Evaluating the English-learning of engineering students using the Grey S-P chart: a Facebook case study in Taiwan

Transcription

1 Volume 3, Number, 0 WIETE 0 Global Joural of Egeerg Educato Evaluatg the Eglsh-learg of egeerg studets usg the Grey S-P chart: a Facebook case study Tawa Bor-Tyg Wag, Ta-We Sheu & Naga Masatake Natoal Tachug Uversty of Educato Tachug, Tawa ABSTRACT: The am of ths artcle was to evaluate the Eglsh-learg of egeerg studets by usg the Grey Studet-Problem chart (GSP chart). The research was coducted wth 37 freshma egeerg studets a prvate uversty cetral Tawa usg Facebook wall posts to make Eglsh seteces, ad the GSP chart to evaluate the egeerg studets Eglsh performace. The two vocabulary tests, based o the Geeral Servce Lst (GSL), wth te questos were gve as the pre-test ad post-test. To ucover the usefuless of learg vocabulary o Facebook, the vocabulary tests scores were compared usg the GSP chart. The fdgs suggest Tawaese egeerg studets performed better o the Eglsh vocabulary examato after practsg makg Eglsh seteces o Facebook. The Web-based tool provdes a ubqutous Eglsh-learg evromet for egeerg studets, ad s beleved to crease studets Eglsh vocabulary acqusto, as well. Keywords: Grey Studet-Problem chart, Facebook, Geeral Servce Lst, ubqutous Eglsh-learg, Eglsh vocabulary acqusto INTRODUCTION Eglsh has become more mportat Tawa gve the tred of globalsato ad teratoalsato. Therefore, the Mstry of Educato (MOE) Tawa put a lot of effort to mprovg studets Eglsh sklls [-3]. The MOE also otced that uversty studets Eglsh performace s worse tha the youger geerato [4]. Hece, ths artcle ams to apply Facebook to Eglsh-learg amog freshma egeerg studets at a prvate uversty cetral Tawa. It s hoped that through makg Eglsh seteces o Facebook, the egeerg studets could ot oly share ther deas, but also crease ther Eglsh vocabulary acqusto. Also, wth ths ubqutous Eglsh-learg evromet, the teacher ad studets ca share ther Facebook posts o matter the tme or locato [5]. Studets were dvded to sx groups, ad eeded to work wth ther groups to fsh the assgmet. Through the collaboratve learg experece, the team members were expected to geerate more deas ad reduce the stress of dog the task [6][7]. Moreover, studets today are dgtal atves, whle teachers are dgtal mmgrats [8]. If teachers could teach studets the way they prefer to lear, t may help mprove attamet [8][9]. Ths project was, therefore, developed to focus o:. How egeerg studets egagemet wth Facebook flueced ther Eglsh vocabulary learg.. Evaluato of egeerg studets Eglsh progress a quattatve way. METHOD Facebook as a Learg Platform Sce early 004, the veto of the Socal Networkg Ste (SNS), Facebook, has flueced learg styles hgher educato [7][9]. Accordg to Kwog, the geeral cocept of Facebook was to crease socal etworkg ad teractos amog Harvard freshma studets, ad later 005, Facebook was opeed up to the people outsde the uversty etwork, so that everyoe the world could use t to frame ther profles, share photos, aouce upcomg evets, jo terestg groups, etc [0]. Compared wth other learg platforms, lke Moodle, whch s more teacher-cetred, Facebook has become a popular platform for studets to post, share or provde feedback wth just oe clck [][]. Some have reservatos that ths kd of socal etworkg platform dstracts studets from ther academc studes, ad, ay evet, the questo s asked: s t useful the formal classroom settg []? However, f teachers could utlse ths coveet ad user-fredly Web tool, desg sutable currcula, ad evaluate studets learg a objectve 5

