1. Introduction

As early as 1946 Allport and Kramer realized that students with no religious affiliation were not as anti-Negro as those who accepted to be protestant or Catholic as Rosenblith did in 1949. The findings of Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) provided some explanations regarding the distinct behaviors of these two groups. They showed that the believers who attended church scored significantly higher on ethnocentrism and authoritarianism than non-attenders. Similar results were reported by scholars such as Gough (1951).

In addition to being closely related to ethnocentrism and authoritarianism, religion was explored in relation to variables such as attitude. Kirkpatrick (1949), for example, found that non-religious people were slightly more humanitarian than religious ones. In other words, compared to non-religious people, his religious participants more punitive attitudes towards criminals, delinquents, homosexuals, prostitutes, and those in need of psychiatric treatment. Similarly, Rokeach (1960) conducted a research project with students and discovered that nonbelievers were consistently less dogmatic, less authoritarian, and less ethnocentric than believers. These findings were also supported through public opinion assessment conducted by Stember (1961).

Stouffer's (1955) findings revealed other aspects of religiosity in individuals’ personal life. They showed that the Americans who attended churches as their regular members were more intolerant of nonconformists or unorthodox groups such as atheists, communists and socialists than those who did not take part in church services. Focusing on these findings Allport and Ross (1967) wondered whether the marked difference between church attenders and non-attenders was due to their educational level. Other researchers’ findings (e.g., Demerath, 1965; Struening, 1963), however, showed that education had no role in religious prejudice. Allport and Ross, therefore, hypothesized that the difference in church attenders and non-attenders’ tolerance of nonconformists might be explained through their religious orientation.

1.1. Religious Orientation

In 1946, Allport and his colleague Kramer conducted a study where they explored the relationship between attending church and prejudice. Their findings revealed that among the students who participated in the survey, those who claimed no religious affiliation were less prone to racism towards the African-Americans than those who declared themselves to be Protestant or Catholic. The study further revealed that the students who had admitted to having strong religious influence in their homes were higher in prejudice than the students who had very small or no religious influence in their homes at all. These findings prompted Rosenblith (1949) to conduct his own study among the students in South Dakota, USA, and his findings, along with those of Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) and Gough (1951) established a similar trend.

Was the trend found in religious influence applicable to other social aspects such as political views and affiliations? Going beyond ethnic prejudice, Stouffer (1955) demonstrated that among a sample of American church members those who had attended churches within the past month were more intolerant of non-conformists (such as socialists, atheists, or communists) than those who had not attended church. His findings seemed to show that on average religious people are generally more intolerant not only toward ethnic but also toward ideological groups.

Stouffer's (1955) findings encouraged scholars such as Kirkpatrick (1949) to investigate the difference between people attending religious gatherings (in particular the church) and those who did not. In effect, they divided their sample populations into two groups: church attenders and non-attenders. However, results from "the Authoritarian Personality" (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950, p. 212) revealed that within the attending members of the church, there seemed to be deviation such that members who attended church “sometimes” or “often” were more inclined towards punitive attitudes and prejudice than those who were “regular” members.

Struening's (1657, 1963) findings echoed the results obtained with the frequency of attending churches. His participants were, however, all faculty members of a large state university. He established a curvilinear relationship when he found non-attenders had lower prejudice scores than any group except the devotees who managed to attend churches 11 or more times a month. In other words, prejudice scores did not increase as church attendance increased. His findings, for example, showed that the faculty members who attended churches two times per month were the most prejudiced followed by those who attended churches once a month.

Holtzman employed a 26-item desegregation scale in three separate studies, i.e., Holtzman (1956), Kelley, Ferson, and Holtzman (1958), Young, Benson, and Holtzman (1960), and the results from all three studies showed a decrease in intolerance with increase in regular church attendance. More evidence of the curvilinear relationship was found by Friedrichs (1959), Tumin (1958), Pettigrew (1959), and Pinkney (1961). These studies confirmed the trend and showed that it holds true regardless of the religion or target of prejudice studied.

Is the teaching of religion itself responsible for being prejudiced towards others? Allport and Ross (1967) gave a negative answer to this very question, they did not, however, mention the prophet Mohammad whose message of peace and rejection of race as irrelevant to human relationships and dignity is well documented. For example, verse 13 of Chapter 49 in the Holy Quran reads,

O men! Behold, we have created you all out of a male and a female and have made you into nations and tribes, so that you might come to know one another. Verily, the noblest of you in the sight of God is the one who is most deeply conscious of Him. Behold, God is all-knowing, all-aware” (Asad, 1980, p. 1015)

In spite of not referring to Mohammad whose teachings are far more recent than Jesus Christ whom Allport and Ross (1967) mention along with other characters such as Mahatma Gandhi and Dr. Martin Luther King, they did contribute to the study of religion in a very positive and academic way when they moved away from the external appearances of behavior and focused on the motivations of individuals who attend church. What differentiates the regular churchgoer from the fringe members in such a way that their actions and thoughts are significantly different? The answer seems to become slightly more apparent when researchers, as Wilson did in his pioneer paper in 1960, look at the religiously orientated people in terms of the extrinsic or intrinsic nature of their orientation. According to Allport and Ross (1967).

