Psychometric properties of the Spanish version of the Exercise Imagery Questionnaire (EIQ)

Co espondence o: Ma ía José, Pé ez-Fabello. Di ección Pos al: Facul ad de Bellas A es, Rúa Maes anza, 2, 36002, Pon e ed a, Spain Tel: 34 986 801 879. Email: [email p o ec ed] Cuade nos de Psicología del Depo e, 20, 3 (julio) Cuade nos de Psicología del Depo e, ol. 20, 3, 41-54 © Copy igh 2018: Se icio de Publicaciones de la Uni e sidad de Mu cia Recibido: 10/05/2019_ Mu cia (España) ISSN edición imp esa: 1578-8423 Acep ado: 04/05/2020 ISSN edición web (h p:// e is as.um.es/cpd ): 1989-5879 Psychome ic p ope ies o he Spanish e sion o he Exe cise Image y Ques ionnai e (EIQ) P opiedades psicomé icas de la e sión en español del Exe cise Image y Ques ionnai e (EIQ) P op iedades psicomé icas da e são em espanhol do Exe cise Image y Ques ionnai e (EIQ) Pé ez-Fabello, M. J.1 & Campos, A.2 1Uni e si y o Vigo; 2Uni e si y o San iago de Compos ela ABSTRACT In his s udy, he psychome ic p ope ies o he Spanish e sion o he Exe cise Image y Ques ionnai e (EIQ) we e e alua ed in a sample o 166 i s -yea unde g adua es o a Bachelo ’s deg ee in Physical Ac i i y and Spo s Sciences, 127 men, and 39 woman; Mage = 20.57, SD = 2.24. Pa icipan s exe cised a leas 4 imes a week. Mo eo e , he Spanish e sions o ou men al image y ques ionnai es we e adminis e ed: MIQ-R, VMIQ, VVIQ, and Go don Tes . Two con i ma o y ac o analyses o he EIQ we e pe o med. Fi s , a h ee- ac o model ( echnique, ene gy, and appea ance image y) he same as he o iginal model, ailed o ob ain he ecommended alues o a good i . Howe e , he second model adjus ed o wo ac o s ( echnique and ene gy image y) was good (CFI = .97, NNFI = .94, RMSEA = .07, SRMR = .04) wi h esul s simila o he o iginal es . Reliabili y was accep able as measu ed by he C onbach’s alpha, bu composi e eliabili y was lowe han ecommended. Howe e , con e gen and disc iminan alidi y was inadequa e. The e was a signi ican co ela ion be ween he EIQ and he Go don Tes . Fu he mo e, Tables ha e been included as an aid he elucida ion o he esul s o he sample o unde g adua es on he Bachelo ’s deg ee in Physical Ac i i y and Spo s Sciences. The esul s a e discussed, and subs an ia ed he Spanish e sion o he EIQ was a alid and use ul es o explo ing wo ypes o image y in pa icula : echnique and ene gy. Keywo ds: isuali a ion; eliabili y; alidi y; physical ac i i y. RESUMEN En es e s udio se e alua on las p opiedades psicomé icas de la e sión en español del Cues iona io de Imágenes de Eje cicio (EIQ) en un g upo de 166 ( 127 homb es y 39 muje es ) es udian es de de p ime año de g ado en Ciencias de la Ac i idad Física y del Depo e (Medad = 20.57, SD = 2,24). Los pa icipan es ealizaban eje cicio al menos 4 eces po semana. Además adminis amos la e sion Española de cua o cues iona ios de imagen men al: MIQ-R, Ci a: Pé ez-Fabello, M. J.; Campos, A. (2020). Psychome ic p ope ies o he Spanish e sion o he Exe cise Image y Ques ionnai e (EIQ). Cuade nos de Psicología del Depo e, 20(3), 41-54 Cuade nos de Psicología del Depo e, 20, 3 (julio) Pé ez-Fabello, Ma ía José & Campos, Al edo 42 VMIQ, VVIQ, Go don Tes . Lle amos a cabo dos análisis con i ma o ios del EIQ, en p ime luga , un modelo con es ac o es (imagen de écnica, ene gía y apa iencia) simila al o iginal, aunque no cumplía los alo es ecomendados pa a un buen ajus e. Sin emba go, el ajus e del segundo modelo con dos ac o es (imagen de écnica y ene gía) ue bueno (CFI = .97, NNFI = .94, RMSEA = .07, SRMR = .04), con esul ados simila es a la p ueba o iginal. La iabilidad medida a a és del al a de C onbach ue accep able pe o la iabilidad compues a ue más baja de lo ecommendable. La alidez con e gen e y disc imina e esul ó adecuada. Encon amos una co elación signi ica i a en e el EIQ écnica y el Go don Tes . También se incluyó una abla de ba emación del g upo es udian es o Physical Ac i i y and Spo s Sciences. Los esul ados se discu ie on, y se obse ó que la e sion Española del EIQ es una p ueba álida y ú il pa a explo a , especialmen e, dos ipos de imagen: écnica y ene gía. Palab as cla e: isualización; iabilidad; alidez; ac i idad ísica. RESUMO Nes e es udo a aliamos as p op iedades psicomé icas da e são em espanhol do Exe cise Image Ques ionnai e (EIQ). Pa icipa am 166 (127 homens