2 way, studets may crease ther motvato for class partcpato [3]. For stace, both the structor ad the studets could check the bullet board, sed e-mals or messages ad share vdeo clps o Facebook. Udoubtedly, ths dgtal atves geerato seems to be more terested the Web.0 evromet (.e. blogs, You Tube, Twtter) tha the tradtoal classroom. Therefore, teachers eed to cosder modfyg ther teachg methods ot oly to ft studets pace of learg, but also to create a studet-cetred learg evromet [4]. To date, the commucato betwee studets ad teachers has bee doe va e-mal, but ths seems to be more prvate ad other studets do ot share deas wth ther peers. I order to crease the teractos betwee structor-studet or studetstudet, educators may wat to apply a ew techology teachg, ad to set up ther ow Facebook stes to keep touch wth the studets a ubqutous Eglsh-learg evromet, where studets ca browse the Eglsh-related artcles provded by the teacher o the Iteret ad the structor ca reply publc [5]. GSP Establshmet Through Grey Relatoal Aalyss Ths secto gves a bref troducto to establshg the GSP chart through Grey Relatoal Aalyss (GRA) [5-].. Raw data establshmet: x = x (), x (),, ; k,,..., m 0 ( = L Referece Vector () x = ( x (), x (), x = ( x (), x (), x = ( x (), x (), x = ( x (), x (), M M =,,..., Comparatve Vector (). Geerato of the grey relato: try to stadardse the data by usg three methods larger-the-better, smaller-thebetter, ad omal-the-better. I ths artcle, the method of larger-the-better s used, ad t s show as follows: x ( k) m x ( k) x ( k) = max x ( k) m x ( k) (3) where max ( k) x meas the maxmum umber j ad m ( k) x meas the mmum umber j. 3. Calculato of the grey relatoal: apply Naga s equato to calculate the grey relato [6-8]. Amog them, the referece vector of the local GRA s X 0, ad the comparatve vector s X j. Whe Γ0 s close to, t meas that X 0 ad X j are hghly correlated. The equato s show as follows: Δ Γ0 = Γ( x o( k), x ( k)) = Δ max max Δ Δ 0 m Δ ρ ρ 0 = 0 = ( [ Δ ρ 0 ( k)] ) k= x (4) where GSP Chart Theory Δ max ad Δ m represet Δ 0 s maxmum ad mmum, respectvely. The Grey Studet-Problem chart (GSP chart) orgated from the Studet-Problem chart (SP chart), whch was proposed by Takahro Sato the 970s. Based o the SP chart, t s possble to aalyse the data a quattatve way, ad to dagose studets learg through the calculato of the Problem Cauto Idex (CP) ad Studet Cauto Idex (CS) []. However, the SP chart oly codes the data as ether 0 (correct aswer) or (correct aswer), but t fals to reflect the actual values the scorg []. Therefore, Naga Masatake combed GRA ad the S-P chart to establsh the GSP chart, whch allows for a more specfc aalyss of the data [6]. Compared to the SP chart, the GSP chart ca deal wth ucerta, multvarate, ad dscrete data through mathematc equatos, ad the preset the results as fgures [6][9-]. The GSP also combes the Rasch model to create measures, ad the equato s show as follows [9-][3]. Pr( X c = θ ) = c + (5) + exp[.7a ( θ b )] 5

