IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

Community learning with a 24/7 chatusing smartphones

Kimmo Vehkalahti1 and Matti Nelimarkka21University of Helsinki, Finland

2Aalto University and University of Helsinki, Finlandhttp://www.helsinki.fi/people/Kimmo.Vehkalahti/

IASE 2015 Satellite ConferenceAdvances in statistics education:

developments, experiences, and assessmentsRio de Janeiro, Brazil | 22–24 July 2015

1 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

Outline

1. Introduction2. Approach

2.1 Course overview2.2 Participants2.3 Data and documentation2.4 Weekly schedule and Moodle2.5 Reports and assessments

3. Presemo software4. Outcomes5. Conclusions6. References

2 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

1. Introduction

I Blended learning combines traditionalclassroom teaching with various forms of ICT(information and communication technology)

I Efficient method for teaching and learning ofData Analysis

I We focus on a Data Analysis course where wehave used Moodle for sharing materials andenabling discussions

3 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

1. Introduction

I Students: ”Moodle is clumsy and too slow”I Faster channels for communications neededI Need to ask ”stupid questions” anonymouslyI Social media (Facebook etc.) problematic:

I Students who are not in Facebook, may not want to sign inI At least open groups might attract outsidersI Regulations of using social media at UniversityI Those platforms are not usually anonymous

4 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

1. Introduction

I We have tried to answer these needs by usingPresemo software

I Allows students to use a regular web page ontheir smartphone, laptop or other internetconnected devices

I Anonymous chat: instantly discuss, askquestions and answer them

I Independent of time and place, no need to log in

5 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2. Approach

2.1 Course overview2.2 Participants2.3 Data and documentation2.4 Weekly schedule and Moodle2.5 Reports and assessments

6 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.1 Course overview

Data analysis with Survo and SPSS (5 ECTScredits), 7 weeks of basic tasks:

I building a data set from a text fileI creating, re-coding and transforming variablesI working with univariate graphs and statisticsI forming and visualizing cross-tabulationsI performing elementary statistical testsI analyzing dependencies with simple regressions

7 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.1 Course overview

I Skills learned and practised using two statisticalsoftware programs:

I Survo (Sund, Vehkalahti & Mustonen 2012, Sund 2011,Mustonen 1992)

I SPSS (IBM 2015)

I Students may choose Survo or SPSS (or both).

I Emphasis mostly on Survo, implemented as anopen source package for R (R Core Team 2013,Ihaka & Gentleman 1996)

I SPSS still supported at the Faculty8 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.2 ParticipantsNowadays, everyone needs Data Analysis!Result: an interesting heterogeneity and richness:

I Typically about 80–100 participantsI Including ca. 20 major students of StatisticsI Most students from Faculty of Social Sciences

I Also: Behavioural Sciences, Arts, Science, Medicine,Agriculture and Forestry, Biological and EnvironmentalSciences

I Computer Science, Archeology, Chemistry, Mathematics,Geography, Physics, Psychology, Education, Languages, . . .

I Mixture of undergraduates and post-graduates9 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.3 Data and documentation

We only use real-world data sets:I Primarily obtained from Finnish Social ScienceData Archive (FSD), incl. questionnaires

I Also selected open-access data:I Prices and earnings (UBS)I Economic Freedom (Heritage Foundation)

Two important large-scale topics emphasized:1. Documentation of the work process2. Reproducible research(Vehkalahti 2005, Gentleman & Temple Lang 2007)

10 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.4 Weekly schedule and Moodle

Course organized on weekly basis (7 weeks):

I Lecture on Tuesday morning:I present the goals and exercises of the weekI demonstrate the required technical phases

I Computer class on Tue/Wed (3 alternatives)I Supervised by two assistant teachers together

I Independent work (rest of the week)I Here: the biggest need for community learning!

