Interaction Equivalency in Self-Paced Online Learning Environments

Chapter 3: Methodology

CHAPTER 3. METHODOLOGY

Statement of the Problem

Interaction has long been identified as a key element to successful online learning programs (Beldarrain, 2006; Moore, 1993). While not the sole indicator of high-quality and effective online education programs, there is significant evidence to suggest that meaningful interaction with other students and the instructor is integral to the development of thriving learning environments (R. E. Brown, 2001; Garrison & Cleveland-Innes, 2005; H. C. Greene, 2005; Joyce Lee et al., 2006; Swan, 2002). Insufficient or ineffective interaction may lead to student isolation, while exorbitant levels may lead to overload or frustration (Berge, 1999; Willging & Johnson, 2004). Yet, little empirical evidence currently exists as to the value that learners place upon the various types of interactions in a self-paced learning environment.

Purpose of the Study

The purpose of this study was to expand upon previous research advocating for the purposeful design of interaction within the online learning experience (C.-W. Chang, 2006; Hirumi, 2002). This study builds upon existent knowledge of online education by specifically examining the composition of the online learning experience of adult learners in a hybrid self-paced learning environment that includes numerous interactive components and instructor facilitation designed to foster an engaging and meaningful learning experience. The goal of this study was therefore to extend the bounds of previous research into interaction within the online learning experience by examining what forms of interaction learners in a self-paced online course value most as well as what impact they perceive interaction to have on their overall online learning experience.

Research Questions

This exploratory study sought to examine the experiences and preferences of adult learners concerning the various interactions that they encounter in a self-paced online course. The following four primary research questions guided data collection and analysis efforts:

1. What forms of interaction do adult learners engage in most in self-paced online courses?
2. What forms of interaction do adult learners value most in self-paced online courses?
3. What forms of interaction do adult learners identify as equivalent in self-paced online courses?
4. What impact do adult learners perceive interaction to have on their self-paced online learning experience?

Research Design

Building upon previous research in human-human interaction in online learning environments, this mixed methods study utilizing a concurrent data collection methodology documented the experiences of adult learners participating in a self-paced online course employing various levels of interpersonal interaction. A brief overview of mixed methods research is warranted in justifying why the selected approach was ideal for this study.

Mixed methods research has been operationally defined in chapter 2 as, “collecting and analyzing quantitative and qualitative data within either a single study or multiple studies.” Mixing such qualitative and quantitative datasets provides richer insight into the phenomenon than if either dataset were to be used alone and provides strengths that offset the weaknesses inherent in each sole approach (Creswell & Plano Clark, 2007; Jick, 1979, December). Rather than being limited to a single ideology, the researcher is free to utilize all possible methods to explore a research problem.

While researchers have been collecting both quantitative and qualitative data in the same studies for years, the notion and acceptability of mixing different data types in the same study to result in a distinct research methodology has emerged only within the past decade (Creswell & Plano Clark, 2007). Disagreements still exist among scholars concerning many fundamental tenets related to the field of mixed methods research, such as: nomenclature and basic definitions used, design issues, issues in drawing inferences, and logistics of conducting mixed methods research (Teddlie & Tashakkori, 2003). Yet, mixed methods research has grown from a subset of traditional quantitative and qualitative approaches to be recognized as a viable and effective methodology that stands alone from other research frameworks (Creswell & Plano Clark, 2007; Tashakkori & Teddlie, 2003). It is a research approach that can bridge the rift that still exists in some venues between quantitative and qualitative research (Onwuegbuzie & Leech, 2005).

The evolution of mixed methods research parallels the following development movements of qualitative research that took place during the twentieth century, characterized by Denzin and Lincoln (1994): the traditional (1950-1970), blurred genres (1970-1986), the crisis of representation (1986-1990), and postmodern or present movements (1990-present). Tashakkori and Teddlie (2003) note the similarities between the development of qualitative research and mixed methods research methodologies, pointing out that a substantial amount of important mixed methods took place as early as the time period from 1900-1950. These early examples of mixed methods research predominantly involved extensive use of interview and observation protocols. Tashakkori and Teddlie further note that the traditional period (1950-1970) was characterized by efforts aimed at debunking positivism while also encouraging multimethod designs. As such approaches gained traction among the research communication and increasing numbers of studies utilizing a mix of methods appeared, the debate surrounding the merits of such studies grew (Tashakkori & Teddlie, 1998). The so-called “paradigm wars” resulted in mixed methods approaches being openly received as an opportunity to discredit strictly positivistic mindsets that had previously dominated for centuries. As mixed methods research has become increasingly popular and has been rigorously scrutinized, it now is considered to be a separate methodology that can leverage the strengths of both the qualitative and quantitative approaches (J. C. Greene, 2008; R. B. Johnson & Onwuegbuzie, 2002).

