Ethnic differences and predictors of racial and religious discriminations among Malaysian Malays and Chinese

Abstract

Studies on racial and religious discriminations in Malaysia tend to be avoided. This is due to their sensitive nature, possibly becoming political ammunition, and individuals being accused of seditious intent. Much that is necessary to discuss discrimination in Malaysia remains unclear. It is not known to what extent contact between groups is undesirable especially as neighbors in Malaysia. This study examined ethnic differences and predictors of racial and religious discriminations among 1200 Malaysians (319 Chinese and 881 Malays). Discrimination was conceptualized through having the attitude of not wanting people of a different race or a different religion as neighbors. Ethnic differences in discriminating against others were found to be significant. Malays showed higher means of not wanting neighbors of a different race or religion compared to Chinese. Demographic factors and beliefs reflecting conservatism were not good predictors, explaining only slight variances. However, a person who discriminates based on religion was consistently a predictor of exhibiting racial discrimination, and vice-versa, with approximately double the odds among Malays. The finding highlights a strong interplay between race and religion, suggesting that Chinese and Malays conflate the two together.

Keywords:

• intergroup contact theory
• social identity theory
• discrimination
• prejudice
• race
• religion

PUBLIC INTEREST STATEMENT

The complex relationship between race and religion has always been a sensitive issue in Malaysia. This is due to the construct of ethnic identity entailing what makes a person a Malaysian—be it a Malaysian Malay or a Malaysian Chinese—in this pluralistic society. However, due to the reluctance on discussing this open secret of discrimination among Malays and Chinese, not much is known on the prevalence, the predictors, and exactly how race and religion fit into the picture. This research found that the Malays have a higher prevalence rate to discriminate compared to the Chinese. Factors predicting discrimination had also been identified. An interesting interpretation of the finding is that both groups seemingly conflated race and religion together. Despite the attempt of the Malaysian government to promote unity in diversity and claiming a harmonious pluralistic society, this may not be the case. There is more to be done.

1. Introduction

Today, discrimination still exists in all societies. Although things have changed for the better in the west. Countries of western democracies—following the September 11 incident—have been observed to treat religious minorities better than those in Asia and the Middle East (Akbaba, 2009). However, reports of discrimination may not reflect the reality of the situation. Various factors can predict the reporting behavior of those who experienced discrimination, including sociodemographic factors, such as race and marital status (Ro & Choi, 2009), and age and education (Gee & Walsemann, 2009). Hence, discrimination in the real world may be underreported.

Discrimination based on race has been linked to various health issues, such as mental health (Assari et al., 2017; Roberts et al., 2004) and physical health (Mouzon et al., 2017). In one study, while discrimination predicts health, the relationship is not found the other way around due to temporal factors (Gee & Walsemann, 2009). It can also predict academic performance (Huynh & Fuligni, 2010). Due to feeling discriminated against, it may result in the person being less involved with physical activities in public and interaction with others (Sharaievska et al., 2010). Similar outcomes have been noted with religious discrimination (Every & Perry, 2014). Hiring decisions in the workplace can also be predicted by an applicant’s race (Lancee, 2019) and religion (Wallace et al., 2014). Furthermore, the religiosity of a person is linked to behavior (Aminnuddin, 2019a, 2019c, 2020a). The interplay of race and religion creates a symbiotic relationship especially if the two are being framed together and legitimized by the state; this may develop a higher level of discrimination toward the minorities in society (Fox, 2000a). However, non-affiliation can also be a cause for being discriminated against. The non-religious can experience discrimination due to identifying as an atheist or an agnostic (Cragun et al., 2012).

At this point, no studies have been done in Malaysia on investigating racial and religious discriminatory attitudes as proxies of actual discrimination. Related studies to this are more on multicultural society and politics, through qualitative approach and narrative nature (Harris & Han, 2020; Neo, 2012; Noor, 2009; Noor & Leong, 2013). Researchers have also investigated hiring decisions (Lee & Abdul Khalid, 2016), discrimination concerning gender and higher education (Aminnuddin, 2020b), and attitudes toward diversity (Brown et al., 2018) using Malaysians as the sample population. There are also studies on discrimination-prone groups such as people with HIV (Tee et al., 2019) and immigrant workers (Noor & Shaker, 2017). But there are no quantitative studies directly assessing racial and religious discriminations in Malaysia.

This is intriguing, considering racial and religious discriminations being prevalently reported with many high-profile incidents (Pusat KOMAS Malaysia, 2019). It needs to be understood that race and religion are deeply rooted in Malaysian society. The two are among the factors forming the ethnic identity of a group having shared values and culture (Liebkind, 2006). This is apparent in Malaysia where race and religion tend to be conflated together as one’s ethnic identity (Ma, 2005). For example, the Malay race is linked to Islam, while the Chinese race is linked to Buddhism, creating a specific ethnic identity. Due to this, Chinese Muslims have a complicated position in Malaysian society. Discrimination toward ethnic identity based on the interplay of race and religion—as legitimized by the state or political actors—can further increase the degree of discrimination even more (Fox, 2000a). This is the case in Malaysia where, for example, being a Malay is equated to being a Muslim (Stewart, 2012), but when a Chinese Buddhist converted to Islam, their status did not really change much. However, for the latter, it is getting better, unlike in the past (Ma, 2005). Historical and political context also support this notion concerning the mobilization of race and religion (J. C. H. Lee, 2018; Noor, 2009). With only having limited studies, much that is necessary to discuss discrimination remains unclear, especially empirically in the country. This needs to be addressed to fully comprehend the current state of racial and religious discriminations in Malaysia, resulting in not only a contribution to the research area but also in further corroborating the findings of previous studies in the Asia region.

