HOW CAN EXPERT CONSENSUS METHODS ENHANCE THE DESIGN OF IMMERSIVE LEARNING PRACTICAL MODELS FOR DEAF OR HARD-OF-HEARING STUDENTS IN TVET?

Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4245 HOW CAN EXPERT CONSENSUS METHODS ENHANCE THE DESIGN OF IMMERSIVE LEARNING PRACTICAL MODELS FOR DEAF OR HARD-OF-HEARING STUDENTS IN TVET? RINI HAFZAH BINTI ABDUL RAHIM1, DINNA NINA BINTI MOHD NIZAM1, NUR FARAHA BINTI MOHD NAIM1 and ASLINA BAHARUM2 1Use Expe ience G oup, Facul y o Compu ing and In o ma ics, Uni e si y Malaysia Sabah, 88400 Ko a Kinabalu, Sabah, MALAYSIA 2 Depa men o Da a Science and A i icial In elligence, School o Enginee ing and Technology, Sunway Uni e si y, Banda Sunway, 47500 Selango , MALAYSIA E-mail: 1RINI_HAFZAH_DI21@ilu .ums.edu.my, 2dinn[email p o ec ed].my, 3 a aha.na[email p o ec ed]u.my, [email protected] ABSTRACT T adi ional audi o y-based eaching app oaches limi he e ec i eness o p ac ical skills acquisi ion o Dea o Ha d-o -Hea ing (DHH) s uden s in Technical and Voca ional Educa ion and T aining (TVET). Despi e inc eased in e es in imme si e echnologies like augmen ed eali y (AR), he ield lacks alida ed, inclusi e ins uc ional models ailo ed o DHH lea ne s. This s udy add esses his gap by in eg a ing he Nominal G oup Technique (NGT) and Fuzzy Delphi Me hod (FDM) o design and alida e an Imme si e Lea ning P ac ical Skills (ILPS) model. The no el y lies in he combined use o NGT and FDM o consensus-building among expe s in AR, gami ica ion, and DHH educa ion—an app oach no commonly applied in inclusi e model de elopmen . Resul s e ealed a s ong expe consensus (>97%) on 15 i ems ac oss h ee co e cons uc s: Lea ning Inpu Medium, P ac ical Skills Module, and AR Gami ica ion Fea u es. This esea ch o e s a eplicable and pa icipa o y model de elopmen p ocess and in oduces a alida ed amewo k o inclusi e imme si e lea ning in TVET. The s udy con ibu es new knowledge by demons a ing how expe -d i en me hods can ope a ionalize inclusi e pedagogy h ough imme si e echnologies. This s udy demons a es how combining FDM and NGT may success ully e alua e inclusi e design elemen s o imme si e lea ning. The esul s suppo he de elopmen o a p ac ical skills model wi h a DHH ocus and p o ide a epea able amewo k o inclusi e cu iculum co-c ea ion. This combina ion s eng hen consensus among 11 panel o expe s and acco ding o he s udy's indings, he NGT and FDM app oach has made i simple and quick o esea che s o con i m c ucial de ails ha should be highligh ed. To help DHH s uden s lea n mo e e ec i ely, i is ad ised ha mo e esea ch be done in collabo a ion wi h cou se designe s. To p o ide a scalable app oach o de eloping imme si e, accessible lea ning en i onmen s in specialized educa ional con ex s, his s udy hopes o demons a e how e ec i ely NGT and FDM collabo a e o inclusi e ins uc ional design. Keywo ds: Educa ional Technology, Teaching And Lea ning, Hea ing Impai ed, Highe Educa ion, Model De elopmen . 1. INTRODUCTION DHH s uden s ace unique challenges in acqui ing p ac ical skills due o limi a ions in con en ional audi o y-based eaching and lea ning me hods [1]. DHH s uden s encoun e signi ican challenges in mas e ing p ac ical skills wi hin TVET se ings due o he limi a ions o con en ional audi o y-dependen eaching me hods. Despi e ad ancemen s in imme si e echnologies such as AR and gami ica ion, he e emains a conspicuous lack o alida ed, inclusi e ins uc ional models ha add ess he unique communica ion and lea ning needs o DHH lea ne s. Fu he mo e, ew s udies ha e sys ema ically applied s uc u ed expe consensus me hods like he NGT and FDM o de elop and alida e such models. This ep esen s a c ucial Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4246 me hodological and pedagogical gap. Wi hou accessible, expe -in o med amewo ks, DHH s uden s isk u he ma ginaliza ion in skill-based lea ning en i onmen s. This s udy is needed o b idge his gap by o e ing a no el, alida ed, and eplicable imme si e lea ning model, one ha le e ages expe insigh s o ensu e ele ance, inclusi i y, and p ac ical e ec i eness in eaching DHH lea ne s essen ial echnical skills. In TVET se ings, whe e p ac ical skills acquisi ion is c ucial, hese challenges a e especially p onounced [2–4]. The absence