Numerous reasons exist as to why researchers choose to employ mixed methods in research. As previously mentioned, one primary reason is to leverage both qualitative and quantitative data in a single research study. Such methods allow for the measurement of, “overlapping but also different facts of a phenomenon, yielding an enriched, elaborated understanding of that phenomenon” (J. C. Greene, Caracelli, & Graham, 1989, p. 258). Green et .al use the analogy of “peeling the layers of an onion” when characterizing this application of mixed methods research when seeking to learn of the different levels of a phenomenon. While a single approach may only yield data at the surface level, utilization of several alternative methods can yield a much deeper level of understanding (McMillan & Schumacher, 2006). This expansion of knowledge from one method to another is a definite strength and often leads to united or substantiated findings from different data sources. Some have gone as far as to claim that the methodological pluralism indicative of mixed methods research frequently results in superior research as compared to monomethod research (R. B. Johnson & Onwuegbuzie, 2002).

The growth of interest in mixed methods research is evident through the vast number of books, journal articles, and funded research projects available today utilizing the mixed methods model of research. In fact, entire journals, such as the Journal of Mixed Methods Research, are now devoted entirely to scholarly dialogue and dissemination of mixed methods research. Examples are not limited to any single discipline, but rather are pervasive among virtually all social science research (Tashakkori & Creswell, 2008). Many previous studies in the literature illustrate the utility of mixed methods approaches to addressing research questions particularly in the field of online education (Franklin, Peat, Lewis, & Sims, 2001; Mehanna, 2004; Oladoke, 2006; Swan & Shih, 2005).

As with any research methodology, inherent challenges exist with mixed methods research that researchers must consider. For one, mixed methods research is complex and involves exploring differing approaches to investigating a research problem. Additional time and resources are often necessary to collect and analyze both quantitative and qualitiative data (Lodico, Spaulding, & Voegtle, 2006). Research procedures are usually more complicated and require additional effort to provide clear presentation of findings. These are all aspects one must consider when deciding on the appropriate research design.

As many research questions can be most effectively answered through mixed research solutions (R. B. Johnson & Onwuegbuzie, 2002), the selection of the mixed methods framework for this research study incorporating a concurrent data collection method was based upon the value of resulting first-hand accounts from participants of their experiences and perceptions of interaction within a unique online learning environment. The resulting data from the mix of quantitative and qualitative interview questions included a combination of rich narrative accounts along with numeric frequencies of engagement. The mixed methods approach was ideally suited for addressing the proposed research questions, as such an approach has been utilized successfully by previous researchers to glean important insights from participants of the dynamics of novel online education initiatives (Mehanna, 2004; Oladoke, 2006; Swan & Shih, 2005). Data was collected in the form of semi-structured, in-depth interviews conducted near the conclusion of the course to understand from the perspective of learners the perceived value of various modes of interaction as well as interaction experiences and preferences. Such rich learner perspectives provided an enlarged and deepened range of immediate experience, adding great value to the body of knowledge by forming a solid foundation for future inductive or empirical studies.

Sechrest and Sidani (1995, p. 78) note that both quantitative and qualitative methodologies, “describe their data, construct explanatory arguments from their data, and speculate about why the outcomes they observed happened as they did.” While this mixed methods study was predominantly qualitative in nature and the majority of the interview questions yielded rich narrative responses from participants regarding their individual experiences, the frequencies of engagement in the different types of interactions provided great insights into the overall experiences and ultimately the value that participants placed upon the various types of interaction that they experienced.

