relationship between student performance and specific online

RELATIONSHIP BETWEEN STUDENT PERFORMANCE AND SPECIFIC ONLINE SUPPORT MATERIALS IN AN OPERATIONS COURSE

by Kenneth J Klassen & Ernest N. Biktimirov

Department of Finance, Operations & IS Faculty of Business, Brock University 500 Glenridge Ave, St Catharines, Ontario, Canada, L2S 3A1

Published in the Journal of the Academy of Business Education 8, Summer 2007, 40-48.

Citation information: Klassen K. J. and E. N. Biktimirov. (2007). “Relationship between student performance and specific online support materials in an operations course,” Journal of the Academy of Business Education 8, Summer, 40-48.

1

Relationship Between Student Performance and Specific Online Support Materials in an Operations Course Kenneth J Klassen Ernest N Biktimirov Faculty of Business, Brock University, Canada This study examines the relationship between students’ use of specific online support materials and their performance in a traditional, face-to-face introductory operations management course. Five measures are used: total hits, hit consistency, accesses to homework solutions and PowerPoint slides, and the number of unique files accessed. Findings indicate that accesses to homework solutions are positively related to students’ performance. In addition, results suggest that access to specific files (rather than access to online course materials in general) is related to student performance in the course. Key Words: Course Web Site, Student Accesses, Student Performance Disciplines of Interest: All Business Disciplines INTRODUCTION The adoption of computer technology for teaching and learning has grown rapidly in recent years. Pedagogical uses range from simple PowerPoint presentations in face-to-face lectures to completely online instruction. The growing number of completely online courses has inspired a large body of literature examining their effectiveness [e.g., DeTure, 2004; Wang & Newlin, 2000, 2002]. While the majority of the courses use traditional face-to-face instruction supplemented with online support materials, research on the relationship of these materials to students’ performance in these “hybrid” courses has been quite limited [Baugher, Varanelli, & Weisbord, 2003]. There have been a few studies into specific aspects, including the one by Berry [2004], who contributes to the improvement of hybrid courses by identifying specific benefits of online instruction and offering recommendations for incorporating these benefits into face-to-face courses with online support. This study examines the value of online support materials that are used to supplement face-to-face instruction in an introductory operations management course. The value of different online course materials is assessed by examining their relationship to the student’s performance in the course. The current study contributes to the literature in three ways. First, this study examines the value of online support materials in a hybrid, introductory operations management course. As stated above, this is still rare. Second, this study uses new, specific measures of online course activity. As far as can be determined, this study is the first to measure student access to each of the following: homework solutions, PowerPoint lecture slides, and the number of unique support files accessed by each student. This differentiation makes it possible to generate conclusions about the value of specific types of support files. Third, this study provides a more precise measure of student total hits, by measuring access to specific support files (i.e., “content hits”). Other studies have measured total hits to the course website [e.g., Baugher et al., 2003]. Students appear to use the homepage very differently from one another. Data used in this study showed that for some students, the number of homepage hits is quite small compared to the number of content hits, for some it’s about equal, and for some the number of homepage hits is more than twice the number of content hits. The following possible explanations can be provided for these observations: in the first case the student downloaded multiple files in the content modules without re-

