FUZZY EVALUATING MANAGEMENT PERFORMANCE AND MARKETING STRATEGIES IN COMMUNITY COLLEGES

Transcription

1 Inernaional Journal of Innovaive Compuing, Informaion and Conrol ICIC Inernaional c01 ISSN Volume 8, Number 10B, Ocober 01 pp FUZZY EVALUATING MANAGEMENT PERFORMANCE AND MARKETING STRATEGIES IN COMMUNITY COLLEGES Yu-Lan Lee 1 Dian-fu Chang and Berlin Wu 3 1 Deparmen of Educaional Policy and Adminisraion, Naional Chi Nan Universiy, Taiwan aerylee@yahoo.com.w Graduae Insiue of Educaional Policy and Leadership, Tamkang Universiy, Taiwan 14066@mail.ku.edu.w (corresponden auhor) 3 Deparmen of Mahemaical Sciences, Naional Cheng Chi Universiy, Taiwan berlin@nccu.edu.w Received December 010; revised Sepember 011 ABSTRACT. Recenly, worry abou he worsen qualiy of communiy colleges has become public concerns in Taiwan. Wihin a high compeiive markeing, he managers in communiy colleges have been expeced for beer performance in heir eaching qualiy o arac poenial sudens. This sudy aims o apply fuzzy measuremen and propose an effecively soluion o improve he performance of communiy colleges. We design a fuzzy model o rerieve managemen feaures in communiy colleges and redefine heir performance indicaors. The resul reveals he imporan facors for beer performance, which are measured by fuzzy, are enrollmen, aendance, and saisfacion. Communiy colleges can selec beer markeing sraegies o evaluae heir curriculum design hrough sudens needs using fuzzy measuremen. Keywords: Performance managemen, Markeing sraegy, Communiy colleges, Performance indicaors 1. Inroducion. Many counries have promoed life-long learning o respond he changeable world for aduls. The relaed measures have also affeced he developmen of communiy colleges in Taiwan. In recen years, he esablishmen of communiy colleges has been viewed as he chance for reshaping he civil sociey. Under his circumsance, he communiy colleges were asked o face he accounabiliy in erms of review heir educaional goals, paricipans needs, and inernal curriculum design o achieve heir missions. However, many communiy colleges have experienced hard o assess 1

2 managemen performance only by hemselves. In his sudy, we apply a new way o evaluae he course performance from he adul learners perspecives o provide new marke-driven sraegies for he communiy colleges. Alhough many aduls ener communiy colleges and ake courses hrough a marke mechanism in Taiwan. We have experienced ha he local communiy colleges have encounered wo kinds of difficulies recenly. One is heir paricipans have less ineress in academic and public affair courses. The oher is heir paricipans over emphasis he pracical courses and lead some marke-driven communiy colleges provided oo many life skills courses. A lo of communiy colleges have faced he dilemma beween mainaining heir ideal core-value o develop civil sociey and meeing paricipans recreaional needs o seek survival. In some cases, communiy colleges have become cram-schools and lose heir original purposes. The quesion is, can we combine he paricipans ineress wih he public good in a communiy college? By doing wha, he paricipans will feel more meaningful wih regard o he services of communiy colleges? The purpose of his sudy is o explore he poenial facors of course performance in which he facors can be emerged and applied o assess he effeciveness of managemen. More precisely, his sudy ries o solve he following wo quesions: 1.How o esablish performance managemen indicaors by fuzzy logics for communiy colleges?. Wha markeing sraegies should we propose for he communiy college?. Managed Performance in Communiy Colleges. Alhough many researchers have menioned he imporance of performance managemen in communiy colleges []. Performance indicaors have also become an accepable ool for measuring insiuional effeciveness in communiy colleges. However, i is sill lack of sudy o deal wih he course performance indicaors in communiy colleges effecively..1 Measuring he performance of communiy college. The so-called performance means a kind of level ha he individuals achieve heir inended goals []. Measuring performance can be simply divided ino wo caegories o explain. One is financial indicaors, he oher is non-financial indicaors. Typically, he communiy colleges are belong o non-profi organizaions in Taiwan. Reviewing relaed lieraure, we found he Onario Colleges have idenified five measuremens o be used as he key performance indicaors. They are graduae employmen, graduae saisfacion, employee saisfacion, suden saisfacion, and graduaion rae []. The idea of key performance indicaors (KPI) is a province-wide accounabiliy ool esablished by he Onario governmen in I is used o measure and reward purposes for college performance in meeing specific goals and objecives. As we know, he communiy colleges are differen from general universiies. Therefore, his sudy needs o base on he feaure of communiy colleges and provide a muli-dimension model for heir performance measuremen.. Markeing Sraegy for he communiy colleges. Glasen proposed esablishing a srong brand and giving cusomers he bes experiences. Each individual seps in a communiy college will encouner a number of ouch poins. Therefore, brand and

