 0.9) (Guilford & Fruchter, 1973). According to this rule of thumb, 
the correlation will be ignored when its strength is less than 0.2. 
This finding implies that changes in religiosity have little impact on 
changes in saving behaviour. Moreover, the low r value was caused 
by the fact that the r values of the observed studies were mostly low 
(see Figure 1), and even two of them were negligible, i.e., Ababio and 

14 
 

 
 

Figure 2  
 

Standardized Residual 

 

Figure 3  
 

DFFITS Values 

 

 
 
 
 
 
 
 
 
 
 
 
 
 

15 
 

Figure 4  
 

Cook's Distances 

 
The current study's goal was to examine the existing empirical evidence on the relationship between 
religiosity and saving behaviour through a meta-analysis approach. The findings show that religiosity has 
a positive, but weak correlation with saving behavior with r = 0.303 (p < 0.001; 95% CI = (0.188; 0.418)). 
As a rule of thumb, the correlation strength is described as negligible (r < 0.2), low (0.2 ≤ r < 0.4), 
moderate (0.4 ≤ r < 0.7), high (0.7 ≤ r < 0.9), or very high (r ≥ 0.9) (Guilford & Fruchter, 1973). 
According to this rule of thumb, the correlation will be ignored when its strength is less than 0.2. This 
finding implies that changes in religiosity have little impact on changes in saving behaviour. Moreover, 
the low r value was caused by the fact that the r values of the observed studies were mostly low (see 
Figure 1), and even two of them were negligible, i.e., Ababio and Mawutor (2015), and Kassim et al. 
(2019). Meanwhile, there were only four studies that provided empirical evidence that religiosity had a 
moderate impact on saving behaviour, namely Mei Teh et al. (2019), Murdayanti et al. (2020), Priyo 
Nugroho et al. (2017),  and Yayeh (2014). On the other hand, Ababio and Mawutor (2015) obtained the 
lowest r value (0.05), showing that religious belief had no effect on household savings. They assumed that 
religious beliefs that promoted values like frugality, hard work, and honesty did not increase the savings 
habits of households in Ghana. Likewise, Kassim et al. (2019) found that religiosity had no effect on 
saving behaviour (r = 0.06). They assumed that an individual's level of religiosity would lead to the 
preference to spend money in God's name rather than saving it. Unfortunately, neither of them tried to dig 
a bit deeper into this finding, such as linking it to the lowest frequency value of the saving questions in 
their questionnaire. In fact, it could be a clue to   crucial evidence showing the low impact of the 
relationship between religiosity and saving behaviour. 

The result obtained also supports the notion that religious people have varying understandings of saving 
behaviour. These understandings stem from two opposing perspectives on saving money taught by 
various religions, including Christianity and Islam. These religions consider saving to be either a positive 
or negative practice (Yayeh, 2014). In terms of saving as a positive practice, Christianity and Islam, for 
example, teach saving when there is enough to go around in times of scarcity (Genesis 37-50; Sahih al-
Bukhari 5357). Religious people who put into practice this belief will also use their saving behaviour to 
support their frugal lifestyle (Agarwala et al., 2019). Meanwhile, in terms of saving as a negative practice, 
religions such as Christianity and Islam, for example, warn against the dangers of hoarding wealth (Luke 
12:16-21; Quran 9:34). Religious people who engage in this practice are motivated by fears that if saving 
money becomes a passion, they will become trapped in a cycle of amassing wealth and being stingy one 



    87      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

Mawutor (2015), and Kassim et al. (2019). Meanwhile, there were 
only four studies that provided empirical evidence that religiosity had 
a moderate impact on saving behaviour, namely Mei Teh et al. (2019), 
Murdayanti et al. (2020), Priyo Nugroho et al. (2017),  and Yayeh 
(2014). On the other hand, Ababio and Mawutor (2015) obtained the 
lowest r value (0.05), showing that religious belief had no effect on 
household savings. They assumed that religious beliefs that promoted 
values like frugality, hard work, and honesty did not increase the 
savings habits of households in Ghana. Likewise, Kassim et al. (2019) 
found that religiosity had no effect on saving behaviour (r = 0.06). 
They assumed that an individual’s level of religiosity would lead to 
the preference to spend money in God’s name rather than saving it. 
Unfortunately, neither of them tried to dig a bit deeper into this finding, 
such as linking it to the lowest frequency value of the saving questions 
in their questionnaire. In fact, it could be a clue to   crucial evidence 
showing the low impact of the relationship between religiosity and 
saving behaviour.