Persons with this [extrinsic] orientation are disposed to use religion for their own ends. The term is borrowed from axiology, to designate an interest that is held because it serves other, more ultimate interests. Extrinsic values are always instrumental and utilitarian. Persons with this orientation may find religion useful in a variety of ways—to provide security and solace, sociability and distraction, status and self-justification. The embraced creed is lightly held or else selectively shaped to fit more primary needs. In theological terms the extrinsic type turns to God, but without turning away from self. (p. 434)

People with intrinsic orientation, however, find their master motive in religion. Other needs, strong as they may be, are regarded as of less ultimate significance, and they are, so far as possible, brought into harmony with the religious beliefs and prescriptions. Having embraced a creed the individual endeavors to internalize it and follow it fully. It is in this sense that he lives his religion. (Allport & Ross, 1967, p. 434)

Feagin (1964) used a more developed scale than Wilson (1960) did in order to measure the intrinsic orientation as well as the extrinsic one. His scales are essentially the same as those discussed by Allport and Ross (1967). Most notably, Feagin concluded that extrinsic and intrinsic motivations represented two independent dimensions and that only the extrinsic orientation was related to intolerance toward African-Americans. These as well as other scholars' finding helped establish three important findings dealing with the relationship between prejudiced attitudes and religious orientations followed by individuals:

1. On average, church attending individuals are more prejudiced than non-attenders.

2. While the above holds true, it can obscure a curvilinear relationship if the results are not sufficiently scrutinized. While most attenders are more prejudiced than non-attenders, a significant minority of them are less prejudiced.

3. The extrinsically motivated are the irregular fringe members who are high in prejudice. The intrinsically motivated are, however, the constant, devout, internalized members who are low in prejudice.

Allport and Ross (1967) presented the finding above in the beginning of their paper - Personal Religious Orientation and Prejudice - and continued to expand upon Wilson’s (1960) work by designing a psychological measure to include a broader and more improved scale of prejudice which was not confined to a certain ethnic group resulting in a refined extrinsic-intrinsic scale. Their scale which consisted of 20 questions and called "Religious Orientation" (p. 436) gained credit among the scientific community and it was implemented in a multitude of studies. The sample to whom Allport and Ross administered their questions was, however, in no sense representative. Graduate-student members of a seminar collected the 309 cases from the following church groups: Group A, 94 Roman Catholic (Massachusetts); Group B, SS Lutheran (New York State); Group C, 44 Nazarene (South Carolina); Group D, 53 Presbyterian (Pennsylvania); Group E, 3S Methodist (Tennessee); Group F, 28 Baptist (Massachusetts). (p. 436)

Allport and Ross (1967) followed an elaborate way of scoring their 20 items by following Peabody (1961) who prepared two versions of the same item, i.e., positively worded and negatively worded. He administered the positively worded version in one session and the negatively worded in another and then compared each individual's responses to differentiate those who were consistently pro or anti toward the content of items. Instead of having positive and negative items Allport and Ross, however, had their intrinsic and extrinsic items on which they analyzed their participants’ responses. The results showed that some people were consistently intrinsic, i.e., had a strong tendency to agree with intrinsically worded items and reject the extrinsically worded ones. Correspondingly, some other people were consistently extrinsic. A large group of people surveyed were, nonetheless, inconsistent and persisted in endorsing any or all items that seemed to favor religion in any sense. Allport and Ross chose to call this group “indiscriminately pro-religious” (p. 437).

Although the relationship between religious orientation and prejudice has been the main topic of investigation, the correlation between religious orientation and a large variety of factors such as marital satisfaction (Hosseinkhanzadeh & Niyazi, 2011) psychological and physiological health (Hunter & Merrill, 2013), happiness (Moltafet, Mazidi & Sadatic, 2010), cancer (Meyer, Altmaier & Burns, 1992), coping strategies (Mostafaei, 2012), pain management (Low, 1997), child abuse (Henderson, 2008), social conformity (Rodriguez & Henderson, 2010) and even economical aspects of life in way of managerial efficiency (DeNoble et al. 2007; Pyle, 1993; Tahir, & Abdul 2013) have been studied by a noticeable number of researchers and remains a hot topic of curiosity and exploration. The majority of these researchers have used some form of the scales influenced by Allport and Ross (1967) to address their variables of interest.

Unfortunately, however, few studies have explored the factorial validity of Allport and Ross’s (1967) conceptualization of religious orientation. They have employed the ROS as a universal measure which allegedly measures its takers’ intrinsic and extrinsic motivations everywhere in the world. Although Brewczynski and MacDonald (2006) critized such an application by announcing that “virtually all of the empirical research done in the psychology of religion in general, and with the ROS in particular, have been done with English-speaking participants” (p. 64), researchers such as Ghorbani et al (2000) have administered it to their Iranian participants who speak Persian as their mother language. Studies based on factor analysis, however, question the validity of approaching religion as a two-factor construct.