2

accessing the homepage, in the second case the student only downloaded one file at a time before exiting the site directly, and in the third case the student downloaded one file at a time but always returned to the homepage before exiting the site. This suggests that variables based on content hits may be more valid than those based on total website hits. The next section reviews prior literature on the use of technology for teaching, focusing primarily on studies that consider the value of online support materials. Then, separate sections on the sample and course description, data collection, and the variables used will provide the background for this study. The analysis and discussion of results will then lead to the concluding remarks. LITERATURE REVIEW Given the increasing use of technology for teaching and learning, a growing body of literature considers the educational benefits of technology. For example, prior research has examined: the effectiveness of technology to support different course activities [e.g., Bergmann & Bergmann, 2003; Bocij & Greasley, 1999], characteristics of students who use and benefit from technology [e.g., DeTure, 2004], performance of students in Web-based versus face-to-face courses [Aragon, Johnson, & Shaik, 2002], and performance of students in Web-based versus face-to-face courses with online support [Goldberg, 1997]. While there is a large and growing proportion of courses that use online support to supplement traditional face-to-face teaching, research studying the relationship of these online support materials to students’ performance is quite limited. One exception is the study by Baugher et al. [2003]. These authors use two measures of web utilization: total hits and hit consistency (how consistently a student accesses web material throughout the course), the latter of which had not been examined before. Their findings suggest that consistency of hits throughout the course is positively related to course performance, while the total number of hits is not related to course grades. In contrast, two studies by Wang & Newlin [2000, 2002] found that total hits (in the first and last week of classes) are related positively to student performance. However, these courses are taught entirely online; it may be that hits are more important in a completely online course than they are in a hybrid course. Adding further ambiguity to the comparison of the above studies is the fact that the Baugher study measures hits to the homepage plus hits to secondary pages, while the Wang & Newlin studies consider only homepage hits. The current study builds on the work of Baugher et al. [2003] by using more measures of web utilization and by differentiating between various types of support materials. There have been a few studies that have considered the value and appropriate use of different support materials; these studies have generally been carried out in an off-line environment. For example, Rayburn and Rayburn [1999] find a positive relationship between homework completion and student performance. On the other hand, Peters, Kethley & Bullington [2002] show that requiring graded homework has a negative relationship to students’ exam performance. The examination of the use of PowerPoint presentations produces similar mixed results. For example, while Szabo & Hastings [2000] find that using PowerPoint presentations does not enhance student learning, Bartsch & Cobern [2003] conclude that PowerPoint can be beneficial. In addition, the question arises whether instructors should make PowerPoint slides available to students before lectures. As DenBeste [2003] suggests as a discussion point, if students have PowerPoint slides available outside of class, they might decide not to attend the lectures. The current study will add insight to this body of literature as well. SAMPLE AND COURSE DESCRIPTION Eighty-three undergraduate students in two sections of an introductory operations management course at a medium-sized Canadian university participated in the study. The same instructor taught both sections in the winter 2004 semester using the same content and pedagogical methods.

3

Introductory operations management is a required course for undergraduate students working toward their bachelor degree in business administration. It is normally scheduled to be taken in the students’ sophomore year, although it can often be taken later without adversely affecting progress through the degree. These two sections included 39 sophomores, 40 juniors, and 4 seniors. The WebCT learning management system [2005] was used to support traditional classroom instruction and to collect data about student online course activity. Students from both sections of the course accessed the same website. The WebCT site for this course contained the course syllabus, PowerPoint slides for use in class lectures, solutions for homework problems, instructions for assignments, and other supporting files (e.g., information for the final exam, student grades). None of the above were handed out in class except the syllabus. PowerPoint slides consisted of lecture outlines – aspects “missing” from the slides were the most crucial and important aspects, including key terms and mathematical procedures. Thus, if a student simply downloaded the slides and then skipped the lecture, there would be reduced learning value. At the same time, if a student did not download the slides but came to the lecture, they had a fairly difficult time keeping up with the pace of the lecture because they would typically have to take notes “frantically” the entire time. Based on the instructor’s observation during the term, most students did print these out, although some still attempted to take all notes in class. Most of the support files were posted to the site at the beginning of the term, but a few were posted later (e.g., instructions for assignments and exams). With the exception of the assignment instructions, all files on the site were “optional” for students, so they could attend class and do all required assignments and exams without using the files offered. DATA COLLECTION Data was collected on students’ online course activity from the WebCT system, and student grades in prior courses were obtained from university records. With the WebCT course management system, an instructor has the option to place files on the site in several ways. For files placed in “Content Modules”, detailed statistics are kept by the WebCT system. As such, for this course all files were set up on the site in Content Modules, and WebCT recorded the date and time each student accessed each file. VARIABLES The course grade consisted of both individual and group components. Using only the individual portion (instead of the entire course grade) provided a more precise measure of student’s performance in the course, and, consequently, stronger results. Thus, Course Grade, the dependent variable, is composed of all the individual course components, which are: final exam (40%), mid-term exam (20%), and class contribution (5%). Class contribution is based primarily on the quality and quantity of participation in class discussions. All regression analyses in this study use the Course Grade as the dependent variable. Content Hits is the total number of times the student accessed any file in the content modules of the course WebCT site. This is similar to the variable “Total Hits” used in prior studies [e.g., Wang & Newlin, 2000, 2002; Baugher et al., 2003], except that Content Hits does not include hits to the homepage of the site. Students must use the homepage initially to access the content pages, but there is nothing on the homepage that can be opened or downloaded. Hit Consistency measures how consistently students access the website. During the 13 week semester, 13 periods can be identified for purposes of analysis. Topics are generally covered in approximately one week time periods. The first period starts at the first lecture and ends one week later at the 3rd lecture (both sections had lectures twice a week), the second period starts at the 3rd lecture and lasts a week until the 5th lecture, etc. until week 13, which results in 12 periods. The 13th period is the time between the last