The result obtained also supports the notion that religious people have 
varying understandings of saving behaviour. These understandings 
stem from two opposing perspectives on saving money taught by 
various religions, including Christianity and Islam. These religions 
consider saving to be either a positive or negative practice (Yayeh, 
2014). In terms of saving as a positive practice, Christianity and Islam, 
for example, teach saving when there is enough to go around in times 
of scarcity (Genesis 37-50; Sahih al-Bukhari 5357). Religious people 
who put into practice this belief will also use their saving behaviour 
to support their frugal lifestyle (Agarwala et al., 2019). Meanwhile, in 
terms of saving as a negative practice, religions such as Christianity 
and Islam, for example, warn against the dangers of hoarding wealth 
(Luke 12:16-21; Quran 9:34). Religious people who engage in this 
practice are motivated by fears that if saving money becomes a 
passion, they will become trapped in a cycle of amassing wealth and 
being stingy one day (Sawyer, 1954). As a result, in studies where 
there is empirical evidence that religiosity has a moderate impact 
on saving behaviour, their respondents are likely to view saving as 
a positive practice. In contrast, respondents in studies with empirical 
evidence that religiosity has a low impact on saving behaviour may be 
less open to the idea of saving as a positive practice.

In addition, the current study discovered heterogeneity at a high level 
(Q(10) = 108.574, p < 0.001; I2 = 89.08%). Heterogeneity may exist, 



88        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

because of differences in study quality (e.g., values for effect size and 
significance), methodology, sample size, demographic factors, and 
respondent characteristics. From a statistical perspective, quantifying 
heterogeneity can be used to determine whether or not population 
effect sizes are likely to be consistent or varying (Borenstein et al., 
2011; Hedges & Olkin, 1985). Meanwhile, the summary information 
shown in Tables 1 and 2 demonstrates that the characteristics of 
studies were not all the same. There were differences even within the 
studies themselves, like variation in religious factors (see Table 2). 
For instance, Ismail et al. (2018); and Yayeh (2014) collected data 
from a variety of religious adherents, whereas other studies only 
collected data from a single religious adherent. As a result, the present 
study was motivated to investigate the cause of heterogeneity using 
a technique known as meta-regression analysis. In meta-regression, 
the effect size of each study as the independent variable is regressed 
on the study characteristics as the dependent variable (Chen & Peace, 
2021). Furthermore, the year of publication, sample size, and Muslim 
religion were investigated to determine the source of heterogeneity. 
They acted as independent variables in the meta-regression analysis 
because they were relatively complete data. The results of meta-
regression analysis then revealed that publication year, sample size, 
and Muslim religion were not statistically significant for the presence 
of heterogeneity (see Table 5).

CONCLUSION AND FUTURE STUDIES

The current study used a meta-analysis to synthesize the findings 
of previous studies to determine the effect of religiosity on saving 
behaviour. The meta-analysis of 11 journal articles and 1,063 
respondents revealed that religiosity has a low impact on saving 
behaviour. The current study also confirmed the notion that religious 
people have two different perspectives on saving behaviour, holding 
the divergent view that saving can be either a negative or positive 
practice. Because the study’s findings indicated that religiosity has 
little influence on saving behaviour, it is possible that people with a 
high level of religiosity have a less rigid perspective on saving as a 
positive behaviour.

The findings have important implications for the development of 
theories in the field of saving behaviour. In researching religiosity and 



    89      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

saving behaviour, this study has pioneered a new analytic approach 
known as meta-analysis. It has provided a retrospective summary 
of the existing literature on the relationship between religiosity and 
saving behaviour. It could help researchers make a more informed 
decision about which variables to use. Furthermore, this study has 
added to the body of knowledge about the true effect of the relationship 
between religiosity and saving behaviour. It could be used as a model 
for future research as well as a tool for additional analysis.

The findings also have practical implications for increasing public 
awareness. This study has provided insights for authorities and financial 
institutions interested in encouraging religious people to save. More 
understanding of the findings can aid them in the improvement of 
plans focused on savings advocacy and savings facilitation. Savings 
advocacy would help religious households and individuals understand 
the importance of saving and resource management. It would be a 
critical step in developing their saving habits. Meanwhile, financial 
institutions can create a type of savings account that corresponds to 
the religious predisposition. It would promote the notion that saving 
money is a good deed. The plans are beneficial in encouraging 
religious people to engage in responsible financial behaviour. 
Collective savings will have a significant impact on the economy if 
these individuals have a strong savings predisposition.

