Cultural Differences in Transactional Distance Preference by Arab and American Distance Learners

Moore (1972, 1973, 1986, 1993) explains that “distance” in distance education should be viewed as a pedagogical distance between the teacher and learner instead of only a geographical distance. He suggests that, in addition to autonomy, this distance is a function of two other variables related to the teaching and learning transaction: dialogue and structure. This is referred to as the theory of transactional distance. However, the literature on both variables in addition to autonomy is not comprehensively researched across cultures. Drawing from cross-cultural literature, the two variables function differently across cultures. Thus, they are expected to affect the way learners perceive their ability to control their learning (self-efficacy) and the way they seek help, both of which are expected to eventually affect their potential to succeed in distance education. Therefore, the purpose of this study was to investigate cultural differences between Arab and American distance learners by focusing on the following questions:

1. Are there any differences between Arab and American distance learners' preferences toward transactional distance: structure and dialogue (interaction with instructor with instructor and interaction with other learners)?

2. What are the relationships between (1) distance learners' preferences toward transactional distance, (2) learner self-efficacy, and (3) help seeking?

From a cross-cultural perspective based on Hofstede (2001), dialogue will depend on the nature of relationships (individualism vs. collectivism) and hierarchy (power distance) in a society. Hofstede suggests that collective cultures try to maintain harmony and face through avoiding confrontations and conflicts, whereas individualistic cultures prefer open and direct discussion of conflict. Students in individualistic cultures are encouraged to speak more than in collective cultures. So, not speaking in class in collective cultures does not necessarily reflect lack of knowledge or ability. As Lewis (2003) explains, this pertains to cultural differences in communication styles. Lewis compares between the different structures for reasoning between Westerners and Easterners. He labels the Western logic as “monocular,” which means that it is based on one side of a proposition and linearly. Dialogue and debate facilitate this type of logic.

By contrast, Lewis (2003) divides Easterners' logic into two types: binocular and polyocular logic. Binocular logic is based on two sides of a proposition. For example, the Chinese try to accommodate another's point of view while keeping their own. So, monologue is a preferred form of communication because silence is a meaningful time for reflection and preparation of a response. Second, polyocular logic is based on a “reactive, circular form of thinking where a multiplicity of points of view is accommodated” (p. 145). For example, the Japanese make an extra effort to save everybody's face by accommodating all points of views. Although the binocular and polyocular logic lead to more ambiguity, it provides Eastern people with more comfort because “different interpretations of a situation facilitate avoidance of conflict and leave options open for future cooperation” (p. 145). This makes Asians avoid direct answers and expect a solution to emerge whereas Americans offer direct answers to problems.

In addition, dialogue is a function of power distance, especially between the teacher and learner. Based on this, Hofstede (2001) suggests that the learning environment in collective cultures is more teacher centered than it is in individualistic cultures. This suggests a more formal, one-way communication and that the strict social order gives the teacher the authority to initiate all communication in the classroom. By contrast, in small power distance cultures, where there is more stress on equality between students and teachers, students may make uninvited interventions. The way cultural views of relationship and hierarchy affect distance learners dialogue is not clearly used in distance education environments.

Similarly, “different cultures have different needs for structure in order to function efficiently” (Mead, 2005, p. 185). The function of structure in cross-cultural literature can be explained through Hofstede's dimensions of power distance and un-certainty avoidance. When uncertainty is high, the teaching process is very structured with “precise objectives, detailed assignments and strict timetables” (Hofstede, 2001, p. 162). In low uncertainty avoidance, there is less structure in the educational process and students are rewarded for originality. In low uncertainty avoidance, there is less structure in the educational process and students are rewarded for originality.

Mead (2005) divides structure into formal and informal. A formal structure stresses rules and laws. It regulates tasks and relationships, so that responsibilities are specialized and roles are clear. Applied to education, this suggests a precise description of educational tasks, including: who will perform the task, how it will be performed, what tasks typically precede it, what resources are needed to perform it, and where and when the task is performed. In addition, clear communication signals are provided, such as, which superior to report to. This type of structure is common in countries with high power distance.