3 markeing have become a chain, brand image is a promise--a promise ha mus be kep [7]. Thinking he markeing sraegies, we will find he key poin is how o mainain he educaional qualiy of communiy colleges o mee paricipan s requess? The markeing sraegies will provide differen hinking abou how o run an idea communiy college. 3. Mehods. 3.1 Research framework. Figure 1 represens a dynamic process of performance managemen in communiy colleges. The idea of communiy colleges inegraed ino naional, communiy college herself, and personal benefi in Taiwan. The sysems followed he Law of Lifelong Learning 00. Typically, he communiy colleges commissioned or self-managemen by local governmen o provide life skills, academics courses, and leisure relaed courses for aduls. I is an open enrollmen for aduls in he communiy. Their survival usually deermined by heir qualiy of educaion in a free marke. In his sudy, we ry o link he goals of communiy colleges o heir course performance and hen o find beer markeing sraegies for he communiy colleges. Goal Orienaion Naional benefi, Communiy college benefi, Personal benefi Civil sociey, Knowledge liberaion, Life skills Personal Selec courses by aduls needs Review Performance Achieve Communiy Colleges Commissioned or self-managemen by local governmen, providing life skills, academics courses, and communiy club Sysems The law of Lifelong Learning 00 Course Performance Analysis Financial (Enrollmen) Cusomers (Saisfacion) Inernal (Aendance) No Review Performance and values Esablished Evaluaion indicaors Yes Markeing Markeing sraegies FIGURE 1. Dynamic process of communiy college in Taiwan 3

4 In his sudy, we propose a research framework following he goal values of communiy colleges o evaluae heir course performance. In our design, he performance indicaors, which use o evaluae he courses performance in communiy college, can be defined by he financial, inernal, and cusomers facors. We invied five expers in his field o help us o selec he main performance indicaors. Finally, we selec enrollmen, aendance, and saisfacion as our performance indicaors o ransform he fuzzy daa colleced from he paricipans (see Figure ). Background variables: Gender, educaion background, age. Environmenal variables: communiy ype, member of classes (daa collecing from Nanou communiy college in Taiwan ) Goal values 1.Goal values of communiy college educaion.reasons for communiy college classes 3.Fuure direcion of he communiy college Values for life regional worldwide Performance Indicaors (seleced by expers advisemen) 1.Financial Facor. Inernal Facor Enrollmen Aendance Courses Performance 3. Cusomers Facor Saisfacion FIGURE. Research framework 3. Fuzzy saisics and weighing. The fuzzy saisics has become a useful ool for measuring ambiguous conceps in science and social science [1,3,,9-10]. Why he radiional numerical model canno explain complex and ambiguous human and social phenomena properly? Previous researchers reminded he risk is oo many limiaions for digial daa and over-inerpreaion [4]. Using fuzzy daa ransform, we ry o avoid such kind of risks. However, he ambiguous daa are consisen wih human logics, we need a powerful way o deal wih he fuzzy daa during compuing process. The conceps of fuzzy se proposed by Zadeh and applied o fuzzy measuremen o deal wih he dynamic environmen. I will give a more reasonable descripion for many daa ransform [1,6,8]. In his sudy, we designed fuzzy quesionnaire o collec daa and ransform he daa o furher inerpreaion. Definiion 3.1 Fuzzy weighing. In order o invesigae he weighs of each impac facors 4