To explore the factorial validity of ROS with non-English speakers, Brewczynski and MacDonald (2006) translated the ROS into Polish and administered it to 385 undergraduate Catholic students in Poland. When they applied the structural equation modeling (SEM) to their collected data, their results showed that their Polish ROS consisted of three factors, i.e., intrinsic, personal extrinsic and social extrinsic, instead of two, i.e., intrinsic and extrinsic. Based on their findings, Brewczynski and MacDonald agreed with Byrne and Watkins (2003) that there are inherent limitations in “establishing the cross-cultural validity of measurement instruments” (p. 63). In the present study, the ROS developed by Khodadady and Baghei (2012) was, therefore, employed because its factorial validity has been established in the language with which its participants converse with each other.

1.2. Multifactorial Nature of Religious Orientation

In order to find out whether the ROS measured two religious orientations within an Iranian context or not, Khodadady and Gholparvar (2011) replicated Brewczynski and MacDonald’s (2006) study. They did not, however, employ SEM as Brewczynski and MacDonald had done, arguing that the results obtained through SEM contradicted those obtained through experimental designs. To support their position Khodadady and Gholparvar quoted Shiotsu and Weir (2007) who employed the SEM in their study of grammar and claimed that syntactic knowledge is relatively more significant than “vocabulary breadth in predicting text reading comprehension test performance” (pp. 123-4). Shiotsu and Weir’s claim was based on the regression and correlation summary of their study three in which they administered three tests to 591 participants to explore the relationship among grammar, vocabulary and reading comprehension ability.

Similar to Shiotsu and Weir (2007), Khodadady, Pishghadam and Fakhar (2010), decided to explore the relationship among syntax, vocabulary and reading comprehension ability. In contrast to Shiotsu and Weir, Khodadady et al., however, designed an experimental study in which they recruited 82 female learners of English as a foreign language (EFL) and divided them into two homogenous groups established via a pre-test. In one group, they taught vocabulary explicitly while the other received explicit instruction in grammar. At the end of the semester they developed two reading comprehension tests, i.e., traditional multiple choice item test (MCIT) and schema-based cloze MCIT, a grammar and vocabulary test and administered them to both groups.

Table 1 presents the results reported by Khodadady, Pishghadam and Fakhar (2010). As can be seen, the correlation coefficients obtained among the traditional MCIT (.77), schema-based cloze MCIT (.43) measuring the reading comprehension ability and the grammar test is noticeably lower than those obtained among the vocabulary test and the traditional MCIT (.82), and schema-based cloze MCIT (.57). Although as measures of reading comprehension ability traditional MCITs and schema-based cloze MCITs are developed on two different rationales (see Khodadady 1997, 1999a, 2014), they both show higher correlations with vocabulary than with grammar and thus challenge Shiotsu and Weir’s (2007) findings based on the SEM. While grammatical knowledge explains only 18% of variance in the schema-based cloze MCIT, the amount of variance explained by vocabulary reaches 32%.

Table 1. Regression and Correlation Coefficients of Language Components Reported in Two Studies

Khodadady and Gholparvar (2011) translated the ROS into Persian and administering it to 329 undergraduate university students. When they applied Maximum Likelihood (ML), Principal Axis Factoring (PAF) and Principal Component Analysis (PCA) methods of factor analysis to their collected data, with the exception of item three, "the prayers I say when I am alone carry as much meaning and personal emotion as those said by me in the presence of people", the other 20 items comprising the ROS loaded acceptably on four latent variables, indicating that the ROS measures four orientations, not two. Their results also provided further evidence to support Khodadady and Hashemi's (2010) findings that PAF provides the most logically related items comprising latent variables which underlie psychological scales when the extracted factors are rotated by Varimax with Kaiser Normalization (VKN).

1.3. Enlarged Religious Orientation Scale

Khodadady and Bagheri (2012) examined the 21-item ROS and realized that it lacked a number of religious practices such as supporting orphans exhorted in various chapters of the Quran such as 2:215, i.e., 'They will ask thee as to what they should spend on others. Say: "Whatever of your wealth you spend shall [first] be for your parents, and for the near of kin, and the orphans, and the needy, and the wayfarer; and whatever good you do, verily, God has full knowledge thereof."' (Asad, 1980, p. 69). They, therefore, added twelve new items to the scale in order to find out whether their addition would bring about the extraction of more rotated factors or not.

Khodadady and Bagheri (2012) administered their enlarged 33-item Persian ROS to 536 undergraduate students majoring in agriculture, architecture, engineering, English language and literature, psychology, Russian language and literature, science and theology at Ferdowsi University of Mashhad and Mashhad Branch of Islamic Azad University to validate the scale. The results of subjecting their collected data to PAF and VKN showed that the addition of new items increased the number of factors from four to seven. Furthermore, similar to Khodadady and Gholparvar’s (2011) study, item three, "the prayers I say when I am alone carry as much meaning and personal emotion as those said by me in the presence of people", did not load on any of the factors established by Khodadady and Bagheri, i.e., Inspirational, Intrinsic, Social, Concessional, Theo-pacific, Humanitarian and Sacrificial.

One of the greatest advantages of the 33-item Persian ROS over its earlier 21-item version is its being based on the religion practiced by Iranian Persian speakers, i.e., Islam. Based on the microstructural approach of schema theory (Khodadady, 1997, 2013), Khodadady and Bagheri (2012), had to analyze all the schemata constituting sentence 15 on the 21-item English ROS, "the prayers I say when I am alone carry as much meaning and personal emotion as those said by me during services" and replace the schema "services" with "people". This is because the schema "services" involves a number of activities such as playing music and singing in churches which are not carried out in Mosques. Muslim worshippers come together to offer everyday salats instead.