3 where θ represets test-takers abltes; a s the tem dscrmato parameter; b s the tem dffculty parameter; c s the tem guessg parameter. Pr( X = θ ) s the probablty of a test taker (ablty =θ ) to aswer the questo correctly []. The tem dffculty parameter, pots o the theta cotuum, correspods to a 50% probablty of edorsg the tem []. RESEARCH DESIGN Partcpats A total of 37 freshma egeerg studets the Freshma Eglsh course took the Eglsh vocabulary pre-test ad post-test (based o the GSL words). The GSL s a set of,000 words selected (based o the frequecy of usage) to be of the greatest geeral servce to Eglsh learers. The egeerg studets ages raged from 8 to 0, ad most of them have leared Eglsh for at least sx years before takg part ths research. Ther Eglsh level s defed as CEF A, after takg the ufed placemet test (Stadford test) at the begg of the academc year. Data Collecto ad Aalyss All of the egeerg studets the same Freshma Eglsh course took the vocabulary pre-test at the begg of the course. As for the Facebook assgmet, the structor had to create her ow profle page o Facebook frst, ad the group page for the egeerg class. The, the egeerg studets were asked to regster o Facebook ad jo the group Egeerg B, whch was admstrated by the structor. Next, the egeerg studets were dvded to sx groups, ad each group was assged wth a 4-word Eglsh vocabulary (based o the GSL), ad each group had to upload pctures related to the meag of the assged words to Facebook, ad make a related Eglsh setece below each pcture every week. At ths stage, studets could choose to take ther ow pctures or get the pctures o-le. The other groups were ecouraged to gve commets o each pcture ad setece o Facebook, or at least they eeded to choose lke or dslke o Facebook. After sx weeks practse, the partcpats were asked to take the vocabulary post-test to check ther Eglsh vocabulary progress. The examples of Facebook wall posts are show Fgure ad Fgure. Fgure : Group work o Facebook (Example I). Fgure : Group work o Facebook (Example II). 53

4 The studets ths research were asked to take the pre-test ad post-test vocabulary tests class to detfy ther profcecy levels, ad check f ay dffereces could be foud betwee the two tests. Moreover, ther test scores were collected ad aalysed usg the GSP chart. Establshg the GSP Chart By codg the examato results of the pre-test ad post-test to the S-P chart, ad the applyg MATLAB 7.0 to aalyse the data [6][9][0], the GSP fgures are show as Fgure 3 ad Fgure 4. Fgure 3: The GSP of Eglsh vocabulary - pre-test. ANALYSIS: RESULTS AND DISCUSSION Fgure 4: The GSP of Eglsh vocabulary - post-test. Based o the Gamma value Fgure 3 ad Fgure 4, egeerg freshma studets ca be clustered to four groups accordg to ther abltes, ad the results are show Table [9][0][3]. Table : Clustered group. Group Examato Gamma value performace A Hgh Gamma B Itermedate-hgh 0.75 Gamma 0. 5 C Itermedate 0.5 Gamma 0. 5 D Low 0.5 Gamma Obvously, there are two types of studets whose Eglsh ablty s above 0. 5 the post-test tha the pre-test. That s, egeerg studets beeft from learg vocabulary o Facebook, ad the progress s show Fgure 3 ad 54

5 Fgure 4. Moreover, the slope Fgure 4 s greater tha Fgure 3, whch ca be terpreted as havg a more dscrmatg a, ad ths meas most of the egeerg studets performed better the post-test. However, ths Web-based learg may ot be deal for low-level studets sce there are stll 6 studets whose abltes (Gamma value) are below 0. 