Sampling Design

Participants

The participants for this study were online adult learners enrolled in a fully-online professional development certificate program for children’s ministry professionals and laity offered by a private, higher education institution located in the northeastern part of the United States. This study was specifically limited in scope to investigating the interaction experiences and preferences of self-paced online learners in order to provide in-depth data concerning the distinctive aspects of interaction within this unique learning ethos. The one-year online certificate program in Children’s Ministry (CMUO) offered by Valley Forge Christian College (VFCC) employs such a self-paced, emergent and flexible design and was selected as an appropriate context in which to explore the dynamics surrounding the interaction preferences of adult online learners. VFCC Academic Affairs granted permission for the study to be conducted with the desired VFCC student body and pledged to provide the necessary access to the selected group of adult learners.

Sample Course Characteristics

The study sample included all learners successfully completing either of two concurrent sections of an undergraduate certificate program course offered by VFCC. This course utilized a unique self-paced format whereby a new cohort of learners began the course with an instructor each month. Participants were afforded the freedom to proceed through the course materials and activities at an individualized pace with the only stipulated deadline being an end-of-course deadline. The researcher happened to also serve as the course instructor for this course and therefore was privy to the intimate details of the wide range of course design, development, and facilitation processes. Explicitly detailed research design and methods were therefore essential to combat the existent opportunity for bias as well as threats to reliability and validity.

The course selected, while constructed within the Blackboard learning management system (LMS) and incorporating a variety of asynchronous computer-mediated communication tools standard in the LMS, also included integration of a personal learning landscape consisting of an assortment of various social networking features such as blogging, tagging of resources, e-portfolios, and learner-driven communities. This social learning landscape entitled “CMUOnet” and branded likewise for consistency with the VFCC certificate program, was powered by the open source social networking platform Elgg and available for learners to access at http://cmuonet.org, as shown in Figure 4.

Figure 4. Elgg Instance – CMUOnet (http://cmuonet.org)

CMUOnet is a learning landscape consisting of a fully featured electronic portfolio, Weblog and social networking system, connecting CMUO learners and creating communities of learning. It allows CMUO learners and instructors to connect both socially and academically. CMUOnet exists to promote collaboration and reflection in an academic and social setting. CMUO learners are able to meet fellow students, either online or face-to-face, without invading their privacy. While communications taking place within the framework of a given CMUO course only last for the duration of the course, learners can use CMUOnet to collaborate and build ministry networks that will last and be accessible long after the course(s) the learner is enrolled in have ended.

Incorporating features of both the LMS and learning landscape, course activities and the resulting interactions were designed to integrate a full complement of asynchronous communication and social networking tools available today in creating a socially-constructed self-paced learning environment. This model displayed in Figure 5 serves as one emerging approach to adult online learning that embraces a socially-constructed framework (Rhode, 2006).

Figure 5. Anatomy of a socially constructed self-paced learning environment

Such a learning environment involves the creation of a personalized learning environment (PLE) to incorporate a cohort-based approach whereby learners could begin any given course at a designated regular timeframe (e.g., monthly, quarterly, weekly, etc.). The learner is placed into a cohort (i.e., class) with other learners beginning the course at that given time and at least one instructor. This small group comprises a restricted Course Community (CC) in which the members of the community are able to interact throughout the duration of the course. For example, during the first week of the course, the instructor could facilitate some introduction and ice-breaker activities in which the learners and instructor(s) could begin to connect. As the course progresses, learners begin to advance at individual rates and may choose to remain connected closely with the cohort or may seek support from other communities (either online or face-to-face).

While self-paced learners should have continual access to the CC throughout the course, a true socially-constructed environment should allow for learners the freedom to join interest/subject matter communities to individually construct the learning networks deemed most beneficial. These self-selected and formed communities should provide learners with the opportunity to create, communicate, and collaborate within a context of ubiquitous connectivity (ie: community members should remain connected throughout the duration of the course). Whereas traditional online learning environments built within a learning management system (LMS) are available to learners only for the duration of the course, a socially-constructed self-paced learning environment (SCSPLE) should allow for access to designated communities after the course is completed.

Figure 5 is an initial attempt to illustrate the complexities of SCSPLE’s. Individuals (I) within course communities (CC) have access to both the restricted CC but also to numerous learner-initiated interest communities. An individual can belong to numerous course and interest communities simultaneously and can join or leave these communities at any time. If learners do not find a community for the desired topic/interest, they can create a new community and invite others to join. Each CC is comprised of a specific and defined population (ie: the learners and instructor(s) for the course) while each interest community (IC) is dynamic, allowing for the members to freely join or leave at any time and engage in a wide assortment of formal and informal interpersonal interactions. The course serving as the self-paced learning environment that was explored in this study incorporated this SCSPLE model for online learning.