Mead (2005) suggests that in these cultures there is a preference for “hierarchical lines of control and communication” (p. 172). In other words, communication is more vertical. An informal structure is reflected in felt but not expressed obligations between the more and less powerful persons. Responsibilities and routines are constantly changing. In this case, it is possible that there is more than one superior to report to. This type of structure is more common in cultures where there is less need to avoid un-certainty and have low power distance. In comparison with the previously mentioned formal structures, informal structures are ambiguous and authority is less precise allowing people to negotiate for influence. Communication is more horizontal between all levels and units (Mead, 2005). White (1999) explains that although novice distance learners in her study expressed no initial uncertainty, during the middle of the course most students felt less sure about either themselves as learners or about their understanding of learning material in this solo learning context. In order to overcome these feelings, learners reported using affective control strategies such as selftalk, taking a break, and acknowledging their feelings. Another strategy was continuing engagement in the task.

Moore (1980) suggests four levels of classifying programs based on the variables of structure and dialogue: –D–S (no dialogue, no structure), –D+S (no dialogue, some structure), +D–S (some dialogue, no structure) and +D+S (some dialogue, some structure. Braxton (2000) suggests classifying programs into nine levels instead of four by adding a middle level to dialogue and structure, so programs can be ranked on each component of transactional dialogue and structure, so programs can be ranked on each component of transactional distance as low, medium or high.

Dialogue is defined as the “extent to which a learner may communicate with his teacher” (Moore, 1973, p. 665). Moore suggests that this dialogue should be “purposeful, constructed and valued” (Moore, 1993, p. 24). The lack of dialogue in programs is an indicator of the lack of responsiveness to individual learner needs (Moore, 1993). In reality, this could be provided not necessarily though a Socratictype dialogue, but most importantly through providing an effective communication system that responds to student needs such as answering specific questions about homework problems (Vogeli, 1999). Moore uses the term “dialogue” to reflect a mutual relationship between the teacher and learner. As he explains (1993, p. 24), “each party is a respectful and active listener; each is a contributor, and builds on the contribution of the other party or parties.” The extent of this dialogue, he explains, will be affected by a number of factors including educational philosophies, content matter, individual personalities and most importantly medium of communication. Web-based technologies allow for more peer interaction that is accessible beyond constrains of time and location (Sung, Chang, Chiou, & Hou, 2005). However, Zimmerman (2002) notes from her qualitative study of self-regulation behavior of distance learners that skillful self-regulators value interaction less than do naïve self-regulators. However, the former do not mind helping the latter.

Structure is defined as “the extent to which the objectives, implementation procedures, and evaluation procedures of teaching program are prepared … to meet specific objectives, implantation plans, and evaluation methods of individual students” (Moore, 1980, p. 21). This determines the flexibility or rigidity of educational programs to respond to individual learner needs.

It is important to note that both dialogue and structure will depend on the type of media used because the differences in the inherent media characteristics may permit for more or less dialogue and/or structure resulting in different degrees of transactional distance. For example, interactive video or audio tend to over-structure material and allow for little dialogue resulting in large transactional distance (Moore, 1993). Jung (2001) analyzed 58 research articles in five leading distance education journals to investigate the pedagogical processes in web-based instruction (WBI) using the theory of transactional distance. He found that WBI provides more “structural flexibility” because it provides a way to “adapt contents to match individual students” (p. 530). By doing so, it increases student interaction. Three types of interaction were encouraged in WBI: academic interaction, collaborative interaction, and interpersonal interaction. Results from this study seem to be inconclusive. The researcher did not provide enough description of the analysis criteria for classifying articles in relation to pedagogical components of transactional distance. A few studies were frequently cited while others were not even mentioned.

Bandura (1991) defines self-efficacy as people's perceived ability “to exercise control over their own level of functioning and over events that affect their lives” (p. 257). Part of the cognitive process of self-efficacy is assessing environmental constrains that may reduce personal control (Bandura, 1991). It was found to be positively related to positive control orientation (Yamaguchi, 2001). Using motivational strategies, learners “initiate and direct their behavior toward desired learning goals” (Bois & Staley, 1997, p. 175). This requires learners to activate their self-evaluation processes so they have a personal standard to judge their progress. Those with high self-efficacy persist in the face of difficulties, while others who have self-doubts abort their effort prematurely (Bandura, 1977, 1991). Self-efficacy beliefs are affected by the regulation of the degree of anxiety and depressive mood one may experience when pursuing goals. The lack of one's belief in his or her ability to control potential threats results in coping deficiency that may lead individuals to magnify “the severity of possible threats” and consequently distress themselves (Bandura, 1991, p. 25).