e 39 mulhe es) alunos do p imei o ano de uma licencia u a em A i idade Física e Ciências do Despo o (Midade = 20.57, SD = 2,24). Os pa icipan es exe ci a am-se pelo menos 4 ezes po semana. Além disso, adminis amos a e são em espanhol de qua o ques ioná ios de imagem men al: MIQ-R, VMIQ, VVIQ, Go don Tes . Fo am ealizadas duas análises con i ma ó ias do EIQ, p imei o um modelo com ês a o es (imagem écnica, ene gia e apa ência), exa amen e como o o iginal, embo a não se a endesse aos alo es ecomendados pa a um bom ajus e. No en an o, o ajus e do segundo modelo com dois a o es ( écnica e imagem ene gé ica) oi bom (CFI = .97, NNFI = .94, RMSEA = .07, SRMR = .04) com esul ados semelhan es ao es e o iginal. A con iabilidade medida a a és do al a de C onbach oi acei á el, mas a con iabilidade compos a oi meno que o ecomendado. A alidade con e gen e e disc iminada oi adequada. Hou e uma co elação signi ica i a en e o EIQ écnica e o es e de Go don. Também oi incluída uma abela de escala do g upo de es udan es de A i idade Física e Ciências do Despo o. Os esul ados o am discu idos, e obse ou-se que o EIQ é um es e álido e ú il pa a explo a , p incipalmen e, dois ipos de imagem: écnica e ene gia. Pala as cha e: isualização; con iabilidade; alidade; ac i idade ísica. INTRODUCTION Men al image y is an e ec i e echnique ha has been used in many di e en ypes o physical and spo s ac i i ies (Cumming & Ramsey, 2009; Giacobbi, Hausenblas, & Pen ield, 2005; Kosse & Mun oe- Chandle , 2007). Using men al image y in ol es ec ea ing an expe ience s o ed in he memo y o c ea ing a new one o he pu pose o imp o ing lea ning, and de eloping ce ain skills aimed a pe o mance op imiza ion (Simons, 2000). As a men al echnique used in he spo s con ex , i is also employed o enhance sel -con idence, o con ol exci emen and anxie y, and aise mo i a ion and goal achie emen in spo speople (Cumming & Williams, 2012; Palmi Gue e o and Rie a Rie a, 2016). The cu en in e es in men al image y in ields such as spo s and physical aining has been well illus a ed by se e al s udies such as Campos, López-A aujo, and Pé ez-Fabello’s (2016) analysis o he ypes o men al image y used in physical and spo s ac i i ies. A wide ange o spo s ha e been ound o bene i om using men al image y isualiza ion aining echniques such as baske ball (Rekik, Khacha em, Belkhi , Bali, & Ja aya, 2019); oo ball, bo h in playe (Pe is- Delcampo, 2019), and eam s udies (Mo eno- Fe nández e al., 2019); gol (F ank, Land, Popp, & Schack, 2014, 2016), ennis (Fo es e al., 2019, Meie , F ank, G öben, & Schack, 2020 ), and olleyball (Fo es e al., 2020). S udies on image y mo emen ha e p ima ily ocused on one speci ic ype o image y, mo o image y. Mo o image y is a dynamic men al ep esen a ion enabling a mo o mo emen o be exe cised in he wo king memo y, wi hou pe o ming he ac ion in eali y (Dece y, 1996). The p ac ice o mo o image y, which Cuade nos de Psicología del Depo e, 20, 3 (julio) Spanish e sion o he EIQ 43 is widely popula and has been esea ched ex ensi ely, is a men al simula ion p ocess, also known as men al p ac ice, in ol ing he sys ema ic use o image y o exe cise a mo emen wi hou pe o ming he ac ion i sel (Di Rienzo e al., 2016). Recen s udies ha e used physiological measu es o assessing men al p ac ice. Lebon e al. (2019) e alua ed mo o image y and i s in ol emen in he p ocesses o p epa a ion and eal-li e execu ion o mo emen s using ansc anial magne ic s imula ion. A ecen s udy using physiological es s, Zabicki e al. (2019), obse ed spa ial pa e ns o neu onal ac i i y in imagined ac ions measu ed by MRI scanning, and signi ican posi i e co ela ions be ween neu onal neu al dissimila i y alues and he pa icipan s’ subjec i e e alua ions o he in ensi y o image i idness. Mo an and O’Shea (2020) sugges s ha he p ac ice o mo o image y d i es ela i ely high le els o he mo o sys em ha e ine and pe ec he men al ep esen a ions o an a hle e’s ac ions du ing he i s s ages o skills acquisi ion. Nume ous s udies on mo o image y ha e been unde pinned by he mo o simula ion heo y ( o u he in o ma ion see: Mo an & O’Shea, 2020; Mo eno-Fe nández e al., 2019; Pe is-Delcampo, 2019). This heo y asse s men al p ac ice is e icacious as an imagined mo emen in ol es he in e nal simula ion o he eal mo emen