Printer Friendly

Online discussion and learning outcomes.

Introduction

As online delivery formats grow in popularity, so does online class size and/or the number of sections per course. The problems of keeping track of the students and the challenges of conducting meaningful discussions seem to increase rapidly with class size. The widely used Learning Management Systems (LMS) have tools for email and discussion boards but their power and accessibility pale in comparison to the tools of social media. Many of these social media have the consequential advantage of students' everyday use; hence their use in the classroom does not force students into an additional electronic portal that they would not naturally use.

Facebook has much to recommend it as an instructional tool for student/faculty communication Bosch (2009). First, for almost all students there is no learning curve. Students, with few exceptions, know how to use Facebook and have an account. Thus there is no need for the instructor to write up an instruction sheet. Students already book mark these sites, or have applications (apps) installed on their mobile devices. Instructors, however, may experience a learning curve. Our paper seeks to make that curve less steep. Second, the messages, once sent, are read and responded to in a timely manner. This result is facilitated by the student behavior of constantly checking their Facebook accounts, and because notification of posts are pushed to their email accounts. This is not a behavior usually attributed to the standard LMS.

When using Facebook for instructional purposes, thought should be given to the settings of the privacy controls. This includes informing users of the optimal privacy settings, as students tend to overlook these settings. On the instructor side we recommend creating a group as a secret Facebook group, which adds a protective layer approaching that of the standard LM system. Also we recommend adding students to the group as "members," not as "friends,". In this way the student privacy settings restrict the instructor to the student profile to the more restrictive "public" instead of the less restrictive "friend,". This approach addresses the "creepy tree house" downside of using social media in an instructional setting described by (McBride 2008).

Studies of the use of Facebook in an educational setting report mixed results. (Kirschner and Karpinski 2010) and (Junco 2012) report that increased time spent on Facebook is associated with reduced learning outcomes. On the other hand, (Pellizzari 2012) reports a positive association of Facebook usage and learning outcomes.

Using Facebook (FB) as a Discussion Forum

In this study, weekly discussion and class Q/A is conducted in a secret Facebook group for each section with students added as group members (not as friends). A discussion thread is initiated each week and a portion of the student grade is based on weekly contributions to the thread. The contributions could be in the form of posing a course related question, responding to a posted question, posting and describing a link to relevant material, or commenting on the linked material. The mobile app feature of Facebook facilitates a 24/7 lively and productive discussion, and helps create a shared community experience of helpful assistance and a thoughtful exchange of viewpoints for the participants.

Privacy Settings

As we become more integrated via the social electronic media, so does the importance of our identity on the Internet. In a social networking site there is a very obvious portrayal of self and users should perform "hygiene" over their medium identity as they do on their person. Because their identity in the form of a Facebook profile is out for the world to see, students need to be cognizant that various employers, teachers, colleagues, etc. could examine the identity. While there is not anything necessarily wrong with not keeping one's personal information privy, it does potentially expose one to issues down the road, namely decisions by a future boss (Chang 2008). In fact, according to a CareerBuilder.com survey, 37 % of employers examine a job candidate's Facebook profile before hiring (Haefner 2012).

To assess student awareness of privacy controls in Facebook, we surveyed the Facebook pages of students enrolled in a Principles of Microeconomics course in Fall 2011 and Winter 2012. The results of the tabulation are shown in Table 1.
Table 1 Student usage of privacy controls

Variable      Obs   Mean    Std.Dev.  Min    Max

view_pictures 171   0.26    0.44      0      1

view_wall     171   0.53    0.28      0      1

view_friends  171   0.74    0.44      0      1

view_tab      171   0.83    0.38      0      1

view_profile  171   0.98    0.15      0      1


The table summarizes student usage of privacy settings for five Facebook tools. The privacy settings are equal to 0 if the setting restricts access to only friends, and equal to 1 if the setting allows all visitors to view the tool. A mean close to 0 indicates a stricter setting than a mean close to 1. Given best practice recommendations, the better privacy setting is on the stricter side, ideally preventing people without "friend" status from viewing the (1), (2), (3), or (4) categories, since these are potentially very personal. However, from our data set, more than 25 % of students had their pictures available for anyone to view (category #1). Nearly 53 % let any user see category (2), which often shows personal conversations with friends, activities recently performed, and narrated thoughts of the user, and over 83 % of students had their personal information (category #4) freely available on the web. These findings show that very few of the students in this sample are tightly protecting themselves and are consistent with (Bosch 2009).