In distance education, how learners maintain their effort and control their anxiety remains largely unexplained. Lee and Witta (2001) found that self-efficacy for online technology and online course content has increased during the course of a semester; however, it was not predictive of learner performance. In another study by Lynch and Dembo (2004), self-efficacy for learning and performance alone was found to explain 7% in the variance in learner grades in a blended online course. Computer self-efficacy was found to predict the likelihood of learners' future participation in web-based courses as well as their satisfaction with web-based courses (Lim, 2001).

From a cross-cultural perspective, views of self-efficacy are expected to be mediated by the external sociocultural beliefs leading to differences in the way self-efficacy beliefs operate in Western and non-Western cultures. From a review of 20 cross-cultural research studies, Klassen (2004) concludes that even when students perform equally well, non-Western collectivist groups report lower self-efficacy than Western individualistic groups. He suggests that collective efficacy works the same way for collective groups as self-efficacy for individualistic groups. Collective efficacy refers to “group's shared belief in its conjoint capabilities to organize and execute the courses of action required to produce given levels of attainments” (Bandura, 1997, p. 477). Lim (2004) reached a similar conclusion in her study of U.S. and Korean online students. She found a significant difference in learning motivation between the two groups, with U.S. students scoring higher in their motivation in relation to the perceived importance, relevancy of learning content, learning process and learner's ability to learn, while Korean students showed more learner control. Lim explains that while Asian student are more orientated toward effort attributions and performance goals, U.S. students are more oriented toward a mastery of learning over time and enjoying the learning process.

Knowles (1975) suggested that self-directed learners are engaged in a process in which they “take the initiative with or without the help of others in diagnosing their learning needs, formulating learning goals, choosing and implementing learning strategies, and evaluating learning outcomes” (p. 18, italics added). Moore (1972) suggests that autonomous learners will turn to teachers for help temporarily surrendering their control over their learning process. However, the type and purpose of help has been established by them in the first place. Students with prior distance education experience are more likely to ask for the instructor's help outside the course (King, Harner, & Brown, 2000). There are qualitative differences noted by Zimmerman (2002) in help seeking between distance learners who persist and those who do not persist. Noncompleters rely more on face-to-face help while completers try to adapt their help-seeking behavior to the online environment.

Kitsantas and Chow (2007) explain that distance education provides electronic, asynchronous means for seeking help, which makes it less threatening compared with traditional education with face-to-face interaction. These researchers found that college students in distance and distributed learning classes felt less threatened to seek help from formal sources such as the instructor and support services. They also sought help more frequently than students in traditional courses. Ryan, Pintrich, and Midgley (2001) suggest, “considering help seeking, not just as an academic self-regulatory strategy, but also as a social interaction with others.” Coming from a collective culture, relationship building is more significant for Arab students than for American students. This is not to say that Arab students are more socially competent than American students. However, because of their multi-active communication style they attach more importance to relationships (Lewis, 2003), which requires frequent interaction.

The study sample came from two distance teaching institutions: The Arab Open University (AOU) (Kuwait and Bahrain branches) and The Pennsylvania World Campus in the United States. The study used an equivalent sample of 95 cases from each group. The following table provides more descriptive statistics for sample demographics.

Other than equality of cases from each group, the two samples are not comparable at all levels because they differed in the type of programs students studied and level of study (graduate, undergraduate). Given these limitations, the researcher believes study results do reflect cultural differences and calls for more research to validate these differences.

Measuring Interaction. Preference for future interaction was measured as a continuous variable according to Bischoff, Bisconer, Kooker, and Woods (1996) by number of times students would prefer to communicate with (1) their instructor and (2) other learners in future courses. In this study they were asked to indicate their preference on a scale from 0 (no communication whatsoever) to 100 (the ultimate amount of communication) to communicate in their current courses.

Measuring Structure. Preference for future course rigidity or flexibility was measured according to Chen and Willits (1998), who measured student preference for flexibility or rigidity of teaching methods, learning activities, pace, attendance, objectives, choice of readings, requirements, deadline of assignment, and grading. They subdivide these into two factors: course delivery (α = .69) and course design (α = .75). In this study, although the measure for structure was found to be reliable with Cronbach's alpha coefficients for internal consistency of more than .80 (0.884 for the Arab sample and 0.824 for the American sample), a confirmatory factor analysis using structural equation modeling did not hold. A follow-up exploratory factor analysis with principal component and Varimax rotation revealed three factors for the American sample with eigen values more than one. The factors explained 70% of the variance in structure and 43% of this variance was explained by the first factor alone. For the Arab sample, only two factors had eigen values more than one and they explained 67.2% of the variance in structure. These results slightly differ from the results reported by Chen and Willits (1998), who originally wrote the scale. The authors found two factors (course delivery and course design) instead of three for structure.

Given the multidimensionality of structure, only the factor indicating “course design” was used in further analysis to avoid lack of measurement invariance. These items were course objectives, choice of readings, requirements, and deadlines of assignments. Items not included in further analysis were course pace, grading, attendance, teaching methods, and learning activities. These items may reflect differences between implantation of distance education programs at the Penn State World Campus and the Arab Open University. To confirm that the course design factor is the same in both groups, a multiple-group measurement model was fitted using confirmatory factor analysis. This model was found to fit well (RMSEA = 0.00, NNFI = 1.01, CFI = 1.00). The model was scaled by fixing the loading of “course requirements” on the factor to 1.00. Table 2 shows maximum likelihood parameters for the confirmatory factor analysis for structure.

Self-efficacy and help seeking were adapted from the MSLQ “The Motivated Strategies for Learning Questionnaire,” which was developed by Pintrich and his colleagues in 1990s at the University of Michigan (Pintrich, Smith, Garcia, & McKeachie, 1993). MSLQ's self-efficacy for learning and performance scale is used to measure a wider range of the dimensions of self-efficacy including expectancy for success, self-appraisal of one's ability, and judgment of one's ability to accomplish tasks. In this study, MSLQ items were modified to measure self-efficacy at the program level instead the original course-level in the scale. In addition, only six of the original eight scale items were used to reduce overall scale size. The six-item scale was found to be reliable with Cronbach's Alpha Coefficients for Internal Consistency of more than .80 (0.891 for Arabs and 0.872 for Americans).

The help-seeking scale in the MSLQ was comprised of four items asking respondents to indicate the likelihood of knowing when they need help and being “able to identify someone to provide them with some assistance” (Pintrich et al., 1993, p. 29). This scale in the original MSQL has low reliability (Cronbach α = 0.52), so one additional item was added: “I use online forums to ask for help from other students.” The item reflects the nature of help seeking in distance education environment. This item was added based on the research of Whipp and Chiarelli (2004) and Zimmerman (2002) in relation to help seeking in web-based distance education.

To investigate differences between the groups, MANOVA was conducted on the three variables of interest in this study: structure, interaction with instructor, and interaction with students. MANOVA resulted in a significant multivariate effect for group, Wilk's λ = 0.689 and p < 0.000 meaning that in the population, Arabs and Americans were not equal on all the criterion distance education variables. This relatively high (= 0.754) Wilk's λ indicates a weak relationship (η² = 0.246) between distance education variables (preference for course structure, preference for interaction with instructor, and preference for interaction with other students) and group. To investigate the multivariate effect, the following table presents the univariate ANOVA for the three variables, two of which were significant. In this analysis, note that the adjusted alpha for each test is 0.017 for the univariate ANOVAs after adjusting for the experiment-wise type I error, for the α level of 0.05.

Differences between Arab and American students in their preferences toward structure were consistent with cross-cultural literature on uncertainty avoidance. The group mean of the Arab sample (m = 13.43) on their preference for structure was significantly higher than the American sample (m = 18.45). Hofstede (2001) explains that Arabs have higher uncertainty avoidance than Americans, and thus we would expect they would prefer more structure. This means that in cultures with high uncertainty, the teaching process is characterized with “precise objectives, detailed assignments and strict timetables” (Hofstede, 2001, p. 162). In addition to this cultural factor, the open university model followed by the Arab Open University lends itself to more structure because course content and structure are prepackaged with little flexibility for tutors to change. The Pennsylvania State University World Campus instructors have more flexibility in their courses.

Similarly, consistent with differences between the collective Arab culture and the individualistic American culture, Arab students (m = 69.06) preferred significantly more interaction with their instructors than did American students (m = 47.75). Only preference for interaction with other students was not found to be significant in the univariate analysis. Although it was not significant, Arabs had a slightly higher mean (m = 59.72) on this variable than Americans (m = 54.18). Within an open university system like the AOU, students already have more interaction with their instructors than in an online system with limited interaction with instructors like the WC. On average AOU provides weekly face-to-face tutorial sessions for 1 hour for courses with three credit hours. In addition, AOU tutors are required to maintain announced weekly office hours (AOU, 2006).

Students' qualitative comments provide some insight about the type of interaction students in each group yearned for. This should be viewed in light of how cultural views of relationship and hierarchy affect the type and source of interaction. Arab students expressed a need for more supportive interaction with instructors. Students wanted instructors' help in learning both about content and technology. However, they expressed a gap between them and their instructors and difficulty communicating with them. Some students felt a few instructors were drawing the line in their relationship with them. Students referred to them as “arrogant” and “don't care about students' struggle with courses.” One of the most frequent themes in students' comments was the lack of qualified teachers at the Open University. Lack of qualification was described in terms of lack of content knowledge about courses and lack of competency in using the distance education system. Students suggested that some instructors taught using traditional teaching methods, and they were not flexible enough to accommodate the new system, let alone help students with it. These characteristics of the relationship between students and instructors are consistent with Hofstede's (2001) description of high-power distance cultures, in which there is high teacher-student inequality and teachers are perceived as educational gurus. So, for students they had to look for more formal structures (Mead, 2005) and ways of communication with their instructors; at many times this would result in a gap between the two because in distance education it is expected that learners would take more responsibility for their learning and initiate communication.

Some American students said that they missed the personal face-to-face communication with their instructors. A number of them suggested alternative ways of communication such as taped lectures instead of written lessons and weekly online live chats between instructors and students. The most frequent request for interaction with instructors was students' need for instructors' feedback on course assignments:

Table 4 presents Pearson's correlation between the transactional distance variables (structure, interaction with instructor, interaction with students), and help seeking and self-efficacy.

For both samples, seeking help was significantly and positively related to preference for interaction with other students. In other words, students who seek more help generally would like to have more interaction with other students, maybe as potential sources of help. For the Arabs, help was also significantly and positively related to self-efficacy. Arab students who ask for help are more likely to have higher self-efficacy and prefer to interact more with other learners. For the American sample, in addition to the correlation between help and interaction with other learners, help was also significantly and positively correlated with preference for interaction with the instructor. This was not the case for the Arab sample, which may reflect the high power distance in the Arab culture between instructor and students consequently affecting the way students perceive instructors as sources of help. This also suggests differences in the sources students seek for help. While American students preferred seeking help from formal sources (i.e., instructors) as well as less formal sources (i.e., other learners), Arab students looked mainly for less formal sources (i.e., other learners) and even to do so they needed to have higher self-efficacy. In addition, their preference toward interaction with their instructors and other students and are correlated indicating a general preference for interaction per se.

An interesting significant negative correlation was found between self-efficacy and structure for the American group. Students with preference toward more structure were more likely to have less self-efficacy. This is consistent with Moore's theory of transactional distance when we perceive self-efficacy as one indicator of learner autonomy. The more autonomous learners have greater capacity to structure their own learning and therefore would prefer less structure in their programs (Moore & Kearsley, 2005).

It may seem many of the results presented in this study are consistent with knowledge and understanding about cultural differences between Arab and American cultures; however, the way these cultural differences are reflected in distance education setting is not well documented or researched. Empirical evidence from this and similar studies should support decisions about instructional design and student support in distance education programs to ensure learner success across cultures. Today's technology allows for flexible structure and more interaction, yet there are variations in learner cultural preference and readiness. However, technology alone cannot lead to effectiveness and quality in education; many times in distance education more dialogue and customization of courses is required (Lemak, Shin, Reed, & Montgomery, 2005). Therefore, when designing and implementing distance education programs internationally, cultural views of relationship, hierarchy, structure, time, and other cultural variations provide insightful guidelines for such customization and modification.