5 the post-test. Ths s probably because the low level studets are easly dstracted by other games, pop-out ads or terestg vdeo clps whle surfg the Iteret. Or they may feel solated, frustrated ad axous whle dog the o-le assgmet []. Therefore, t s suggested that the teacher could supervse these studets ad provde clearer structos to let them uderstad the advatages of o-le learg. CONCLUSIONS The results of the study dcated that most egeerg studets mproved ther Eglsh vocabulary, after sharg ther posts o Facebook. Some studets eve corrected others errors, ad gave feedback o Facebook. Studets teded to provde ad share more deas o ths socal etworkg ste tha the tradtoal Eglsh taught classroom. I the classroom, whe the structor asked questos, few studets were wllg to aswer. However, whe there were ew posts o Facebook, there were commets gve uder the posts rght away. Most studets repled o Facebook, ad showed that they ejoyed the flexblty of the o-le learg evromet, ad more ovatve deas were smulated, whch also facltated learg []. Accordg to Marato, Facebook emphasses categores ad aspects of popular culture that teeagers fd mportat [4]. Hece, teachers could combe Facebook wth Eglsh learg to crease studets learg motvato. However, teachers eed to provde clear structos to esure the academc usage of Facebook stead of the socal oe. The study used the GSP chart to aalyse egeerg studets performaces a quattatve way. Studets abltes were clustered as show clearly the fgures, whch are better tha the orgal SP chart. Wth the GSP chart, ucerta, multvarate, ad dscrete data could be coducted [6]. Also, t s easer for teachers to check dvdual studets progress usg the fgures ad, the, teachers could ot oly provde remedal structos, but also adjust the currculum desg to sut the egeerg studets eeds [5][6]. Fally, there were oly 37 egeerg studets volved ths research, ad t s suggested that future research could use a larger pool of partcpats. Ths stude oly used quattatve measures to evaluate the data, but t could be better, f there was a qualtatve approach, such as tervews or questoares to exame the studets profcecy levels. REFERENCES. Bray, M. ad Lee, W.O., Educato ad Poltcal Trasto: Themes ad Expereces East Asa. Hog Kog: Hog Kog Uversty Press (00).. Tawa Mstry of Educato, Educato ma Polcy (008), 0 Jue 0, 3. Vohra, P., Kasuba, R. ad Vohra, D.P., Preparg egeers for a global workforce through currcular reform. Global J. of Egg. Educ., 0,, 4-48 (006). 4. Chag, W.C., Promotg Eglsh profcecy at colleges ad uverstes Tawa: polces ad strateges. J. of Educatoal Resources ad Research, (007). 5. Hseh, H.C., Che, C.M. ad Hog, C.M., Cotext-aware ubqutous Eglsh learg a campus evromet. Proc. 7th IEEE Iter. Cof. o Advaced Learg Techologes, (007). 6. Göl, Ö ad Nafalsk, A., Collaboratve learg egeerg educato. Global J. of Egg. Educ,,, (007). 7. Madge, C., Meek. J., Welles, J. ad Hooley, T., Facebook, socal tegrato ad formal learg at uversty: t s more for socalsg ad talkg to freds about work tha for actually dog work. Learg, Meda ad Techol., 34,, 4-55 (009). 8. Presky, M., Dgtal atves, dgtal mmgrats. O the Horzo, 9, 5, -6 (00). 9. Seddg, Z.S., Capretz, L.F. ad House, D., A multcultural comparso of egeerg studets: mplcatos to teachg ad learg. J. of Socal Sceces, 5,, 7- (009). 0. Kwog, V., Reach out to your studets usg MySpace ad Facebook. Idaa Lbrares, 6, 3, (007).. Roblyer, M.D., McDael, M., Webb, M., Herma, J. ad Wtty, J.V., Fdgs o Facebook hgher educato: a comparso of college faculty ad studet uses ad perceptos of socal etworkg stes. Iteret ad Hgher Educato, 3, (00).. Tolver, F., My studets wll Facebook me but wo t keep up wth my o-le course: the challeges of ole structo. Amerca Commucato J., 3, 59-8 (0). 3. Tsao, C., Che, I. ad Tsa, J., A acto research o promotg ursg studets Eglsh-learg motvato through Hybrd Web-Based Istructo. 00 Iteratoal Symp. o ESP, -8 (00). 4. Pccol, G., Ahmad, R. ad Ives, B., Web-based vrtual learg evromets: a research framework ad a prelmary assessmet of effectveess basc IT sklls trag. MIS Quarterly, 5, 4, (00). 5. Huag, S.J., Itegrato of the grey relatoal aalyss wth geetc algorthm for software effort estmato. Europea J. of Operatoal Research, 88, 3, (008). 6. Naga, M., Hadouts of Graduate Isttute of Educatoal Measuremet ad Statstc at NTCU (00) (upublshed). 55