The results suggest the need to include in introductory materials for using Facebook information about the importance of the privacy settings. (Munoz and Towner 2009) has a good discussion of recommended privacy settings.

How to Create and Moderate Threaded Discussions in FB

We create the discussion board in Facebook as a closed group. To complete the process of creating a closed group the moderator has to invite a "friend" to join. We use a "John Doe" friend account for this purpose. Students are invited to join as members and the administrator does not accept "friend" requests. In the group settings, the options to allow only the administrator to create threads and to send push notification to members are selected. After all the students are signed in as members the group status is changed from "closed" to "secret".

The moderator starts the weekly discussion thread with a post, and students post comments to the thread. To post a link just paste the URL in a comment and Facebook automatically generates the code for a hyper link. Readings that are on the web can easily be linked to as well as ones that arc password protected. With password protected links the user is prompted for the password and then the page opens. Posting the links this way makes access easier and thus can stimulate conversation in the thread.

Another idea we use is to have a menu of discussion activities for the student to choose from. Some students like to answer questions, others like to ask them. Some students learn from video explanations, others like written explanations, still others like graphical expositions. In this approach we give many points for activities, say a possible 300, and let the student choose the preferred activities that will total a maximum of 100. Here is an example list of possible activities: posting a question, posting an answer, posting and describing a link to a resource that clarifies the material, and reporting typos in the lecture notes. These activities can be done multiple times. The activities can also be graded for performance. Simply describing the value of a link can be 25 points, or the moderator can evaluate the quality of the link description and assign less than 25 points. A well-known problem is that of posting at the last minute. So the point value can be increased by 25 % by a cutoff date to encourage early posting.

Navigation of the Threads

The thread most recently posted to always appears at the top of the wall. Sometimes students are catching up and post to threads from a couple weeks prior, so to offset that the instructor can post a "bump" to the thread that they want at top, and then delete the "bump" post to clear the thread of this book keeping activity. On the other hand, there may be occasions when the instructor has reason to bring a past thread to the class's attention. To search for that thread the key strokes of "crtl" and "F" will bring up a search box, and entering the appropriate phrase will bring up the sought for thread.

Examples of Learner Engagement

The below exchange is an example that illustrates levels of learner engagement by the speed of response, and peer-to-peer learning. The example is a transcript of the exchange in a thread on a Faccbook wall for one of the Econ 1201 courses included in this study.

* Student 1: for number 7 on Aplia how do we know what is the most likely long run supply? (9/12 at 9:00 pm)

* Student 2: for question fourteen I am confused because I am coming up with a positive number (9/12 at 9:03 pm)

* Student 3: (responding to Student 2) I have a similar problem when calculating, but just do the quantity of one divided by the price of the other. The price of ice cream is positive and the quantity of syrup drops and is negative, therefore it is negative, and complements. Think it through logically to check. Would it make sense for them to be substitutes? They are complements since when one price goes up, the quantity of the other goes down. (9/12 at 9:05 pm)

* Student 4: (responding to student 1) the elasticity of supply is typically greater the longer the period of adjustment, so the curve should reflect that when you calculate the elasticity. For this problem, you will see that the longer the period of time that passes the greater the ability to increase the quantity supplied in response to a higher price. (9/12 at 9:05 pm)

* Student 1: (responding to student 4) thank you! (9/12 at 9:08 pm)

Also, on the Facebook wall the engagement level can be apparent by the number of "likes" for a post, in the number of "comments" to a thread, and finally, in the number "Ha Ha" and "LOL" and "(: Thx" posted within a thread. Anecdotally, we have noted many students became "friends" during the class, and many set up study groups.