in designing an ac ion plan, p edic ing he ou come, and in p epa ing he means o achie e he goals. Mo eo e , he e is a unc ional equi alence be ween simula ion and execu ing an ac ion. This heo y d aws on he axonomy o Pai io (1985), subsequen ly ex ended by o he au ho s (see, Mun oe-Chandle , & Gammage, 2005), who ha e p oposed i e image y unc ions: Cogni i e Speci ic (e.g., skills), Cogni i e Gene al (e.g., s a egies), Mo i a ional Speci ic (e.g., se ing objec i es), Gene al Mo i a ional-Anxie y (e.g., exci a ion), and Gene al Mo i a ion -Dominio (e.g., sel -con idence). Bo h simula ion heo y and Pai io’s (1985) axonomy unde pin he s udies o Hall (1995), who was one o he i s au ho s o poin ou he bene i s o image y in spo s, bo h in compe i i e and ec ea ional, and link exe cise and physical ac i i y. Mo eo e , Hall sugges ed ha exe cise image y enhanced sel -con idence o ask pe o mance by p o iding people wi h a sense o achie able success. Hall’s indings pa ed he way o new esea ch on exe cise image y, gi ing ise o he Exe cise Image y Ques ionnai e (EIQ; Hausenblas, Hall, Rodge s, & Mun oe, 1999). Though his measu emen ins umen was designed on he p emise ha image y has cogni i e and mo i a ional unc ions, he ques ionnai e does no co e ed he en i e ange o image y unc ions ( o a e iew o he model see, Mun oe-Chandle , & Gammage, 2005). Pa icipan s in he Hausenblas e al.'s s udy (1999) we e olun ee ae obic p ac i ione s who belonged wo di e en uni e si y campuses (144 unde g adua es 89% women in he i s campus, and 267 unde g adua es, 97.6% women in he second). The i ems on he es we e d awn om he i s phase o he s udy ollowing he pa icipan s’ esponse o h ee basic ques ions: When do exe cise s use image y? Why do exe cise s use image y? and Wha do exe cise s image? Con en alidi y was de e mined by six expe s ( h ee exe cise p o essionals and h ee exe cise pa icipan s). A e a con i ma o y ac o analysis, Hausenblas e al. (1999) ound h ee ac o s co esponding o h ee ypes o image y commonly used by use s: (i) appea ance image y, which e e s o imagining onesel becoming heal hie and imp o ing one’s physical appea ance; (ii) ene gy image y, which is ela ed o imagining onesel ull o ene gy and eady o exe cise; and (iii) echnique image y, which implies imagining he co ec execu ion o he echnique. The analysis o he fi indexes indica ed excellen model fi (RMSEA = .05; AGFI = .93; CFI = .97). C onbach’s alphas o he subscales we e accep able (α om .71 o .85). Finally, Hausenblas e al.’s (1999) s udy ound mos pa icipan s egula ly exe cising used image y a di e en imes and si ua ions, and ha he use o men al image y inc eased wi h exe cise. Gammage, Hall and Rodge s (2000) del ed deepe in o he s udy o exe cise image y by applying he EIQ o a sample o 577 exe cise s (312 women) who pa icipa ed in a wide a ie y o physical ac i i ies. The esul s showed ha gende , exe cise equency, and he ype o physical ac i i y in luenced he use o exe cise image y. Appea ance image y was used mos o en, ollowed by echnique, and ene gy image y, espec i ely. Gende also in luenced he use o exe cise image y, wi h signi ican di e ences ound be ween men and women in appea ance, and echnique image y. Women used appea ance image y mo e o en han men, bu men used echnique image y mo e o en han women. Mo eo e , each speci ic ype Cuade nos de Psicología del Depo e, 20, 3 (julio) Pé ez-Fabello, Ma ía José & Campos, Al edo 44 o physical ac i i y was ound o de e mine p e e ences in he use o exe cise image y. La e se e al EIQ s udies ha e co obo a ed he asse ion ha he equency o exe cise image y posi i ely p edic s exe cise beha iou and in en ion ( o a e iew see Kosse & Mon oe-Chandle , 2007; Mun oe- Chandle & Gammage, 2005). Gammage e al. (2000) ha e conjec u ed di e en ypes o exe cise image y in luence mo i a ion o exe cise, and ha images based on physical appea ance play a i al mo i a ional ole in spo speople. In a mo e de ailed s udy on a sample o 150 ac i e exe cise s (193 women) engaged in an a ay o exe cise ac i i ies, S anley, Cumming, S andage, and Duda (2012) ound ha , despi e being associa ed o mo i a ion as poin ed ou by Gammage e al. (2000), appea ance image y was ac ually mo i a ion con olled, in o he wo ds, i deal wi h an ex e nal mo i a ion ocused p ima ily on