1.1. Contextual background of racial and religious discriminations in Malaysia

In the Malaysia Racial Discrimination Report 2018, racial politics is identified as a major concern in the country (Pusat KOMAS Malaysia, 2019). Throughout the report, incidences of racial discrimination have been argued as being used as a political tool to garner support among the public. Although the focus of the report is on racial discrimination, religion is also a recurring theme. In Malaysia, it is difficult not to discuss race without touching on religion. This is because of how Malays are being equated to being Muslims (Stewart, 2012). The situation is even more worse for Chinese Muslim migrants, who are in limbo between identifying themselves with either non-Muslim Chinese community or non-Chinese Muslim community (Ma, 2005).

The population in Malaysia is dominantly Malays (including the indigenous people) (69.1 percent), followed by Chinese (23 percent), with a small percentage of Indians (6.9 percent) (Department of Statistics Malaysia, 2019). Although the Chinese population is a minority, the number is significant enough to sway public opinions and for political purposes. Due to the pluralistic nature of the society—some argue polarized—caused by historical reasons such as colonization and migration (Abd Rahman et al., 2019), discriminations based on race and religion are controversial topics. With many incidents framed as being offensive, the Malays are usually labeled as being the victims, while the minorities are the perpetrators. It is also common for the Malays to claim that their rights, which are being constitutionally protected, had been infringed. However, in many cases, it is politicization of race and religion with political motives (J. C. H. Lee, 2018).

Discrimination in Malaysia, in some cases, is not a very apparent issue, considering it is too deeply ingrained in the society and the government. Ethnocentric positive discriminatory policies also exist in Malaysia. For example, the New Economic Policy (NEP) has been practiced since 1971, aiming to address income inequality and economic disparity. It was a reaction to racial riots in 1969 involving Chinese and Malays. Hence, the NEP came into existence to address the economic disadvantage of the Malays (J. C. H. Lee, 2018). However, since its implementation, it has strayed from the original objective of addressing economic inequality. Tension between ethnic groups can even be seen when the government formalized the Malay language as the national language (Wan Husin, 2013). All things considered, the Malaysian society has become racialized: socially, politically, and economically (Fee & Appudurai, 2011).

With mistrusts between the two ethnicities, political leaders have used ethnic identity as a political theme in their agenda for more than half a century, even more so by and among Malay politicians (Noor, 2009). It was a way for them to hold on to power. During the Malaysian general election in 2018, a better inter-ethnic relationship is seen, as evident to and resulting in the change of government. However, it was also due to this that the ruling party at that time tried to use religious politics, fearing the masses who were warming up to candidates of a different ethnicity than their own (J. C. H. Lee, 2018).

Due to the interplay between race and religion in Malaysian society has been ongoing for decades, discrimination based on race and religion are normal occurrences. However, unless they are high profile incidents (see Pusat KOMAS Malaysia, 2019), it is difficult to determine the actual statistics of day to day discriminations experienced by Malaysians. In the public sector, it is not unheard of that government jobs and contracts are given out through allocation based on ethnicity being Malay (Stewart, 2012). However, it is difficult to determine if religion is or is not a factor because of how the two overlaps. Similarly, the influence of ethnicity is also a factor in hiring decisions in the private sector (H. A. Lee & Abdul Khalid, 2016), although it is not Malay applicants that receive favorable treatment. It is Chinese applicants that tend to receive more call-backs from employers. Another issue concerning discrimination in Malaysia is the conflation of race with other concepts such as religion and animals. The Malay race is equated with the religion Islam (Stewart, 2012). While pigs are attributed to the Chinese race as a form of beastly racialization (Neo, 2012).

The situation where not much is known quantitatively concerning discrimination is not exclusive to only Malaysia but also to Singapore, a neighboring country. The issue of discrimination such as race is also seen as a white elephant in the room in Singapore (Velayutham, 2017). In one recent review paper, only four quantitative studies in Singapore on racial discrimination had been found (Chew, 2018). There were other studies done but were usually more on using the narrative approach (Chew, 2018). Since then, there is research that addressed this (Aminnuddin, 2019b), that focused not only on racial discrimination but also on other discrimination including based on religion and sexuality. The population in Singapore is also similar to Malaysia, although in the former, Chinese (74.3 percent) is the majority, followed by Malays (13.4 percent) and Indians (9 percent) (Singapore Department of Statistics, 2018). Indians are the minority in both countries. However, the research did not look at ethnic differences between the Chinese and Malays in Singapore. It is not known if ethnic differences in discrimination are significant between the two ethnic groups. The situation concerning discrimination being less studied or under-research in Singapore is more or less similar in Malaysia, with most research done through a qualitative lens.

Many avoid discussing discrimination because it can be manipulated by certain parties and deemed as seditious in Malaysia. Because of this reluctance, researchers tread cautiously around the topic of discrimination in Malaysia while attempting to be subtle or with prudence, trying not to sound seditious. In the words of Khoo Kay Kim, a prominent scholar on Malaysia’s social and political issues, “Very often, when we have a problem, it becomes a racial issue. And then you cannot proceed from there, you cannot further discuss it, because it [is] already sensitive” (J. C. H. Lee, 2018).

1.2. Theoretical background and the present study

Research on discrimination tends to be discussed within the framework of intergroup contact theory and social identity theory. The origin of the former theory can be traced to Allport’s contact hypothesis (Allport, 1954). Its central argument is on the lack of contact between two groups, resulting in prejudicial attitudes and negative sentiments. For social identity theory, a group’s identity can foster pride and provide legitimacy of superiority over another group (Tajfel et al., 1971). All these—lack of contact, having pride, and perceived legitimacy of superiority over another group—may result in discrimination. Recent studies have shown that facilitating exposure and contact of positive nature between groups can nullify prejudice; for example, an experimental study had been done among Indonesian Muslims and Christians that showed reduced negative sentiment (Kanas et al., 2015). Hence, if contact can be promoted between two different ethnic groups, it may facilitate a more positive attitude and understanding, which may then help to negate the pride and legitimacy a person holds over another.