Facebook for most students has no learning curve. But it can be a problem for the older student less used to social media. Be prepared for a small minority of older adults that would prefer not to use Facebook. One may feel inclined to make an accommodation using a combination of private email and announcements.

Data

Data were collected from three sections of a course in Principles of Microeconomics taught by the same instructor in Fall 2011. Participating students signed a consent form, approved by our Institutional Review Board, allowing these data to be used for research purposes. Descriptive statistics for learning outcomes and student academic, demo-graphic and learning style characteristics are shown in Table 2. The four variables for class rank are indicator variables. The class rank of the students was comprised of predominately freshman and sophomores (68 %), and 42 % of the students were female. The student majors were comprised of 55 % economics or business, and 21 % math or sciences. Fifty percent of the students held jobs and worked on average 12.75 h a week. The average Math SAT score was 583, and Verbal was 535. Average GPA entering the course was 3.08. The average final exam score, which is used as the measure of learning outcome, was 67.46. Weekly homework was assigned from the Aplia electronic service and the mean score for the semester was 71.
Table 2 Descriptive statistics of sample

                               N       Mean      S.D.

Freshman (=1)                  78      0.15      0.36

Sophomore (=1)                 78      0.53      0.50

Junior (=1)                    78      0.17      0.38

Senior (=1)                    78      0.10      0.31

Other                          78      0.03      0.16

Female (=1)                    78      0.42      0.50

Econ, bus. Major (=1)          78      0.55      0.50

D_MSE (=1)                     78      0.21      0.41

Have job (=1)                  75      0.51      0.50

Weekly hours worked            75      12.75     16.67

Math SAT                       78      583.46    76.48

Verbal SAT                     78      535.26    91.21

GPA at beginning               72      3.08      0.56

Average score on all Aplia HW  94      86.89     21.42

Learning style email (Aural= 1)71      0.15      0.36

Final exam score               88      67.46     12.80

Observations                   94


Information on learning styles was collected because previous empirical studies of learning outcomes in Principles of Economics courses have reported a correlation between exam scores and student characteristics of learning styles and personality types (Boatman et al. 2008; Borg and Shapiro 1996; Charkins et al. 1985; Durham et al. 2007; Emerson and Taylor 2007). The VARK survey of student learning style preferences was used because its four categories of learning are directly applicable to course design issues addressed by instructors of principles of economics classes (Boatman et al. 2008). The VARK survey assigns preferences for learning style modalities based on student responses to 16 questions. The four learning style modalities are Auditory, Read/Write, Visual, and Kinesthetic. Auditory is described as a preference for learning from information that is "heard or spoken"; Visual is described as a preference for learning from information that is presented in graphs, charts, and diagrams; Read/Write is described as a preference for learning from information that is "displayed as words"; and Kinesthetic is described as a preference for learning by doing (VARK 2013). Students are characterized as having a dominant learning style if the responses are skewed to one modality, or are characterized as having a mix of learning styles if the responses are distributed across two, three or four of the modalities. Based on the student responses we constructed the indicator variable "Learning Style Email". By the reasoning that Auditory learners prefer learning by conversation, and because Facebook has a conversational flavor to it, "Learning Style Email" was assigned the value of "1" if the student has a predominant preference for Auditory, or a preference for a mix of 2 modalities where Auditory is one of the modalities. Based on this definition, 15 % of the students in the sample were assigned the value of one for the variable "Learning Style Email".

Table 2 reports the average number of posts per discussion thread. The average student made 11 posts per discussion thread. About 8 of these posts had substantial content (i.e. more that the conversational comments such as "Thank you" and "LOL". Posts were grouped into 3 categories. "Posts of an answer" were most frequent (58 %) followed by "posts of a question" (37 %), and the remainder was "posts with links" (5 %). The former category has the lowest frequency. A possible explanation is that though posting links in a Facebook thread is mechanically quite simple, the linked resource had to be located and described, which is more involved.