o adequa e and inclusi e ins uc ional models o en hinde s DHH s uden s om ully engaging in lea ning ac i i ies, impac ing hei skill p o iciency and long- e m ca ee p ospec s. The e is a g owing need o imme si e, accessible lea ning models ha le e age cu ing-edge echnologies o b idge his gap, os e ing an en i onmen whe e DHH s uden s can acqui e p ac ical skills mo e e ec i ely [5–7]. Imme si e lea ning echniques, pa icula ly when combined wi h echnology-d i en app oaches, hold conside able po en ial o imp o ing accessibili y and lea ning ou comes o DHH s uden s. Techniques such as AR and gami ica ion ha e been shown o enhance engagemen and comp ehension by c ea ing an in e ac i e and isually ich lea ning expe ience [8,9]. Al hough AR and gami ica ion a e inc easingly used in educa ional esea ch [10], ew s udies ha e e ec i ely add essed he needs o DHH s uden s, pa icula ly in p ac ical skill-based lea ning con ex s. P io wo ks [11] and [12] ha e explo ed assis i e echnologies and sign language sys ems ye o en lack a comp ehensi e model o imme si e lea ning ailo ed o DHH lea ne s in TVET en i onmen s. Mo e c i ically, hese s udies gene ally omi s uc u ed consensus me hods o model alida ion. The absence o pa icipa o y echniques such as he NGT and FDM unde mines he igo and inclusi eness o p io model de elopmen e o s. Add essing hese gaps, he cu en s udy p oposes and alida es an ILPS model h ough a dual-me hod consensus app oach, con ibu ing a no el, empi ically g ounded amewo k o inclusi e imme si e educa ion. Howe e , o design an e ec i e imme si e lea ning model o DHH s uden s, i is essen ial o in eg a e inpu om subjec ma e expe s and s akeholde s. The NGT and FDM a e pa icula ly aluable o his pu pose, enabling a sys ema ic consensus-building p ocess ha inco po a es expe opinions and p io i izes model elemen s acco ding o eal-wo ld ele ance and e icacy [13,14]. NGT sessions usually consis o i e o en pa icipan s and las be ween one and hal o wo hou s [12,13,14]. Acco ding o Lloyd-Jones, Fowell, and Bligh (1999), he esea che 's job in NGT is o acili a e and adminis e , which minimizes e ec on he da a [18]. In many esea ch me hodologies whe e he esea che 's p econcep ions a e en o ced h ough ques ion aming and answe coding, Lomax and McLeman (1984) e e o he "omniscience o he esea che " [19]. In NGT, his is a oided since g oup membe s o ganize, classi y, and p io i ize he eplies. Howe e , he e ec i eness o he app oach depends on how well he s imulus ques ion is o mula ed, and i is impe a i e ha he esea che is clea abou he in o ma ion hey hope o ob ain om he p ocedu e. In hei 1975 s udy, Delbecq, Van de Ven, and Gus a son con as ed NGT wi h Fuzzy FDM [20]. The e ised measu emen scales a e e i ied using a FDM. The use o uzzy Delphi is based on he p ac ical esul s o he many IT/IS esea ch a ia ions o e ed by. This ool is a e y help ul me hod when a g oup o expe s mus accep a gi en le el o esea ch. Addi ionally, he Fuzzy Delphi p ocedu e is an in e es ing me hod o g oup decision-making conce ning he ague no ions o expe opinion alignmen [21]. In o de o ensu e accu acy and consis ency o opinion, su ey me hodologies a e employed in conjunc ion wi h lowe expenses, which allows expe s o ully exp ess hei hough s wi hou ea o misunde s anding and allows hei esul s o be implemen ed apidly. The pu pose o his p ojec is o c ea e an ILPS ha is especially sui ed o DHH s uden s in TVET se ings. By using NGT and FDM, we wan o de e mine he essen ial elemen s o an inclusi e model, e alua e i s iabili y, and make su e i sa is ies he unique equi emen s o DHH s uden s. In o de o p o ide mo e equi able educa ional oppo uni ies o he de elopmen o p ac ical skills o ma ginalized lea ne s, his pape examines he me hodology used, he esul s ob ained om expe inpu s, and he implica ions o his imme si e lea ning model on he skill acquisi ion o DHH s uden s. Gi en ha DHH s uden s equi e p ac ical skills o e ec i e imme si e lea ning, ou goals we e o: (1) desc ibe he cha ac e is ics o ILPS o help hem lea n mo e e ec i ely; and (2) p io i ies hose ILPS o in e en ion, keeping in mind ha DHH s uden s equi e p ac ical skills o e ec i e imme si e lea ning. By combining NGT and FDM Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4247 analysis, we we e able o each a consensus o expe opinion in o de o accomplish hese goals. 