physical imp o emen in spo s. Howe e , echnique image y was posi i ely ela ed o au onomous mo i a ion (in insic mo i a ion), which consis ed o using images o a co ec execu ion o an exe cise. This app oach e lec s an in insic ocus o he use o echnique image y as a psychological s a egy. Al hough S anley e al. (2012) ound no signi ican ela ion be ween ene gy image y and au onomous mo i a ion, hese au ho s assumed he mo i a ional ole o his image y would be ans o med in ime, being mo e in insically g a i ying in he long e m. Likewise, Mun oe-Chandle and Gammage (2005) ha e sugges ed he na u e o a use ’s image y may a y h oughou he p ocess o exe cising, ini ia o y images may in ol e esul s-based con en (such as hose ela ed o appea ance) in o de o ini ially se he in en ion o exe cise. La e on, du ing he aining session, use s may ake a di e en app oach using mo e ene gy images so as o keep hei ene gy and exci emen up. Hall, Rodge s, Wilson, and No man’s (2010), analysis o a sample o 470 pa icipan s (252 women), o whom 202 we e egula exe cise s, 138 non-exe cise who in ended o exe cise, and 130 non- exe cise wi h no in en ion o exe cising. They ound in e en ions wi h men al image y we e e icacious in egula exe cise s, and non- egula exe cise s, al hough in ended on exe cising, bu he e icacy o men al image y in non-exe cise s wi h no in en ion o exe cising was low. Rodge s, Hall, Blancha d, and Mun oe (2001) ound images ep esen ed 20% o he a iance in obliga o y exe cise using he EIQ be o e and a e a 10-week exe cise p og am o 243 adul exe cise s. Appea ance- ela ed images did no signi ican ly p edic obliga o y exe cise, and ene gy- ela ed images we e he s onges p edic o . In a la e s udy, Roge s, Mun oe, and Hall (2002) ound ha appea ance image y signi ican ly p edic ed exe cise in en ion bu ailed o p edic exe cise beha iou . Kosse and Mun oe-Chandle ’s (2007) e iew concluded ha exe cise image y could be an e icien ool o imp o ing pe o mance and adhe ence o physical ac i i y p og ams. Ajibua and Peculia (2016) con i med his esul on a sample o 150 unde g adua es (81 women). Bea ing in mind ha isualiza ion o physical ac i i y in non-compe i i e spo has ecei ed sca ce a en ion in he scien i ic li e a u e, IEQ can be conside ed a good gene al measu e o he men al image y o exe cise. Thus, he aim o he p esen s udy was o ansla e he EIQ in o Spanish and o assess he psychome ic p ope ies o he Spanish e sion o he Exe cise Image y Ques ionnai e (EIQ; Hausenblas e al., 1999). I was pos ula ed ha he Spanish e sion o he EIQ would ha e a simila numbe o ac o s as he English e sion (Hausenblas e al., 1999), wi h good alidi y and eliabili y. We also wo king wi h he hypo hesis was he e would be signi ican co ela ions be ween EIQ and ou men al image y es s ansla ed in o Spanish, he psychome ic p op ie ies o which ha e been analysed in p e ious s udies. Two o he mo emen image ies ques ionnai es applied in his s udy ha e been ex ensi ely used in physical and spo s ac i i ies: The Spanish e sion (Campos & González, 2010) o he Mo emen Image y Ques ionnai e-Re ised (MIQ-R; Hall & Ma in, 1997), and he Spanish e sion (Campos & Pé ez, 1990) o he Vi idness o Mo emen Image y Ques ionnai e (VMIQ; Isaac, Ma ks, & Russell, 1986). The gene al image y i idness es used was he Spanish e sion (Campos, González, & Amo , 2002) o he Vi idness o Visual Image y Ques ionnai e (VVIQ; Ma ks, 1973), as well as he Spanish e sion o he men al image y con ol es (Pé ez-Fabello, & Campos, 2004) o he Go don Tes o Visual Image y Con ol (Go don Tes , Richa dson, 1969). Cuade nos de Psicología del Depo e, 20, 3 (julio) Spanish e sion o he EIQ 45 MATERIAL AND METHOD Pa icipan s The sample consis ed o a o al o 166 i s -yea unde g adua es on he Bachelo ’s deg ee in Physical Ac i i y and Spo s Sciences a a Spanish s a e Uni e si y, 127 men and 39 women; age ange 18 o 26 yea s, mean age 20.57 yea s, (SD = 2.24). Pa icipan s we e egula exe cise s, exe cising 4 imes weekly, and p ac icing a b oad ange o spo s: i ness, a hle ics, kayak, su , ennis, sailing, owing, swimming, oo ball, ugby, and olleyball. Ma e ials The Spanish ansla ion o he Exe cise Image y Ques ionnai e (EIQ; Hausenblas e al., 1999). The EIQ consis s o 9 i ems ela ed o he equency o use o men al images while exe