However, a number of important limitations of the study have to be 
considered. First, the current meta-analytic method has synthesized 
only a few studies. Although the limited sample used in the present 
study was acceptable, the findings would be more reliable if a larger 
number of studies could be synthesized. Second, of the 11 previous 
studies that were synthesized, Islam was the religion which had the 
highest proportion of adherents, with only a small proportion of 
adherents from other religions. Although that was the nature of the 
meta-analytic samples, the current pattern of results could be more 
generalizable if the proportion of adherents was more balanced. 
Third, the current study synthesized 11 studies had used primary data 
and measured variables with a questionnaire. This technique opens 
the possibility of a person’s proclivity to respond to a questionnaire 
with a positive self-image, also known as a socially desirable response 
(Van de Mortel, 2008). Although bias has been reduced by selecting 
studies from reputable journals, it is still possible to be biased when 
answering sensitive questions, such as religious ones.



90        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Furthermore, more research is needed to examine the relationship 
between religiosity and saving behaviour, more specifically, the 
importance of using longitudinal studies at multiple points in 
time. This type of research could shed light on the minor impact 
of the relationship and contribute to a critical explanation for this 
phenomenon.

ACKNOWLEDGMENT

This research received no specific grant from any funding agency in 
the public, commercial, or not-for-profit sectors.

REFERENCES

Ababio, A. G., & Mawutor, G. (2015). Does religion matter for 
savings habit of households in Ghana? Singaporean Journal 
of Business Economics and Management Studies, 4(8). https://
doi.org/10.12816/0019680

Abdullah, N., & Abd. Majid, M. S. (2001). Saving behaviour in 
Islamic framework: The case of international Islamic University 
Malaysia. Jurnal Syariah, 9(2), 61-84. https://mjs.um.edu.my/
index.php/JS/article/view/22905. 

Agarwala, R., Mishra, P., & Singh, R. (2019). Religiosity and consumer 
behavior: A summarizing review. Journal of Management, 
Spirituality & Religion, 16(1), 32-54. https://doi.org/10.1080/
14766086.2018.1495098

Ajzen, I. (1991). The theory of planned behavior. Organizational 
Behavior and Human Decision Processes, 50(2), 179-211. 
https://doi.org/10.1016/0749-5978(91)90020-t

Allport, G. W., & Ross, J. M. (1967). Personal religious orientation 
and prejudice. Journal of Personality and Social Psychology, 
5(4), 432-443. https://doi.org/10.1037/h0021212

Arruñada, B. (2010). Protestants and Catholics: Similar work ethic, 
different social ethic. The Economic Journal, 120(547), 890-
918. https://doi.org/10.1111/j.1468-0297.2009.02325.x

Azzi, C., & Ehrenberg, R. (1975). Household allocation of time and 
church attendance. Journal of Political Economy, 83(1), 27-56. 
https://doi.org/10.1086/260305

Bandura, A. (1977). Social learning theory. Prentice-Hall.



    91      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

Barro, R. J., & McCleary, R. M. (2003). Religion and economic 
growth across countries. American Sociological Review, 68(5), 
760. https://doi.org/10.2307/1519761

Basedau, M., Gobien, S., & Prediger, S. (2018). The multidimensional 
effects of religion on socioeconomic development: A review of 
the empirical literature. Journal of Economic Surveys, 32(4), 
1106-1133. https://doi.org/10.1111/joes.12250

Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). 
Cross-cultural psychology: Research and applications: 
Cambridge University Press.

Bilen, M. (2016). Global wealth inequality: Islamic economic 
perspective for market or state oriented solution. Turkish 
Journal of Islamic Economics, 3(1), 17-35. https://doi.
org/10.15238/TUJISE.2014.3.1.17-35

Borenstein, M., Hedges, L. V., Higgins, J. P. T., Rothstein, H. R., & 
Higgins, D. J. P. T. (2011). Introduction to meta-analysis. John 
Wiley & Sons.

Chen, D.-G. D., & Peace, K. E. (2021). Applied meta-analysis with R 
and Stata. CRC Press.