2. LITERATURE REVIEW P e ious s udies in he ield o educa ional echnology ha e widely documen ed he po en ial o AR and gami ica ion o enhance lea ning engagemen , pa icula ly in STEM and highe educa ion con ex s [10]. Addi ionally, esea ch a ge ing DHH s uden s has la gely cen e ed a ound assis i e ools such as sign language ecogni ion sys ems o oice- o- isual ansla ion pla o ms [11] and [12]. Howe e , hese e o s o en emain isola ed echnological in e en ions wi hou in eg a ion in o a pedagogically sound o alida ed ins uc ional model. Mo e impo an ly, ew s udies ha e sough o de elop inclusi e lea ning amewo ks speci ically ailo ed o DHH s uden s in TVET, whe e hands-on skill acquisi ion is pa amoun . Fu he mo e, he use o consensus- based me hods such as he NGT and FDM is la gely absen in p io li e a u e, esul ing in models ha may lack bo h p ac ical ele ance and inclusi i y. Mo i a ed by hese gaps, he p esen s udy in oduces a alida ed ILPS model o DHH lea ne s, co-de eloped h ough s uc u ed expe consensus. Unlike p e ious wo k, his s udy does no simply e alua e echnological a o dances, bu con ibu es a scien i ically g ounded, scalable amewo k o inclusi e p ac ical skills aining an inno a ion in bo h me hodology and educa ional applica ion. 3. METHODOLOGY The design and de elopmen o an imme si e lea ning model o p ac ical skills a ge ed o DHH s uden s equi e a igo ous and inclusi e app oach o ensu e ha he model is bo h e ec i e and mee s he speci ic lea ning needs o hese s uden s. This s udy employs a mixed-me hod app oach, u ilizing he NGT and he FDM o ga he , analyze, and p io i ize expe inpu . These me hods enable a s uc u ed, consensus-d i en p ocess ha inco po a es di e se pe spec i es, acili a ing he design o a lea ning model ha is ailo ed o he unique equi emen s o DHH s uden s in TVET con ex s. 3.1 Phase 1: Iden i ying Key Model Componen s using he Nominal G oup Technique (NGT) NGT is a me hodical echnique ha inds a g oup's common iewpoin s on a gi en subjec [22]. Delbecq, Van de Ven, and Gus a son desc ibed social planning scena ios as ollowing: explo a o y esea ch; ci izen engagemen ; use o in e disciplina y specialis s; and p oposal assessmen [15]. O iginally, i was hough o as a "pa icipa ion echnique o social planning si ua ions". Since hen, he me hod has been used in many di e en g oup con ex s, including social science empi ical esea ch. Al hough i has been u ilized in educa ion esea ch o some deg ee [23– 25], i seems o be mo e equen ly employed in he ield o heal h s udies when i comes o social science esea ch. The NGT p ocess is qui e egimen ed and consis s o ou main s ages: 1. Coming up wi h ideas on i s own in esponse o a p omp . 2. Round- obin sha ing (and lis ing) o hese concep s wi hou deba e. 3. Making each concep clea on i s own and assembling ela ed concep s in o g oups. 4. Indi iduals o e o choose which ideas come i s . The e is deba e on he op imal sample size o employ when using NGT echniques in esea ch. Acco ding o ce ain schola s, NGT may be conduc ed on a la ge g oup o a single coho [19,26,27], bu i can also be b oken up in o smalle g oups o acili a e e ec i e communica ion, depending on he needs o he s udy. Because o his, he sample sizes shown in Table 1 ha ha e been used in p e ious esea ch p ojec s a e as ollows: Table 1: NGT Sample Size [28]. Au ho Sample Van de Ven dan Delbecq (1971) 5 – 9 expe s/pa icipan s Ho on (1980) 7 – 10 expe s/pa icipan s Ha ey dan Holmes (2012) 6 – 12 expe s/pa icipan s Abdullah & Islam (2011) 7 – 10 expe s/pa icipan s Ca ney e al (1996) Min. 6 expe s/pa icipan s Because o he a o emen ioned e e ence, he esea che chose 11 expe s o ake pa in he Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4248 s udy's NGT and FDM me hod. This sum is deemed sui able o his in es iga ion gi en he exis ing ci cums ances ha es ic in e ac ions. 3.2 Phase 2: Valida ion and Expe Consensus Using Fuzzy Delphi Me hod (FDM) The p ocedu e is b oken down in o a numbe o p ima y phases o inish he me hod. P io o using he Delphi Fuzzy echnique, he ini ial s age in ol es de eloping he i ems ha equi e expe app o al. Selec ing quali ied expe