I Reports submitted to Moodle before MondayI Assessments (self+peer) on Monday

11 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.4 Weekly schedule and Moodle

Moodle is used throughout the course:I Delivering data sets, lecture files, and exercisesI Sharing students’ weekly reportsI Sharing summaries of self+peer reviewsI Reminding the students about deadlinesI Informing of extra topics related to the contentsI Discussing the course matters on a Q&A forum(not much used – too slow!)

12 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.5 Reports and assessments

Weekly reports: results and documentationI 6–8 pages of text, tables, graphs, and syntaxI Describe the work process so that otherparticipants could follow and repeat

I Reports openly available for the participantsafter submit

13 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.5 Reports and assessmentsAssessments: self+peer review

I Self-assessment of own reportI Peer-review of another reportI Instructions and criteria in MoodleI Peer reviewers chosen weekly by teacherI Fully transparent assessment (no blind review)I Technically implemented using e-forms (Eduix)I Numeric grades (0–5) and textual reviews:

I Six-stage, non-proportionate scale: 5 (excellent), 4 (verygood), 3 (good), 2 (satisfactory), 1 (adequate), and 0 (fail)

I Grade 0 (fail) used only if report not submittedI Grades justified by a free-form explanation

14 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.5 Reports and assessments

Summaries of assessment data shared weekly:

I Students’ assessments constitute a weekly data:I Two numerical grades (self+peer), see previous pageI Textual supplements of both grades

I Transformed from e-form (Excel format) intocompact summary by teacher (using Survo)

I Summary shared in Moodle on Wednesday(about 10 pages per week)

Final grades based on assessments and reports.

15 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

2.5 Reports and assessmentsView from assessment summary (here: no names)Helsingin yliopisto, sosiaalitieteiden laitos

Data-analyysi Survo- ja SPSS-ohjelmistoilla, kevät 2015

Itse- ja vertaisarvioinnit, työviikko 7 i = itsearviointi, v = (saatu) vertaisarviointi______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

i A1 (name removed) ( 5 ) Raportointi ja ajot onnistuivat, vaikka kaikkia tarvittaviakäskyjä en ehtinytkään suorittaa harjoitustunnilla. Tein tarvittavat ajot myöhemmin jamielestäni ne onnistuivat kivasti.

v A1 (name removed) ( 5 ) Raporttia on helppo seurata, ja tarvittavat tehtävät ja analyysiton tehty erinomaisesti. Kirjoittaja selvästi ymmärtää hyvin tuloksena ja SPSS:n salat.

i A3 (name removed) ( 4 ) Normiraportti, lisänä pari kuvaa selkeyttämään.

v A3 (name removed) ( 2 ) Raportin analyysit olivat omien sanojenkin mukaan hiemanpuutteellisia, ulkoasussa oli parannettavaa. Raportti eteni johdonmukaisesti, muttaulkoasun, tilastojen tulkinnan ja puutteiden vuoksi annan arvosanaksi vain tyydyttävän.

i A4 (name removed) ( 4 ) Tein työn huolellisesti, tuli joitakin tulkintavirheitä, jotkakyllä korjasin vielä viimeiseen versioon. Tein pyydetyt kuvat ja käytin syntaksia ohjeidenmukaisesti. Kuten aiemmin, pyrin valitsemaan omia muuttujia (en ohjeissa valmiiksiannettuja vaihtoehtoja). Pohdin tuloksia ja niiden merkitystä. Tällaista analyysia onkiinnostavaa tehdä jatkossakin!

v A4 (name removed) ( 5 ) Raporttia oli kiva lukea. Käskytkin näyttivät tutuilta, koskaitsekin tein samalla ohjelmalla. Tehtävät näyttivät olevan kaikki tehty ja raportointi olionnistunutta.

i A5 (name removed) ( 4 ) Raportti ja analyysit ovat hyvät ja kattavat. Viimeistäanalyysia olisi pitänyt vähän jatkaa.

v A5 (name removed) ( 4 ) Hyvä ja selkeä esitys. Sopivan vakavalla otteella tehty.

