Development of Muslim Identity Expression Scale

This study aimed to investigate the psychometric properties of Muslim identity expression scale, a measurement instrument used to evaluate various manifestations. Muslim identity obtained expression in a multitude of domains, including the private sphere, politics, economics, socio-culture, legal matters, and education. The expression was categorized into two distinct functions, namely consolidation and mobilization. In the context of Indonesia, consolidation was defined as an expression that enhanced identity of individuals as Muslim. Meanwhile, mobilization changed the condition of Muslim, often related to power and politics, such as participating in political actions to defend Islam or supporting the establishment of political parties. This study comprised two distinct phases, namely the development of the scale and the examination of the structural validity. Based on a review of relevant literature and results from an open-ended questionnaire, 55 items were compiled and administered to 546 subjects. The results showed that 34 items met the criteria consisting of four dimensions, namely personal, interpersonal, collective action, and political participation expression. Confirmatory factor analysis (CFA) with 34 items and four dimensions matched the structure offered ( ꭓ 2 = 1793.805 df = 521, p < .01; RMSEA = 0.066, 90% CI = 0.062-0.069; CFI = 0.945; TLI = 0.94). The results provided preliminary evidence to support Muslim identity expression scale as a valid measuring tool with sound conceptual and empirical psychometric bases.


Introduction
Religion is an identity that offers several benefits to its adherents according to psychological study.Individuals with relatively strong religious identities tend to show lower depressive symptoms, improved self-confidence, and better self-control in avoiding risky behavior (Koenig, 2009;McCullough & Willoughby, 2009;Sedikides & Gebauer, 2010) while fulfilling the need for meaning in life.Furthermore, adherents are inclined to maintain, strengthen, or defend the religion by expressing identity (Klein et al., 2007), which depends on expression (Haslam & Reicher, 2007).
The definition of Muslim identity expression is any form of behavior aimed at showing Muslim identity, such as social action, physical appearance (e.g., the use of group symbols or attributes), and verbal expression describing group attitudes.Identity can be expressed in various domains, such as private, political, economic, socio-cultural, law, and education.The scale typically focuses on activities such as performing prayers, reading the Qur'an, visiting mosques, and wearing the headscarf (Abu-Rayya et al., 2009;Abu Raiya et al., 2008;Francis et al., 2008;Jasperse et al., 2012).Islam is known as an orthopractic religion, implying that differences in the level of diversity can be observed from the application of the teachings in daily life (El-Menouar, 2014).This religion can also be expressed in public spaces, such as politics, economics, socio-culture, law, and education (Ansor, 2016;Fealy & White, 2008;Hasan, 2009).
According to social identity performance theory, identity is expressed in various ways and grouped into two categories based on consolidation and mobilization functions (Klein et al., 2007).In the context of Islam, the two functions correspond with categorizing the religious manifestations as Cultural and Political (Azra, 2012;Effendy, 2011;Tibi, 2005).Furthermore, using the framework of social identity performance theory, this study constructs Muslim identity expression instrument with diverse choices.
There is some urgency in developing a scale for the expression of Muslim identity.First, the scale measures the expression of identity as Muslim.In the context of research and psychological studies, it understands variations in the concept and the relationship to factors influencing the expression of religious identity, such as contextual, personality, or religious beliefs.Second, this instrument can be used by the government and related institutions to understand the forms of Muslim identity expression.The formulated policies are more inclusive and responsive to various groups.Third, the scale provides information needed to increase internal and inter-religious awareness among diverse communities.By understanding the expression, greater tolerance between different groups can be built.The information generated can overcome stereotypes from ignorance or lack of understanding of the diverse ways of articulating religious identities.