Therefore, although this present study is based on intergroup contact theory and social identity theory, the focal point does not concern to what extent contact exists between groups, nor to what extent they have pride and hold the notion of superiority over another. Instead, this study examines whether or not these groups would want to have intergroup contact, and how these groups vary in exhibiting discrimination. Existing relevant studies include discrimination in the housing sector (Auspurg et al., 2019; Carlsson & Eriksson, 2015; Sawert, 2019). However, these studies focused more on determining the extent of racial discrimination, and the other groups involved tend to be immigrants, foreign students, and expatriates. Hence, it is not known to what extent the Malays would want the Chinese as their neighbors and vice versa. If this can be determined, then there will be empirical evidence on predictors of discrimination, and if ethnic differences between the two exist when it concerns discrimination.

The present study examines ethnic differences and predictors of exhibiting racial and religious discriminations by Malays and Chinese in Malaysia. Few studies have examined discrimination in Malaysia, especially of quantitative nature. Previous research relevant to race in Malaysia, and to a certain extent, religion, is either on hiring decision (H. A. Lee & Abdul Khalid, 2016), choice of school types for children (Ismail, 2018), attitudes toward diversity (Brown et al., 2018), immigrant workers (Noor & Shaker, 2017), or asylum seekers (Cowling & Anderson, 2019). Other research includes national unity between ethnic groups (Azlan et al., 2018; Azlan, 2019) and the colonial construction of race (Reddy & Gleibs, 2019). Limited papers had been found that attempted to measure discrimination directly and to further analyze it in-depth; most of them have been done through a narrative approach, citing high-profile incidents (Pusat KOMAS Malaysia, 2019).

Therefore, it remains unclear concerning racial and religious discriminations being exhibited by Malays and Chinese, especially if questions on discrimination toward others are explicitly asked. Ethnic differences and the predictors are also not known. For issues such as discrimination, it is imperative for researchers to first assess to what degree a person discriminates another, followed by ethnic differences, especially in a pluralistic society such as in Malaysia, and then to determine the variables that can be predictors. Only by having the necessary information, which can provide insights into the situation at the most basic level yet significant, that future scholars can have a basis to further explore it.

In this present study, discrimination is operationalized through the attitude of not wanting a neighbor of a different race or a different religion. The assessment of attitude as a proxy for discrimination is both reliable and valid due to its predictive value for the latter (Carlsson & Eriksson, 2017). A higher prevalence of discrimination exhibited by Malays is expected due to the ethnicity being the slightly more dominant group, with more power and voice in the society compared to Chinese (J. C. H. Lee, 2018). It is also anticipated that other factors might be able to predict discrimination, as previous research suggests, such as age (Franssen et al., 2013), education (Wang & Froese, 2020), religion (Brown et al., 2018; Cowling & Anderson, 2019), and other discriminatory attitudes (Aminnuddin, 2019b; Haney, 2016).

Therefore, several hypotheses are proposed: the prevalence of exhibiting discrimination will be higher among Malays compared to Chinese (hypothesis 1); demographic factors—ethnicity (Malay), gender (male), age (older), education (less educated), employment (unemployed), and marital status (married)—will positively predict discrimination (hypothesis 2); conservatism—specifically importance of religion and tradition to oneself—will positively predict discrimination (hypothesis 3); and religious discrimination will positively predict racial discrimination and vice versa (hypothesis 4). This study is the first of its kind to quantitatively examine racial and religious discriminations, in the context of ethnic differences and predictors, in the Asia region.

2. Materials and methods

2.1. Data and sample population

This study used the data for Malaysia from the World Values Survey Wave 6 for this research (Inglehart et al., 2014). The sample population in Malaysia was from three ethnic groups: Chinese, Malay, and South Asian (operationalized in the data to include Indian, Hindu, Pakistani, and Bangladeshi). The number of respondents within the South Asian group was too few for analysis to be used in this research; hence, it was excluded. However, this was also due to the proportion of ethnic groups in the country. Probability proportional to size was used. The sampled population was stratified first based on states, and then further stratified by rural and urban population, race, gender, and age to reflect the entire adult population in Malaysia. Questionnaires were translated and back-translated into the English language, and then a pilot test was conducted. Data collection was done between March and June 2012.

Data was collected from 1200 respondents. The Chinese population consisted of 319 respondents (49.8 percent females and 50.2 percent males), ranging in age from 18 to 76 years old (M_age = 40 years, SD = 13.877). For the Malays, the population consisted of 881 respondents (48.2 percent females and 51.8 percent males), ranging in age from 18 to 80 years old (M_age = 40.04 years, SD = 13.866). The combined percentages for both ethnicities in terms of gender were 48.7 percent females (M_age = 39.34 years, SD = 13.208) and 51.3 percent males (M_age = 40.67 years, SD = 14.438). Age ranged between 18 and 90 years old, with a combined mean age of 40.03 years (SD = 13.863).

2.2. Variables and procedure

Demographic variables consisted of ethnicity, gender, age, education, employment, and marital status. Except for age and education, all the demographic variables were treated as ordinal variables in binary form: ethnicity (Chinese = 0, Malay = 1), gender (female = 0, male = 1), employment (employed = 0, unemployed = 1), and marital status (not married = 0, married = 1). Employment and marital status were coded into the stated binary form. No changes were made on age. For education, it was originally coded from 1 (no formal education) to 9 (university-level education, with degree). However, it was reverse coded for this study for a higher number to reflect having less education. It is expected that people having less education would be more prejudice.