i A7 (name removed) ( 3 ) Sain tehtävät pääpiirteissään tehtyä ja ymmärsin mitä olintekemässä. Mielestäni analysoin tekemistäni melko onnistuneesti ja raportti on selkeä. Enkuitenkaan kaikkia R-komentoja toimimaan, jotta olisin pystynyt tekemäänvarianssianalyysin jatkotehtävät loppuun saakka, ja koska edellisviikolla koin ongelmiaPrices and Earnings -tehtävässä, en pystynyt tällä viikolla jatkamaan siitä pisteestäjohon olisin normaalitilanteessa jäänyt, ja näin ollen tehtävä jäi nyt siltä osinvajaaksi.

v A7 (name removed) ( 4 ) Selkeä raportti ja aika lailla kaikki vaiheet tehty. Kiinnostavaaoli lukea vaihteeksi SURVOlla tehtyä analyysiä. Vaikutti aika loogiselta. Kuvia/taulukoitaolisi voinut olla pari enemmän (vaikkapa esimerkkejä eri vaiheista), mutta kyllä työ jotällaisenaan oli informatiivinen. 16 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

3. Presemo software

Activities, schedule and deadlines: much needfor discussion inside and outside classroom!

I Moodle is not so good for (quick) discussions(students do not like it’s Q&A forum)

I Idea: use an anonymized chat insteadI Instantly discuss, ask questions and answer them

Take advantage of their smartphones!

17 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

3. Presemo softwareWeb-based participation system:

I Different activities (poll, voting, chat)

I Easy: no registrations or loginsI Anonymous: users are not identifiedI Simple: works on web browser on any device(smartphones, laptops, . . . )

18 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

3. Presemo softwareViews from Presemo chat of the 2015 course:

19 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

3. Presemo software

20 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

4. Outcomes

Interesting difference of two courses:

I 2014: almost 1000 messages (in 7 weeks)(mostly short, some very long)

I 2015: only about 100 messages (in 7 weeks)(almost all quite long!)(see the views on two previous pages)

21 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

4. OutcomesDistribution of daily messages in 2014:

I Activity follows the weekly scheduleI Chat most active just before the deadlinesI Two Sunday peaks: most demanding weeks

22 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

4. OutcomesMessages aggregated by hour in 2014:

I 24/7 nature of the course can be clearly seenI Chat most active from 9–10am (peer-reviewdeadline on Tuesdays)

23 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

4. Outcomes

Rough classification of message contents:

1. Questions related to Data Analysis or software2. Answers to the questions (few by the teachers)3. Relaxed messages for lifting spirits

(important for learning community!)

24 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

5. Conclusions

Presemo turned out to be a success:I Simple anonymized chat was a useful aidI Enabled discussions in fast pace 24/7I Decrease the stress before deadlinesI ”Stupid” questions easy to ask without shameI Easy to answer other students’ questionsI Community learning, encouragement, peersupport, students helping each other

”It’s not what the teacher does, it’s what thestudent does” (Biggs, 2003)

25 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

References (page 1/4)Biggs, J. (2003). Teaching for Quality Learning at University (2nd ed.). The

Society for Research into Higher Education & Open University Press.

Birnholds, J., Hancock, J. & Retelny, D. (2013). Tweeting for Class:Co-Construction as a Means for Engaging Students in Lectures. InProceedings of the SIGCHI Conference on Human Factors in ComputingSystems – CHI ’13, 797–800. doi:10.1145/2470654.2470767.

Bradner, E., Kellogg, W.A. & Erickson, T. (1999). The Adoption and Use of’BABBLE’: A Field Study of Chat in the Workplace. In Proceedings ofthe Sixth European Conference on Computer Supported CooperativeWork, 139–158. doi:10.1007/978-94-011-4441-4_8.http://dl.acm.org/citation.cfm?id=351698.

