Embedding formative assessment features into LabView interface

Embedding o ma i e assessmen ea u es in o LabView in e aces REV 2007 - www. e -con e ence.o g 1 Embedding o ma i e assessmen ea u es in o LabView in e aces Max Con adi1, Reinha d Langmann1, An ónio M. Ca doso2, J. M. Fe ei a2 1 Uni e si y o Applied Sciences, Dusseldo , Ge many 2 Faculdade de Engenha ia, Uni e sidade do Po o, Po ugal Abs ac — Vi ual ins umen s (VI), namely hose designed using LabView, a e inc easingly used o enable access o emo e ins umen s o e he web. In an academic con ex , such VIs may be used o build a emo e wo kbench ha enables he s uden s o ca y ou a labo a o y ask om hei homes. Remo e wo kbenches, o simula ion en i onmen s, a e no mean o eplace eal labs, bu a he o complemen hem. They enable he s uden s o con inue hei wo k in he lab, e en a e lea ing he uni e si y campus. In mos cases, a emo e wo kbench cons i u es a eplica o wha he s uden sees in a eal wo kbench. Howe e , VI in e aces may be designed o p o ide a much iche educa ional expe ience, namely by embedding o ma i e assessmen ea u es, as desc ibed in his wo k-in- p og ess pape . Index Te ms— e-lea ning, o ma i e assessmen , online labs, LabView. I. INTRODUCTION The objec i e o his wo k is o design embedded assessmen ea u es in o common VIs, used o build emo e wo kbenches in academic con ex s. An embedded assessmen ea u e may be seen as a VI ex ension ha cap u es and s o es o ma i e assessmen in o ma ion, desc ibing how a s uden handles an ins umen . This in o ma ion will be made a ailable o he s uden and o i s eache , and may be used o iden i y a eas whe e he s uden skills / knowledge need o be imp o ed. The pic u e below illus a es a emo e wo kbench ha was de eloped o suppo labo a o y assignmen s in digi al / analogue elec onics. This emo e wo kbench comp ises an oscilloscope, a wa e o m gene a o , and a mul ime e , besides li e images om he lab. Figu e 1. A emo e wo kbench suppo ing basic elec onics expe imen s The s uden s unning an expe imen will ha e o adjus he wa e o m gene a o and he oscilloscope. I his was done in a eal lab, wi h a u o nea by, he/she would be able o assess i he s uden knew how o handle hese ins umen s, jus by wa ching how he s uden ca ied ou his i s s ep. The objec i e o his wo k is o s udy how a he same conclusion can be de i ed by au oma ically moni o ing how he s uden handles he VIs. II. REFERENCE MODEL The PTSE lab a FEUP al eady uses a ious VIs ha a e used o design emo e expe imen s in elec onics. All such VIs we e de eloped using LabView. The p oposed app oach consis s o analysing he LabView sc ip s o such VIs and de ising ways o iden i y and cap u e he ele an embedded assessmen in o ma ion. In he ini ial phase o his wo k, a simple VI ep esen ing a basic ins umen (e.g. a ol me e ) was used o look o an app op ia e de elopmen model and i s implemen a ion s a egy. Only a ew s uden ac ions a e ele an om he assessmen poin o iew. One o he i s ques ions o be conside ed was he e o e how o iden i y wha ac ions should be logged in o he assessmen epo . On he o he hand, a gi en ac ion may o may no be ele an , depending on he ope a ing condi ions a ha momen , and on wha he s uden may be ying o achie e. The ules ha iden i y when a gi en ac ion is ele an o assessmen pu poses a e he e o e dynamic and mus be con inuously upda ed as he expe imen p og esses. Le ’s conside he example o an oscilloscope VI whe e one channel is being used o obse e a gi en inpu signal. The V/di and ime base bu ons de ine he display condi ions, and need o be adjus ed by he s uden in acco dance wi h he ampli ude and equency o he inpu signal. A 5 V / 1 KHz sine wa e may be isualised by adjus ing he display channel o 2 V/di and he ime base o 100 μs/di , al hough he e a e o he se ings ha migh as well be selec ed in his case (e.g. he same V/di scale and a 50 μs/di ime base, o any o he neighbou ing ampli ude and ime base scales). The accep able anges o V/di and ime base se ings a e he e o e dependen on he inpu signal, and may be ep esen ed as shown in igu e 2, whe e each ange is de ined by a lowe and a highe limi , de ining a do ed a ea ha ep esen s he success space. No ice ha he success space may be n-dimensional in mo e complex examples, o uni-dimensional in e y simple cases (such as he ol me e VI e e ed abo e). Embedding o ma i e assessmen ea u es in o LabView in e aces REV 2007 - www. e -con e ence.o g 2 Ampli ude scale Time base scale Time / di lowe bounda y Time / di highe bounda y V / di highe bounda y V / di lowe bounda y Figu e 2. Gene ic ep esen a ion o a possible success space An assessmen e en is igge ed when he s uden s a e asked o calib a e he oscilloscope channel. Reaching he success space (and s aying he e) in less han 20% o he ime a ailable o aking mo e han 100% o ha ime o ge he e, will gene a e o ma i e assessmen in o ma ion, which will be added o he assessmen epo . Any e en ha adds in o ma ion o he assessmen epo is a alid assessmen e en . When he s uden eaches he goal o he assessmen e en wi hin 20%-100% o he ime a ailable, no assessmen in o ma ion is p