s wi h backg ounds in academia and indus y is he second s age. In o ma ion om chosen expe s mus be ga he ed wi hin a speci ied ime ange o he hi d s ep. In o de o gene a e signi ican co ela ions be ween he wo de ined ou comes, cus ome engagemen , and he la en a iable o assessmen , he inal s age is an analy ic echnique. Mo e in o ma ion abou FDM s eps is p o ided in Table 2. Table 2: FDM S eps. S ep Fo mula ion 1.Expe selec ion Fo his esea ch, a o al o ele en expe s we e in ol ed. Se e al expe s we e in i ed in o de o de e mine he impac o he e alua ion c i e ia on he a iables ha would be in es iga ed u ilising linguis ic a iables. Among o he hings, hese p o ide desc ip ions o possible issues wi h he i em. 2.De e mining linguis ic scale This me hod yields uzzy iangle numbe s, o iangula uzzy numbe s, om all linguis ic a iables. The linguis ic a iables a e con e ed a his s age by adding uzzy numbe s o hem [26] (m1, m2, m3) is he iangula uzzy numbe , which ep esen s he alues m1, m2, and m3. m1 and m2 ep esen he lowes and mos easonable alues, espec i ely, whils m3 ep esen he highes alues. To ansla e linguis ic a iables in o uzzy numbe s, on he o he hand, uzzy scales a e made using iangula uzzy numbe s. Odd digi s ep esen he numbe o le els on he uzzy scale. Figu e 1 show T iangula uzzy numbe . Figu e 1: T iangula uzzy numbe 3.The De e mina ion o Linguis ic Va iables and A e age Responses Once he selec ed specialis esponds, he esea che needs o con e all Like scales o uzzy scales. This p ocess is also known as de e mining he a e age esponse o each uzzy numbe [30] . 4.The de e mina ion o h eshold alue "d" When de e mining he deg ee o ag eemen among specialis s, he h eshold alue is c ucial [28]. The ollowing o mula is used o ge he dis ances o each uzzy numbe , m = (m1, m2, m3) and n = (m1, m2, m3): Figu e 2: T h eshold alue "d" 5.Iden i y he alpha cu agg ega e le el o uzzy assessmen Each objec is assigned a uzzy numbe a e expe consensus [32]. Fuzzy alues a e calcula ed and de e mined using he ollowing o mula: Amax=(1) Ú4 (m1 + 2m2 + m3). 6.Di uzzica ion p ocess This p ocess uses he o mula Amax = (1) Ú4 (a1 + 2am + a3). When he esea che uses a e age uzzy numbe s o a e age answe s, a sco e numbe be ween 0 and 1 is gene a ed [32]. A = 1/3 * (m1 + m2 + m3), A = 1/4 * (m1 + 2m2 + m3), and A = 1/6 * (m1 + 4m2 + m3) a e he h ee o mulas ha a e used in his ope a ion. The median alue o "0" and "1" is α-cu = (0 + 1) / 2 = 0.5; his is he A-cu alue. I he esul ing A alue is less han he α-cu alue = 0.5, he i em will be ejec ed because his does no imply expe ag eemen . The alpha cu alue mus be mo e han 0.5 [33] . 7.Ranking p ocess The placemen app oach selec s i ems based on de uzzi ica ion alues and Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4249 expe ag eemen ; he elemen wi h he highes alue is de e mined by he mos impo an posi ion [34] . In o de o ge a high deg ee o expe consensus on he ea u es ha bes p omo e he de elopmen o p ac ical skills in DHH s uden s, he FDM p ocess e ec i ely assis ed in alida ing and e ining he sugges ed model elemen s. These esul s show he e ec i eness o FDM in inclusi e design, gua an eeing ha he inished model is p ac ical and ele an especially o he a ge s uden s who a e DHH s uden s. 4. RESULT The esul s o his s udy illus a e he p ocess and ou comes o designing an imme si e lea ning p ac ical skill model o DHH s uden s in TVET. Th ough he NGT and FDM, a se o key elemen s was iden i ied, alida ed, and e ined o c ea e a model ha emphasizes accessibili y, engagemen , and p ac ical skills acquisi ion. This sec ion p esen s indings om each phase o he s udy, including expe inpu analysis, consensus me ics, and p elimina y pilo es ing ou comes wi h DHH s uden s. 4.1 NGT Findings: Table 3: Main Cons uc s. I ems/ Elemen s Cons uc 1 Cons uc 2 Cons uc 3 Vo e 1 6 7 7 Vo e 2 7 7 7 Vo e 3 6 7 7 Vo e 4 7 7 6 Vo e 5 7 7 7 Vo e 6 7 7 6 Vo e 7 7 7 7 Vo e 8 6 6 6 Vo e 9 7 7 6 Vo e 10 6 7 7 Vo e 11 7 7 7 To al Coun 73 76 73 Pe cen age 94.81 98.7 94.81 Rank P io i y 2 1 2 To al Consensus Sui able Sui able Sui able Table 4: Cons uc 1: DHH Lea ning Inpu Medium. I ems/ Elemen s Tex Pic u e Video Sign Language Vo e 1 6 5 7 7 Vo e 2 6 6 5 7 Vo e 3 6 5 7 7 Vo e 4 7 6 7 7 Vo e 5 7 7 7 7 