Empirical Model

Because we are interested in how participation in the Facebook discussion activity affected learning outcomes, as a first cut we estimated a model that has a focus on variables related to discussion activity:

[y.sub.i] = [[beta].sub.1][x.sub.1i] + [[beta].sub.2][x.sup.2i] + [[epsilon].sub.i], where i = 1, 2, ... N; (1)

where [y.sub.i] is final exam score, [x.sub.1i] is participation in discussion, [x.sub.2i] is propensity for participation in discussion," i" is the ith student, [[epsilon].sup.i], is the error term, and N is the total number of students. As in many other studies, final exam score is used as the measure of learning outcome (Anstine and Skidmore 2005; Coates and Humphreys 2004; Figlio et al. 2010; Gratton-Lavoic and Stanley 2009; Marburger 2006). The variables used to measure participation in discussion are average per week posts, posts with substantial academic input, posts with links, posts with questions, and posts with answers (as shown in Table 3). The variables used to measure propensity for participation in discussion are average score on the weekly homework assignments in Aplia and a learning style measure. On the reasoning that students are more likely to participate because they have more knowledge about the topic of the thread, we chose the average score on the weekly homework assignments in Aplia. On the reasoning that participation may depend on preferences and skills, we chose a measure based on learning style preferences. As explained above, from the responses to the VARK survey we constructed the indicator variable "Learning Style Email "as a measure of the propensity to participate in the discussion activity.
Table 3 Descriptive statistics of Facebook posts

                                               N

Posts (Avg. Per Week)                          94

Posts with substantial content (Avg. Per Week) 94

Posts with links (Avg. Per Week)               94

Posts of an answer (Avg. Per Week)             94

Posts of a question (Avg. Per Week)            94

Observations                                   94

                                    Mean

                                    10.95

                                    7.82

                                    0.50

                                    5.34

                                    3.38

                                    S.D.

                                    13.41

                                    9.42

                                    1.42

                                    6.39

                                    5.38


Because our course has 3 hourly exams, and the weekly discussion threads and homework assignments are uniquely tied to each of the 3 hourly exams, the data can also be viewed as a panel. Using the panel we can test whether the effect of discussion on hourly exam scores varies systematically over the semester. Specifically we use the data arrayed as a panel to test whether the effect of a student's increased experience in using the discussion board over the semester is reflected in their hourly exam score. For this purpose we created a panel based on each of the three hourly exams and the usage measures for participation in the discussion threads that correspond to each of the exams.

Following (Stanca 2010) the panel data were modeled as:

[y.sub.it] = [[beta].sub.1][x.sub.1it] + [[beta].sub.2][x.sup.2it] + [[epsilon].sup.it], where i = 1, 2, ... N; t = 1, 2, ... T. (2)

where [y.sub.it], is hourly exam score, [x.sub.1it] is participation in discussion, [x.sub.2it] is propensity for participation in discussion," i" is the ith student," t" is the tth hourly exam, [[epsilon].sub.it] is the error term, N is the total number of students, and T is the total number of hourly exams.

Results and Discussion

The estimates of the first empirical model are reported in Table 4. The first column reports the estimates for the model with measures of participation in the discussion forum and measures of the propensity to participate. The number of observations with valid responses is 70 and the adjusted [R.sup.2] 0.28. The coefficients of the first three variables measuring participation: posts, posts with substantial content, and posts with links, are not statistically significant at the 10 % level. However, the coefficient of the variable "posts of a question" is positive and statistically significant at the 10 % level, and the coefficient of the variable "posts of an answer" is positive and statistically significant at the 5 % level. The two variables measuring propensity to participate are "average score on all Aplia HW" and "learning style email". The coefficient of the former is positive and statistically significant at the 5 % level and the coefficient of the latter is statistically insignificant at the 10 % level.
Table 4 OLS regression results full sample

                                   (1)

                                   OLS



Posts (Avg. per week)              -1.267

                                   (1.660)

Posts with substantial content     -0.581

(Avg. per week)                    (0.747)

Posts with links (Avg. per week)   2.568

                                   (1.627)

Posts of a question (Avg. per week)1.884 (*)

                                   (1.717)

Posts of an answer (Avg. Per Week) 2.103 (*)

                                   (2.090)

Average Score on all Aplia HW      0.770 (***)

                                   (4.690)

Learning Style Email (Aural= 1)    -1.854

                                   (0.504)