As described above, the very inclusion of schemata such as "services" has rendered the 21-item ROS religion specific, i.e., Christianity. Surprisingly, however, many researchers have employed it in Iran to explore the relationship between religious orientations and their variables of interest. Chavoshiet al. (2008), for example, administered the ROS along with the General Health Questionnaire developed by Goldberg and Williams (1988) to 150 Iranian university employees and 150 nonemployees. Based on their results, they claimed that “the religious orientation of employees was more internal and that of nonemployees was more external” (p. 149) without providing any correlations between religious orientations and mental health, i.e., “state of emotional well-being in which a person is able to function comfortably within his society” (Bhatia, 2009, p. 260)!

Khodadady and Bagheri’s (2012) study, however, provides the latest ROS which contains statements describing what Iranians do in their every day practice of Islam in order to lead a religious life. In contrast to the 21-item English ROS measuring two orientations within a Western English speaking society, i.e., intrinsic and extrinsic, their 32-item ROS addresses seven motivations underlying religious practices, i.e., Inspirational, Intrinsic, Social, Concessional, Theo-pacific, Humanitarian and Sacrificial. [The ROS consists of 33 items. Similar to Khodadady and Golparvar’s (2011) study, item three, “the prayers I say when I am alone carry as much meaning and personal emotion as those said by me in the presence of people,” did not, however, load acceptably on any of the seven factors extracted by Khodadady and Bagheri. It is does not, therefore, contribute to any of the seven genera explored in this study.]

1.3.1. Inspirational Genus

Five linguistic sentences represent five cognitive species which underlie the inspirational genus. (The species will be described in the discussion section shortly.) Inspirationally oriented students participate in Qadir and Qorban Salats, visit the Prophet Mohammad’s or [Shiite] Imams’ descendents on Eid Qadir, participate in ceremonies celebrating the birthdays of the Innocent Imams, attend not only the ceremony of the Night of Qadr but also their Imams’ mourning ceremonies.

The inspirational genus established by Khodadady and Bagheri (2012) is a unique factor because it appears as the first latent variable underlying the university students’ religious orientation and correlates the highest with the last genus, i.e., sacrificial (r = .614, p<.01). Among grade three senior high school (G3SHS) students, however, the inspirational genus was extracted as the fourth factor by Khodadady, Mousavi, and Sarraf (2012). It correlated the highest with the G3SHS students’ social factor (r = .55, p<.01). These results reveal the ever-evolving nature of cognitive genera, i.e., factors, and thus support Khodadady’s (2013) conceptualization of schemata as the basic units of cognition which represent specific dynamic concepts such as salats and ceremonies. Based on their common features they enter into a hierarchical relationship with each other to generate the broader concepts of genus and domain, i.e., inspiration genus and religious orientation, respectively.

1.3.2. Intrinsic Genus

The intrinsic genus underlying the ROS consists of four species. Intrinsically orientated university students consider religion very important because it answers their many questions about the meaning of life. They try hard to carry their religion over into all other dealings in life. Their religious beliefs are in fact what really lie behind their whole approach to life. It is also important for intrinsically motivated students to spend periods of time in private religious thought and meditation.

The intrinsic genus has a unique composition for university students because it does not exist for G3SHS students. This is because the four species which constitute university students’ intrinsic genus load along with five other species on G3SHS students’ social genus, indicating that as students move from secondary to tertiary education their genera of religion changes in their constituting species as well as the degree of relationship they hold with other genera. The university students’ intrinsic genus correlates the highest with their social factor (r = .56, p<.01), indicating that part of their inspirational and intrinsic genera are social in nature.

1.3.3. Social Genus

As the third factor underlying ROS, social genus consists of six species. Socially oriented students prefer the Quran Study and religious groups to others. They also study the literature about faith. The primary reason for their interest in religion is that their mosques provide them with congenial social activities and they attend them if they are not prevented by unavoidable circumstances. Being a mosque member helps them establish themselves in the community and formulate good social relationships.

In addition to correlating strongly with the inspirational and intrinsic genera (both r = .56, p<.56), social genus correlates significantly with theo-pacific (r = .398, p<.01), humanitarian (r =.348, p<.01), and sacrificial (r =.445, p<.01) genera, emphasizing the important role mosques play in formulating social relationships. Through networking university students get involved in social activities whose nature can become humanitarian such as visiting patients on various occasions. They are also fulfilled socially by not only sacrificing animals but also consuming the meat of sacrificed animals. Khodadady and Bagheri’s (2012) results, thereforem show that university students’ social interactions are defined within a religious context backed by humanitarian and sacrificial genera.

1.3.4. Concessional Genus

Five statements constitute the fourth factor, i.e., concessional genus, measured by the ROS administered in this study. Individuals having this particular motivation pray chiefly because they have been taught to pray. It doesn’t matter so much what they believe so long as they lead a moral life. Although they are religious persons, they refuse to let religious considerations influence their everyday affairs. Occasionally they find it necessary to compromise their religious beliefs in order to protect their social and economic well-being. And finally, although they believe in their religion, they feel that there are many more important things in their lives.