cising. Pa icipan s mus alue he equency o images on a 9-poin scale (1 = ne e and 9 = always). The ques ionnai e is composed o h ee subscales: echnique, ene gy, and appea ance. The echnique subscale ep esen s a cogni i e unc ion o images and is cen ed on execu ion skills and co ec ly execu ed echniques. Fo example, “When I hink abou exe cising, I imagine my o m and body posi ion” is one i em ha assesses echnique image y. The ene gy subscale is belie ed o play a mo i a ional ole and linked o being exci ed o ene ge ic while exe cising. An example o an i em is “To ake my mind o wo k, I imagine exe cising”. The hi d subscale, appea ance, is also ela ed o a mo i a ional unc ion cen ed on i ness ha includes i ems such as “I imagine a i e -me om exe cising”. The da a in he p esen s udy ob ained a C onbach’s alpha o .78 o he h ee- ac o EIQ o al, and .70 o he wo- ac o EIQ o al, as well as .78 o he echnique scale, .75 o he ene gy scale, and .78 o he appea ance scale. The C onbach’s alphas o he EIQ subscales we e below he le els ound in p e ious s udies: Hausenblas e al., 1999: echnique = .86, ene gy = .90, and appea ance = .84; Rodge s e al., 1999: echnique = .90, ene gy = .88, and appea ance = .87; Gammage e al., 2000: echnique = .86, ene gy = .85, and appea ance = .87; Gammage, Hall, & Ginis, 2004: echnique = .90, ene gy = .81, and appea ance = .88; Roge s e al., 2002: echnique = .83, ene gy = .87, and appea ance = .78; Wilson, Rodge s, Hall, & Gammage, 2003: echnique = .85, ene gy = .87, and appea ance = .88; and Hall e al., 2010: anging om .74 o .89. The Spanish e sion (Campos & González, 2010) o he Mo emen Image y Ques ionnai e-Re ised (MIQ- R; Hall & Ma in, 1997). The es comp ises 8 i ems, 4 on he isual scale and 4 on he kinaes he ic scale. Each i em in ol es mo ing an a m, leg, o he whole body. I akes ou s eps o comple e each i em: i s , a posi ion o a mo emen is desc ibed and he pa icipan is asked o ca y i ou . Secondly, a mo emen is desc ibed and he pa icipan mus pe o m i . Thi dly, he pa icipan is asked o go back o he ini ial posi ion and hen o imagine he - o himsel making he mo emen (o he i ems o he isual scale), o o eel he mo emen she o he has jus made (in he i ems o he kinaes he ic scale) wi hou ac ually making he mo emen . Finally, he pa icipan is asked o e alua e he di icul y wi h which she o he imagined o el he mo emen on a 1–7-scale, 1 being “ e y easy o imagine o eel” and 7 “ e y di icul o imagine o eel”. In his s udy he C onbach’s alpha was .86 o he o al scale, .84 o he isual subscale, and .92 o he kines he ic subscale. Vadocz, Hall, and Mo i z (1997) ound a C onbach’s alpha o .79 o he isual scale, and .79 o he kines he ic scale. Campos and González (2010) ob ained a C onbach’s alpha o .84 o he o al scale, .80 o he isual scale, and .84 o he kines he ic scale. Addi ionally, he MIQ-R ende ed a co ela ion o -.34 (p < .001) wi h he VMIQ, and -.26 (p < .001) wi h he VVIQ. The Spanish e sion (Campos & Pé ez, 1990) o he Vi idness o Mo emen Image y Ques ionnai e (VMIQ; Isaac, Ma ks, & Russell, 1986). The VMIQ consis s o 24 i ems designed o assess he abili y o isually and kines he ically imagine an a ay o mo emen s. When comple ing he VMIQ, pa icipan s a e equi ed o imagine and e alua e (1 = pe ec ly clea and as i id as no mal ision, and 5 = no image, you only know ha you a e hinking o he skill) each i em was e alua ed wice: i s , by imagining wa ching somebody else pe o m he mo emen , and second, by imagining pe o ming he mo emen hemsel es. Thus, he e a e a o al o 48 esponses wi h low sco es co esponding o mo e i id image y. The 24 i ems all in o six g oups, wi h ou i ems in each g oup. The g oups a e as ollows: i ems ela ing o basic body mo emen s (I ems 1–4, o example, “Walking”); i ems ela ing o basic mo emen s wi h mo e p ecision (I ems 5–8, o example, “Reaching o some hing on ip oe”); i ems ela ing o mo emen wi h con ol bu some unplanned isk (I ems 9–12, o example, Cuade nos de Psicología del Depo e, 20, 3 (julio) Pé ez-Fabello, Ma ía José & Campos, Al edo 46 “Falling o wa ds”); i ems ela ing o mo emen con olling an objec (I ems 13–16, o example, “Ca ching a ball wi h wo hands”); i ems ela ing o mo emen s ha cause imbalance and eco e y (I ems 17–20, o example, “Running downhill”); and i ems ela ing o mo emen s