Sophomore (=1)



Junior (=1)



Senior (=1)



Econ, Bus. Major (=1)



D_MSE(=1)



Weekly hours worked



Math SAT



Verbal SAT


Female (=1)



Constant                           -3.472

                                   (0.236)

Observations                       70

[R.sup.2]                          0.355

Adjusted [R.sup.2]                 0.282

F                                  4.879


                                   (2)

                                   OLS



                                 -1.184

                                 (1.552)

                                 -1.076

                                 (1.203)

                                  1.882

                                 (0.920)

                                  2.467 (*)

                                 (2.116)

                                  2.141 (*)

                                 (2.078)

                                  0.787 (**)

                                 (3.570)

                                  0.00473

                                 (0.00130)

                                 -2.881

                                 (0.916)

                                  7.694 (*)

                                 (1.862)

                                 -4.912

                                 (0.856)

                                 -0.921

                                 (0.280)

                                  6.265

                                 (1.625)

                                  0.105

                                 (1.060)

                                  0.0383

                                 (1.644)

                                  0.0404 (*)

                                 (1.907)

                                  3.663

                                 (1.515)

                                -52.07 (*)

                                 (2.513)

                                 54

                                 0.717

                                 0.595

                                 5.861

Absolute t statistics in parentheses
(+) p<010,(*) p<0.05,(**) p<0.01,(***) p<0.001


Column 2 of Table 4 displays estimates for a model that includes the variables in the column 1 model and adds additional variables to control for student academic and demographic characteristics. As compared to the column 1 model results, the number of observations with valid responses dropped to 54, because some observations did not include complete information (the largest number of which are students that took the ACT instead of the SAT exams) and the adjusted [R.sup.2] rose to 0.59. The results for the estimated coefficients in the column 1 model are not substantially affected. As in the column 1 model, the coefficients of the first three variables measuring participation are not statistically significant at the 10 % level. As in the column 1 model, the coefficient of the variable "posts of a question" remains positive, but the level of statistical significant increased to the 5 % level, and the coefficient of the variable "posts of an answer" remains positive and statistically significant at the 5 % level. The results for the two variables measuring propensity to participate arc unchanged from the first model.

Using the panel based on each of the three hourly exams and the corresponding measures of discussion participation, we estimated model 2 for each of the three hourly exams. The OLS estimation results are reported in Table 5. The adjusted [R.sup.2] statistic for the regression models range from 0.11 for exam 2, to 0.48 and .35 for exams 1 and 3, respectively. For exam 1 and exam 2 the estimated coefficients of the five variables for participation in the discussion forum are consistently statistically insignificant at the 10 % level. For exam 3, however, the estimated coefficients of four of the five variables for participation are statistically significant at the 5 % level. This is consistent with the explanation that using the discussion forum in a way to have a significant impact on learning outcomes takes time because of the time it takes time to become accustomed to how the forum works and time to build up relationships and a sense of community.
Table 5 OLS regression results: by hourly exam

                                (1)           (2)

                                Exam 1        Exam 2

Total posts (Avg. per week)     0.261         -1.714

                                (0.0373)      (0.247)

Posts with substantial content  -0.475        -0.722

   (Avg. per week)              (0.0757)      (0.157)

Posts with links                5.591         7.459

   (Avg. per week)              (0.359)       (0.700)

Posts of a question             -7.527        2.660

   (Avg. per week)              (1.158)       (0.236)

Posts of an answer              2.917         5.880

   (Avg. per week)              (0.428)       (0.804)

Aplia HW score                  1.770***      0.468*

                                (6.926)       (2.611)

Learning style email (Aural = 1)-0.438        4.368

(0.118)                         (0.988)       (0.504)

Constant                        -88.97***     25.07

                                (3.832)       (1.630)

Observations                    65            69

[R.sup.2]                       0.539         0.201

Adjusted [R.sup.2]              0.482         0.110

F                               9.509         2.198

                        (3)

                        Exam 3



                        -5.101

                        (0.950)

                        -12.21*

                        (2.037)

                        34.23*

                        (2.117)