Vo e 6 7 6 7 6 Vo e 7 7 7 7 7 Vo e 8 6 6 6 6 Vo e 9 7 7 7 7 Vo e 10 6 5 7 7 Vo e 11 6 6 5 7 To al Coun 71 66 62 65 Pe cen age 92.21 85.71 93.51 97.4 Rank P io i y 3 4 2 1 To al Consensus Sui able Sui able Sui able Sui able Table 5: Cons uc s 2: P ac ical Skills Lea ning Module. I ems/ Elemen s Sa e y S ep by S ep Demons a ion P ac ical Skill Vo e 1 7 4 7 7 Vo e 2 6 6 5 7 Vo e 3 7 5 7 7 Vo e 4 6 7 7 7 Vo e 5 7 7 7 7 Vo e 6 5 6 6 7 Vo e 7 7 7 7 7 Vo e 8 6 6 6 6 Vo e 9 7 7 7 7 Vo e 10 7 4 7 7 Vo e 11 6 6 5 7 To al Coun 71 65 71 76 Pe cen age 92.21 84.42 92.21 98.7 Rank P io i y 2 3 2 1 To al Consensus Sui able Sui able Sui able Sui able Table 6: Cons uc 3: AR Gami ica ion Fea u es. I ems/ Elemen s Use In e ace 3D Model Gameplay Le el Di icul y Challenge Rewa d Sco e Vo e 1 5 7 6 6 6 7 5 Vo e 2 6 6 7 5 6 6 7 Vo e 3 6 7 7 5 6 6 6 Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4250 Vo e 4 7 7 6 5 7 7 7 Vo e 5 6 5 6 5 4 5 7 Vo e 6 6 4 6 6 7 6 6 Vo e 7 7 7 7 6 6 7 7 Vo e 8 6 5 5 5 5 4 5 Vo e 9 7 7 7 6 5 7 7 Vo e 10 6 7 6 6 6 7 5 Vo e 11 6 6 7 6 6 6 7 To al Coun 68 68 70 61 64 68 69 Pe ce n age 88.3 1 88.3 1 90.9 1 79.2 2 83.1 2 88.3 1 89.6 1 Rank P io i y 3 3 1 5 4 3 2 To al Conse nsus Sui able Sui able Sui able Sui able Sui able Sui able Sui able Table 3 o Table 6 displays he model's o e all ag eemen and e alua ion sco es. This esea ch shows ha all model build concen a ions a e wi hin he ideal ange. I is now necessa y o he p opo ion o exceed 70% in ligh o he esul s o hese in es iga ions. E e y i em abo e 70% expe consensus, acco ding o he s udy o expe app o al da a. A ew s udies ha bols e his idea include Mus apha e al. (2022) and Deslandes, Mendes, Pi es (2010). This enables he esea che s o d aw he conclusion ha he model's essen ial componen s a e p ac ical and well-liked by he in ended audience. The leng hy ounds o expe judgemen needed by he Delphi me hod migh be eplaced wi h a as e al e na i e, he modi ied NGT me hodology [25,30,32]. 4.2 FDM Findings: Valida ion and Consensus on Model Componen s Table 7: P ac ical Skills Lea ning Module. De uzzi ica ion Repo Resul Sa e y S ep by S ep Demons a ion P ac ical Skill Expe 1 0.05249 0.1837 0.05249 0.02099 Expe 2 0.00525 0.04724 0.12072 0.02099 Expe 3 0.05249 0.06823 0.05249 0.02099 Expe 4 0.00525 0.10497 0.05249 0.02099 Expe 5 0.05249 0.10497 0.05249 0.02099 Expe 6 0.12072 0.04724 0.05249 0.02099 Expe 7 0.05249 0.10497 0.05249 0.02099 Expe 8 0.00525 0.04724 0.00525 0.03674 Expe 9 0.12072 0.06823 0.12072 0.15221 Expe 10 0.05249 0.1837 0.05249 0.02099 Expe 11 0.00525 0.04724 0.12072 0.02099 S a is ics I em1 I em2 I em3 I em4 Value o he i em 0.04772 0.09161 0.0668 0.03435 Value o he cons uc 0.06012 I em < 0.2 11 11 11 11 % o i em < 0.2 100% 100% 100% 100% A e age o % consensus 100% De uzzi ica ion 0.90909 0.81818 0.90909 0.96364 Ranking 2 3 2 1 S a us Accep Accep Accep Accep Table 8: DHH S uden Lea ning Inpu Medium De uzzi ica ion Repo Resul Tex Pic u e Video Sign Language Expe 1 0.02624 0.09972 0.03674 0.0105 Expe 2 0.02624 0.01575 0.13646 0.0105 Expe 3 0.02624 0.09972 0.03674 0.0105 Expe 4 0.03149 0.01575 0.03674 0.0105 Expe 5 0.03149 0.07348 0.03674 0.0105 Expe 6 0.03149 0.01575 0.03674 0.04724 Expe 7 0.03149 0.07348 0.03674 0.0105 Expe 8 0.02624 0.01575 0.02099 0.04724 Expe 9 0.03149 0.07348 0.03674 0.0105 Expe 10 0.02624 0.09972 0.03674 0.0105 Expe 11 0.02624 0.01575 0.13646 0.0105 S a is ics I em1 I em2 I em3 I em4 Value o he i em 0.02863 0.0544 0.05344 0.01718 Value o he cons uc 0.03841 I em < 0.2 11 11 11 11 % o i em < 0.2 100% 100% 100% 100% A e age o % consensus 100% De uzzi ica ion 0.94545 0.87273 0.93636 0.98182 Ranking 2 4 3 1 S a us Accep Accep Accep Accep Table 9: Imme si e Lea ning Augmen ed Reali y Gami ica ion Fea u es De uzzi ica ion Repo Resul Use In e ace 3D Model Gameplay Le el Di icul y Challenge Rewa d Sco e Expe 1 0.12 072 0.06 823 0.01 575 0.05 249 0.03 149 0.06 298 0.11 547 Expe 2 0.00 525 0.01 05 0.04 199 0.06 298 0.03 149 0.00 525 0.05 774 Expe 3 0.00 525 0.06 823 0.04 199 0.06 298 0.03 149 0.00 525 0 Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4251 Expe 4 0.05 249 0.06 823 0.01 575 0.06 298 0.08 923 0.06 298 0.05 774 Expe 5 0.00 525 0.10 497 0.01 575 0.06 298 0.19 945 0.11 022 0.05 774 Expe 6 