5 in communiy college, we apply he fuzzy se heory and is survey echnique o collec weighing daa. We propose he process of eniy fuzzy weighing as follows: Sep1: Firs, deermine he effecive facors A = { A 1, A,..., Ak } for he real impac facor assessmen; Sep: Le m be he membership of imporance of facor j for he i of he reviewers; ij Sep3: Cumulae he fuzzy weighing w j of A j by w n m ij j k n m j 1i 1 In his sudy, we invied five expers o weigh he influenial facors of performance indicaors relaed o Nanou communiy college, he resul shows as Table 1. TABLE 1. Fuzzy weighs of he influenial facors of performance indicaors facors Number of regisered sudens Rae of aendance Saisfacion expers Toal membership weigh The weighs are followed he formulas o compue: m33 w 3 3 = m k = 0.. ij m11 w 1 3 = m k = 0.,, 3.3 Ranking fuzzy daa. We proposed a mehod no only relaively easy in compuing bu also fewer limiaions o fuzzy daa. The following definiion is o ransform he fuzzy daa o a inernal fuzzy number for ranking purpose. Definiion 3. Defuzzificaion for a inerval fuzzy number. Le A [ a, b] be a inerval on U wih is cenric c ( a b)/. Then he defuzzificaion number RA of A [ a, b] is defined as follows: ln(1 r) RA c (1 ), r Where r b a is he lengh of he inerval [3]. 3.4 Nonparameric es for fuzzy daa. The use of nonparameric mehods was inroduced o es he hypoheses of median which idenical in he Wilcoxon rank sum es. This

6 suggess he es of saisics W = he sum of he rank in he combined samples associaed wih X observaions. If W is less han he criical value he null hypohesis of he same median, i will be rejeced. Anoher quesion is he inconsisency of he scores, which is hough o be he bigges barrier o reliable assessmen [4]. To answer his quesion, we use he nonparameric mehod o es he hypoheses of median which idenified in he Kruskal-Wallisi one-way analysis of variance. We can le N = k n i be he oal number of observaions in he k reamens. When we assigned he rank 1 o he smalles of pool scores, hen he rank o he nex, and so on o he larges one, which was given he rank N. The Kruskal-Wallis es used in his sudy is defined as follows: 1 k Ri K 3( N 1) N( N 1) n Here R i is he sum of he ranks assigned o observaions in he k-h reamen. Since K follows a (k-1) disribuion, he null hypohesis of equal means (median) is rejeced when K exceeds he criical value. Example 3.1 A random sample of paricipans, i was separaed ino hree age groups, who have regisered in a communiy college. Each paricipan is fied a repor abou he rae of aendance in relaed courses in pas year. The daa and calculaions are summarized in Table. A he 0. 0 level of significance, can we say here is a difference in erms of aendance rae exising in hese groups? i TABLE. The rae of aendance in communiy college in differen age groups Observaions R Group 1 (,4) (3,4) (,3) (3.,4) (,6) (.,7) (3,3.) (1.,) (.,4) Group (,6.) (3.,4.) (4.,6.) (1.,.) (,7) (4.,6) (4,6) Group 3 (3.,4) (6.,7) (3.,6.0) (6.,8) (6,8) (6,7) K ( ) 3( 1) =8.3> 0.0() =.99. ( 1) Under significance level 0. 0, > ( k 1), hen we rejec H An Empirical Sudy. 4.1 Research samples. This sudy used a fuzzy quesionnaire o collec daa from Nanou Communiy College in 009. There are en affiliaed insiuions in Nanou Communiy College, and heir locaions are in he cenral area of Taiwan. The quesionnaire included 6