                        17.89*

                        (2.497)

                        17.07*

                        (2.241)

                        0.771***

                        (5.714)

                        1.863

                        3.407

                        (0.281)

                        67

                        0.417

                        0.347

                        6.016

Absolute t statistics in parentheses

(+) p<0.10,(*) p<0.05,(**) p<0.01,(***) p<0.001


The coefficient of the variable "posts with substantial content" is negative (12.21) and statistically significant at the 5 % level. An interpretation is that frequent posting to the discussion forum is a sign that the student is distracted from the course content and hence the frequency of conversation on the forum is negatively influencing their learning, at least as measured by the exam assessment instrument. This result is consistent with the negative correlation between GPAs and Facebook usage reported in (Kirschner and Karpinski 2010). They found that frequent Facebook users tend to be social extroverts engaged in extracurricular activities, and academics are of lower priority for them than for non-Facebook users. Similarly, (Junco 2012) a report that time spent on Facebook "chatting" is negatively related to time spent preparing for class.

The coefficients of the variables "posts with links", "posts of a question", and "posts of an answer" are positive (34.23, 17.89, and 17.07, respectively) and statistically significant at the 5 % level. For each the numerical effect is larger in absolute value that the negative effect of the estimated coefficient of "posts with substantial content" (12.21). An interpretation is that by the last third of the semester students had developed ways to use Facebook discussion to enhance their learning outcomes, at least as measured by the exam assessment instrument. These positive coefficients are consistent with the positive association repotted in Pellizzari (2012). The estimated coefficient of "posts with links" (34.23) is numerical greater that "posts of a question" (17.89). The former activity involves a relatively more complex activity of searching the Internet for a useful educational resource and then describing how it is useful to understanding the course material. The latter activity is a relatively simpler task of posing a question. Both results are suggestive that time spent on very focused activities have a larger impact, in absolute value, on learning outcomes than time spent in less focused "chatting".

Summary

Facebook has many advantages to recommend its use as a discussion forum for academic instruction. Among the advantages are that all students, with few exceptions, have a Facebook account and know how to use it, and Facebook users have easy and frequent access through apps installed on their mobile devices.

Using Facebook in an academic setting can have disadvantages. Some instructors may experience a steep learning curve. Our paper seeks to flatten that curve. We explain the "how to" of creating discussion threads on a secret group wall. We show how the discussion thread on the group wall allows the instructor to have the posts easily grouped and allows easy access to review and evaluate student contributions to the discussion topic. Also, when using Facebook for instructional purposes, students should be coached on the appropriate settings for the privacy controls. Best practice suggests that the privacy setting for items with high personal information content such as discussion on the wall and the personal information tab be set for viewing by "friends only". However our data show that more than 50 % of the student respondents do not invoke this privacy setting.

Overall our empirical estimates provide cautious support for the hypothesis that active participation in the discussion board has a positive effect on exam score at a statistically significant level. The OLS estimates of the effect of posts related to question and answer dialogue show a positive impact on the cumulative final exam score at a 5 % level of statistical significance. This result is consistent with the view of that using Facebook in academic instruction can be an effective tool for assisting the average student to resolve questions about the course material and for promoting peer-to-peer learning. An analysis of Facebook usage over the semester shows that the positive effect does not take hold immediately. In our data the positive effects were only exhibited in the last third of the semester. This finding suggests a need for upfront coaching of how to participate in. and benefit from the discussion. A disconcerting finding reported for the last third of the semester was that a high volume of posting was negatively associated with the hourly exam scores. This finding is consistent with the behavior of general "chatting" distracting students from studying. Though the numerical effect was smaller in absolute value that the other positive effects, it is a reason to coach students to focus on the academic purpose of the discussion forum.

The findings are suggestive of both the promise and peril of using social media in the academic classroom. The sample size of this study is relatively small and further research with larger samples is needed to determine if these findings will hold up in different institutional settings of higher education.

Published online: 23 December 2013

[c] International Atlantic Economic Society 2013

References

Anstine, J., & Skidmore, M. (2005). A small sample study of traditional and online courses with sample selection adjustment. Journal of Economic Education, 36(2), 107-127.