0.00 525 0.22 044 0.01 575 0.05 249 0.08 923 0.00 525 0 Expe 7 0.05 249 0.06 823 0.04 199 0.05 249 0.03 149 0.06 298 0.05 774 Expe 8 0.00 525 0.10 497 0.13 122 0.06 298 0.08 398 0.22 569 0.11 547 Expe 9 0.05 249 0.06 823 0.04 199 0.05 249 0.08 398 0.06 298 0.05 774 Expe 10 0.00 525 0.06 823 0.01 575 0.05 249 0.03 149 0.06 298 0.11 547 Expe 11 0.00 525 0.01 05 0.04 199 0.05 249 0.03 149 0.00 525 0.05 774 S a is ic s I em 1 I em 2 I em 3 I em 4 I em 5 I em 6 I em 7 Value o he i em 0.02 863 0.07 825 0.03 817 0.05 726 0.06 68 0.06 107 0.06 299 Value o he cons uc 0.05617 I em < 0.2 11 10 11 11 11 10 11 % o i em < 0.2 100 % 90% 100 % 100 % 100 % 90% 100 % A e age o % consens us 97% De uzzi ica ion 0.90 909 0.88 182 0.92 727 0.80 909 0.84 545 0.89 091 0.9 Ranking 2 5 1 7 6 4 3 S a us Acc ep Acc ep Acc ep Acc ep Acc ep Acc ep Acc ep Table 7 o Table 9 show esul s o FDM, ollowing da a p ocessing, he da kened h eshold alue is highe han he 0.2 h eshold alue (> 0.2) (see able 9). In o he wo ds, he e a e expe whose iews do no acco d o e en coincide on some issues. The a e age alue o all Ne Collabo a i e Lea ning cons uc s and componen s, on he o he hand, displays he a e age h eshold alue (d) < 0.2, o 0.08625. The i em has a high deg ee o expe ag eemen i he h eshold (d) a e age alue is less han 0.2 [35,36]. In he meanwhile, he o al expe ag eemen pe cen age is 100%; 100%; 97%, which is mo e han >75% and sa is ies he equi emen s o expe ag eemen on his issue. Fu he mo e, he a e age uzzy answe , o Alpha- Cu de uzzi ica ion alues, all su pass α-cu => 0.5. The alpha cu alue should be mo e han 0.5 and should be disca ded i i is less han 0.5 [29,30,32]. 5. STUDY CONTRIBUTION Mos p io esea ch on imme si e lea ning o DHH s uden s has explo ed he use o AR o gami ica ion indi idually, o en wi hou a s uc u ed app oach o model de elopmen o alida ion. These s udies a ely in eg a e inclusi e design me hodologies ha sys ema ically inco po a e expe opinion ac oss disciplines. Mo eo e , alida ion p ocesses in ea lie s udies ypically o e look he speci ic needs o DHH lea ne s in p ac ical skill se ings, pa icula ly wi hin he TVET domain. This s udy di e s by in eg a ing he NGT and FDM, a me hodological combina ion no p e iously applied in his con ex o de elop and alida e an ILPS model. I es ablishes expe consensus on h ee co e cons uc s (Lea ning Inpu Medium, P ac ical Skills Module, and AR Gami ica ion Fea u es), iden i ying 15 key elemen s alida ed wi h o e 97% ag eemen . This app oach no only ensu es con en alidi y bu also e lec s inclusi e and accessible pedagogical design. The e o e, he key con ibu ion o his s udy lies in i s pa icipa o y, consensus-d i en de elopmen p ocess ha esul s in a scalable and eplicable model o inclusi e skill-based educa ion o DHH s uden s. 6. CHALLENGE AND OPEN RESEACH ISSUES E en hough his s udy used NGT and FDM o e ec i ely cons uc and es an ILPS model o DHH s uden s in TVET, he e a e s ill a numbe o obs acles and un esol ed esea ch ques ions. 6.1 Implemen a ion and Usabili y Tes ing Despi e being app o ed by consensus o expe s, he concep has no ye been pu o he es in ac ual class oom se ings. Ac ual lea ning esul s, lea ne engagemen , and usabili y a e s ill no quan i ied. Pilo s udies wi h DHH s uden s should be used in u u e s udies o u he hone and con ex ualize he model. 6.2 Lea ne -Cen ic Model Re inemen The cu en model is d i en by expe s. Al hough his imp o es con en alidi y, DHH s uden s hemsel es did no di ec ly con ibu e o i . I s e icacy and inclusi eness would be imp o ed by inco po a ing lea ne inpu ia use expe ience esea ch o pa icipa o y design. 6.3 Scalabili y Ac oss Disciplines Al hough he model is designed o TVET, i has no been es ed o i s applicabili y o o he ields (such as science, hospi ali y, o au omo i e Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4252 enginee ing). In o de o e alua e he ILPS model's adap abili y, u u e esea ch migh in es iga e domain-speci ic modi ica ions. 6.4 In eg a ion wi h B oade Pedagogical F amewo ks Aligning he model wi h well-known educa ional heo ies like Cogni i e Load Theo y and Uni e sal Design o Lea ning (UDL) migh be ad an ageous. To assess how he app oach wo ks wi h o complemen s hese inclusi e educa ion amewo ks, mo e esea ch is equi ed. 6.5 Technological In as uc u e and Teache Readiness Adop ing imme si e echnology like AR needs ins i u ional suppo , aining, and app op ia e in as uc u e. Fu u e implemen a ion s udies mus add ess he un esol ed ques ion o whe he widesp ead adop ion in si ua ions wi h limi ed esou ces is easible. In addi ion o imp o ing he sugges ed ILPS pa adigm, esol ing hese un esol ed p oblems will u he he con e sa ion a ound inclusi e ins uc ional design and imme si e lea ning ools o special educa ion needs. 