This course was also a unique learning environment selection for this study in that it mandated the use of both the Blackboard LMS and CMUOnet by each of the learners as part of the required learning activities. This made possible a unique discovery of learner preferences for different types of interactions fostered by these varying suites of online communication tools. Such emergent collaborations included blogging, podcasting, social bookmarking, and informal search for additional supplementary resources. In addition, the course sections selected were equally unique, as one section had only a single learner as compared to fourteen learners in the other. One might expect that such a disparity of course sizes would adversely affect results as a course scenario with a lone learner would be limited and therefore eliminate all possible interpersonal interactions beyond one-on-one contact with the course instructor. However, due to the integration of a course community and course blog that included learners in multiple sections of the same course, the lone learner in the August 2007 course section had opportunity to interact with others outside of the formal learning space in Blackboard. Therefore, unique comparisons and contrasts were gleaned between the perspective of the learner in a course section where formal interaction with other learners was limited as compared to the perspectives of learners in a simultaneous course section where formal interactions with other learners was much more feasible.

Participant Selection

Adult online learners having currently enrolled in at least one self-paced online course in the online children’s ministry certificate program of Valley Forge Christian College (VFCC) served as the convenience sample for this study. A common approach within education research, the convenience sampling technique, is often utilized when the characteristics of a specific group of individuals matches the attributes of the phenomenon being studied (McMillan & Schumacher, 2006). The study sample included all learners enrolled in two concurrent sections of a VFCC undergraduate certificate program course. The sample size was restricted to this select group of learners as they were the most current cohort of learners to be actively enrolled in a self-paced online course at the institution at the time of the study and had the freshest memory of their experience, considered vital by researchers who claim that the farther removed a participant is from the direct experience, the more problematic recall becomes (Fink & Kosecoff, 1998). The course enrollment for the August 2007 and September 2007 sections of the course totaled fifteen learners, with one learner enrolled in the August 2007 section and fourteen learners enrolled in the September 2007 section. Of the fifteen learners who began in either of the selected course sections, only eleven successfully completed at least three of possible four units of the course to qualify to participate. Invitations to participate were sent via e-mail to the eleven learners who successfully completed 75% or more of the course, of which 10 responded and agreed to participate in the study. The lone learner from the August section was among those who successfully completed the course and agreed to participate. Therefore, the resulting study sample size study was 10 (n=10), a remarkable response rate of 91%.

Participant Demographics

Of the 10 participants in this study, 7 were female and 3 were male. The age range of participants varied from 18-25 to 56-65, representing a truly diverse cross-section of adult learners. The majority of the participants (n=6) were in the 36-45 age range which also happens to represent the largest age group of all CMUO learners. Table 1 displays the demographic characteristics gathered regarding the participants.

Table 1

Two participants confirmed that this course was the first online course with CMUO that they had completed, while the remaining eight learners noted that they had previously enrolled in a self-paced CMUO course. Table 2 depicts the range of previous online learning experiences as noted by the participants.

Table 2

The context of the participants’ experience is certainly an important factor worth noting. When participants were asked how many courses they were taking presently, responses were split exactly in half, with 50% currently taking just the single course while the remaining 50% confirming that they were simultaneously enrolled in another course.

No further demographic or background data was reviewed. Interview questions were largely qualitative in nature and addressed various aspects of the types of interactions built into the course and participants’ preferences for each. Chapter 4 provides an overview of the rich experiences of participants that was captured. While the resulting data was primarily qualitative in nature consisting of rich narrative experiences, graphical representations of frequency are provided where appropriate to further illustrate the quantitative responses received.

Time Frame

The nature of the self-paced online learning experience raises a unique set of variables for consideration. New CMUO course sections begin monthly, allowing learners to commence and conclude their learning experiences when most convenient. While courses have specific starting and ending dates, the self-paced nature of the course learning activities further individualizes the overall learning experience, as learners are able to progress through the course at an individualized pace. Therefore, the experiences of the collective body of learners enrolled within the program are unique to each individual, as learners are routinely at different stages of course completion. In order to collect data that most accurately reflects the overall experiences and perspectives of learners at the time of study, it was conducted with learners enrolled in the August 2007 and September 2007 sections of the course near the end of their learning experience. Such summative reflections from learners provided valuable insights into the overall online learning experience and the perceived equivalency of various interactive attributes of VFCC’s self-paced online learning program.