Beliefs to reflect conservatism were assessed using items asking on the importance of religion and tradition. For assessing the former, respondents were asked how important religion is in their life. This was rated with 1 (very important) to 4 (not at all important). However, in assessing the importance of tradition, respondents were provided with a statement concerning a person’s values. The respondents were first told that the interviewer would present a description of values that are important to an unnamed fictional individual. They were presented with the statement “Tradition is important to this person” and asked to what extent the person is similar to him or her. Ratings of 1 (very much like me) to 6 (not at all like me) were provided. Other values assessed in the questionnaire include being successful, being rich, and being creative; but these were not included in this present study.

The difference in the way of assessing the importance of tradition, relative to assessing the importance of religion, is due to the original questionnaire focusing on evaluating the importance of tradition through the angle of how similar the respondent is with a fictional individual’s values. This method of assessment was not done in evaluating the importance of religion. Nonetheless, both still reflects whether or not religion or tradition is important to the person. However, similar to education, these two were reverse coded for this study. It is expected that people who are more conservative—in terms of religion and tradition—would be more prejudiced of others outside their own groups. These two variables were coded to reflect a higher number indicating the person holds more importance on religion and tradition, and a lower number indicating having less importance.

Finally, indicators on existing discriminatory attitudes or behaviors were assessed using items that asked if a specific group is not desirable as neighbors. These items, although of attitude nature, are reliable proxies of discrimination (Carlsson & Eriksson, 2017) and have been employed in a similar manner in previous research (Aminnuddin, 2019b; Haney, 2016; Manalastas et al., 2017). In the survey, respondents were provided with a list of groups. They were asked to indicate which groups they would not want as neighbors. For this present study, the focus is on people of a different race and people of a different religion. Not wanting to be neighbors is operationalized to reflect discrimination. Those who did not select people of a different race are coded as 0 (does not discriminate), while if they stated they do not want this group as a neighbor, they were coded as 1 (discriminate). Similar coding was done for responses concerning people of a different religion.

Further details on the questionnaire can be accessed online (Inglehart et al., 2014). The coding and treatment of variables were similar to previous studies on discrimination in the United States and the Netherland (Haney, 2016) and Singapore (Aminnuddin, 2019b).

2.3. Statistical analysis

This research examined associations between variables, followed by determining ethnic differences in exhibiting racial and religious discriminations. Independent samples t test was conducted to assess whether or not Malays showed more racial and religious discriminations toward others compared to Chinese. Then predictors of exhibiting discrimination through hierarchical logistic regression analysis were tested using SPSS 25. Hierarchical logistic regressions were conducted based on three groups: Chinese, Malays, and combined.

Variables were entered in blocks for a total of four blocks for the hierarchical regression. Using the combined group as an example, the first block consisted of gender and age. In the second block, education, employment, and marital were entered. In the third block, two variables were added: the importance of religion and tradition. Finally, in the fourth block, discrimination against a group was added; in this block, if the outcome variable was racial discrimination, then the variable added here would be religious discrimination and vice versa. For analyzing the other groups, Chinese and Malays, the variable ethnicity was not added in the first block. Entering blocks by blocks of variables would show how predictors in each block explain the variances.

3. Results

3.1. Descriptive statistics

The means, standard deviations, counts, and percentages of sample population characteristics are presented in Table 1. Among 1200 Chinese and Malays, a majority of them were married and employed. Over 40.7 percent completed technical and vocational school, 12.3 percent completed university-preparatory education, and 7.9 percent were at the university level or already graduated. On religion, 96.7 percent claimed it is important for them. Similarly, most of them leaned on the side of perceiving themselves viewing tradition as important. Concerning discrimination, 30.5 percent of respondents did not want people of a different race as neighbors, while 30 percent did not want their neighbors to be people of a different religion.

Table 1. Descriptive Statistics of Study Variables

	Chinese (N = 319)		Malay (N = 881)		Combined (N = 1200)
Variable	M	SD	M	SD	M	SD
Age	40.0	13.877	40.04	13.866	40.03	13.863
Education (1 to 9)	4.68	2.019	5.07	1.834	4.97	1.892
Religion is important (1 to 4)	3.48	0.717	3.94	0.25	3.82	0.474
Tradition is important (1 to 6)	4.71	1.134	4.85	1.162	4.81	1.156
	Count (%)
Male	160 (50.2%)		456 (51.8%)		616 (51.3%)
Unemployed	58 (18.2%)		233 (26.4%)		291 (24.3%)
Married	202 (63.3%)		628 (71.3%)		830 (69.2%)
Racial discrimination	55 (17.2%)		311 (35.3%)		366 (30.5%)
Religious discrimination	59 (18.5%)		301 (34.2%)		360 (30%)

3.2. Ethnic differences on racial and religious discriminations

Independent samples t tests were then conducted to compare the means of racial and religious discriminations between Chinese and Malays. Results indicated that Malays (M = 0.35, SD = 0.478) showed more racial discrimination than Chinese (M = 0.17, SD = 0.378), t(706.813) = −6.786, p <.001, two-tailed, Cohen’s d = 0.418. Similarly, Malays (M = 0.34, SD = 0.475) showed more religious discrimination than Chinese (M = 0.18, SD = 0.389), t(681.753) = −5.801, p < .001, two-tailed, Cohen’s d = 0.369. Both results support hypothesis 1 on racial and religious discriminations.