7. CONCLUSION In conclusion, his s udy success ully de eloped and alida ed an ILPS model speci ically designed o DHH s uden s in TVET. The scien i ic con ibu ion o his esea ch lies in i s no el in eg a ion o he NGT and FDM o sys ema ically elici , e ine, and alida e expe consensus on c i ical componen s o imme si e, inclusi e ins uc ional design. This me hodological inno a ion add esses a majo gap in he li e a u e by p o iding a alida ed model ha p io i izes accessibili y, engagemen , and pedagogical ele ance. The ou come is a scien i ically g ounded amewo k comp ising 15 key elemen s ac oss h ee alida ed cons uc s: Lea ning Inpu Medium, P ac ical Skills Module, and AR Gami ica ion Fea u es. This esea ch no only ad ances he me hodological igo in inclusi e educa ion model de elopmen bu also o e s p ac ical guidance o u u e implemen a ions in imme si e lea ning en i onmen s aimed a ma ginalized lea ne popula ions. This s udy adds subs an ially o he exis ing body o knowledge by add essing a c i ical and unde explo ed a ea, de eloping alida ed imme si e lea ning models o DHH s uden s in echnical and oca ional educa ion. While p e ious esea ch has examined AR and gami ica ion in gene al educa ional se ings, ew ha e a ge ed accessibili y in model de elopmen , and none ha e applied he combined use o he NGT and FDM o his con ex . This dual-me hodological app oach no only enhances he igo o model alida ion bu also ensu es inclusi i y by inco po a ing expe consensus in o e e y s age o design. Consequen ly, his s udy con ibu es a no el, empi ically suppo ed amewo k ha can be adap ed o use ac oss inclusi e educa ional echnologies. I ills bo h a me hodological and pedagogical gap and o e s ac ionable insigh o esea che s and p ac i ione s commi ed o equi able and imme si e echnical educa ion. Conside ing ha DHH s uden s need p ac ical skills o success ul imme sion lea ning, his s udy app op ia ely ou lines he ai s o ILPS o aid in hei lea ning and anks hose ILPS o assis ance. The esul s show ha he o e all expe ag eemen pe cen age is g ea e han 75% and mee s he c i e ia o expe ag eemen on his ma e . The combined use o NGT and FDM o e s a obus , e icien pa hway o inclusi e model alida ion. This app oach educes bias and ensu es ha educa ional models o DHH s uden s a e bo h pedagogically sound and p ac ically alida ed. Fu u e cu iculum de elope s can adap his me hod o di e se lea ne needs in echnical educa ion. Using FDM ins ead o he adi ional Delphi echnique allowed his s udy o be inished quickly and wi h expe consensus, which was one o i s main s eng hs. Fu he mo e, ce ain specialis s who could ha e s onge opinions han o he s we e no able o domina e he discussion due o he o ganized echnique o NGT. This migh occu , o ins ance, when a senio expe o ocus g oup membe has a epu a ion o being asse i e o dominee ing, which in luences he opinions o o he expe s. This app oach will undoub edly lowe he possibili y o bias by gua an eeing anonymi y, encou aging he expe s' opinions o uncon en ional iewpoin s, and allowing esponses o be en i ely independen wi hou he ea o c i icism om o he pa icipan s, which is ypically p esen in any egula g oup discussions o mee ings. One o he me hod's d awbacks, hough, is ha he expe s mus be eminded epea edly o p o ide hei answe s. Because i imp o es anspa ency and expedi es he p ocess, using NGT and FDM oge he o es ablish expe consensus is s ongly ad ised. The anking o he socioecological isk a iables was con i med by he FDM analysis's Jou nal o Theo e ical and Applied In o ma ion Technology 31s May 2025. Vol.103. No.10 © Li le Lion Scien i ic ISSN: 1992-8645 www.ja i .o g E-ISSN: 1817-3195 4253 obus ness. The e o e, i is s ongly ad ised o include FDM