Islam in Indonesia
Indonesia, the country with the fourth largest number of Islam followers in the world, holds religion as a significant aspect of the culture (Hamdi Muluk et al., 2018).Several studies showed that the majority of Muslim strongly believed in the religion and actively participated in worship (Fossati et al., 2017;Hassan, 2006;Pepinsky et al., 2018).Since the collapse of the New Order regime in 1998, there has been an increase in Muslim identity expression in different forms.In Indonesia, identity is expressed through religious practices and other aspects of life (Ansor, 2016;Fealy & White, 2008;Hasan, 2009).
The manifestation of Islam can be categorized into two Cultural and Political aspects (Azra, 2012;Effendy, 2011;Tibi, 2005).Cultural Islam is influenced by Nurcholish Majid and Abdurahman Wahid, who view Islam as separate from political power.It advocates for the practice in daily life by paying attention to local culture, and guiding principles for national and state affairs (Burhani, 2013).This is because the imposition of a single religious teaching has the potential to create division in the nation.Therefore, it is important to also consider the needs of other religious groups in Indonesia.According to the perspective of cultural Islam, Muslim identity expression should be strengthened.This expression can JP3I (Jurnal Pengukuran Psikologi dan Pendidikan Indonesia), 12(2), 2023 110-126 http://journal.uinjkt.ac.id/index.php/jp3iThis is an open access article under CC-BY-SA license (https://creativecommons.org/licenses/by-sa/4.0/)appear in the form of freedom in worship, use of religious symbols, Islamic education, and an increased presence of themes in the public spaces, such as the growing number of Islamic shows on television, without demanding more power from Muslim.
The idea promoted encourages Islam to rule in all fields (Fauzia et al., 2011;Sakai & Fauzia, 2014).The expression of identity mobilization is carried out by individuals to achieve group goals.In this context, several objectives are debated and pursued by various Islamic groups.The first is to demand an Islamic-based political system or the establishment of a caliphate (Muhtadi, 2009).The second is the complete and comprehensive implementation of Islamic law (Bush, 2008;Munabari, 2017).The third is to demand greater representation of Muslim in state leadership institutions (Hicks, 2012;Muhtadi, 2012).

Social Identity Performance Theory
The theoretical framework used to explain Muslim identity expression is the social identity performance theory, which is an extension of the social identity deindividuation effect (SIDE).The model (SIDE model;Reicher et al., 1995) is rooted in Social Cognitive Theory (Turner & Reynolds, 2012), explaining the behavior of individuals in groups.Furthermore, it offers a distinctive perspective, suggesting that the experience of deindividuation consciously conforms to the expected group norms.
The explanation showed the cognitive dynamics of the SIDE theory.Klein et al., (2007) expanded the discussion of identity display strategies.Identity performance is a behavior that aims to express or suppress social identity.This is relevant to the norms applicable to prominent social identities and takes various forms, such as social action, physical appearance (e.g., the use of group symbols or attributes), as well as verbal expression describing group attitudes.The three points obtained from this definition are, first, behavior is carried out to express social identity, distinguishing the concept from those related to habitual factors.Second, the behavior should enable others to recognize that the individual performing the act is part of a particular group.This includes behavior that may not conform to group norms but still reflects identity performance.Third, the behavior is carried out for the benefit of the group, rather than individual.
Identity performance has two functions, namely identity consolidation and mobilization (Klein et al., 2007;Scheepers et al., 2003).Identity consolidation is aimed at strengthening and confirming group identity and this occurs when individuals feel uncertain or insecure about the membership.Identity consolidation may be pursued when individuals perceive that the group existence is threatened or recognized in the larger social system.Meanwhile, identity mobilization has been reported to initiate change in social reality.

Methods
The first analysis of this study aims to identify themes and forms of expression of Muslim identity.Questionnaires were distributed with open questions to mention important things related to the expression as Muslim online.The second examines the factor structure of Muslim identity expression scale using Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).In addition, the convergent and discriminant validity of the scale was measured with an instrument related to the Political Ideology construct.The entire study process was approved by the Research Ethics Committee at the Faculty of Psychology, University of Indonesia, Number 107/FPsi.Komite Etik/PDP.04.00/2022.

Participants and Instruments
The first study aimed to create a set of items for Muslim identity expression.The questionnaire was distributed online with questions, such as "Mention 10 things to express identity as a Muslim?This should be answered according to what comes to the mind."Concerning the first part of the questionnaire, participants were provided with information related to the aim.In the subsequent two parts, questions were answered regarding demographic information and work with instruments.Invitations with a brief description of the study and a link to the online questionnaire were disseminated through various social media networks and the Internet.This was conducted for two weeks recruiting participants asked to share the link with the contacts.
In the second study, EFA was used to extract the factors from a set of identity expression items.Furthermore, CFA was carried out to confirm the factors.The second phase included 564 Muslim, ranging in age from 17 to 66 (M age = 32.4SD age = 9.92), and comprising 247 males and 317 females.Among participants, 3, 143, 290, and 125 had Junior High School, Senior High School, Undergraduate, and Postgraduate education.Approximately 125, 51, 100, and 15 individuals were identified as part of Nahdlatul Ulama, Muhammadiyah, Tarbiyah movement, and Salafi, respectively.Meanwhile, 273 participants were not affiliated with any organization.
The questionnaire was administered online and comprised of three parts.The first provided participants with information on the purpose of the study and seeks the consent to participate.The subsequent 2 parts included questions related to demographic data and work on instruments.Over a period of two weeks, invitations containing a brief description of the study and a link to access the online questionnaire were disseminated through various social media networks and the Internet.Participants were also promoted to share the questionnaire link with the contacts.In this context, participation was voluntary, and compensation was provided for the first 50 individuals.
The instruments used in the second phase of data collection consisted of two scales.The first was Muslim identity expression, which comprised 55 items.The response options ranged from 1 (strongly disagree) to 7 (strongly agree).The second scale was the political ideology scale, which consisted of 12 items and response options ranging from 1 (strongly disagree) to 7 (strongly agree) (Muluk et al., 2019).
Political ideology is defined as a set of interrelated attitudes related to state organization.The concept is measured using the Indonesian version of the scale, which includes religious, social, and economic dimensions (short version 12 items) (Muluk et al., 2019).The scale has 7 response options (1 = Liberal, 7 = Conservative) and (1 = Strongly Disagree, 7 = Strongly Agree).Examples of items from religious, social, and economic dimensions include 'In electing national leaders, one should be of the same religion,' 'Indonesia does not need to recognize LGBT rights,' and 'The government should provide cheap staple goods for the people', respectively.

First Study Results
The data were obtained based on behavior, physical appearance, or verbal expression aimed at showing Muslim identity to other parties.Table 1 shows the grouping of themes according to the forms of expression mentioned by participants.A total of 806 responses described ways of expressing Muslim Identity.Eating manners Eating and drinking with the right hand, eating in a sitting position, and praying before eating 5 1%

Environmental manners
Maintaining cleanliness and using water sparingly 4 0%

Islamic cultural art
Listening to Islamic songs, installing calligraphy at home, and reading Islamic books 3 0% According to Table 1, worship is the most frequent form of identity expression cited by participants, followed by morality and appearance.In the subsequent stage, 129 items were compiled using the expression themes in the first study.Some of the items originated from the forms of expression mentioned by participants.The forms of expression below 5% distinguish a Muslim from followers of other religions.Even though Islamic teachings do not mandate the concept, Muslim may be expressed in this typical way.Furthermore, interviews were conducted with psychometric, social psychology, and Islamic studies experts.The items were reduced to 55 in total, through careful consideration.Some items, such as obligatory worship, were excluded due to the inability to describe identity variations.In addition, a readability test was conducted with five undergraduate Muslim students, who found all items easy to understand with no corrections needed.The next stage included the distribution of questionnaire.

Exploratory Factor Analysis Results
EFA was conducted using SPSS version 26 and the distributions of scores for most of the items were leptokurtic and skewed.Therefore, data analytic strategies known to perform better with non-normal data were selected (Fabrigar et al., 1999).Principal axis factor (PAF) analysis was conducted on the 55 items to determine the suitability of the Pearson correlation matrix for factor analysis.The Kaiser-Meyer-Olkin index (.95) and the Bartlett roundness test [χ2(1485) = 16695.557,p < .001]showed that the matrix was in a suitable state.The number of factors to be extracted was determined using several criteria, namely latent Kaiser, scree plots, and parallel analysis (Worthington & Whittaker, 2006).The scree plots supported 7 factors, while the Kaiser latent criteria produced 10.In this context, the parallel analysis used a code developed by O'Connor (2000), which generated 100 random permutations of the original data set, showing a 5-factor extraction.PAF analysis was used to extract 5, 6, and 7-factor solutions as the final stage in determining the number of factors to be retrieved.However, the analysis results showed that the solutions with 5,6,7 factors were not satisfactory since a 4-factor solution was attempted.
This study selected the 4-factor solution closely consistent with the theoretical conceptualization.Furthermore, Promax rotation was performed to select items with 4 factors assumed to be correlated (Kahn, 2006).The items were selected based on cross-loading, commonalities, and factor loading below 2, 0.2, and 0.4, respectively (Hair et al., 2019).
Promax rotation resulted in the removal of 15 items as the factor loading was <0.4 and was removed as the commonalities were below 0.2.Furthermore, the fourth items were removed and were substantially different from the same factors.The second rotation obtained better factors, and no items were deleted after meeting the criteria.After 2 rotations, the best number of factors was found to be 4, with a total of 34 items.Muslim identity expression scale with 34 items in the Indonesian version was found in Individual expression consists of items that describe individually performed identity.The concept includes expression in various fields, such as religious practices, Islamic economics, and entertainment.Interpersonal expression describes community relations, such as keeping clean and inviting others to perform good deeds.The expression of collective action consists of items related to various forms advocating for stakeholders about issues of Muslim community.Meanwhile, political participation consists of items regarding Islamic politics.