Cheung, M. W.-L., Ho, R. C. M., Lim, Y., & Mak, A. (2012). 
Conducting a meta-analysis: Basics and good practices. 
International Journal of Rheumatic Diseases, 15(2), 129-135. 
https://doi.org/10.1111/j.1756-185X.2012.01712.x

Cochran, W. G. (1954). Some methods for strengthening the 
common χ2 tests. Biometrics, 10(4), 417-451. https://doi.
org/10.2307/3001616

Cook, R. D. (1977). Detection of influential observation in 
linear regression. Technometrics, 19(1), 15-18. https://doi.
org/10.2307/1268249

Davutyan, N., & Öztürkkal, B. (2016). Determinants of saving-
borrowing decisions and financial inclusion in a high middle 
income country: The Turkish case. Emerging Markets Finance 
and Trade, 52(11), 2512-2529. https://doi.org/10.1080/154049
6X.2016.1187596

Field, A. P., & Gillett, R. (2010). How to do a meta-analysis. British 
Journal of Mathematical and Statistical Psychology, 63(3), 
665-694. https://doi.org/10.1348/000711010X502733

Filipova, L. (2012). Are attitudes conducive to economic growth 
stronger in protestants than in others? https://nupi.brage.
unit.no/nupi-xmlui/bitstream/handle/11250/277240/WP--804-
Lenka-R.pdf?sequence=3



92        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Geertz, C. (1973). The interpretation of cultures: Selected essays 
(Vol. 5019): Basic books.

Gerbing, D. (2014). Influence of data points on the estimated model. 
School of Business Administration Portland State University.  
http://web.pdx.edu/~gerbing/515/Resources/Influence.pdf

Glock, C. Y. (1962). On the study of religious commitment. 
Religious Education, 57(sup4), 98-110. https://doi.
org/10.1080/003440862057S407

Guilford, J. P., & Fruchter, B. (1973). Fundamental statistics in 
psychology and education (5th ed.). McGraw-Hill.

Guiso, L., Sapienza, P., & Zingales, L. (2003). People’s opium? 
Religion and economic attitudes. Journal of Monetary 
Economics, 50(1), 225-282. https://doi.org/10.1016/S0304-
3932(02)00202-7

Hayward, R. D., & Kemmelmeier, M. (2011). Weber revisited: A 
cross-national analysis of religiosity, religious culture, and 
economic attitudes. Journal of Cross-Cultural Psychology, 
42(8), 1406-1420. https://doi.org/10.1177/0022022111412527

Hedges, L., & Olkin, I. (1985). Statistical methods for meta-analysis. 
Academic Press.

Higgins, J. P., & Thompson, S. G. (2002). Quantifying heterogeneity 
in a meta-analysis. Statistics in Medicine, 21(11), 1539-1558. 
https://doi.org/10.1002/sim.1186

Iannaccone, L. R. (1998). Introduction to the economics of religion. 
Journal of Economic Literature, 36(3), 1465-1495. 

Inglehart, R. (2018). Culture shift in advanced industrial society.
Princeton University Press.

Ismail, S., Khairuddin, N. S., Alias, N. E., Loon-Koe, W., & Othman, 
R. (2018). An empirical analysis of saving behavior among 
Malaysian employees. International Journal of Academic 
Research in Business and Social Sciences, 8(10), 1070-1080. 
https://doi.org/10.6007/ijarbss/v8-i10/4822

Iyer, S. (2016). The new economics of religion. Journal of 
Economic Literature, 54(2), 395-441. https://doi.org/10.1257/
JEL.54.2.395

Kassim, A. A. M., Mohamed, J., Jahari, N. A., Nasyikin, N., & 
Zain, M. (2019). Personal saving behavior among Muslim 
employees. International Journal of Academic Research in 
Business and Social Sciences, 9(11). https://doi.org/10.6007/
ijarbss/v9-i11/6545

Keister, L. A. (2003). Religion and wealth: The role of religious 
affiliation and participation in early adult asset accumulation. 



    93      

The International Journal of Banking and Finance, Vol. 18, Number 1 (January)  2023, pp: 67–94

Social Forces, 82(1), 175-207. https://doi.org/10.1353/
sof.2003.0094

Klaubert, A. (2010). Striving for savings: Religion and individual 
economic behavior Working Paper Series in Economics. 
Lüneburg: Leuphana Universität Lüneburg, Institut für 
Volkswirtschaftslehre.

Law, K. S. (1995). The use of Fisher’s Z in Schmidt-Hunter. Type 
meta-analyses. Journal of Educational and Behavioral 
Statistics, 20(3), 287-306. https://doi.org/10.2307/1165404

Lenhard, W., & Lenhard, A. (2016). Calculation of effect sizes.   
https://www.psychometrica.de/effect_size.html

Lenski, G. (1961). The religious factor: A sociological study of 
religion’s impact on politics, economics, and family life. 
Garden City.