The results reported by Khodadady and Bagheri (2012) question the empirical validity of concessional orientation in terms of its contribution to the domain of religion. This is because it is the only factor which does not correlate significantly with the ROS (r = -0.03, ns). Neither does it correlate with theo-pacific (r = 0.01, ns) and humanitarian (r = -0.06, ns) genera. The concessional orientation does, however, show its strongest relationship with intrinsic orientation though in a negative direction (r = -.395, p<.01), followed by inspirational (r = -.254, p<.01) and social (r = -.207, p<.01) genera.

1.3.5. Theo-pacific Genus

Similar to concessional genus, five species comprise the theo-pacific genus extracted from Khodadady and Bagheri’s (2012) ROS. Theo-pacifically oriented students are quite often keenly aware of the presence of God. Religion helps them keep their life balanced and steady in exactly the same way as their citizenship, friendships, and other memberships do. Their primary purpose of prayer is to gain relief and protection on the one hand and secure a happy and peaceful life on the other. Religion offers them comfort when sorrows and misfortune strike.

Theo-pacific genus correlates the highest with the intrinsic genus (r = .498, p<.01) followed by the inspirational (r = .457, p<.01) and social (r = .398, p<.01) genera. These findings explain why Shiites commemorate the martyrdom of their Imams in general and Imam Hossein in particular. By comparing their everyday problems with what Imam Hossein, his family and disciples suffered in Ashura simply because they did not accept to swear allegiance to the imposed Caliph by force, university students pacify themselves and gain enough motivation to get involved in social activities.

1.3.6. Humanitarian Genus

Humanitarian genus consists of five species. They describe humanitarian individuals as the believers who donate on religious occasions, consider visiting patients as a religious duty, do charitable work like supporting orphans, devote some money to help charitable institutes and visit the deprived areas in their cities to help the settlers. Humanitarian genus relates significantly to inspirational (r = .56, p<.01), theo-pacific (r = .56, p<.01), intrinsic (r = .37, p<.01), and social (r = .35, p<.01) orientations, indicating that the more humanitarian university students are in their religious practices, the more inspired they become to get peace from their religious commitment. They motivate themselves intrinsically and socially to practice Islam. It does not, however, relate to concessional genus. The same lack of relationship holds true for G3SHS students, indicating that G3SHS and university students who compromise their religious principles lose their sympathy towards other fellow human beings as well.

1.3.7. Sacrificial Genus

Sacrificial Orientation involves loving to sacrifice an animal on the day celebrating the Festival of Sacrifices (ID AL-ADHA). Sacrificially oriented students also love consuming the meat of sacrificed animals. It is the only factor which shows the strongest relationship with the inspirational genus (r = .56, p<.01), explaining 38% of variance in each other. The sacrificial genus also correlates significantly with humanitarian (r = .513, p<.01), social (r = .445, p<.01), intrinsic (r = .440, p<.01) and theo-pacific (r = .349, p<.01) genera as well. Furthermore, it relates significantly but negatively to concessional orientation (r = -.142, p<.01) as do inspirational(r = -.254, p<.01), intrinsic (r = -.395, p<.01), and social (r = -.207, p<.01) genera.

According to the preamble of the amended Constitution of the Islamic Republic of Iran (1989), "the cultural, social, political, and economic institutions of Iranian society [are] based on Islamic principles and norms, which represent an honest aspiration of the Islamic Ummah [community]". Islam has been practiced as the official religion in the country for over 35 years (i.e., from 1979 to 2014). An important variable which is missing in the preamble is the relationship of Islamic principles and norms to the educational outcomes pursued in primary schools, high schools and universities. There should be research findings to allow educators in general and policy makers in particular to find out whether the budget allocated to religious organizations have born any fruits in the specified educational centers.

To the best knowledge of present researchers no research projects have, however, been conducted to explore the relationship between religion and education in general and Islamic orientation and educational outcomes in particular. The present project is, therefore, designed to find out whether the teaching of Islamic principles and norms by various Iranian governments in the past 35 years has had any effect on learning English as a foreign language (EFL) so that the fulfillment of objectives pursued by a number of organizations such as the Ministry of Culture and Islamic Guidance (MCIG) can be assessed objectively. The main problem addressed in the project is thus to find out whether religious orientation as a cognitive domain and its constituting genera relate significantly to the EFL learning at tertiary education in Iran.

2. Methodology

2.1. Participants

Two hundred sixteen, 186 female (86.1) and 30 male (13.9), university students took part in the present project voluntarily. Their age ranged between 20 and 50 (mean = 27.26, SD = 7.71). Almost half of them were married (n = 106, 49.1%). They were majoring in English language and literature (n = 116, 53.7), English translation (n = 6, 2.8%) and teaching EFL (n = 94, 43.5%) at Ferdowsi University of Mashhad (n = 70, 32.4%) Khayyam University (n = 94, 43.5%) and Imam Reza University (n = 50, 23.1%) at undergraduate (n = 144, 66.7) and graduate (n = 72, 33.3%) levels. They had all learned English at various private institutes before being accepted in the university. They spoke Persian as their mother language.