3.3. Hierarchical logistic regression

Results of hierarchical logistic regressions with predictors to predict racial discrimination are presented for the following groups: Chinese (see Table 2), Malays (see Table 3), and combined (see Table 4). The results for all three groups showed that demographic factors and beliefs concerning conservatism are not reliable predictors of racial discrimination. Hence, hypothesis 2 and hypothesis 3 are not supported. However, exhibiting religious discrimination is a positive predictor of racial discrimination, supporting hypothesis 4.

Table 2. Racial Discrimination—Chinese—Model 1

	Model 1.1	Model 1.2	Model 1.3	Model 1.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Male	0.52* [0.28–0.95]	0.58 [0.31–1.07]	0.58 [0.31–1.09]	0.50 [0.22–1.12]
Age	0.98 [0.96–1.00]	0.96** [0.927–0.984]	0.95** [0.92–0.98]	0.96 [0.93–1.00]
Less educated		1.18 [0.99–1.41]	1.18 [0.99–1.41]	1.08 [0.87–1.35]
Unemployed		2.01 [0.98–4.15]	1.97 [0.94–4.10]	1.46 [0.57–3.75]
Married		2.10 [0.98–4.51]	2.20* [1.01–4.79]	1.55 [0.59–4.07]
RI			1.64* [1.01–2.67]	1.42 [0.77–2.63]
TI			0.91 [0.69–1.19]	1.12 [0.78–1.60]
ReligiousD				33.49*** [14.86–75.45]
Nagelkerke R^2	.040	.095	.117	.501
ΔR^2		.055*	.022	.384***
Model p	.338	.118	.336	.748

RI = Religion is important. TI = Tradition is important. ReligiousD = Religious discrimination.

* p < .05. ** p < .01. *** p < .001.

Table 3. Racial Discrimination—Malays—Model 2

	Model 2.1	Model 2.2	Model 2.3	Model 2.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Male	1.19 [0.90–1.1.56]	1.25 [0.94–1.66]	1.23 [0.92–1.64]	1.50 [0.96–2.34]
Age	0.99 [0.98–1.00]	0.98** [0.97–1.00]	0.98* [0.97–1.00]	0.99 [0.97–1.01]
Less educated		1.10* [1.01–1.20]	1.09* [1.01–1.19]	1.00 [0.87–1.13]
Unemployed		1.21 [0.87–1.67]	1.22 [0.87–0 = 1.69]	1.27 [0.77–2.10]
Married		1.20 [0.83–1.73]	1.20 [0.83–1.74]	2.13* [1.18–3.83]
RI			1.69 [0.88–3.26]	1.13 [0.44–2.93]
TI			0.85** [0.75–0.95]	0.96 [0.80–1.15]
ReligiousD				65.97*** [41.79–104.13]
Nagelkerke R^2	.006	.017	.032	.641
ΔR^2		.011	.015**	.609***
Model p	.376	.414	.190	.250

RI = Religion is important. TI = Tradition is important. ReligiousD = Religious discrimination.

* p < .05. ** p < .01. *** p < .001.

Table 4. Racial Discrimination—Combined—Model 3

	Model 3.1	Model 3.2	Model 3.3	Model 3.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Malay	2.63*** [1.90–3.63]	2.43*** [1.76–3.37]	2.02*** [1.41–2.87]	1.80* [1.09–2.99]
Male	1.02 [0.79–1.31]	1.09 [0.84–1.41]	1.08 [0.83–1.39]	1.15 [0.79–1.67]
Age	0.99* [0.98–1.00]	0.98*** [0.97–0.99]	0.98*** [0.97–0.99]	0.99 [0.97–1.00]
Less educated		1.11** [1.03–1.20]	1.11** [1.03–1.20]	1.01 [0.90–1.13]
Unemployed		1.31 [0.97–1.76]	1.31 [0.97–1.77]	1.21 [0.78–1.87]
Married		1.34 [0.97–1.86]	1.35 [0.97–1.88]	1.82* [1.12–2.96]
RI			1.66** [1.14–2.43]	1.39 [0.82–2.36]
TI			0.86** [0.77–0.96]	0.99 [0.85–1.17]
ReligiousD				53.56*** [36.48–78.63]
Nagelkerke R^2	.050	.066	.081	.620
ΔR^2		.016**	.015**	.539***
Model p	.119	.067	.109	.910

RI = Religion is important. TI = Tradition is important. ReligiousD = Religious discrimination.

* p < .05. ** p < .01. *** p < .001.

Model 1.1 predicted 82.8 percent cases correctly, and the percentage increased to 89 percent in Model 1.4. Predictors in the first three blocks only explained the variances slightly (4 percent to 11.7 percent). In the fourth block, the model explained 50.1 percent of the variance with the inclusion of religious discrimination as a predictor.

Model 2.1 predicted 64.7 percent cases correctly, and the percentage increased to 89.1 percent in Model 2.4. Predictors in the first three blocks only explained the variances slightly (0.6 percent to 3.2 percent). In the fourth block, the model explained 64.1 percent of the variance with the inclusion of religious discrimination as a predictor.

Model 3.1 predicted 69.5 percent cases correctly, and the percentage increased to 89.2 percent in Model 3.4. Predictors in the first three blocks only explained the variances slightly (5 percent to 8.1 percent). In the fourth block, the model explained 62 percent of the variance with the inclusion of religious discrimination as a predictor.

Results of hierarchical logistic regressions with predictors to predict religious discrimination are presented for the following groups: Chinese (see Table 5), Malays (see Table 6), and combined (see Table 7). As with previous results concerning predictors of racial discrimination, demographic factors and conservatism are not reliable predictors of religious discrimination. Therefore, hypothesis 2 and hypothesis 3 are not supported. However, again, hypothesis 4 is supported whereby exhibiting racial discrimination is a positive predictor of religious discrimination.