Boatman, K., Courtney, R., & Lee, W. (2008). See how they learn: the impact of faculty and student learning styles on student performance in introductory economics. American Economist, 52(1), 39-48.

Borg, M. O., & Shapiro, S. L. (1996). Personality type and student performance in principles of economics. Journal of Economic Education, 27(1), 3-25.

Bosch, T. E. (2009). Using online social networking for teaching and learning: Facebook use at the University of Cape Town. Communication, 35(2), September 2012-185-200. Retrieved from doi: 10.1080/02500160903250648.

Chang, G. (2008). Facebook can be a factor in admissions, employment--Hopkins uses social sites in admissions decisions. The Johns-Hopkins Newsletter, 9/24, 11/11/2013.

Charkins, R. J., O'Toole, D. M., & Wetzel, J. N. (1985). Linking teacher and student learning styles with student achievement and attitudes. Journal of Economic Education.

Coates, D., & Humphreys, B. R. (2004). No significant distance between face-to-face and online instruction: evidence from principles of economics. Economics of Education Review, 23(6), 533-546.

Durham, Y., McKinnon, T., & Schulman, C. (2007). Classroom experiments: not just fun and games. Economic Inquiry. 45(1), 162-178. doi: 10.1111/j.1465-7295.2006.00003.x.

Emerson, T. L., & Taylor, B. A. (2007). Journal of Economic Education, 38(1), 18-35.

Figlio, D. N., Rush, M., & Yin, L. (2010). Is it live or is it internet? Experimental estimates of the effects of online instruction on student learning. (Working Paper No. 16089). NBER Working Paper Series.

Gratton-Lavoie, C., & Stanley, D. (2009). Teaching and learning principles of microeconomics online: an empirical assessment. Journal of Economic Education, 40(1), 3-25.

Haefner, R. (2012). More employers screening candidates via social networking sites. Message posted to http://www.carcerbuilder.com/Article/CB-1337-Getting-Hired-More-Employers-Scrceening-Candidates-via-Social-Networking-Sites.

Junco, R. (2012). The relationship between frequency of Facebook use, participation in Facebook activities, and student engagement. Computers and Education, 58(1), 162-171.

Kirschner, P. A., & Karpinski, A. C. (2010). Facebook and academic performance. Computers in Human Behavior, 26(6), 1237-1245.

Marburger, D. R. (2006). Docs mandatory attendance improve student performance? Journal of Economic Education, 37, 148-155.

McBride, M. (2008). Classroom2.0: avoiding the creepy treehouse. Retrieved 9/20, 2012, from http://melanicmcbride.net/2008/04/26/creepy-treehouse-v-digital-literacics/.

Munoz, C. & Towner, T. (2009). How to use Faccbook in the college classroom. In L. Gibson (Ed.), Proceedings of society for information technology & teacher education international conference 2009 (pp. 2623-2627). Chesapeake.

Pellizzari, P. (2012). Facebook as an academic learning platform: A case study of mathematics. (Working Paper No. 01). Ca' Foscari University of Venice, Italy: Department of Economics Working Paper.

Stanca, L. (2010). The effects of attendance on academic performance: panel data evidence for introductory microeconomics. Journal of Economic Education, 37(3), 251-266.

VARK, (2013.). The VARK modalities. Message posted to http://www.vark-learn.com/english/page.asp?p=categories.

DOI 10.1007/s ll294-01 3-9453-9

O. R. Harmon (*)* W. T. Alpert * J. Histen

Department of Economics, University of Connecticut, 365 Fairfield Way, Unit 1063, Storrs, CT 06269-1063. USA

e-mail: harmon@uconn.edu
COPYRIGHT 2014 Atlantic Economic Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:R. Harmon, Oskar; T. Alpert, William; Histen, Joseph
Publication:International Advances in Economic Research
Article Type:Report
Geographic Code:1USA
Date:Feb 1, 2014
Words:5358
Previous Article:Is it Abenomics or post-disaster recovery? a counterfactual analysis.
Next Article:Does labor market status influence self-assessed health?
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters