Examining the Trust Factor in Online InstructorLed College Courses

1
Examining the Trust Factor in Online
Instructor-Led College Courses

• Shalin Hai-Jew, Ed.D.
• Kansas State University (KSU),
• Office of Mediated Education (OME)
• April 22, 2006, Seattle
• Association of American Colleges Universities
  (AACU)
• Learning and Technology Implications for
  Liberal Education and the Disciplines

2
Introduction Online Learning

• Online learners meet for 10 weeks in an
  academic quarter in high-interactive
  instructor-led online college classrooms. They
  interact with each other as virtual peers (on
  virtual teams) through Web-based courseware.
  Andragogy and constructivism assume
  inter-relationships as bases for adult learning.

3
Introduction (cont.)

• Often, human interactions online are
  asynchronous. Learners and instructors interact
  through a non-human technology through mostly
  text and occasionally graphics. Theres often no
  face-to-face time. There are no body language or
  tonal cues. The emotional affect tends to be
  flat. Emotions are conveyed through words and
  emoticons. ?

4
Initial Questions

• What is the role of trust in such a virtual
  circumstance?
• How is trust (a multi-dimensional construct)
  operationalized and manifested in such online
  classrooms?
• Is trust important for effective learning in such
  an online classroom situation?
• How may trust be enhanced in this circumstance?

5
Purposes of the Study

• Define the roles of trust in the following online
  relationships
• Student to instructor (reciprocal)
• Student to student (reciprocal)
• Student to curriculum (one-way)
• Student to oversight organizations (reciprocal)
• Student to technology (one way)

6
Operationalizing Trust

• Trust (trust) n. 1. Firm reliance on the
  integrity, ability, or character of a person or
  thing. 2. Custody care. 3. Something
  committed into the care of another charge. 4.
  a. The condition and resulting obligation of
  having confidence placed in one. B. One in
  which confidence is placed. 5. Reliance on
  something in the future hope. 6. Reliance on
  the intention and ability of a purchaser to pay
  for in the future credit.

7
Theoretical Underpinnings Trust and Distance
Learning / eLearning(Literature Review)

• Trust is an essential part of human relations and
  cooperation as a key enabler. Trust allows for
  risk-taking, a necessary component in learning.
  Trust enables people to make changes. Trust
  promotes sharing and mutual learning, under the
  relational constructivist model.

8
Theoretical Underpinnings Trust and DL
(cont.)

• Trust appears as a construct in law, psychology,
  sociology, political science, economics,
  business, religion studies, philosophy,
  anthropology, history, computer science,
  sociobiology, organizational development,
  education, management literature, and others. It
  has been studied in interdisciplinary ways, too.
  

9
Theoretical Underpinnings of Trust and DL (cont.)

• Trust has a moral dimensionof ethically
  justifiable behavior as expectations. General
  trusters tend to have a higher ethical
  sensibility (Uslaner, 2000 01, p. 579)
• Swift trust tends to be unstable, fragile.
  High initial trust has been linked to mitigating
  the effects of later disappointment (as a
  positive construct).
• Three-Part Relation to Trust (1) truster
  properties (2) person being trusted, (3)
  specific context (Hardin, 1992, as cited by
  Kramer, 1999, p. 574)

10
Ways to Conceptualize Trust

• Individual attribute
• Behavior
• Situational feature
• Institutional arrangement (Sitkin and Roth, 1993,
  as cited by Bigley and Pearce, July 1998, p. 405)
  
• Trust is seen as a cultural construct.
• Trust represents choice behavior. It is based
  on cognition and emotion.

11
Three Types of Trust

• Calculus-based trustbased on mutual calculations
  on what each party may gain from the other (also
  deterrence-from-risk based trust, transactional
  trust)
• Knowledge-based trustbased on mutual
  interactions and experiences (also
  cognition-based trust)
• Identification-based trustbased on similarity
  and agreement between individuals, shared values,
  also affective-based trust (Dibben, Harris and
  Wheeler, Apr. 2003, p. 6 Lewis Weigert, 1985,
  p. 970)

12
(Another) Three Types of Trust

• Personal trusthonesty, ethics, follow-through,
  intentions, handling of confidential information,
  straightforwardness
• Expertise trusta persons standing in his/her
  field, datedness of knowledge, credible
  information use to support ideas, application of
  expertise to situations
• Structural trustbased on a persons role and
  responsibilities (Joni, March 2004, pp. 84 85)
  

13
Factors that Lead to Mistrust

• Ambiguity
• Caution
• Deceit
• Editing or screening
• Limiting channels
• Secrecy
• Indirection
• Gimmicks

• Hostile humor
• Lack of emotion (Harvey, 1983, as cited by
  Fairholm, 1994, p. 139)
• Absence of faith in others (Mirowsky and Ross,
  1983, as cited by Ross, Mirowsky Pribesh, Aug.
  2001, p. 568)
• Not a necessarily negative valence (Lewicki,
  Mcallister, Bies, 1998, p. 455)

14
Current DL Challenges

• Whole student learning
• Retention
• Academic dishonesty
• Technological challenges, standardization
• Personalization (vs. standardization, automation,
  AI, simulations, boxed courses)

15
Statement of the Problem

• The virtual aspects of high-interactive,
  instructor-led online learning may impede or
  preclude the building of trust between
  individuals. Trust is a crucial social glue that
  allows people to take risks and to build
  learning, make changes, essential components of
  constructivism, which is said to guide the
  andragogy of online learning.

16
Five Research Questions

• How is trust manifested in an online classroom?
• What does a high-trust online learning classroom
  and community look like?
• What factors contribute to trust or mistrust,
  and how are these elements related?

17
Five Research Questions (cont.)

• How can trust as an asset be protected and
  leveraged in a virtual learning environment?
• Is there a relationship between high-trust and
  the effectiveness of student online learning (as
  measured by the proxies of student
  retention/persistence, course grades, and student
  perceptions)?

18
Research Methodology

• Literature review (Trust, DL, virtual teaming)
• Initial informal survey of online learners
• Creation of Online Trust Student Survey (OTSS)
  using Likert-type measures of both the importance
  of the item and student experienced measure of
  that item in their online classroom

19
Research Methodology (cont.)

• Pre-testing of OTSS survey on subgroup (DL
  students, faculty and administrators) for
  construct validity survey revision
• Online launch of survey for quantitative analysis
  (N 630) factoral analysis, comparisons
  between means, ANOVA and MANOVA of descriptive
  factors with output component variables,
  regressions, and correlations
• Collection of post-survey interview data for
  qualitative analysis from online learners (both
  high-trust and low-trust), online instructors and
  distance learning administrators

20
Null Hypothesis

• There is no correlation at the p lt .05 level
  between learners trust level in an online
  instructor-led classroom and their effective
  learning.

21
Population and Sample

• 630 WAOL learners per quarter for freshman and
  sophomore-level courses
• Fully online learning via Blackboard courseware
  technologies
• Random sample from online learners who opt-in to
  take part in a 20-minute online survey in Winter
  2005 via Perennial Survey

22
WashingtonOnline Virtual Campus (WAOL-VC)

• WashingtonOnline Virtual Campus represents a
  consortium of the 34 community colleges of
  Washington State. It offers some 21,000 FTEs of
  credits annually to learners from around the U.S.
  and the world.
• The courses are built by Washington state
  community colleges instructors, who work as teams
  (lead instructor and two supporting instructors)
  to develop courses. The lead instructor creates
  the courses, and other instructors may teach
  them.

23
Perennial Survey Screen Shot
24
Survey Instrument (OTSS)

• There were nine categories of trust factors
• (47 items)
• Individual trust propensities
• Communications
• Instructor
• Organizations
• Peer-to-peer relations

• Policy macro-structure
• Student empowerment
• Curriculum
• Technologies

25
Post-Survey DL Administrator Interview

• How do you influence how instructors teach in the
  program?
• How do you influence the online curriculum?
• How important is trust between a student and an
  instructor in an online learning environment?
  Why?
• How important is trust between a student and
  other students in an online learning environment?
  Why?
• How important is trust between a student and the
  curriculum in an online learning environment?
  Why?
• How important is trust between a student and
  courseware technologies in an online learning
  environment? Why?
• What aspects of leadership in administration
  contribute to learner trust?

26
Post-survey Online Instructor Interview

• Is trust an important factor in successful online
  learning? If so, how? If not, why not?
• How important is trust between a college student
  and instructor in an online learning environment?
  Why? How do you see this trust manifested?
• How important is trust between college students
  (peers) in an online learning environment? Why?
  How do you see this trust manifested?
• How important is trust between student and
  curriculum in an online learning environment?
  Why? How do you see this trust manifested?

27
Online Instructor Interview (cont.)

• How important is trust between student and
  courseware technologies in an online learning
  environment? Why? How do you see this trust
  manifested?
• Is there a certain time when trust solidifies
  in an online classroom? If so, when? If never,
  why?
• What aspects of the online classroom contribute
  to building trust?
• What aspects of the online classroom contribute
  to creating distrust?
• In a case of mistrust, how can a class
  reestablish trust?

28
Post-Survey Online Student Interview (High-trust
group, Low-trust group)

• What personality indicators do you use to know
  whether or not to trust an instructor?
• How can an instructor come across as real in an
  online space? Please give some from-life
  examples.
• Do you consciously build others (students and
  instructors) trust in you when you participate
  in an online class? If so, how? If not, why not?
  
• Have you ever felt like your trust was violated
  in an online class by an instructor? Please
  explain what happened. Please share as many
  experiences as possible.
• Have you ever felt like your trust was violated
  in an online class by a fellow student? Please
  explain what happened. Please share as many
  experiences as possible.

29
Descriptors of Survey Respondents(Frequency and
Percentage Distribution)

• Frequency Percentage
• OTSS Respondents
• 630 100
• Gender
• Female 520 83
• Male 110 16
• Year in College
• Freshman 174 27
• Sophomore298 47
• Junior 90 14
• Senior 22 3
• Fifth Year46 7
• Age Range
• 15 19 103 16
• 20 29 266 42
• 30 39 125 19
• 40 49 102 16
• 50 59 32 4
• 60 69 2 - -

30
Descriptors of Respondents (cont.)

• Cumulative GPAs As (50), Bs (45), Cs (3),
  Ds (0 but there were a few in number), and Fs
  (0 with none)
• Racial Breakdown 84 Caucasian, 3 Hispanic
  American, 2 Asian American, 1 African American,
  1 Native American, 5 as other, 1 Unknown

31
Descriptors of Respondents (cont.)

• Reasons for Taking an Online Course
• Academic schedule (24)
• Convenience (23)
• Work (18)
• Family (13)
• Commute (9)
• Other (4)
• Health, academic advisor suggestion, course
  reputation, and instructor reputation (1 each)

32
Attitudes Towards Online Learning at the
Beginning of a Course

• Positive expectations (39)
• Enthusiastic (16)
• Neutral attitude (25)
• Negative (2)
• Skeptical (16)

33
Prior Experiences with Online Courses

• 1-5 prior online courses (50)
• 0 prior online courses (37)
• 6-10 prior online courses (8)
• 11 15 prior online courses (1)
• 16 20 prior online courses (1)
• 66 was said the prior online learning was
  effective 30 had mixed results, and 4 found
  the online learning ineffective

34
Familiarity with Subject Matter of Analyzed WAOL
Course

• 47 had no prior experience from either high
  school or college in the subject matter of the
  course about which they were describing
• 21 had had a quarters worth
• 13 had had two quarters of experience
• 8 had an academic years worth of experience
• 2 had four quarters worth
• 1 had five quarters worth of experience
• 2 had six quarters worth
• 2 had 7 quarters or more of prior experience

35
Factor Analysis ResultsLevel of Importance to
Online Learning

• PROSOLIDLevel of professionalism of oversight
  organizations, solidity of the curriculum
• AUTHENAuthenticity of learning, instructor
  supportiveness
• INSPRESInstructor ethics, presence,
  boundary-setting
• PEERINTPeer interactions, full expressiveness
• PROBRESTimely resolution of learner problems

36
Factor Analysis ResultsLevel of Agreement with
the Student as a Learner

• INSEFFETInstructor effectiveness
• STRUINTEStructural integrity of overseeing
  organizations
• TECHNORETechnological responsiveness and
  stability
• STUDEMPOStudent empowerment
• INFOVALIInformational validity
• SOCLIFESocial life of online learners
• REALHONReality in simulations, honesty in
  co-learning, real-world learning

37
Survey Respondent Descriptors and Variables
(ANOVA and MANOVA)

• For the MANOVAs, the year in college showed a
  high frequency on the TECHNORE (technological
  responsiveness) factor (F 3.158, p .014).
• The age descriptor connected with STRUINTE
  (structural integrity) with an F 3.273, p
  .006. Gender and age interacted for a
  statistically significant F 6.312 and p .000
  with STRUINTE as well.
• No other statistically significant issues were
  surfaced through the MANOVA between these
  descriptor variables and these four factors based
  on learner responses to the OTSS 47 variables
  about their online learning experiences related
  to trust.

38
Survey Respondent Descriptors and Variables
(ANOVA and MANOVA)

• In terms of gender differences, in a test between
  subjects, high Fs existed for STRUINTE
  (structural integrity) (F 21.437, with a p
  .000). STUDEMP (student empowerment) also showed
  a significant difference (F 10.565, p .001).
  Lesser differences were observed for INSEFFECT
  (instructor effectiveness) with F 8.787, p
  .003, and TECHNORE (technological responsiveness)
  with F 6.685, p .010). This said, the
  statistical imbalance between females to males (N
  520 to N 110) should be considered.

39
Other Highlights from Findings

• INDIVIDUAL TRUST PROPENSITIES Learners
  identified their own trust propensities as the
  most salient of the three factors here and their
  sense of self-motivation and focus next, followed
  by the (reverse-phrased) threat to sense of
  well-being as not important.

40
Main Highlights from Findings

• COMMUNICATIONS Communications with the
  instructor are critical to learners in terms of
  their perception of online trust. The
  instructors responses need to be appropriate.
  His/her sense of ethics has to be strongly
  expressed, without any apparent conflict of
  interest and imbued by a sense of good will and
  flexibility.

41
Main Highlights from Findings (cont.)

• INSTRUCTOR The instructor needs to follow
  his/her official role, foremost. Of second
  importance is his/her respect for learner
  privacy, then instructor enthusiasm and then
  professional credentials. The least important
  aspect was that of extra-role behavior offered by
  instructors such as letters of recommendation,
  contacts with professionals in the field, and
  facilitation of internship opportunities.

42
Main Highlights from Findings (cont.)

• ORGANIZATIONS Learners find trust of their home
  institutions professionalism is more critical
  than their view of WAOL or their particular
  academic field.
• PEER-TO-PEER RELATIONS Learner anonymity was
  defined as the most important factor followed by
  the perception of the need to learn from peers,
  the amount of planned interactivity in the online
  classroom, and the encouragement of all peers to
  participate. Having shared values with peers was
  deemed the least important.

43
Main Highlights from Findings (cont.)

• POLICY MACRO-STRUCTURE WAOL-VC respondents
  identified the close adherence to stated policy
  as the most critical, with the timeliness of
  instructors posting guidelines as the next most
  important. Having access to the classroom before
  the quarter started, following a routine and
  having accurate academic advising about online
  courses seemed to be less critical to
  respondents.

44
Main Highlights from Findings (cont.)

• STUDENT EMPOWERMENT Linking grades to actual
  learning was a critical factor in student
  empowerment. Learners also expressed the
  importance of instructor encouragement of
  learners to be proactive. Instructor control
  over student messages and whether learners had
  control to make changes to the learning in the
  online classroom both seemed less salient.

45
Main Highlights from Findings (cont.)

• CURRICULUM This category had many highly-ranked
  scores. The most critical variable was the need
  to have complete lectures and course materials.
  Having real simulations online was important as
  well as having responsive handling of learning
  problems. Having clear directions was important
  as well as knowledge that the curriculum was
  college-level material. Of lesser importance was
  the up-to-datedness of curricular materials,
  clarity that no cheating or plagiarism was
  occurring in the classroom, and the offering of
  prior student work examples for perusal.

46
Main Highlights from Findings (cont.)

• TECHNOLOGIES Respondents identified their level
  of technological trust as the most important of
  the three variables, with timely solving of
  technological issues and reliable courseware of
  high importance as well (albeit at slightly
  lesser scores based on the factorial analysis).

47
Paired Samples Statistics

• Most of the 47 factors showed higher importance
  of rating than the actual perceived ranking of
  that particular variable through paired samples
  statistics. In other words, they valued the
  factors higher than their perception of the
  presence of that factor in their actual studies.
  
• Yet, the variations were slight, with small mean
  differences.

48
Q1 How is trust manifested in an online
classroom?

• For 47 of respondents, trust develops by the
  middle of the course.
• 41 suggest that trust exists from the beginning
  as a given, a form of swift trust.
• 8 suggest that trust never develops.
• 2 suggest that trust develops at the conclusion
  of the course.

49
Q1 How is trust manifested in an online
classroom? (cont.)

• Trust manifests in timely communications mutual
  respect among learners high ethics, fairness,
  grading transparency and professionalism of the
  instructor integrity in educational
  institutions sincere, substantive, and timely
  postings by peers clear and enforced policies
  proactive and empowered learners appropriate
  comprehensive college-level curriculum and
  accurate online simulations, and stable
  technologies with ready 24/7 support.