6 7. Yamaguch, D., G.D. ad Naga, M., New grey relatoal aalyss for fdg the varable structure ad ts applcatos. J. of Grey System, 8,, (005). 8. Yamaguch, D., G.D. ad Naga, M., Verfcato of effectveess for grey relatoal aalyss models. J. of Grey System, 0, 3, 69-8 (007). 9. Sheu, T.W., Wag, B.T, Lag, J.C., Tzeg, J.W. ad Masatake, N., Applyg grey S-P chart evaluatg the mportace of college Eglsh wrtg lexcal usage. Proc. of INTED 0 Cof., (0). 0. Sheu, T.W., Tzeg, J.W., Lag, J.C., Wag, B.T. ad Masatake, N., A proposal of three parameters logstc model based o Rasch model GSP. Proc. 3rd Cof. o Techol. ad Mathematcs Educ., (0).. Sheu, T.W., Tzeg, J.W., Lag, J.C., Wag, B.T. ad Masatake, N., A study of the mportat dcator of e-learg courses usg GSP chart. Iter. Cof. o Iformato Maagemet (0).. Yu, S.C. ad Yu, M.N., The relatoshps amog Idces of dagostc assessmets of kowledge structures ad S-P Chart aalyss. J. of Educ. & Psychology, 9,, (006). 3. Sheu, T.W., Wag, B.T, Lag, J.C., Tzeg, J.W. ad Masatake, N., Applyg grey S-P chart to aalyze the Eglsh lsteg performaces amog college studets. Proc. 5th Cof. o Grey System Theory ad Applcatos, A07-A4 (00). 4. Marato, G. ad Barto, M., Paradox ad promse: MySpace, Facebook, ad the socopoltcs of socal etworkg the wrtg classroom. Computers ad Composto, 7, (00). 5. Uzak, J., Oladra, M.T. ad Moalos, R., Revewg ad developg course descrptos to comply wth accredtato requremets. Global J. of Egg. Educ.,,, (00). 6. Sheu, T.W., Tzeg, J.W., Lag, J.C., Wag, B.T. ad Masatake, N., A study of the mportat dcator of e-learg courses usg GSP chart. Iter. Cof. o Iformato Maagemet (0). BIOGRAPHY Bor-Tyg Wag s a lecturer the Foreg Laguage Ceter at Feg-Cha Uversty. She s curretly a PhD studet the Graduate Isttute of Educatoal Measuremet ad Statstcs at the Natoal Tachug Uversty of Educato, Tachug, Tawa. Her research terests clude educatoal measuremet ad Eglsh laguage teachg. She s the drector of the TKIA (Tawa Kase Iformato Assocato). Ta-We Sheu receved a PhD degree Mathematcs from the Natoal Osaka Uversty, Japa, 990. He s the Dea of the College of Educato ad a professor at the Graduate Isttute of Educatoal Measuremet at the Natoal Tachug Uversty of Educato, Tachug, Tawa. Hs studes focus o Item Respose Theory (IRT), educatoal measuremet ad e-learg. He s the drector of the TKIA (Tawa Kase Iformato Assocato). Masatake Naga receved hs Master s degree egeerg from Touka Uversty Japa 969. He was also a researcher (Dr Matsuo research) at the Tohoku Uversty, whle workg toward hs PhD egeerg. From 989, he worked at the Tekyo Uversty Departmet of Scece ad Egeerg as a assstat professor ad, evetually, as a egeerg professor. Now he s the Char Professor the Graduate Isttute of Educatoal Measuremet at the Natoal Tachug Uversty, Tachug, Tawa. Hs research terests clude approxmato, strategy system egeerg, formato commucato etwork techology, aget, Kase formato processg, grey system theory ad egeerg applcatos. He s also a member of Isttute of Electrocs, Iformato ad Commucato Egeers (IEICE). 56