in decision-making esea ch and p ocedu es. While his s udy success ully de eloped and alida ed an imme si e lea ning p ac ical skills model o DHH s uden s using expe consensus, se e al limi a ions mus be acknowledged. Fi s , he eliance on expe opinion, hough igo ous ia NGT and FDM, excludes di ec eedback om DHH lea ne s, limi ing insigh s in o use -cen e ed design p e e ences. Second, he ela i ely small and con ex -speci ic expe panel may no cap u e b oade educa ional di e si y, a ec ing gene alizabili y. Thi d, he model’s e ec i eness has no ye been empi ically es ed in class oom se ings, meaning i s pedagogical impac emains heo e ical. Finally, logis ical aspec s such as esou ce a ailabili y, eache eadiness, and ins i u ional adop ion ba ie s we e beyond he scope o his s udy. Recognizing hese limi a ions no only enhances he anspa ency o his esea ch bu also p o ides a oadmap o u u e in es iga ions ocused on model implemen a ion and lea ne ou comes. Fu u e esea ch migh employ o he echniques by including DHH s uden s and a mo e comp ehensi e s udy se ing ha sugges s his ILPS ac oss di e en ields and cu icula. Based on he esul s o his s udy, u u e esea ch can also c ea e a pa icula module. Pe haps a specialized e e ence ma e ial o lec u e s o use while c ea ing and o ganizing hei classes will be c ea ed in he u u e wi h he de elopmen o a unique imme si e lea ning echnology usage module. REFERENCES: [1] Halley, M. and Connelly, L., 2022. “I Made Me Think Like a Nu se”: Vi ual Simula ions Wi h In e p e e s. Jou nal o Nu sing Educa ion 61(6): 334–7, Doi: 10.3928/01484834-20220404-06. [2] Meadow-O lans, K.P. and Sass-Leh e , M., 1995. Suppo Se ices o Families wi h Child en Who A e Dea : Challenges o P o essionals. Topics in Ea ly Childhood Special Educa ion 15(3): 314–34, Doi: 10.1177/027112149501500305. [3] Nkolola-Wakumelo, M. and Manyando, M., 2013. A si ua ional analysis o he use o sign language in he educa ion o he Dea in Zambia: A case o Magwe o and S Joseph’s schools o he Dea . Language Ma e s 44(3): 69–88, Doi: 10.1080/10228195.2013.839733. [4] Skel on, T. and Valen ine, G., 2003. “I eels like being Dea is no mal”: An explo a ion in o he complexi ies o de ining D/dea ness and young D/dea people’s iden i ies. Canadian Geog aphe 47(4): 451–66, Doi: 10.1111/j.0008-3658.2003.00035.x. [5] Abdul Rahim, R.H., Mohd Nizam, D.N., Mohd Naim, N.F., and Baha um, A., 2024. Imme si e Lea ning Cons uc s o Dea and Ha d-o -Hea ing S uden s in TVET P ac ical Skill. 2024 In e na ional Visualiza ion, In o ma ics and Technology Con e ence, IVIT 2024, Uni e si i Malaysia Sabah, Use Expe ience G oup, Facul y o Compu ing and In o ma ics, Sabah, Ko a Kinabalu, Malaysia: Ins i u e o Elec ical and Elec onics Enginee s Inc. p. 192–9. [6] Riaz, T., Anjum, A., Anjum, A., and Khan, M.A., 2023. Echoes in Silence: A Technological Leap o Pakis an Sign Language T ansla ion and Recogni ion. 2023 IEEE 20 h In e na ional Con e ence on Sma Communi ies: Imp o ing Quali y o Li e using AI, Robo ics and IoT, HONET 2023, Ins i u e o In o ma ion Technology, Quaid-i-Azam Uni e si y, Islamabad, Pakis an: Ins i u e o Elec ical and Elec onics Enginee s Inc. p. 204–9. [7] Nunes, J., Ve ma, A., and A una, M., 2024. Con e ing Voice Signal o Visual Indian Sign Language (ISL). In: M., P., M., B., S., G.W., and K., K.A.S., edi o s. AIP Con e ence P oceedings, ol. 3075. Depa men o Compu ing Technologies, Facul y o Enginee ing and Technology, College o Enginee ing and Technology, SRM Ins i u e o Science and Technology, Tamil Nadu, Ka ankula hu , 603203, India: Ame ican Ins i u e o Physics. [8] El Hassan, A., Nabou, A., Jaoua , M., Ez- Za zou i, H., Qazda , I., and Bel a ji, K., 2024. Towa d a Gami ied App oach Based on Augmen ed Resou ces in Educa ion. ACM In e na ional Con e ence P oceeding Se ies, Compu e Sys ems Enginee ing Labo a o y (LISI), Cadi Ayyad Uni e si y, Ma akech, Mo occo: Associa ion o Compu ing Machine y. [9] Zhao, Y. and Wang, Q., 2024. Applying augmen ed eali y mul imedia echnology o cons uc a pla o m o ansla ion and eaching sys em. Heliyon 10(7), Doi: 10.1016/j.heliyon.2024.e28700. [10] A eş, H. and Pola , M., 2025. Le e aging augmen ed eali y and gami ica ion o enhanced sel - egula ion in science educa ion.