demanding con ol in ae ial si ua ions (I ems 21–24, o example, “Jumping in o wa e ”). The C onbach’s alpha o he VMIQ was .98 o he ques ionnai e o al, .97 o imagining onesel making he mo emen , and .97 imagining somebody else making he mo emen . VMIQ es – e es eliabili y has been demons a ed o e a 3-week pe iod wi h a g oup o physical educa ion s uden s, = .76 (Isaac e al., 1986). The VMIQ has also demons a ed adequa e concu en alidi y wi h he Vi idness o Visual Image y Ques ionnai e wi h no ice, expe ienced, and in e na ional-le el ampolinis s. The co ela ions we e .75, .45, and .65 espec i ely (Isaac e al., 1986). The Spanish e sion (Campos e al., 2002) o he Vi idness o Visual Image y Ques ionnai e (VVIQ; Ma ks, 1973) is composed o 16 i ems ela ing o di e en si ua ions, o example, “Visualise a ising sun. Conside ca e ully he pic u e ha comes be o e you mind’s eye”. Pa icipan s mus isualise and e alua e he i idness o each gi en image, o ins ance, in he p e ious si ua ion: “The sun is ising abo e he ho izon in o a hazy sky”. Image i idness mus be assessed on a 5-poin scale, 5 meaning “no image, you only know wha you a e hinking o he skill”, and 1 meaning “a pe ec ly clea image and as i id as no mal ision”. Thus, high sco es in he VVIQ indica e low image i idness. Pa icipan s comple e he ques ionnai e i s wi h hei eyes open and hen wi h hei eyes shu . The C onbach’s alpha o his sample was .96. Campos e al. (2002) ob ained a C onbach’s alpha o .88. Pé ez-Fabello and Campos (2004) ound ha he Vi idness o Visual Image y Ques ionnai e co ela ed om -.40 o -.24 wi h he Go don Tes o Visual Image y Con ol (Go don es ; Richa dson, 1969). In a ecen s udy, Pé ez-Fabello and Campos (2020) ob ained a C onbach’s alpha o .93. The Spanish e sion (Pé ez-Fabello, & Campos, 2004) o he Go don Tes o Visual Image y Con ol (Go don Tes , Richa dson, 1969) is comp ised o 12 i ems in which pa icipan s a e asked o imagine a mo o ca and hen asked o a e on a 3-poin scale (0 = no, 1 = unsu e, 2 = yes) whe he hey can imagine i in a ious di e en colou s, posi ions, and s a es o mo ion, o example “Can you see a ca s anding in he oad in on o a house?” To al sco es ange om 0 o 24, whe e high sco es indica e be e image-con ol. The C onbach’s alpha o his sample was .74. In s udies p e ious, he C onbach’s coe icien o he Go don Tes was es ima ed a .69 (Pé ez-Fabello & Campos, 2004). Pé ez-Fabello and Campos (2004) ob ained a co ela ion o -.40 (p < .001) wi h he Vi idness o Visual Image y Ques ionnai e (VVIQ; Ma ks, 1973). P ocedu e The s udy was conduc ed in acco dance wi h e hical ules con ained in he Decla a ion o Helsinki o 2000, and was app o ed by he e hics commi ee o he Uni e si y o he i s au ho . All unde g adua e s uden s olun ee ed o pa icipa e in he s udy. The ansla ion p ocess o he EIQ (Hausenblas e al., 1999) was pe o med in ou s eps. Fi s , he i s au ho , who is luen in English and Spanish, ansla ed he EIQ in Spanish. Then, he second au ho , who was also luen in English and Spanish back- ansla ed he EIQ back o English wi hou e e ing o he o iginal e sion. Thi d, he bo h au ho s d a ed he inal e sion o he EIQ. Finally, bo h au ho s and a p o essional English o Spanish ansla o , who is an expe in psychology, edi ed he syn ax o he i ems, spelling, and any g amma ical e o s o he inal e sion o he EIQ (see Appendix I). A he gym o he Facul y o Physical Ac i i y and Spo s Sciences, he ollowing ques ionnai es we e adminis e ed in g oups o app oxima ely 20 pa icipan s: he EIQ, MIQ-R, VMIQ, VVIQ, and he Go don Tes . The o de o es s was coun e balanced. Da a Analysis S a is ical analysis was pe o med using he IBM SPSS 20.0 so wa e p og am and IBM SPSS Amos 20 (2011). In he p elimina y analysis, a ypical alues on he scale we e de e mined by he Mahalanobis dis ance. The baseline o de ining a case as a ypical was se a he e y conse a i e le el o .001, conside ing he indica ions o Hai , Ande son, Ta ham, Black (1999). The uni a ia e no mali y was assessed wi h he skewness and ku osis whe e indexes close o ze o and less han 2 indica e he Cuade nos de Psicología del Depo e, 20, 3 (julio) Spanish e sion o he EIQ 47 simila i y wi h he no mal cu e o uni a ia e da a (Bollen & Long, 1993; Nu iala e al., 2012). Ma dia's coe icien was used o mul i a ia e no mali