Table 5. Religious Discrimination—Chinese—Model 4

	Model 4.1	Model 4.2	Model 4.3	Model 4.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Male	0.75 [0.42–1.33]	0.86 [0.47–1.55]	0.84 [0.46–1.52]	1.18 [0.54–2.59]
Age	0.99 [0.97–1.01]	0.96** [0.94–0.99]	0.96** [0.93–0.99]	0.98 [0.94–1.01]
Less educated		1.22* [1.02–1.44]	1.21* [1.02–1.44]	1.17 [0.94–1.47]
Unemployed		2.27* [1.13–4.55]	2.31* [1.14–4.71]	1.94 [0.75–5.01]
Married		2.17* [1.03–4.57]	2.24* [1.05–4.79]	1.81 [0.68–4.78]
RI			1.52 [0.96–2.40]	1.24 [0.69–2.21]
TI			0.79 [0.61–1.02]	0.74 [0.53–1.02]
RacialD				32.87*** [14.67–73.65]
Nagelkerke R^2	.009	.082	.108	.485
ΔR^2		.073**	.026	.377***
Model p	.051	.519	.295	.396

RI = Religion is important. TI = Tradition is important. RacialD = Racial discrimination.

* p < .05. ** p < .01. *** p < .001.

Table 6. Religious Discrimination—Malay—Model 5

	Model 5.1	Model 5.2	Model 5.3	Model 5.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Male	1.04 [0.79–1.38]	1.08 [0.81–1.44]	1.05 [0.79–1.41]	0.76 [0.49–1.20]
Age	0.99* [0.98–1.00]	0.98** [0.97–0.99]	0.98* [0.97–1.00]	0.99 [0.97–1.01]
Less educated		1.14** [1.04–1.24]	1.13** [1.04–1.24]	1.14 [1.00–1.31]
Unemployed		1.14 [0.82–1.58]	1.14 [0.82–1.59]	0.95 [0.57–1.59]
Married		0.85 [0.59–1.22]	0.85 [0.59–1.23]	0.48* [0.27–0.85]
RI			1.87 [0.95–3.67]	1.88 [0.65–5.48]
TI			0.82** [0.72–0.92]	0.84 [0.70–1.01]
RacialD				66.13*** [41.87–104.46]
Nagelkerke R^2	.010	.026	.047	.653
ΔR^2		.016**	.021**	.606***
Model p	.245	.668	.582	.876

RI = Religion is important. TI = Tradition is important. RacialD = Racial discrimination.

* p < .05. ** p < .01. *** p < .001.

Table 7. Religious Discrimination—Combined—Model 6

	Model 6.1	Model 6.2	Model 6.3	Model 6.4
Predictor	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]	OR [CI = 95%]
Malay	2.30*** [1.68–3.15]	2.15*** [1.56–2.96]	1.81*** [1.28–2.57]	1.20 [0.73–1.98]
Male	0.98 [0.76–1.25]	1.03 [0.80–1.34]	1.01 [0.78–1.32]	0.90 [0.61–1.32]
Age	0.99** [0.98–1.00]	0.98*** [0.97–0.99]	0.98*** [0.97–0.99]	0.99 [0.97–1.01]
Less educated		1.15*** [1.07–1.24]	1.15*** [1.06–1.24]	1.15* [1.02–1.28]
Unemployed		1.29 [0.96–1.74]	1.30 [0.96–1.76]	1.14 [0.73–1.78]
Married		1.04 [0.75–1.43]	1.05 [0.76–1.46]	0.69 [0.43–1.12]
RI			1.61* [1.11–2.32]	1.31 [0.79–2.17]
TI			0.81*** [0.73–0.91]	0.81** [0.69–0.95]
RacialD				53.56*** [36.48–78.65]
Nagelkerke R^2	.042	.061	.083	.622
ΔR^2		.019***	.022***	.539***
Model p	.141	.485	.956	.762

RI = Religion is important. TI = Tradition is important. RacialD = Racial discrimination.

* p < .05. ** p < .01. *** p < .001.

Model 4.1 predicted 81.5 percent cases correctly, and the percentage increased to 89 percent in Model 4.4. Predictors in the first three blocks only explained the variances slightly (0.9 percent to 10.8 percent). In the fourth block, the model explained 48.5 percent of the variance with the inclusion of racial discrimination as a predictor.

Model 5.1 predicted 65.8 percent cases correctly, and the percentage increased to 89.1 percent in Model 5.4. Predictors in the first three blocks only explained the variances slightly (1 percent to 4.7 percent). In the fourth block, the model explained 65.3 percent of the variance with the inclusion of racial discrimination as a predictor.

Model 6.1 predicted 70 percent cases correctly, and the percentage increased to 89.2 percent in Model 6.4. Predictors in the first three blocks only explained the variances slightly (4.2 percent to 8.3 percent). In the fourth block, the model explained 62.2 percent of the variance with the inclusion of racial discrimination as a predictor.

4. Discussion

The discussion for the results of this present study is challenging due to having not many similar quantitative studies in literature concerning people who discriminate in Malaysia based on race and religion. Hence, the primary motivation of this study is to address the issue by contributing to literature with information on the actors in Malaysia who discriminate based on race and religion. This study has contributed to the literature concerning racial and religious discriminations in Malaysia, with emphasis on Malays and Chinese. First, this study determined ethnic differences in exhibiting discrimination toward others between Malays and Chinese (hypothesis 1). Second, this study identified predictors of exhibiting racial and religious discriminations in each group, and when the two groups were combined (hypothesis 2 to hypothesis 4). Exhibiting discrimination is operationalized through the attitude of not wanting people of a different race and people of a different religion as neighbors. Finally, the finding suggests there is discrimination toward ethnic identity, especially based on the conflation of race and religion.