2.2. Instrumentation

Three instruments were employed in this study: a Biodata Questionnaire, 32-item Religious Orientation Scale and a disclosed English language proficiency test.

2.2.1. Biodata Questionnaire

In order to obtain the required demographic information a biodata questionnaire containing one short-answer question and six multiple choice items was developed in this study. They dealt with the participants’ age, educational background, gender, the institutes where they had learned the EFL, the university where they were studying, marital states and mother language.

2.2.2. Religious Orientation Scale

The Religious Orientation Scale (ROS) designed and validated by Khodadady and Bagheri (2012) was employed in this study. It consists of 32 sentences describing religious practices with which the participants were required to completely agree, agree, express no idea, disagree or disagree completely. The first sentence, for example, read, "I try hard to carry my religion over into all my other dealings in life". It measures Iranian university students' religious orientation consistently because its alpha reliability coefficient is .89 as shown in Table 2. Among its seven underlying factors, Inspirational genus enjoys the highest reliability, i.e., α = .89, followed by Intrinsic α = .85, and Sacrificial, α = .81, genera. The alphas of other genera range from .79 (social) to .73 (humanitarian), providing some of the most reliable genera among the latent variables established by similar scales such as Spiritual Intelligence Self Report Inventory (SISRI) designed by King (2008). Khodadady and Moosavi (2014) translated the inventory into Persian and validated it with G3SHS students in Mashhad, Iran. Out of seven factors underlying the inventory, the alpha of its first factor, i.e., Purposive genus, did not exceed .76, while the first factor of the ROS had an alpha of .89.

Table 2. Descriptive Statistics of ROS and Its Seven Factors

The seven factors underling the 32-item ROS relate differently to the domain of religion orientation. As can be seen in Table 3, Khodadady and Bagheri (2012) reported the strongest relationship of the domain with its Inspirational (r = .833, p<.01), Theo-pacific (r = .713, p<.01) and Intrinsic (r = .711, p<.01) genera, respectively. The domain did not, however, show any significant relationship with the Concessional genus (r = -.030, ns), questioning its relevance to religious orientation as a cognitive domain. The Concessional genus, however, correlates significantly but negatively with Inspirational (r = -.254, p<.01), Intrinsic (r = -.395, p<.01) and Social (r = -.207, p<.01) and Sacrificial (r = -.142, p<.01) genera, indicating that the more concessional people become in their religious orientation, the less inspirational, intrinsic, social and sacrificial motivations they derive from their religious practices.

Table 3. Correlations among the 32-Item ROS and Its Seven Factors

2.2.3. English Language Proficiency Test

Similar to Khodadady and Namaghi (2013), the disclosed 50-item English Language Proficiency Test (ELPT) developed by Hale et al. (1988) was employed in this study to measure participants’ English language proficiency. To design the ELPT, Hale et al. chose six passages from disclosed Tests of English as a Foreign Language (TOEFL) and developed 150 cloze multiple choice items (MCIs) on these passage. For designing the ELPT, they deleted 150 words from those constituting the passages and replaced them with a table in which four numbered choices were presented. They administered the passages and their cloze MCIs to learners of English as a second language in three local American universities and from among them chose three passages whose constituting items had proved to be the most well-functioning, i.e., they had acceptable item facility and item discrimination indices.

The original format of the ELPT consisted of five pages on which Hale et al. (1988) presented each of their fifty MCIs within a table. Since using a table to present the alternatives of cloze MCIs occupies a lot of space, the format designed by Khodadady and Namaghi’s (2013) was employed in this study. Instead of providing each keyed response along with its three alternatives in a vertical column, they replaced each keyed response with a numbered blank space and offered it along with its alternative in single row consisting of five columns. The table appeared immediately below the paragraph in which the blanks had been given as shown in Table 4.

Table 4. Two Examples of Cloze MCIs Used in the ELPT

Hale et al. (1988) administered the 50-item ELPT along with a standard TOEFL test consisting of listening, structure and written expressions, vocabulary and reading comprehension ability subtests to 11, 290 test takers among whom 476 were Persian speakers. Their results showed that the difficulty level of items matched the range of difficulty encountered in typical TOEFL tests. The mean, standard deviation, and alpha reliability coefficients reported for the test were 28.55, 9.19 and 0.89, respectively.

Table 5 presents the correlation coefficients obtained between the ELPT and the TOEFL as well as its five subtests taken by Farsi speaking participants in Hale et al.’s (1988) study. As can be seen, the ELPT correlates significantly with the TOEFL (r = .89, p<.1). According to Hatch and Lazaraton (1991), when two tests correlate at .80 and higher they measure the same construct and can, therefore, be used interchangeably. In other words the ELPT measures the English language proficiency as best as the TOEFL does. It is, however, much shorter than the TOEFL and does not take more than 33 minutes to be completed.

Table 5. Correlations between the ELPT, the TOEFL and Its Subtests

2.3. Procedures

After printing and copying the ELPT, its answer sheet, and the ROS in adequate numbers, the instructors of English at Ferdowsi University of Mashhad, Khayyam University and Imam Reza University were contacted and their approval was gained to administer the instruments in their classes. On previously agreed and set dates, the second researcher attended the classes, explained the research project and asked the participants to take the instruments as best as they could. They were informed that the research was conducted as part of a graduate program requirement and nobody could have access to their responses. To secure their anonymity further, a code was assigned to each participant and they were asked to write the codes on their scales as well as the answer sheet of the ELPT so that the researcher could run correlational analyses appropriately.