oduced. I he s uden is asked o calib a e he display channel bu is no able o selec V/di and ime base scales wi hin a gi en ime pe iod, his may mean ha his/he skills need o be imp o ed. On he o he hand, i hey each hei goal e y quickly (e.g. in less han 20% o he ime a ailable), hen his may be indica i e o a e y good s uden . In bo h cases, he assessmen e en should p oduce assessmen in o ma ion. The ac ions pe o med by he s uden s (i.e. any in e media e se ings selec ed while ying o calib a e he oscilloscope) du ing a alid assessmen e en shall also be eco ded in he assessmen epo . Ampli ude scale Time base scale Time / di lowe bounda y Time / di highe bounda y V / di highe bounda y V / di lowe bounda y S a poin Fi s ac ion (ano he V/di se ing selec ed) Las ac ion be o e 100% o ime elapsed Figu e 3. Failing o each a success ul ou come The example on igu e 3 illus a es a case whe e he s uden ails o achie e he goal o calib a ing he oscilloscope channel, wi hin he ime a ailable o do so. No ice ha his conclusion is ques ionable, since he/she eached he success space in ime, and migh jus ha e modi ied he V/di scale o show some hing o a colleague. Howe e , and since many assessmen e en s will be igge ed as he expe imen p og esses, imp obable si ua ions as he one jus desc ibed a e no expec ed o be s a is ically meaning ul. III. TECHNICAL IMPLEMENTATION The cases conside ed in he p e ious pa ag aphs indica e he need o de elop embedded assessmen e e ence models, able o p o ide an abs ac ep esen a ion o he equi ed assessmen ea u es. Such e e ence models will be speci ic o each VI, desc ibing hei own n-dimensional success spaces, indica ing wha (dynamic) in o ma ion de e mines he success space bounda ies, he ime pe iod associa ed wi h each assessmen e en , he in o ma ion o be added o he assessmen epo in case o a alid ins ance, e c. Once a e e ence model ha desc ibes he embedded assessmen ea u es o a gi en VI is eady, i is possible o s a i s echnical implemen a ion. Since he VIs ha a e a ailable a he PTSE lab we e buil using LabView, he eques ed embedded assessmen ea u es is bes done in he o m o a LabView VI assessmen ex ension. Assessmen ex ensions should be de eloped in modula o m, meaning ha each VI may exis wi h o wi hou hem. Embedded assessmen ex ensions a e no necessa ily isible a he VI in e ace le el, bu he s uden should always be awa e o hei p esence. Figu e 4. A VI and i s assessmen ex ension Ou i s implemen a ion goal consis ed o eco ding all s uden ac ions and ma king all s uden -d i en e en s, as shown in igu e 4. The assessmen ex ension adds a con ol panel a he o iginal VI bu on, enabling he u o o iden i y hose momen s in ime whe e meaning ul e en s may be loca ed. The u o is able o eplay a no mal speed he whole expe imen , o o as o wa d a a ious speeds. He/she may also jump di ec ly o he momen whe e a gi en e en is ma ked. IV. CONCLUSION The cu en implemen a ion is able o eco d all s uden ac ions and o iden i y he possible meaning ul momen s o assessmen pu poses, bu is s ill missing he au oma ic assessmen ea u es. The inal p o o ype will be able o help u o s and s uden s o ob ain o ma i e assessmen in o ma ion, which can be used o iden i y any a eas whe e u he aining is equi ed. No ice ha eco ding s uden ac ions may aise e hical issues which ha e o be discussed be o ehand and he s uden s mus be awa e o i s p esence. In pa icula , i is impo an o ensu e ha all s uden s a e awa e o he embedded assessmen ea u es and in ag eemen wi h hei use. Embedding o ma i e assessmen ea u es in o LabView in e aces REV 2007 - www. e -con e ence.o g 3 REFERENCES [1] h p://www.ni.com/lab iew/ ( isi ed on Ap il 25 h 2007) [2] I. Gus a sson, J. Zack isson, H. Åkesson, L. Håkansson, I. Claesson, T. Lagö, "Remo e Ope a ion and Con ol o T adi ional Labo a o y Equipmen ." In e na ional Jou nal o Online Enginee ing (iJOE), a ailable a h p://www.i-joe.o g/ojs/ iewa icle.php?id=43 ( isi ed on July 4 h 2006). [3] Ko, C.C.; Chen, Ben M.; Chen, Jianping, C ea ing Web-based Labo a o ies (Ad anced In o ma ion and Knowledge P ocessing), Sp inge , ISBN 1852338377, 2004. [4] J. Ma, J. V. Nicke son, “Hands-On, Simula ed, and Remo e Labo a o ies: A Compa a i e Li e a u e Re iew,” ACM Compu ing Su eys, Vol. 38, No. 3, A icle 7, Sep embe 2006. [5] D. Gille , A. V. Nguyen, Y. Rekik, “Collabo a i e Web-Based Expe imen a ion in Flexible Enginee ing Educa ion,” IEEE T ansac ions on Educa ion, Vol. 48, No. 4, No embe 2005. AUTHORS M. Con adi is wi h he Uni e si y o Applied Sciences, Depa men o Elec ical Enginee ing, Jose -Gockeln- S asse 9, D-40474 Duesseldo (e-mail: maxcon [email protected] ). R. Langmann is wi h he Uni e si y o Applied Sciences, Depa men o Elec ical Enginee ing, Jose - Gockeln-S asse 9, D-40474 Duesseldo , (email: einha d.langmann@ h-duesseldo .de). A. M. Ca doso, is wi h he Faculdade de Engenha ia, Uni e sidade do Po o, Rua D . Robe o F ias, 4200-465 Po o, Po ugal (e-mail: admp se@ e.up.p ). J. M. Ma ins Fe ei a, is wi h he Faculdade de Engenha ia, Uni e sidade do Po o, Rua D . Robe o F ias, 4200-465 Po o, Po ugal (e-mail: jm @ e.up.p ).