7 eigh iems, he paricipans were asked o answer by fuzzy forms in erms of selec heir possible responses wih an inerval of 1 o scale. The sudy disribued 100 quesionnaires, wih 7 copies reurned, oaling 64 valid samples. Table 3 shows he sample disribuion in his case sudy. TABLE 3. The disribuion of seleced paricipans in Nanou Communiy College Gender Female 41(64.1 %), Male 3(3.9 %) Age 0-40 years old years old 61 year old or over Educaional background 6 (9.4%) 49 (76.6%) 9 (14.1%) High school or under high school level College level Graduae school level 7 (4.%) 37 (73%) 0 (0%) 4. Fuzzy weighing and calculaion. In his sudy, we focus on financial, inernal, and cusomer perspecive o evaluae managemen performance. We consider he number of regisered sudens, rae of suden aendance, and sudens saisfacion as main facors which will impac on he managemen performance of communiy colleges. We calculae he fuzzy weighed values o reflec he real values in he communiy college. The fuzzy esing formula for evaluaing values (EV ) and is meanings are as follows: EV = R 0. A 0.3 S 0. R =0.8+(x-0)*0.0, x= number of regisered sudens < 40 A = rae of suden aendance S = degree of saisfacion According o he weighs of influence indicaors reviewed by he five expers, we found ha numbers of regisraion would be accouned for 0. power, he rae of suden aendance was accouned for 0.3 power, and saisfacion wih courses would be accouned for 0. power (see Table 1). The seleced performance indicaors are applied o evaluae he communiy college courses which a leas provided for hree consecuive years, ha is six semesers in our sysem. According he 64 paricipans and heir six semesers aending he hree courses, we collec he fuzzy daa and ransform heir values. The average of Calligraphy Ar, course is belong o communiy club domain, is The average of Life and Law, which is belong o academic domain, is The average of Ballroom Dancing, which is belong o life skill domain courses, is (see Table 4). In hese hree kinds of courses, he performance indicaors of Calligraphy Ar in communiy club domain is he highes, he nex is Life and Law in academic domain, and he Ballroom Dancing (life skill domain courses) is in he las. The curren fuzzy calculaion of he variances in 7

8 Performance indicaors differen courses lised in Table 4. The differences of he hree courses are lised in Figure 3. TABLE 4. Descripive saisics of performance indicaors evaluaing in Nanou Communiy College Saisics/Courses Calligraphy Ar (communiy club Life and Law (academic domain) Ballroom Dancing (life skill domain) Average domain) Minimum Max Sandard Deviaion Variance Noe: Courses have been keeping running for hree consecuive years Performance indicaors of communiy college courses for hree consecuive years Calligraphy (c ommuniy ) Life and Law (academic ) Ballroom Dancing (life skill) Semes ers FIGURE 3. Fuzzy evaluaing he hree courses in hree consecuive years (six semesers) 4.3 Fuzzy saisical analysis he values of communiy colleges. Paricipans indicaed he core values of communiy colleges including prepare for civil sociey, liberal knowledge, and life skills. Their fuzzy modes are 9., 3, 31.7 and heir fuzzy memberships are.1,.36, and.49 respecively. The resul reveals ha he communiy college sudens ranked he life skill courses wih he highes value. 8

9 4.4 Message for creaing markeing sraegies. The findings in Table show ha he iem oo busy (.41) is he main reason o dropou classes and hen no ineresing or no one goes wih (.3). The finding also shows ha propaganda/adverisemens, recommended by friends, and hrough Inerne are he main channels ha paricipans can ge informaion from he communiy college. TABLE. Paricipans perspecives provide for communiy colleges o se heir markeing sraegies Responses Fuzzy memberships Fuzzy mode The reason why Never dropou No ineresing or no one goes wih, oo far, or have No response Too busy No response (3.) paricipans dropou classes 0.16 had bad experiences How paricipans selec communiy colleges By friends By TV, promoion car, newspapers, radio or recommended by he chief of village Propaganda/ adverisemen Inerne Propaganda adverisemens (4.) 4. Tesing he hypohesis wih nonparameric mehods. We use Wilcoxon rank sum es and he Kruskal Wallis (one-way ANOVA for nonparameric mehod) o es he difference of sudy purposes among differen groups [4]. We consider he variables such as paricipans (gender, age, and educaion background), school locaion (ciy or couny), and hree caegories of courses (civil sociey courses, academics courses, and life skill courses). According o he Table 6, we lis he relaed findings as follows: (1) Aduls who sudy in ciy communiy college show more accepable he idea of civil sociey; () Aduls wih lower educaion background end o more accepable he idea of civil sociey han do hose wih higher educaion background; (3) Aduls wih higher educaion background are more likely o ake life skill relaed courses; (4) Men express more waned o join academics courses han do women; () Adul learners, in 0-40 years old, hope o obain diploma; (6) Adul learners, in years old, end o selec vocaional raining. 9