2.4. Data Analysis

The descriptive statistics of items comprising the ROS and ELPT were calculated to determine their functioning. The item facility (IF) and item discrimination (ID) indices of items on the ELPT were also estimated to determine its internal validity. The reliability of the scale, its underlying factors and test was estimated via Cronbach’s (1951) alpha. Pearson-point correlations were utilized to explore the relationship between the ROS and ELPT measuring religious orientation and EFL proficiency. The same correlations were used to find out whether the factors underlying the ROS relate significantly to the ELPT. All statistical analyses were conducted via the IBM SPSS Statistics 20 to test the two hypotheses below.

H1. There is no significant relationship between religious orientation and EFL proficiency.

H2. There is no significant relationship between the factors underlying religious orientation and EFL proficiency.

3. Results

Table 6 presents the descriptive statistics and reliability estimates of the ROS and its underlying factors. As can be seen, the scores on the ROS range from 44 to 145 with a mean of 102.6 which is very close to that of Khodadady and Bagheri (2012) i.e., 102.04. Similarly, the alpha reliability coefficients of the scale for both studies is .89, indicating that the 32-sentence ROS measures the university students’ religious orientation reliably from samples to samples. The alpha of the seven factors ranged from .58 (concessional) to .87 (inspirational and intrinsic). As can be see, the alpha coefficient of concessional factor is the lowest which might be because of its relatively heterogeneous items as reflected in its standard deviation (3.6). Although humanitarian factor comprises the same number of items, i.e., five, its standard deviation is lower (3.2), resulting in its relatively higher alpha coefficient, i.e., .65.

Table 6. Descriptive Statistics and Reliability Estimates of the ROS and Its Seven Factors (N = 216)

Table 7 presents the descriptive statistics and reliability estimates of the ELPT in three studies.

Table 7. Descriptive Statistics and Reliability Estimates of the ELPT

As can be seen, the mean score obtained in this study is 25.41, indicating that the test was more challenging to its participants than to those of Khodadady and Namaghi (2013) and Khodadady (2004), i.e., 26.5 and 30.7, respectively. The challenging nature of the test is further supported by the mean IF (.51) which is noticeably lower than that of Khodadady (2004), i.e., .61. The difference lies in the fact that while the participants of this study did not receive any rewards for their participation, those of Khodadady received a bonus of two scores added to final examination scores in any course they passed with the researcher.

Table 8 presents the correlations between the ELPT, ROS and its underlying factors. As can be seen, the ROS does not correlate significantly with the ELPT (r = -.09, ns). This result confirms the first hypothesis that there is no significant relationship between religious orientation and EFL proficiency. It is in line with Moosavi’s (2014) study in which she could not find any significant relationship between spiritual intelligence and G3SHS students’ English language achievement (r = -.003, ns), either.

Table 8. Correlations between the ELPT, ROS and Its Underlying Factors (N = 216)

Among the seven factors underlying the ROS, only humanitarian (r = -.147, p<.05) and sacrificial (r = -.156, p<.05) factors correlate significantly but negatively with the ELPT, partially rejecting the second hypothesis that there is no significant relationship between the factors underlying religious orientation and EFL proficiency, indicating that the more proficient the university students become in their English language, the less humanitarian and sacrificial they become in their religious orientation.

4. Discussions and Conclusions

The results of this study did not reveal any significant relationship between the domains of religious orientation and English language proficiency because the latter plays a minor role, if any, in university students’ everyday life. This finding is in line with what Moosavi (2014) found in her study. After translating the SISRI into Persian and validating it with 344 female G3SHS students, she correlated their scores on the SISRI and final English language examination and found no significant relationship between the two (r = -.003, ns), indicating that similar to English language proficiency which shows no significant relationship with university students’ religious orientation, G3SHS students’ English language achievement does not relate to their spiritual intelligence.

The linguistic analysis of the schema types constituting the ROS and ELPT explains the reason behind the lack of any significant relationship between the cognitive domains of religious orientation and English language proficiency. As shown in Table 9, 470 schema types constitute the language of ELPT and ROS. The highest number of schema types which is common to both measures, i.e., 10, quantifies determiners representing the concepts of “a”, “all”, “an”, “many”, “one”, “other”, “some”, “such”, “that” and “the”. These concepts play a syntactic role in that they specify nouns and depend on them to express any meaning. Out of 168 noun schema types, however, only one is common to both ELPT and ROS, representing the concept of “things”.

Table 9. Linguistic genera specific and common to the ELPT and ROS

The only noun schema which is common to the ELPT and ROS, i.e., “things”, does not, nonetheless, represent the same concept in the two measures because of the species of which it forms a part in the ELPT, i.e., “Things at different depths are superimposed causing confusion to the viewer” and ROS, i.e., “Although I believe in my religion, I feel there are many more important things in my life”. While the schema “things” has been specified by “at different depths” in the ELPT, it is left to the takers of the ROS to figure out what “things” might refer to.