This study was conducted during a six-week time period during the fall of 2007. The course studied was self-paced, affording learners the flexibility to complete the course in as short as 1 week or up to a full 8 weeks. Therefore, all learners enrolled in the selected course were contacted via e-mail once they had begun the final unit of the course and therefore finished at least 75% of the course and were invited to participate in the interview. Reminder e-mails were sent at 1-week and 2-week time periods for those who did not initially respond to the e-mail invitations. With each interview lasting approximately one hour, interviews were scheduled over a 4-week time period. Consistent with recommendations from the literature concerning transcription and data analysis, an estimated five hours was allotted for transcription of each one hour of interview. Actual transcription time frames remained consistent with estimates provided in the literature. Therefore, a goal of conducting 3 interviews per week was set and successfully adhered to.

Measures

A convenience sample of learners having recently completed a self-paced online course was selected to participate in in-depth interviews to share first-hand accounts of their interaction experiences and preferences in an effort to glean a holistic understanding of the dynamics of and learner preferences for interaction within less structured online learning environments. Such rich learner perspectives provided an enlarged and deepened range of immediate experience, adding great value to the body of knowledge by forming a solid foundation for future inductive or empirical studies. Transcripts of the interviews with learners were analyzed and coded for emerging themes. Self-reported frequencies of engagement in the various interactive components of the course were collected from participants during the interviews and analyzed as well. The resulting findings include both the collective and comprehensive perspectives of VFCC adult learners as they share their experiences and preferences for interaction within the self-paced online education environment.

The semi-structured open-ended interview method was selected for data collection as it provides a consistent and yet flexible inquiry framework (McMillan & Schumacher, 2006; Patton, 1990). At the core of this process is the desire to understand adult learners’ perceptions, perspectives, and understandings of interaction within self-paced online learning. This mixed methods approach facilitated the examination of experiences from multiple perspectives involving data in multiple forms, leading to the formation of generalizations concerning the true essence of the experience from an insider’s point-of-view (Leedy & Ormrod, 2001). Interview questions addressed the three main types of interaction described in the literature (e.g., student-student interaction, student-instructor interaction, and student-content interaction) in regards to both formal and informal learning activities and also explored the tenets of Anderson’s (2003) Interaction Equivalency Theorem. The interviews yielded rich narrative descriptions of learners’ interaction experiences, preferences, and frequencies of interaction engagement in an attempt to understand the interaction dynamics and optimal integration within an online self-paced adult learning program.

Data Collection Procedures

This study involved semi-structured open-ended interviews with each learner agreeing to participate (n=10). Permissions to conduct this study was granted by Capella University via submission of appropriate documentation to the Institutional Review Board (IRB) as well as Valley Forge Christian College Department of Academic Affairs. Since sampled learners were distributed across North America, interviews were conducted over the phone rather than in person. Interviews were comprised of semi-structured, open-ended interview questions concerning learner perceptions and preferences for interaction in the self-paced online learning environment and were conducted to gain deeper insight into learners’ experiences engaging in the various forms of interaction within the course (see Appendix A). All interviews were recorded digitally and then manually transcribed and coded as part of the data analysis process. The QuickTap telephone handset tap, in conjunction with a laptop computer and the audio recording software Audio Hijack Pro was used to record each phone interview. Interviews were then manually transcribed using the software package HyperTRANSCRIBE and then coded for emerging themes. Member checking strategies were utilized and involved sending transcripts of interviews to the corresponding interviewees to confirm accuracy prior to coding.

Each interview consisted of a total of ninety-four pre-determined questions. Interview questions were derived from a review of premier themes within interaction literature and addressed the three main types of interaction described in the literature (e.g., student-student interaction, student-instructor interaction, and student-content interaction) in regards to both formal and informal learning activities. The questions also explored the tenets of Anderson’s (2003) Interaction Equivalency Theorem. Questions were provided to all participants prior to the interview to allow learners to reflect on the questions and subsequently give more substantive responses. At times throughout the interview, additional follow-up questions were asked to help clarify or expanded upon responses. The complete list of interview questions is found in Appendix A.