Deng, L. & Tavares, N.J. (2013). From Moodle to Facebook: Exploringstudents’ motivation and experiences in online communities. Computers& Education 68, 167–176.

Du, H., Rosson, M.B. & Carroll, J.M. (2012). Augmenting ClassroomParticipation through Public Digital Backchannels. In Proceedings of the17th ACM International Conference on Supporting Group Work –GROUP ’12, 155–164. doi:10.1145/2389176.2389201.http://dl.acm.org/citation.cfm?id=2379057.2379081.

26 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

References (page 2/4)Eduix (2015). Browser based application for creating and managing e-forms.

https://e-lomake.fi/web/briefly-in-english.html

EU (2015). European Credit Transfer and Accumulation System (ECTS).http://ec.europa.eu/education/tools/ects_en.htm

FSD (2015). Finnish Social Science Data Archive, Tampere, Finland.http://www.fsd.uta.fi/en/

Gentleman, R. & Temple Lang, D. (2007) Statistical Analyses and ReproducibleResearch. Journal of Computational and Graphical Statistics, 16, 1–23.

Harry, D., Green, J. & Donath, J. (2009). Backchan.nl. In Proceedings of the27th International Conference on Human Factors in Computing Systems– CHI 09, 1361–1370. New York, New York, USA: ACM Press.doi:10.1145/1518701.1518907.http://dl.acm.org/citation.cfm?id=1518701.1518907.

Heritage Foundation (2015). Index of the Economic Freedom.http://www.heritage.org/index/

HIIT (2015). Presemo: a participation platform.http://www.hiit.fi/presemo

27 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

References (page 3/4)IBM (2015). SPSS: a predictive analytics software.

http://www.ibm.com/software/analytics/spss/

Ihaka, R. & Gentleman, R. (1996). R: A Language for Data Analysis andGraphics. Journal of Computational and Graphical Statistics, 5, 299–314.

Kohen, A. (2014). Tumbling Political Theory. Politics,doi:10.1111/1467-9256.12051.

Moodle Community (2015). Moodle: an open-source learning platform.https://moodle.org

Mustonen, S. (1992). Survo – An Integrated Environment for StatisticalComputing and Related Areas. Survo Systems, Helsinki. http://www.survo.fi/books/1992/Survo_Book_1992_with_comments.pdf

Nelimarkka, M., Kuikkaniemi, K., & Jacucci, G. (2014). A Field Trial of anAnonymous Backchannel Among Primary School Pupils. In Proceedingsof the 18th International Conference on Supporting Group Work –GROUP ’14, 238–242. New York, New York, USA: ACM Press.doi:10.1145/2660398.2660399.http://dl.acm.org/citation.cfm?doid=2660398.2660399.

28 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

References (page 4/4)R Core Team (2013). R: A language and environment for statistical computing.

R Foundation for Statistical Computing, Vienna, Austria.http://www.R-project.org/

Sund, R. (2011). Muste - the R implementation of Survo. In the yearbook ofFinnish Statistical Society, pp. 133–146. http://www.survo.fi/muste/publications/sund2011_muste_yearbook.pdf

Sund, R., Vehkalahti, K. & Mustonen, S. (2012). Muste – editorial computingenvironment within R. Proceedings of COMPSTAT 2012, 20thInternational Conference on Computational Statistics, 27–31 August2012, Limassol, Cyprus. pp. 777–788.http://www.survo.fi/muste/publications/sundetal2012.pdf

UBS (2012). Prices and earnings. http://www.ubs.com/global/en/wealth_management/wealth_management_research/prices_earnings.html

Vehkalahti K. (2005). Leaving useful traces when working with matrices.Research Letters in the Information and Mathematical Sciences 8,143–154. Proceedings of the 14th International Workshop on Matricesand Statistics (IWMS 2005).

29 / 30

IASE

2015

Satellite

Conference

|Rio

deJaneiro

,Brazil|

22–2

4July

2015

Thank you for your attention!

30 / 30