4

lecture and the final exam. In this term the 13th period was two weeks long. Also, Reading Week (i.e., Spring Break) is one week in the middle of the term during which there are no lectures, and thus the period incorporating Reading week is also 2 weeks long. As in Baugher et al., [2003], Hit Consistency is calculated by assigning a zero if the student did not access the site during a period, and a one if they did access it. The sum of these resulted in a variable that ranges from 0 to 13. Unique Files Accessed is the number of different files accessed by a student. A student could have a high number of Content Hits, but only access very few unique files through the term, while another student could access every file once and download it immediately, having a low number of Content Hits. There were 30 unique files on the course website, resulting in a variable that ranges from 0 to 30. PowerPoint Slides Accessed measures students’ access to eleven PowerPoint files on the website. These slides are in outline format – there are many “blanks,” and no mathematical portions are provided online. Thus, students are expected to download the slides and fill them in during class. Students have responded very positively to this method of note-taking (based on comments on the course evaluations). PowerPoint Slides Accessed captures both how soon during the term a student accessed these files, and the number of different files accessed. Specifically, it calculates the number of days between the first access to a PowerPoint file and the final exam and then adds these for all eleven files. A PowerPoint file that is never accessed does not contribute towards the value of PowerPoint Slides Accessed. If the day of the first class is day 1, and the day of the final exam is the last day of the term (in this case, it is day 107), then: 11

PowerPoint Slides Accessed =

∑ (107 − P ), i = 1, 2,...11 i

(1)

i =1

where: Pi = day of access to PowerPoint file i. The maximum value for this variable is 1177, which is computed by multiplying 107 days in the term by the eleven possible PowerPoint files. Homework Solutions Accessed measures students’ access to eight homework solution files, measured in the same manner as PowerPoint Slides Accessed. Thus, Homework Solutions Accessed is calculated as: 8

Homework Solutions Accessed =

∑ (107 − H ), i = 1, 2,...8 i

(2)

i =1

where: Hi = day of access to homework solution i. The maximum value for this variable is 856, which is computed by multiplying 107 days in the term by the eight possible homework files. Note that these homework solutions are not graded or discussed in class, but instead answers are posted on WebCT. The homework is directly related to the material on the exams and thus doing the homework should improve exam grades. In contrast, there is not a direct relationship between the homework and most group portions of the course. Finally, the variable GPA is the student’s university average for all courses taken prior to the winter 2004 term. This variable ranges from 0 to 100. GPA has consistently been shown to have significant positive correlation with student performance in different business courses [e.g., Borde, 1998; Chan, Shum & Wright, 1997], and, specifically, in an introductory operations management course [Peters et al., 2002].

5

Many other variables were also considered but then not included. For instance, gender and university level (sophomore, junior, senior) were tested, but were found to have no significant correlation results, and their coefficients were insignificant in the regressions. ANALYSIS Table 1 presents descriptive statistics for all variables in the analysis. The mean Course Grade is 57.54%, which seems quite low. Recall that this includes only the individual portions of the grade; the actual course average with group work included was 65.34% (a “C”), which is lower than the mean GPA of 71.72%. Thus, these students struggled in this course compared to other courses, but performed better on the group portions than individually. The average Content Hits was 50.51. This suggests that, on average, students referred to course content almost 4 times a period (50.51 content hits / 13 periods). [Place Table 1 about here] The mean and standard deviation of Hit Consistency, 8.95 and 2.48, respectively, are close to the same statistics for hit consistency (9.05 and 2.65), reported for the introductory management class by Baugher et al., [2003]. The mean for the Unique Files Accessed indicates students accessed on average 23 out of 30 available content files. Both Hit Consistency and Unique Files Accessed range from 0 to their possible maximums of 13 and 30, respectively, suggesting that the use of web-based course materials varied considerably among students. Whereas the numbers of homework solutions and PowerPoint files were not very different, 8 and 11 respectively, the mean Homework Solutions Accessed of 243.23 is almost three times smaller than the mean PowerPoint Slides Accessed of 687.82. This considerable difference suggests that students accessed PowerPoint slides much earlier than Homework Solutions over the semester. This is logical, as a typical student may access lecture slides before the lecture, but may not access homework solutions until later when they are studying the material. Table 2 displays the correlations between the variables. GPA has the highest correlation with the Course Grade, 0.739, which is significant at the 1% level. Among the five variables that measure online course activity, only Homework Solutions Accessed and PowerPoint Slides Accessed have significant positive correlations of 0.289 and 0.192, respectively, with the Course Grade. While GPA does not have significant correlations with any independent variable, Homework Solutions Accessed has positive significant correlations with all measures of online course activity except Hit Consistency. PowerPoint Slides Accessed has positive significant correlations with all measures of online course activity, except Content Hits. [Place Table 2 about here] Various regression analyses were carried out for the dependent variable, the Course Grade. Initially, stepwise regression was used. All 3 stepwise procedures (general stepwise, forward selection, and backward elimination) produced the same results; only GPA and Homework Solutions Accessed remained in the final model. Residual analysis was performed to ensure these results were not a consequence of outliers or extreme values. For example, the analysis was re-run without eight students who did not access any homework solutions and four who scored above 600. Their removal did not change the regression results. In order to address more research questions, other regressions were also carried out; Table 3 reports a number of regressions that were run, including the final stepwise results (Regression VI). [Place Table 3 about here]

6

Having the highest correlation with the Course Grade, the variable GPA yields an R2 of 0.51 in Regression I, and it is the most significant predictor of Course Grade in all eight regressions. The F-test is used to estimate incremental gains in the predictive power of regressions from introducing each online course activity variable in addition to GPA. Thus, the ‘R2 Change’ and ‘F-test for R2 Change’ compare each model to Regression I. First examining regressions II through VI, it is not surprising that the results are similar to those from the correlation analysis. Specifically, only Homework Solutions Accessed and PowerPoint Slides Accessed have a significant positive relationship with the Course Grade among all online activity variables. Moreover, according to the F-test, only Homework Solutions Accessed and PowerPoint Slides Accessed provide additional predictive power over that provided by GPA alone. In contrast, additions of three other measures of online activity, Content Hits, Hit Consistency, and Unique Files Accessed, do not significantly improve the predictive power of regressions II, III, and IV, respectively, compared to regression I. At this point a few questions are of interest. First, would including all variables improve the model? Second, if all variables are included, will PowerPoint Slides Accessed and Homework Solutions Accessed still have a significant impact on Course Grade? Regression VII shows that the Adjusted R2 improves slightly, and that these two variables remain significant. However, PowerPoint Slides Accessed is only significant at the 0.10 level. Along with Regression VIII, this outcome helps explain the results of the step-wise regression. Regression VIII shows a model including only GPA, PowerPoint Slides Accessed, and Homework Solutions Accessed (this was part of the earlier stepwise regression analysis). It shows that only GPA and Homework Solutions Accessed remain significant. Comparing all the models, it is apparent that regressions VII and VIII result in higher Adjusted R2, but regression VI has the strongest results for the F-Test. Thus, the model that best describes individual course grades is a ‘simpler’ model, including GPA and Homework Solutions Accessed. DISCUSSION The results of the correlation analysis suggests that accesses to homework solutions and PowerPoint slides are associated with higher course grades, while the results of the regression analyses suggest that only students’ accesses to homework solutions are associated with course performance. This evidence suggests the use of online homework solutions in order to improve student performance, which supports the (off-line) findings of Rayburn and Rayburn [1999]. One reason for this result is likely that the homework solutions were not discussed in class. Thus, in order to obtain benefit from them, students had to access them. Moreover, this research suggests that homework problems do not need to be submitted for grading in order to provide learning benefits. The correlation analysis shows a weak positive relationship between the course grade and accesses to lecture slides. This supports the findings of Bartsch and Cobern [2003], while the lack of relationship in the regression analysis supports the findings of Szabo and Hastings [2000]. Thus, this research does not support either viewpoint. It may not be surprising that a positive relationship exists, since the slides are directly related to exam material. It also may not be surprising that the relationship is weak, because PowerPoint slides were covered in class and thus some value could be gained from attending class even without the slides. Future research may uncover more insights, and results may differ depending on the nature of the slides provided; recall that in this study the lecture slides were not complete – they were in outline format. It is interesting that Unique Files Accessed does not have a significant relationship with the Course Grade. This result suggests that accessing different materials in addition to homework solutions and PowerPoint slides may not affect students’ performance. Also, similar to the results for total hits in Baugher et al. [2003], a significant relationship between Content Hits and the Course Grade is not found.

7

A possible explanation for this may be that the best students access files once and download them, whereas weaker students are not as efficient, returning to the site numerous times. In contrast to Baugher et al. [2003], a significant relationship is not found between Hit Consistency and the Course Grade. A possible explanation for this difference may lie in the fact that in the current study most course materials were available to students at the beginning of the semester. Therefore, a student could download all materials earlier in the semester, which would greatly reduce the need to regularly access the course web site. This supposition is supported by the prior result that, on average, students did access PowerPoint slides earlier in the term. Indeed, the average value for PowerPoint Slides Accessed in Table 1 is relatively large (688) compared to its maximum possible value (1177). CONCLUSION While the number of completely web-based courses keeps increasing, a majority of courses use traditional face-to-face format, using online materials to support the classroom environment. However, research on the learning benefits of the students’ use of online materials has been limited. The value of online support materials is examined in a traditional face-to-face introductory operations management course supported with a class web site. Two measures of online course activity that have been used in the literature are used here (total hits and hit consistency), along with three specific measures that have not been analyzed before; access to homework solutions, access to PowerPoint slides, and the number of unique files accessed. After limitations are outlined, the results will be reviewed and future research questions discussed. There are some limitations in this study. Using a single instructor, the same course outline, and the same materials prevented the confounding of results that could occur with multiple instructors. However, this means it is not possible to make generalizing conclusions about the value of online homework solutions and PowerPoint slides for learning based on this one study. Additional research examining the value of online support materials in other courses is needed. Another aspect that likely affected results is that all course materials were available at the beginning of the term. If they had instead been posted as they were needed throughout the term, it may be expected that the results (at least for hit consistency) would be different. Thus, the precise definition of variables in these types of studies is very important. Despite these limitations, and despite the fact that GPA is such a strong indicator of course grade that it can be difficult to find other significant relationships in a regression analysis, new evidence supporting the value of specific types of files was found. After including students’ GPA, it is found that accesses to homework solutions are positively related to students’ performance in the course. In contrast, whereas accesses to online lecture slides have a significant correlation with course performance, these accesses do not contribute to the explanation of course performance when GPA and accesses to homework solutions have been included in the model. In addition, no significant relationship between students’ performance and the following variables is found: number of content hits, consistency of hits during the term, and number of unique files accessed. As mentioned above, further research is needed in other courses and in other settings. In addition, testing hit consistency in a course where materials are posted week-by-week could shed light on the importance of this variable. It is important in future work to denote how important various online components are to the course. For instance, in this study the optional homework was not discussed in class at all; students had to access the website in order to get the solutions. Also, the completeness of the lecture slides is important in terms of how much information the students receive in class versus how much is included online. Enough studies including these details could eventually lead to a categorization of online courses based on the degree of online activity (from 0% to 100%) and thus a better understanding of how important online materials are at these different levels. In addition, it would be interesting to study

8

whether students with prior online or hybrid courses do better than those that have no prior experience, and whether hits or specific types of hits are more important in an online course than they are in a hybrid course. REFERENCES Aragon, S. R., S. D. Johnson, & N. Shaik. “The Influence of Learning Style Preferences on Student Success in Online Versus Face-to-Face Environments,” American Journal of Distance Education, 16 (December, 2002), 227-244. Bartsch, R. A., & K. M. Cobern. “Effectiveness of PowerPoint Presentations in Lectures,” Computers & Education, 41 (2003), 77-86. Baugher, D., A. Varanelli, & E. Weisbord. “Student Hits in an Internet-Supported Course: How Can Instructors Use Them and What Do They Mean?” Decision Sciences Journal of Innovative Education, 1 (Fall, 2003), 159-179. Bergmann, T. J., & M. A. Bergmann. “Application of Internet Technology to Facilitate Student Team Project,” Journal of the Academy of Business Education, 4 (Fall, 2003), 85-102. Berry, G. R. “Lessons from the On-Line Teaching Experience: Suggestions for Enhancing the Face-toFace MBA Classroom,” Journal of the Academy of Business Education, 5 (Spring, 2004), 88-97. Bocij, P., & A. Greasley. “Can computer-based testing achieve quality and efficiency in assessment?” International Journal of Educational Technology, 1 (No. 1, 1999). http://smi.curtin.edu.au/ijet/v1n1/bocij/index.html. Borde, S. F. “Predictors of Student Academic Performance in the Introductory Marketing Course,” Journal of Education for Business, 73 (No. 5, 1998), 302-306. Chan, K. C., C. Shum, & D. J. Wright. “Class Attendance and Student Performance in Principles of Finance,” Financial Practice and Education, (Fall/Winter, 1997), 58-65. DenBeste, M. “Power Point, Technology and the Web: More Than Just an Overhead Projector for the New Century?” The History Teacher, 36 (August, 2003), 491-504. DeTure, M. “Cognitive Style and Self-Efficacy: Predicting Student Success in Online Distance Education,” American Journal of Distance Education 18 (March, 2004), 21-38. Goldberg, M. W. “CALOS: First Results from an Experiment in Computer-Aided Learning for Operating Systems,” ACM SIGCSE Bulletin, 29 (March, 1997), 48-52. Peters, M., B. Kethley, & K. Bullington. “The Relationship between Homework and Performance in an Introductory Operations Management Course,” Journal of Education for Business, 77 (July/August 2002), 340-344. Rayburn, L. G. & J. M. Rayburn. “Impact of Course Length and Homework Assignments on Student Performance,” Journal of Education for Business 74, (July/August, 1999), 325-331. Szabo, A. & N. Hastings. “Using IT in the Undergraduate Classroom: Should we replace the blackboard with PowerPoint?” Computers & Education 35, (2000), 175-187.