Chapter 4 includes a summary of the process and substance of data collection efforts followed by analysis of the research questions in relation to study interview findings. Where appropriate, responses from participants are included to support conclusions reached from the data. Pseudonyms are used in place of the actual names of participants to protect their identities.

Pilot Testing

The necessity for pilot testing any research protocol or instruments prior to distribution or deployment within a research study is widely recognized and stressed throughout the literature (Leedy & Ormrod, 2001; McMillan & Schumacher, 2006). Such pilot testing practices are helpful in identifying needed changes to an instrument or set of instructions and provide the researcher with not only an estimate of the amount of time it will take for the data collection intervention but also an initial view of the pattern of likely responses. Interview instructions and questions were pilot tested with an adult learner representative of the target learner population and who completed at least one course within the online program. However, this learner’s participation did not fall within the target range and therefore was not eligible to be officially included in the study. The field test included a review of instrument clarification, technological presentation, recording and transcription protocols, and data analysis testing. Specific feedback concerning improvements or clarifications was documented and a few minor adjustments were subsequently made prior to implementation.

Ethical Issues

The ethical veracity of any study is of utmost importance, as the protection of participants is among the most important characteristics of reputable and reliable research. Every effort was made throughout this study to ensure that the privacy of each participant was protected. All eligible learners enrolled in the selected course sections were contacted via e-mail and invited to participate in the study. Those agreeing to participate in the interview were asked to return to the researcher a signed copy of the informed consent form and to contact the researcher either via the e-mail or phone contact information provided to set up a mutually convenient time to conduct the interview via phone. Participants were informed that interviews were recorded confidentially for the purposes of the study only. Once transcribed, the resulting interview transcription was sent via e-mail to the participant to ensure accuracy. Interview data, including recorded audio files and transcribed text data, was securely stored electronically using assigned identification codes in place of any participant names or other identification information. Pseudonyms were used in place of participants’ actual names in the data analysis and results to ensure that identities of the participants are protected.

Data Analysis Procedures

At the core of many mixed methods research approaches is the quest for collecting and analyzing a mix of quantitative and qualitative data within a single study in order to grasp the truest sense of participants’ first-hand experience of phenomena. In fact, it has been argued that such approaches to empirical inquiry affording methodological pluralism frequently result in superior research as compared to monomethod research (R. B. Johnson & Onwuegbuzie, 2002). Consistent with inductive research, data analysis efforts involved recognizing categories of description as a primary outcome of the research activities (Marton, 1988). In a more simplistic sense, the goal of such a study is ultimately to take the perspectives of participants experiencing the phenomenon first-hand and categorize such individual experiences into collective groupings where conclusions can then be drawn from. To that end, the data analysis process involved identifying emergent themes from the data that will serve as foundational schema for further data organization and analysis.

While there is no single approach for coding all qualitative data, certain techniques have been shown effective in organizing and classifying data for further analysis. In particular, the coding scheme that includes selection procedure based upon criteria of relevance provides a solid foundation for organizing, coding, and categorizing data (Charles & Mertler, 2002; McMillan & Schumacher, 2006). Transcripts of learner interviews were reviewed and coded to determine emergent themes. Responses were grouped by question as part of the initial analysis process to aid in the comparison of responses to similar questions. As additional patterns and categories were identified, successive codes emerged that were utilized to further describe the data. The qualitative data analysis software HyperRESEARCH was used to aid in the data analysis process. The triangulation of data through multiple sources, including verbatim transcripts, comparisons of qualitative responses to quantitative data from interview questions, and reviewer notes in conjunction with member checks of transcribed interviews aided in strengthening and validating findings.

Select interview questions yielded quantitative data, particularly in terms of the frequencies in which learners engaged in the various types of interactions in the course. As such resulting data is used solely to more fully understand the experiences of the sample of participants rather than making any attempt to generalize findings, statistical analysis of such numerical data was unnecessary and therefore minimized. Where appropriate, responses to questions that either utilized Likert-scale or “yes/no” questions are noted in numeric form.

Expected Findings

Based on a review of the literature concerning interpersonal interaction, it was expected that one or more types of interaction would surface as being preferred for adult learners in the self-paced online learning environment. While others have hypothesized that learners may value formal, learner-instructor interaction highest, little evidence is available in the literature to support or refute the notion of interaction equivalency among interaction modalities. And, while learners may in fact value one mode of interaction over others, it is certainly possible that they may value other types of interaction as equal.