For the first time, it is shown that exhibiting discrimination is much higher among Malays by not wanting people of other races and religions to be their neighbors. It was lower among Chinese. Hard data is provided as evidence of the findings, compared to previous research that discussed the existence of discrimination in Malaysian society through historical and narrative discourse. Despite the Malaysian government promoting harmony and unity in a multicultural society, discrimination still exists. The numbers are not considered small, with almost one-fifth to over a third showed tendencies to discriminate. The prevalence is substantially high considering the government attempts to promote pluralism through unity in diversity. The differences in discrimination between Chinese and Malays are statistically significant with a medium effect size. For racial discrimination, Malays showed over double in terms of prevalence compared to Chinese. For religious discrimination, it was almost double the prevalence. Therefore, the first hypothesis is supported. The prevalence of exhibiting discrimination is higher among Malays compared to Chinese. This contribution to literature is important, considering researchers have no reference point initially for conducting research on statistical information of ethnic differences on racial and religious discriminations. This present study provides a starting reference point for future research to expand further.

Although previous studies have determined demographic factors being predictors of discrimination (Cowling & Anderson, 2019; Franssen et al., 2013; Wang & Froese, 2020), in contrast, this present study has only found them as being inconsistent and need to be further examined. Some demographic variables were initially significant predictors in the early blocks of the models, such as ethnicity but then became nonsignificant after the inclusion of discrimination toward a specific group was added as predictors. This is with the exception of predicting racial discrimination in a combined sample population. Similar findings were observed with beliefs concerning conservatism, specifically viewing religion and tradition as important. Hence, the second hypothesis and the third hypothesis are not supported. Demographics and beliefs concerning conservatism, particularly religion and tradition, are not predictors of racial and religious discriminations.

Generally, rather than demographic factors and beliefs concerning conservatism, it is discriminatory attitude itself that can be a predictor. For example, exhibiting discrimination toward people of a different religion can be a good predictor of racial discrimination. Similar results are observed in predicting religious discrimination using racial discrimination as a predictor. This is consistent with previous studies that have observed discriminatory attitudes toward various groups as reliable predictors for discrimination (Aminnuddin, 2019b; Haney, 2016). Therefore, the fourth hypothesis is supported. Discrimination itself can predict discrimination.

A key finding in this research is the conflation of race and religion. In interpreting the results of racial discrimination predicting religious discrimination, and religious discrimination predicting racial discrimination, it is argued there is a strong association between, even a conflation of, race and religion in Malaysia. This is further strengthened by the exponential increase in variances explained compared to other variables. In general, society in Malaysia views there is an overlap between the race and the religion (Noor, 2007, 2009; Ma, 2005), This act of attribution is not foreign. Being a Malay and being a Muslim have also been conflated together for political mobilization purposes (Stewart, 2012). It is not uncommon to attribute pigs—deemed to be impure by Islam—toward a certain ethnic group such as Chinese (Neo, 2012). Even language attributed to a specific ethnic group can be a factor (Wan Husin, 2013). All these are examples of the conflation of race and religion in Malaysia. This present study contributes novel quantitative evidence of the association between race and religion. However, it is important to emphasize that this is an interpretation of the results. It warrants further empirical investigation to directly examine the conflation.

Overall, the findings that racial and religious discriminations are considerably high can be explained through the intergroup contact theory (Allport, 1954) and the social identity theory (Tajfel et al., 1971). As seen in the context of Malaysia, there is a high prevalence rate of Chinese and Malays not wanting the people of a different race and religion as neighbors. This can be argued to be an outcome of various racial-based policies such as specific school types based on race and quota for higher education. Many of these policies are protectionist by nature to support the Malays, and they affect the Chinese negatively (Kee, 2010; H. A. Lee, 2016). As explained by the social identity theory, it is due to the Malays being the majority group in Malaysia, creating pride, self-esteem, and superiority attitude toward the Chinese minority. This creates negative sentiments linked to homogeneous contact preferences, which worsen understanding and empathy between groups, as viewed through intergroup contact theory (Allport, 1954).

Due to the rigid construct of ethnic identity in Malaysia (Gabriel, 2015; Ma, 2005), especially one that is based on race and religion (Fox, 2000a, 2000b), it is not surprising to observe the outcome of negative sentiments and prejudicial attitudes, further expediting less intergroup contact and intercultural experiences, and resulting in a high prevalence of discrimination among the Malays and the Chinese. This present study provides a prime example of how the two theories—intergroup contact theory and social identity theory—illustrate discriminations where there are a majority group and a minority group in society.

As highlighted earlier, demographic factors and beliefs concerning conservatism are not consistent predictors. This suggests that for policymakers, these may not be as important as compared to other factors. It is discrimination that predicts further discrimination. This implies that the conflation of race and religion is a key issue that needs to be addressed. This may be due to differences in norms and culture based on race and religion being feared. If a person is not open to differences, it may even facilitate intolerance and then discrimination. This is evident with many such incidents of discrimination, resulting in unrest and protest (Pusat KOMAS Malaysia, 2019), supporting past research that found religious discrimination does promote dissent (Akbaba & Taydas, 2011).