Confirmatory Factor Analysis Results
The factor structure of Muslim identity expression scale is tested using confirmatory analysis in Mplus 8 (Muthén & Muthén, 1998-2018) with weighted least squares mean and variance (WLSMV).The model includes four latent variables, namely individual, interpersonal, collective action, and political participation expression.This comprises 34 observable indicators, including 18 individual, 8 interpersonal, 4 collective action, and 4 political participation expression items.
Comparative Fit Index (CFI) and Tucker-Lewis Index (TLI) criteria were used to evaluate the model, with values higher than 0.90 and 0.95, showing an acceptable and very good fit, respectively.The Root Mean Square Error of Approximation (RMSEA) with a value below 0.08 shows an acceptable fit.Meanwhile, a 90% Confidence Interval (CI) of RMSEA with an upper limit value lower than 0.10 showed a good model fit (Byrne, 2012).CFA results showed that the 4-factor model did not fit well according to the chi-square test, even when other indicators were considered good ꭓ 2 = 1793.805df = 521, p < .01;root mean square error of approximation (RMSEA) = 0.066, 90% confidence interval for RMSEA = 0.062 to 0.069; comparative fit index (CFI) = 0.945; and Tucker-Lewis index (TLI) = 0.94.
Table 2 shows the comparison between models, with the 4-factor correlated model being the best.The items have significant loading factors, and detailed results are presented in Table 3.The relationship between the factors is shown in Table 4, with the highest correlation observed between individual and collective action expression (r=0.75, p <.001).Meanwhile, the lowest correlation was identified between interpersonal and political participation expression (r=0.45,p <.001).The blueprint of Muslim identity expression scale based on CFA can be seen in Table 6.

Reversed item
Convergent and discriminant validation tests were conducted by linking the constructs of Muslim identity expression with political ideology, using a scale of 3 dimensions, namely religious, political, and economic.The results showed a positive correlation between political ideology and participation, as well as choice of action (Caprara et al., 2017;Choma et al., 2020).This study reported that Muslim identity expression had a stronger relationship with the religious dimension, as evidenced by a higher value (r=.782, p <.001) compared to the social (r=.540, p<.001) and economic dimensions (r=.407, p < .001).