y. Acco ding o Bollen (1989), mul i a ia y no mali y exis s when Ma dia’s coe icien is less han p (p + 2), whe e p is he numbe o a iables obse ed. To assess, i he hypo hesis gene a ed o he o iginal s udies was con i med, we pe o med a con i ma o y ac o analysis (CFA) ha gi es model- i ing indica o s (Jö eskog & Sö bom, 1993, 1999). Global i o models we e assessed using six indexes: he χ2 o i s deg ees o eedom (d ) a io—because his index alone is e y sensi i e o sample size (Jö eskog & Sö bom, 1993)— he goodness o i index (GFI), he compa a i e i index (CFI), he non-no med i index (NNFI), he oo mean squa e e o o app oxima ion (RMSEA), and he s anda dized oo mean squa e esidual (SRMR). Values o he χ2 o d a io be ween 0 and 3 a e ecommended o a good i (Bollen & Long, 1993). GFI alues highe han .90 a e ecommended, whe eas alues equal o .95 o highe a e ecommended o CFI and NNFI (Hu & Ben le , 1999; Jö eskog & Sö bom, 1993, 1999). Values equal o .08 o lowe a e ecommended o RMSEA and SRMR (B owne & Cudeck, 1993). The C onbach’s alpha (α) and composi e eliabili y (CR) we e calcula ed o e alua e he in e nal consis ency o he ac o s, and a cu -o alue o .70 was applied as ecommended by Nunnaly (1978) and Hai , Black, Babin, and Ande son (2014). As o con e gen alidi y analysis ( o con i m he i ems we e ela ed o hei espec i e ac o ), he a e age a iance ex ac ed (AVE) was calcula ed using he ecommended (AVE ≥.50) e e ence alue (Hai e al., 2014). In ela ion o disc iminan alidi y ( o de e mine he ac o s we e no ela ed o each o he a e), he squa e o he ac o s co ela ions we e con i med o be below he AVE o he same (Hai e al., 2014). As a measu e o concu en c i e ion alidi y, he Pea son p oduc –momen co ela ion coe icien was used o co ela e he EIQ o he o he image y es s o compa able measu es. Finally, he pe cen iles o he EIQ Subscales in bo h models (2- ac o and he 3- ac o EIQ) a e shown. RESULTS In he p elimina y s udy, alues o .001 we e ound du ing he calcula ion o he Mahalanobis dis ance, and in acco dance wi h he conse a i e c i e ion o Hai e al. (1999), a baseline alue o .001 was es ablished o de ining cases as ou lie , wi h he emo al o 11 obse a ions. As o uni a ia e no mali y, he skewness and ku osis indexes o he ques ionnai es we e nea ze o and below he alue o 2. In addi ion, he uni a ia e no mali y was calcula ed by he skewness and ku osis o each i em on he EIQ, ob aining alues ha in mos i ems we e close o ze o o less han he 2, which was abo e he ecommended alues (Bollen & Long, 1993; Nu iala e al., 2012). Mul i a ia e no mali y was con i med by Ma dia's coe icien 17.26 o he h ee- ac o e sion, and 6.51 o he wo- ac o e sion o he EIQ (Bollen, 1989). The da a no mali y ob ained jus i ied he use o he maximum likelihood me hod. Figu e 1 shows s anda dized coe icien s o he p oposed model, wi h alues anging om .55 (I em 2) o .94 (I em 8). All alues we e s a is ically signi ican (p < .001). As es ima ed by he model, he ela ions among he h ee ac o s we e .29 be ween echnique and ene gy, .43 be ween ene gy and appea ance, and .44 be ween echnique and appea ance. Figu e 1. Th ee- ac o model o Exe cise Image y Ques ionnai e (EIQ) The ini ial model did no adjus sa is ac o ily o he da a χ2 (68.65) and d (24), and he a io was 2.86 (p < .001). As he esul o a Chi-squa e is pa icula ly sensi i e o sample size, i is usually accompanied by he a io be ween he alue and he deg ees o eedom. This esul was be ween 0 o 3 ha can be conside ed a good model i o da a (Jö eskog & Cuade nos de Psicología del Depo e, 20, 3 (julio) Pé ez-Fabello, Ma ía José & Campos, Al edo 48 Sö bom, 1993, 1999). The o he Index alues we e: GFI (.91), CFI (.90), and NNFI (.86). The RMSEA and SRMR alues we e .11 and .08. Though he da a in he p esen s udy ailed o each he ecommended alues abo e .95 o CFI and NNFI (Hu & Ben le , 1999; Jö eskog & Sö bom, 1993, 1999), he GFI ob ained a alue highe han he ecommended .90 (Hu & Ben le , 1999; Jö eskog & Sö bom, 1993, 1999). Whils he SRMR was wi hin ecommended alues, he RMSEA ailed o ul il he ecommenda ions o B owne and Cudeck (1993) as i exceeded he ecommended alue o .08 o lowe o RMSEA and SRMR. As i em 2 explained only 30% o he la en ac o , and conside ing each dimension had 3 i ems, a model wi hou he appea ance ac o was assessed wi h