50
Q2 What does a high-trust online learning
classroom and community look like?

• 89 of respondents ranked in the high-trust
  category (defined as those with scores of
  5Somewhat agreeand above). The mean score
  was 6.0155 for this high-trust group.
• 10.5 of respondents ranked in the low-trust
  category (defined as those with scores of
  below-5Neutral and below). The mean score was
  4.4460 for this low-trust group.
• Both had low Adjusted R-Squares in regressions
  but significance in ANOVAs.

51
Q3 What factors contribute to trust or
mistrust, and how are these elements related?

• To address potential issues of multicollinearity
  between these variables, a Pearson Product Moment
  (PPM) Correlation procedure was done to see if
  any of the variables had a correlation of 0.70
  or higher. Based on the PPM, the highest
  potential correlation was between the
  completeness of course materials and lectures
  (IM40ACompleteLectures) with the assigning of
  fair grades (IM39AFairGrdes) with a moderate .689
  Pearson Correlation. The second highest
  potential correlation was a moderate .616 between
  IM40ACompleteLectures and IM37AClearWorkAssign
  .

52
Q4 How can trust as an asset be protected and
leveraged in a virtual learning environment?

• Structure mutual dependencies between learners in
  online classes.
• Have a clear instructor telepresence early in the
  class and throughout the quarter.
• Maintain high ethics by the instructor and
  oversight organizations.
• Protect learner privacy.
• Surface the issue of trust early in the quarter
  as a learner issue. Harness the early trust.
• Maintain high quality curriculum.

53
Q4 How can trust as an asset be protected and
leveraged in a virtual learning environment?
(cont.)

• Post messages regularly.
• Support the posting of substantive and sincere
  peers messages and interactions.
• Maintain consistency in teaching.
• Keep transparency and fairness of grading.
• Have an organizational presence in online
  classes, particularly for the local colleges.
• Support and adhere to stated policies.
• Strengthen learner empowerment via access to
  information, encouragement and nurturing, and
  course decision-making.
• Maintain reliable technologies, with 24/7 learner
  support.

54
Q5 Is there a relationship between high-trust
and the effectiveness of student online learning
(as measured by the proxies of student
retention/persistence, course grades, and student
perceptions)?

• The null hypothesis suggests that theres no
  linear relationship between the variables of the
  online Trust Number and that of Student Success.
  This simple regression with a p lt .05 contrasts
  with the adjusted R Squared (used because it
  adjusts for the inflation found in the R Squared)
  shows that there are statistical grounds for
  possibly rejecting the null hypotheses with a p lt
  .001 according to the ANOVA table. With a high F
  score of 50.620, these statistical results may be
  significant in showing a low positive correlation
  between online trust and student success.

55
Q5 Is there a relationship between high-trust
and the effectiveness of student online learning
(as measured by the proxies of student
retention/persistence, course grades, and student
perceptions)? (cont.)

• The low Adjusted R Square suggests that the
  variation in the independent variable of the
  online Trust Number measure accounts for 7.3 of
  the variation in the dependent variable Student
  Success. That may be expected given the
  complexity of other variables that affect student
  academic success. That said, this finding may
  suggest that the role of online trust is a factor
  to be considered in overall learner success.

56
Interpretation

• Trust seems to be one of many factors that have
  some influence on student success (as measured by
  the proxies of retention, grade outcomes, and
  student perceptions).
• Operationalizing trust is a complex endeavor, and
  many factors apparently affect this construct.

57
Recommendations for Future Research

• More mixed methods explorations into trust in
  online learning would be insightful.
• Further testing of the OTSS instrument would aid
  its development.
• Applying these findings to applied course design,
  online instruction and distance learning
  administration would enhance eLearning.

58
Thanks

• Thanks to the Way, the Truth and the Life Rodin
  Max my dissertation chair Dr. Daisy
  Arredondo-Rucinski committee members Dr.
  David Marshak and Dr. Mark Roddy Connie
  Broughton and Mark Carbon of WAOL Dr. Roberto
  Peña and Dr. John Jacob Gardiner, of Seattle
  Universitys EDLR Program George Fisher of
  Perennial Survey John Backes, Dr. Ann
  Garnsey-Harter, and Jim James at Shoreline
  Community College, and the many students,
  instructors and administrators who contributed
  their insights to this research.

59
Conclusion

• Thanks for your attention.
• Comments? Suggestions? Questions?
• ---
• Eruditio Loginquitas _at_ Instructional Design Open
  Studio (IDOS blog) http//ome.ksu.edu/id/blog/
• shalin_at_ksu.edu and [phone]
• Axio Learning Management System
  www.axiolearning.org