Therefore, policymakers should emphasize more on tolerance and acceptance without dismissing differences. If differences are less emphasized and only similarities are being promoted to nurture unity, then the initial aim of promoting tolerance and acceptance may backfire. This is also the case if the opposite approach is employed, such as in Malaysian history, where differences were being used for fear-mongering. Assimilation may be a solution. However, it is only if the minority is too small of a number that assimilation would be feasible and effective, for example, minorities in Britain (Platt, 2014). But in Malaysia, the Chinese minority still makes up a significant portion of the entire population (Department of Statistics Malaysia, 2019), which will cause assimilation difficult, with them retaining their own identity. This is unlike the situation in the Chinese-dominated Singapore where the Malay minority is too few (Singapore Department of Statistics, 2018).

Hence, a moderate and well-balanced approach is an excellent way to foster harmony in a multicultural and pluralistic society such as the Malaysian society. Improving intergroup contact and intercultural experiences can help reduce discrimination (Kanas et al., 2015). Nevertheless, it remains a problem in the country to have these experiences, which is further aggravated by certain legitimizations such as the establishment of specific school types based on race and quota for higher education. It is only by experiencing the other group’s cultural experiences that can help to foster tolerance and unity through both similarities and differences. Yet, the current environment is not engineered to facilitate this.

Unity in diversity is the key and has been the propagated slogan in Malaysia. However, the question remains, how effective is it? Is it merely to instill tolerance at a superficial level or should it be more than just that? Where is unity in diversity if Malaysians do not want neighbors of a different race and religion? The findings in this study highlight the ineffectiveness of Malaysia’s current effort to counter discrimination and disunity, at least within the neighbor context, which is supported and explained by the intergroup contact theory and social identity theory. There is still plenty to be done.

This study has several limitations. The first limitation is the time of the data collection. The World Values Survey Wave 6 is the most recent data collected for Malaysia. It can be viewed as being dated. However, arguably, it is difficult to conduct studies on exhibiting discrimination, especially to collect quantitative data, on race and religion in Malaysia. Many discussed discrimination from a historical perspective and with a narrative approach and discourse nature (J. C. H. Lee, 2018; Noor, 2007, 2009), with some examining relevant variables quantitatively (H. A. Lee & Abdul Khalid, 2016; Azlan et al., 2018; Noor & Shaker, 2017) but not directly on exhibiting discrimination, especially the interplay of race and religion. Second, discrimination is operationalized from two items that only assessed not wanting people of a different race and people of a different religion as neighbors, to reflect racial discrimination and religious discrimination, respectively. This assessment is only at the attitude level, rather than actual discrimination. However, it can be reliable and valid as a proxy of actual discrimination (Carlsson & Eriksson, 2017). Third, this may not assess discrimination comprehensively or with various dimensions. A single-item measure can still be reliable and valid (Bergkvist & Rossiter, 2007; Cheung & Lucas, 2014; Littman et al., 2006; Milton et al., 2011; West et al., 2012). However, while acceptable, a measure with multiple items or even dimensions is still more preferable (Fisher et al., 2016). Fourth, Indians were not included in this study due to the small number of respondents. If an analysis was done on using this sample, the results would not be reliable nor valid. Hence, it was excluded. However, being the next minority group after the Chinese in Malaysia, findings based on results concerning Indians will prove to be insightful if enough sample population can be obtained, considering the population is 6.9 percent of the entire population (Department of Statistics Malaysia, 2019), which makes this ethnic group truly a minority even in comparison to the Chinese.

These limitations can be addressed in future research. For the next data collection, Wave 7 of the World Values Survey, researchers can examine people exhibiting racial and religious discriminations using time series analysis to observe changes. Using this present study as a starting reference point, researchers can design their studies to examine the study variables from this present study, add more variables, and further refine the research design. Indians in Malaysia should also be considered in the research population. Variables that can be relevant include racial identity, nationalism, patriotism, and a sense of belonging and unity. Another suggestion is that rather than using not wanting a specific group as neighbors to reflect discrimination, researchers may use existing scales measuring discrimination specifically. This will prove to be more in-depth due to multiple items on discrimination being assessed. A person who discriminates may not exhibit his or her prejudice toward other people, and this need be considered, such as implicit and explicit discrimination, and the various dimensions of discrimination. Examining ingroup and outgroup can also provide more insights.

As a conclusion, this present study has provide, for the first time, evidence on the prevalence of racial and religious discriminations in Malaysia, with ethnic differences examined. Predictors are also identified. It is found that the Malays have more than double the likelihood to discriminate based on race and religion compared to the Chinese. Demographic factors and beliefs reflecting conservatism are not consistent predictors. Reliable predictors are discrimination itself: racial discrimination predicts religious discrimination, and religious discrimination predicts racial discrimination. This suggests a conflation of race and religion among Malays and Chinese in Malaysia.

Conflicts of Interest

The author declares no conflict of interest.

Acknowledgements

The author would like to extend gratitude toward Stephen C. Druce (Academy of Brunei Studies, Universiti Brunei Darussalam) and Harris Shah Bin Abd Hamid (Department of Educational Psychology and Counselling, Faculty of Education, Universiti Malaya) for providing feedback during the writing of the manuscript.

Additional information

Funding

The author received no direct funding for this research.

Notes on contributors

Nur Amali Aminnuddin

Nur Amali Aminnuddin is a researcher at the Academy of Brunei Studies, Universiti Brunei Darussalam. He is a member of the Malaysian Psychological Association and the Asian Psychological Association. Having a wide range of interests, he holds a bachelor’s degree in political science, a master’s degree in industrial and organizational psychology, and a Ph.D. in Brunei Studies. His research interests revolve around psychology of religion, social psychology, and behavioral sciences. Religion, specifically Islam, is a common theme in his research. His current research focuses on Islamic religiosity and personality of young Muslims. He had published in the areas of work ethics in organizations, discrimination against marginalized groups, education, personality, and behavior.