9

Wang, A. Y., & M. H. Newlin. “Characteristics of Students Who Enroll and Succeed in Psychology WebBased Courses,” Journal of Educational Psychology, 92 (No. 1, 2000), 137-143. Wang, A. Y., & M. H. Newlin. “Predictors of Web-Student Performance: The Role of Self-Efficacy and Reasons for Taking an On-Line Class,” Computers in Human Behavior, 18 (2002), 151-163. WebCT “WebCT Learning without Limits,” Retrieved May 26, 2005, from http://www.webct.com.

10

Table 1. Descriptive Statistics

VARIABLE

Mean

Median

Minimum

Maximum

Standard Deviation

Course Grade

57.54

58.52

33.75

82.31

11.98

Content Hits

50.51

48.00

0.00

121.00

19.78

8.95

9.00

0.00

13.00

2.48

23.27

24.00

0.00

30.00

4.95

PowerPoint Slides Accessed

687.82

692.00

0.00

1010.00

174.84

Homework Solutions Accessed

243.23

240.00

0.00

730.00

178.59

71.72

71.61

60.39

87.90

6.01

Hit Consistency

Unique Files Accessed

GPA

11

Table 2. Correlations Between the Variables

VARIABLE

Content Hit Unique Files Hits Consistency Accessed

PowerPoint Homework Slides Solutions Accessed Accessed

Course Grade

0.013

Content Hits

Hit Consistency

GPA

0.078

0.095

0.1921

0.2893

0.7143

0.4383

0.5893

0.131

0.2943

0.038

0.4113

0.3003

0.113

0.112

0.4703

0.6113

0.030

0.3943

0.018

Unique Files Accessed PowerPoint Slides Accessed Homework Solutions Accessed 1 2 3

0.082

Significant at the 0.10 level. Significant at the 0.05 level. Significant at the 0.01 level.

12

Table 3. Regression Results REGRESSIONS + VARIABLE Constant GPA

I

II

III

IV

V

VI

VII

VIII

-44.5603 (0.00)

-44.2093 (0.00)

-44.4883 (0.00)

-48.3923 (0.00)

-52.5813 (0.00)

-45.6253 (0.00)

-44.1753 (0.00)

-50.1243 (0.00)

1.4233 (0.00)

1.4253 (0.00)

1.4243 (0.00)

1.4193 (0.00)

1.4173 (0.00)

1.3863 (0.00)

1.3863 (0.00)

1.3893 (0.00)

-0.016 (0.78)

-0.008 (0.86)

Content Hits

-0.013 (0.97)

Hit Consistency

-0.024 (0.96) 0.178 (0.35)

Unique Files Accessed PowerPoint Slides Accessed

-0.340 (0.25) 0.0122 (0.02)

Homework Solutions Accessed

0.0102 (0.10)

0.007 (0.19)

0.0163 (0.00)

0.0183 (0.00)

0.0132 (0.02)

R2

0.5095

0.5097

0.5095

0.5149

0.5418

0.5629

0.5875

0.5722

Adj R2

0.5034

0.4974

0.4972

0.5028

0.5304

0.5520

0.5550

0.5560

R2 Change

0.0002

0.0000

0.0054

0.0323

0.0534

0.0781

0.0628

F-test for R2 Change

0.0318

0.0011

0.8964

5.64532

9.77683

2.87742

5.79523

+ p-values in brackets 1 Significant at the 0.10 level. 2 Significant at the 0.05 level. 3 Significant at the 0.01 level.

13