The results of this study, therefore, show that the domains of religious orientation and EFL proficiency deal with two different constructs for several reasons. First, while religious orientation reveals itself in known schemata and species represented by words and sentences in the participants’ first language, the EFL proficiency requires grappling with the unknown. In other words, while religious orientation deals with university students’ personal life as they live it, the

English language proficiency requires reading and understanding schemata and species which are brought up by the anonymous authors of the three texts constituting the ELPT.

The parasyntactic domain of the ELPT and ROS, for example, consists of 17 and 9 distinct names, respectively. The former includes the schemata “Ben Sands”, “Blues”, “Cherokee”, “Chisholm Trail”, “Craighead Caverns”, “Guinness Book of World Record”, “Home on the Range”, “Jazz”, “John A Lomax”, “Kansas”, “Lomax”, “Lost Sea”, “New Orleans”, “San Antonio”, “Southeastern Tennessee”, “Texas”, and “The Council Room”. None of these share any semantic feature with the nine names employed in the ROS, i.e., “Eid Qadir”, “Eid Qorban”, “God”, “Imams”, “Night Of Qadr”, “Qadir”, “Qorban”, “Quran” and “Salats”.

The second reason for the lack of significant relationship between the domains of religious orientation and EFL proficiency is the former’s dependence on field-dependency while the latter requires field-independency. The species comprising religious orientation require attending social gatherings in which students usually socialize in order to establish appropriate networks. As Wyss (2002) convincingly argued field dependents achieve a higher degree of success in everyday language situations beyond the constraints of the classroom and are capable of performing tasks which require interpersonal communication skills. EFL proficiency, however, requires learning English in classes and comprehending various types of genera represented by texts whose authors shared few schemata, if any, with the participants of this study. The ELPT developed by Hale et al. (1988), for example, included three texts dealing with literature, i.e., “Folk Songs”, geography, i.e., “the Lost Sea” and medicine, i.e., “Tomography”.

And finally, no official attempt is made to motivate the EFL learners in Iran to attain EFL proficiency measured by the ELPT. As the results reported by Khodadady and Ashrafborji (2013) indicated three types of motivation underlie EFL learning in Khorasan Language Institute, i.e., Intrinsic, Extrinsic, and Communicative. None of these motivations, however, is related to the learners English language achievement. Contrary to the researchers’ expectation even the extrinsic motivation did not relate significantly to the learners achievement though its description reveals its field-dependent nature, i.e.Extrinsically oriented EFL learners think that they will be considered as a poorly educated person if they don’t know English and they will disappoint those closest to them if they fail to learn English. They also think that not learning English will have a negative impact on their life in that they cannot find a suitable job. However, mastering English will not only increase their social status but also get them reward from their parents/family. Furthermore, being able to speak English makes the learners feel superior to others. (Khodadady and Ashrafborji, 2013, p. 5)

Similar to motivations underlying the EFL learning in Iran which does not relate significantly to its achievement, religious orientation does not show any significant relationship with EFL proficiency because the Iranian authorities do not provide EFL learners with any officially endorsed external motivation for EFL learn ring such as rewards, public recognition, receiving scholarship from state or private organization and placing EFL proficiency as a determining factor in getting employed in state organizations. Religious orientation is, however, directly and indirectly supported by the theocratic government at local, provincial and national levels.

The present government headed by President Rouhani, for example, submitted its proposed budget for year 1394, new Iranian year starting in March 2015, to the Iranian parliament for approval. The university students following the late leader Ayatollah Khomeini’s line issued a communiqué on December 24, 2014 in which the government was asked to cut the budget allocated for certain private institutes. It has, for example, proposed 14 and two billion tomans (approximately 4,000,000 and 285,714 USD with an exchange rate of 3500 tomans per dollar) for Imam Khomeini Institute (IKI) and Sadra Foudation for Islamic Philosophy (BONYAD HEKMAT ESLAMI SADRA [BHES]), respectively. The proposed budget for the cultural activities of Universities of Medical Sciences throughout Iran is, however, just 3 billion tomans (approximately 857,143 USD).

As the students’ communiqué published by Farhangiyan News (2014) brings up, oil price dropped to half of its previous value worldwide in December 2014. The Iranian public is, according to the communiqué, suffering from the pressure exerted by the US-led sanctions on the one hand and economic ruins left from the previous government on the other resulting in the budget reduction of various governmental organizations. However, the two private IKI and BHES run by Ayatollah Mesbah Yazdi and Ayatollah Seyyed Mohammad Khamanei, respectively, are given liberal budgets! The objectives followed by the IKI might show why religious orientation does not relate to EFL proficiency.

According to the public relations office of IKI (2014), the institute is following three main objectives: 1) specifying and applying Islamic sciences to various fields of human sciences, 2) educating authoritative researchers in Islamic and human sciences who have the necessary scientific qualification to stand against distorted and mixed thoughts, and 3) educating experienced and committed academic members for Howzeh and universities! None of these objectives deal with teaching EFL or sponsoring those organization which offer EFL programs.

It is suggested the study be replicated with another sample of university students majoring in English and reward them appropriately to find out whether it will result in finding a significant relationship between religious orientation and EFL proficiency. It would also be illuminating if the university students' achievement in various courses are explored in relation to their religious orientation.