Mei Teh, G., Palil, M. R., Mohd Zain, N. H., & Mamat, N. H. (2019). 
Predicting private saving within Malay community. Asian 
Academy of Management Journal, 24(Supp.1), 1-17. https://
doi.org/10.21315/aamj2019.24.s1.1

Murdayanti, Y., Ulupui, I., Gurendrawati, E., & Purwohedi, U. (2020). 
An assessment of saving behavior in the religious education 
institution. Journal of Southwest Jiaotong University, 55(4). 
https://doi.org/10.35741/ISSN.0258-2724.55.4.27

Nyhus, E. K. (2017). Saving behaviour: Economic and psychological 
approaches. In R. Ranyard (Ed.), Economic Psychology (pp. 
206). John Wiley & Sons.

Pargament, K. I. (1999). The psychology of religion and 
spirituality? Yes and no. The International Journal for the 
Psychology of Religion, 9(1), 3-16. https://doi.org/10.1207/
S15327582IJPR0901_2

Pew Research Center. (2015). The future of world religions: Population 
growth projections, 2010-2050. https://assets.pewresearch.
org/wpcontent/uploads/sites/11/2015/03/PF_15.04.02_
ProjectionsFullReport.pdf

Prastiwi, I. E. (2021). Analysis of the influence of religiosity, 
environment, and reputation factors on customer decisions on 
saving at KSPPS BMT amanah ummah. International Journal 
of Economics, Business and Accounting Research (IJEBAR), 
5(1). https://doi.org/10.29040/IJEBAR.V5I1.2237

Priyo Nugroho, A., Hidayat, A., & Kusuma, H. (2017). The influence 
of religiosity and self-efficacy on the saving behavior of the 
Іslamic banks. Banks and Bank Systems, 12(3), 35-47. https://
doi.org/10.21511/bbs.12(3).2017.03



94        

The International Journal of Banking and Finance, Vol. 18, Number 1 (January) 2023, pp: 67–94

Renneboog, L., & Spaenjers, C. (2012). Religion, economic attitudes, 
and household finance. Oxford Economic Papers, 64(1), 103–
127. https://doi.org/10.1093/oep/gpr025

Rosenthal, R. (1995). Writing meta-analytic reviews. Psychological 
B u l l e t i n , 11 8 ( 2 ) , 1 8 3 . h t t p s : / / d o i . o r g / 1 0 . 1 0 3 7 / 0 0 3 3 -
2909.118.2.183

Satsios, N., & Hadjidakis, S. (2017). The effect of religiosity and self 
mastery on saving behaviour of Pomak households in thrace. 
Journal Sustaibable Development, Culture, Traditions, 17-27. 
https://doi.org/10.26341/issn.2241-4002-2017-1b-2

Sawyer, W. (1954). Christ’s teachings on the use of money.  https://
scholarworks.harding.edu/wyatt-sawyersermons/281

Singh, G., Sharma, S., Sharma, R., & Dwivedi, Y. K. (2021). 
Investigating environmental sustainability in small family-
owned businesses: Integration of religiosity, ethical judgment, 
and theory of planned behavior. Technological Forecasting 
and Social Change, 173, 121094. https://doi.org/10.1016/j.
techfore.2021.121094

The Jamovi Project. (2021). Jamovi (Version 1.6) [Computer 
Software].  https://www.jamovi.org

Teshome, H. M., Ayalew, G. D., Shiferaw, F. W., Leshargie, C. T., 
& Boneya, D. J. (2018). The prevalence of depression among 
diabetic patients in Ethiopia: A systematic review and meta-
analysis. Depression Research and Treatment, 6135460. https://
doi.org/10.1155/2018/6135460

Tjahjono, G. (2014). Religiosity and the intention to buy luxury goods 
among young Indonesian Muslims in Jakarta (Unpublished 
master’s thesis). University of Western Sydney.   

Van de Mortel, T. F. (2008). Faking it: Social desirability response 
bias in self-report research. Australian Journal of Advanced 
Nursing, 25(4), 40-48. 

Weber, M. (1905). The Protestant ethic and the spirit of capitalism. 
Charles Scribner’s Sons.

Wijaya, I. F., Hakim, A. R., Saputro, N., & Mulyadi, M. (2019). 
Religiosity level and saving decisions in Baitul Maal wat 
Tamwil: The case of Indonesia. Journal of Islamic Marketing, 
11(6), 1465-1483. https://doi.org/10.1108/JIMA-09-2018-0160

Yayeh, F. A. (2014). The impact of religion on household saving 
behavior. Microeconomics: Intertemporal Consumer Choice & 
Savings eJournal. https://doi.org/10.2139/ssrn.2518129

Zuckerman, M., Silberman, J., & Hall, J. A. (2013). The  relation between 
intelligence and religiosity: A meta-analysis and some proposed 
explanations. Personality and Social Psychology Review, 
17(4), 325-354. https://doi.org/10.1177/1088868313497266