10 TABLE 6. Wilcoxon rank-sum es & Kruskal-Wallis es ranks N=64 Courses Group variables N Tesing wih nonparameric saisics Decision =.0 Civil sociey ciy 3 Rank sum of W= 137 Z=.6>Z 0.0 =1.6, p=0.03 couny 9 Rejec H 0 Liberal knowledge Female Male 41 3 Rank sum of W = 1194 Z=-1.94<Z 0.0 =-1.6, p= Rejec H 0 Civil sociey High school 7 College 37 Rank sum of W =100. Z=1.7>Z 0.0 =1.6, p=0.036 Rejec H 0 Life skill High school 7 Rank sum of W = 74. College 37 Z=-.08<Z 0.0 =-1.6, p= Rejec H 0 Life skill 0-40 years old 6 K=0.< 0.0() =.991, p= 0.88 Accep H ~ 9 Diploma 0-40 years old 6 K=6.76> 0.0() =.991, p=0.034 Rejec H ~ 9 Vocaional 0-40 years old 6 K=10.166> () raining 0.0 =.991, p=0.006 Rejec H ~ 9. Conclusions. This sudy examines he of goal values in communiy colleges wih respec o life skills, civic sociey, regional culures, and he world perspecives. Using a fuzzy quesionnaire and is saisical ransformaion, his sudy addressed he membership funcions of core values in communiy colleges according o paricipans demands, aendances, and heir saisfacions. The resuls are as follows: (1) Mos of paricipans are no ineresed in civil sociey courses; () There is a endency o over-value he life skill courses in he communiy college; (3) The life skill courses are he mos popular courses and he 40-year-old group was he main paricipans; () Adverisemen is he larges source for aduls o reference o selec heir communiy college and he relaed informaion go from friends pu in he second. Based on he findings, his sudy provides he following suggesions for he communiy college: (1) Offer life skill courses need o balance wih civil sociey courses. () Enhance local characerisics in courses o rebuild regional values. (3) Design special courses for sudens over he age of 40 years old. (4) Increase public budges for communiy colleges o subsidize seleced courses and more focused on he civil sociey courses. Analyzing he markeing sraegy, his sudy found he vocaional raining has become an imporan componen in communiy colleges. The resul reveals ha preparing sudens second experise for a beer job, designing field-relaed vocaional raining, and preparing special courses for fuure are very imporan. Following hese views o make changes, he communiy college has showed more clear heir goals for fuure and he enrollmen is increased significanly in he following semeser. Finally, we may indicae he idea of fuzzy evaluaing is a useful ool which considered he hree imporan weighing dimensions: financial, inernal, and cusomer in communiy colleges. The model also focuses on criical facors in communiy colleges, such as 10

11 enrollmen, sudens aendance, and heir saisfacion which could creae a pracical performance indicaor sysem o reboo culure in communiy colleges. REFERENCES [1] C. M. Sun and B. Wu, New saisical approaches for fuzzy daa. Inernaional Journal of Uncerainy, Fuzziness and Knowledge-based Sysems, vol.1, no., pp , 007. [] Cenennial College, Key performance indicaors, Rerieved from hp:// Nov. 6, 010. [3] H. Hsu and B. Wu, An innovaive approach on fuzzy correlaion coefficien wih inerval daa. Inernaional Journal of Innovaive Compuing, Informaion and Conrol (IJICIC), vol.6, no.3(a), pp , 010. [4] H. Nguyen and B. Wu, Fundamenals of saisics wih fuzzy daa. Heidelberg: Springer, 006. [] J. F. Chang, Fuzzy inference for assessing process lifeime performance, Inernaional Journal of Innovaive Compuing, Informaion and Conrol (IJICIC), Vol.3, No.6(B), pp , 007. [6] L. A. Zadeh, Fuzzy ses, Informaion and Conrol, vol.8, no.3, pp , [7] S. Glasen, Seps o brand building. Rerieved from hp://sbinfocanada.abou.com/od/markeing/a/brandbuildingsg.hm, 008. [8] S. W. Wang D. F. Chang and B. Wu, Does echnologies really help digial naives? A fuzzy saisical analysis and evaluaion of sudens learning achievemen, Inernaional Journal of Innovaive Managemen, Informaion & Producion, vo.1, no.1, pp.18-30, 010. [9] T. Samasu, K. Tachikawa and Y. Shi, GUI form for car rerieval sysems using fuzzy heory, ICIC Express Leers, vol., no.3, pp.4-49, 008. [10] T. Samasu K. Tachikawa and Y. Shi, Image processing for car shapes in he fuzzy rerieval sysem, ICIC Express Leers, Par B, vol.1, no.1, pp.1-7,