Discussion
This study aimed to develop a comprehensive consolidation represented by aspects of individual and interpersonal expression and mobilization represented by collective action and political participation.The early evidence of convergent and discriminant validity was observed from a higher correlation in ideology on the religious dimension than social and economic.Muslim identity could also be expressed in social and political life.The expression scale differed from the general religious behavior measure focused on worship practice.The scale was used to map and explain the religious behavior of Muslim from a social psychology approach.
EFA is performed to determine the number of factors that cause the correlation of a set of items.In this analysis, information is condensed to explain the variations by a smaller number of items, as well as defining substantive matters from the content or meaning of the latent variables (DeVellis, 2012).EFA provides evidence of the internal structure of the content test (Lambie et al., 2017) and construct validity (Peterson et al., 2017).The assessment should be used and interpreted with various populations (Hays & Wood, 2017).Furthermore, this study adopted Promax rotation to determine the factor structure formed, where the rotation did not limit the correlation.From the Promax rotation, four factors were considered following the theoretical concept.This resulted in the deletion of 14 items with a factor loading <0.40.The factors comprised individual, interpersonal, collective action, and expression of political participation.
In this case, the Promax rotation is considered appropriate and in line with the development of measurement tools to determine several dimensions.The possibility of a correlation was considered between factors as the primary explanation of the distribution of individual factors, interpersonal, collective action, and expression of political participation.In determining factors with EFA, the assumption of normality with the Barlett Test and the Kaiser-Meyer-Olkin index was met.For the Bartlett Test, the values obtained have fulfilled the normality assumption, which is <0.05.Furthermore, the Kaiser-Meyer-Olkin index obtained a value of 0.95 or close to 1.The index is a computation to determine the comparison between the number of squared partial and total correlation.The partial correlation, which forms the basis of factor analysis, greatly determines the required number.The smaller the partial correlation, the more confident the presence of minimal confounding variables.Kaiser-Meyer-Olkin was also defined as sample adequacy and sufficient.
After obtaining several factors found through EFA, analysis using CFA method is essential to obtain statistical evidence regarding the model fit to the data.Therefore, it is important to avoid fixating on the four-factor structure suggested by EFA.The re-evaluation of a one-factor and four-factor model should be conducted to examine the correlations between the factors.In this context, this current study should be carried out to test the structure of other factors in obtaining a fit model.
Based on the analysis with CFA, the results obtained did not fit in the general factor model with an RMSEA value of 0.108 > 0.05 (527) and a 4-factor model.Therefore, the factors are not correlated with each other under RMSEA 0.198 > 0.05 (527).A 4-factor model with correlated factors was found to be 0.066 > 0.07 (521), having an acceptable fit model (Hair et al., 2019).The test supports the Promax rotation when determining which items follow certain factors (Wood et al., 1996).
The statistical fit of the items was determined after obtaining a good model among several factors tested.The factor loadings obtained range from 0.601 to 0.905.The lowest factor load was the item "When there is an opportunity, as Muslim, I do not mind campaigning for a non-Islamic party with the vision and mission which I like."The item has a minor contribution to the factor or 0.6012 = 36% and JP3I (Jurnal Pengukuran Psikologi dan Pendidikan Indonesia), 12(2), 2023 119-126 http://journal.uinjkt.ac.id/index.php/jp3iThis is an open access article under CC-BY-SA license (https://creativecommons.org/licenses/by-sa/4.0/)67% described the other intent.Meanwhile, the item "I will take part in rallies against groups considered insulting or harassing Muslim" is the best item with a loading factor of 0.9052, or contributing 81.9% of the intended factor.
The information about individual factors, interpersonal, collective action, and expression of political participation was also obtained with a correlation of 0.459 to 0.751.Considering the results, there were no high-order factors because the correlation was low.
The scale was developed for Muslim to determine the different levels of identity.Several Islamic groups have different interpretations of teachings, specifically regarding the role of Islam and politics.Therefore, the scale obtained items with common themes in all Islamic groups and various responses, such as selecting candidates for Muslim leaders and Islamic political parties (Baswedan, 2004) and other domains (Fealy & White, 2008).Further study should determine the differences in expression between Islamic groups.
Similar to other scientific investigations, this study has several limitations considered when interpreting the results.This was conducted on a single sample, with a doubtful outcome.Therefore, further studies should be performed on more extensive and diverse samples from various Islamic organizations.It is also necessary to conduct a test-retest to analyze the temporal stability of Muslim identity expression scale.The other estimators should be used besides WLSMV, namely Maximum Likelihood with Robust, to determine the fit model built and provide an overview of the coefficient comparison between items.Future studies can also conduct further analysis by investigating the invariance adequacy of Muslim identity expression measuring instrument.Fulfillment of invariance in measuring instruments should be conducted.The Multigroup CFA method or the Multiple Indicator Multiple Causes (MIMIC) Model (Watanabe et al., 2022) detects Differential Item Functioning, which is the cause of the failure of invariants.Despite the limitations, this study provided preliminary evidence regarding the psychometric strength of Muslim identity expression scale.In addition, it contributed to the understanding of religious identity from a social psychology perspective.

Conclusion
In conclusion, this study was reported to show four types of Muslim identity expression, namely individual, interpersonal, collectivist, and political participation expression.The first phase conducted the scale development by asking about the actions of participants to express Muslim identity.The second phase was related to performance, where the scale development started with 54 items from the openended question in the first.This was reduced to 34 items after exploratory factor analysis was performed and CFA ensured the theoretical construct.From RMSEA value, the 4-correlated factor model was a fit.This study assumed that Muslim identity expression had a stronger relationship with the religious dimension from the political ideology scale (r=.782, p <.001).Despite its weaknesses, initial evidence of the psychometric strength of Muslim identity expression scale was provided.This was different from the scale of religious behavior, limiting the practice of worship.Islamic identity could also be expressed in social and political life.This was specifically important in Muslim societies, where religious expression was part of daily life.
The scale should be used to map the religious behavior of Muslim and explain the community using a social psychology approach.Further studies should be carried out to analyze the predictors of Muslim identity expression or the effect on the psychological well-being of its adherents.

Table 4 .
Correlation between Factors

Table 5 .
Blueprint of Muslim Identity Expression Scale based on CFA