a sho wo- ac o e sion o he EIQ o model imp o emen . Figu e 2 shows s anda dized coe icien s o he p oposed model, wi h alues anging om .65 (I em 1) o .81 (I em 5). All alues we e s a is ically signi ican (p < .001). As es ima ed by he model, he ela ion be ween bo h ac o s was .43 be ween echnique and ene gy. Figu e 2. Two- ac o model o Exe cise Image y Ques ionnai e (EIQ). These alues sugges ed a good i o he model χ2 (14.95) and d (8), a io was 1.87 (p = .06). Index alues we e: GFI (.97), CFI (.97), and NNFI (.94). The RMSEA and SRMR alues we e .07 and .04. In e nal consis ency as measu ed by he C onbach’s alpha, and composi e eliabili y a e shown in Table 1. Mos o he C onbach’s alphas we e abo e .70 as sugges ed by Nunnaly (1978), and Hai e al. (2014). Howe e , composi e eliabili y was below ecommended alues. Table 1 C onbach’s Alpha, Composi e Reliabili y o he h ee-Fac o EIQ, and wo-Fac o EIQ EIQ h ee ac o s EIQ echniqu e EIQene g y EIQappea anc e EIQ o a l C onbach’ s alpha .78 .75 .78 .78 Composi e Reliabili y CR) .64 .34 .59 .73 EIQ wo ac o s C onbach’ s alpha .70 Composi e Reliabili y (CR) .41 .30 .34 The analysis o con e gen alidi y e alua ed se e al c i e ia. Fi s , all o he s anda dized ac o ial loads we e abo e .6, excep o i em 2 o he h ee- ac o model. The alues in bo h o he EIQ models p oposed we e abo e 1.96, con i ming he alidi y o he indica o s used o measu ing he cons uc s. Second, he esul s indica ed ha all he ac o s eached he ecommended cu -o alues (AVE ≥.50) (Hai e al., 2014). The disc iminan alidi y o bo h o he models p oposed showed he AVE alue ( igu es in bold) o each ac o we e highe han he squa e o he co ela ion be ween he cons uc and each o he o he s (Hai e al., 2014) (see Table 2). Table 2 Con e gen and Disc iminan Validi y o he h ee-Fac o EIQ, and wo- Fac o EIQ EIQ h ee ac o s EIQ echnique EIQene gy EIQappea ance EIQ echnique .54 EIQene gy .08 .50 EIQappea ance .19 .18 .80 EIQ wo ac o s EIQ echnique .53 EIQene gy .18 .50 AVE = numbe s in bold Among he co ela ions be ween he EIQ scales and he o he image y es s, i is wo h no ing he co ela ion be ween he EIQ echnique and he Go don Tes (.26, p < .01). The EIQ ene gy subscale also co ela ed signi ican ly -.16 wi h he isual MIQR (p < .05), bu he co ela ion was small as shown in see Table 3. The pe cen iles a e included in Table 4. Cuade nos de Psicología del Depo e, 20, 3 (julio) Spanish e sion o he EIQ 49 Table 3 Pea son Co ela ions be ween he Ques ionnai es and he EIQ Subscales Ques ionnai es 1 2 3 4 5 6 7 8 9 1 VVIQ 2 Con ol Tes -.34** 3 VMIQ .37** -.13** 4 MIQR isual -.19** .03 -.29** 5 MIQRkines he ic -.31** .08 -.16* .51** 6 MIQR o al -.30** .07 -.24** .81** .92** 7 EIQ echnique -.11 .26** -.10 .15 .13 .16* 8 EIQene gy .03 .06 -.08 -.16* .02 .06 .21** 9 EIQappea ance .05 -.03 -.06 -.04 .07 .03 .42** .36** 10 EIQ o al .06 .13 -.10 -.05 .09 .04 .80** .65** .77** *p < .05, **p < .01 Table 4 Pe cen iles, S anda d Sco es, and Di ec Sco es o he EIQ Subscales and he wo- ac o EIQ o al, and he h ee- ac o EIQ o al EIQ echnique EIQene gy EIQappea ance EIQ o al (2- ac o ) EIQ o al (3- ac o ) P Z DS Z DS Z DS Z DS Z DS 5 -1.81 13 -1.69 5 -1.74 12.5 -1.54 24 -1.52 41.5 10 -1.11 16 -1.19 8 -1.20 15 -1.16 27 -1.28 44 20 -.65 18 -.70 11 -.77 17 -.78 30 -.71 50 30 -.42 19 -.38 13 -.34 19 -.40 33 -.42 53 40 .05 21 -.05 15 .09 21 -.15 35 -.14 56 50 .28 22 .28 17 .31 22 .10 37 .24 60 60 .51 23 .44 18 .53 23 .48 40 .43 62 70 .74 24 .77 20 .745 24 .73 42 .82 66 80 .97 25 .09 22 .96 25 1.23 46 1.10 69 90 1.44 27 1.42 24 1.40 27 1.49 48 1.48 73 P = Pe cen iles, Z = S anda d sco es, DS = Di ec sco es DISCUSSION The p ima y aim o his s udy was o alida e he Spanish e sion o he Exe cise Image y Ques ionnai e (EIQ, Hausenblas e al., 1999) in a sample o Spanish unde g adua es on he Bachelo ’s deg ee in Physical Ac i i y and Spo s Sciences, he esul s e ealed he ini ial model, simila o he o iginal s udy (3- ac o / 9 i ems), ailed o sa is ac o ily adjus o he da a. Howe e , as p e iously men ioned abo e, he model was adjus ed by elimina ing he ac o appea ance (wi h h ee i ems), and he new wo- ac o model was adequa e. The C onbach’s alphas o he wo p oposed EIQ models we e accep able, abo e .70. Howe e , composi e eliabili y was weak, no eaching he abo e .70 c i e ion (Hai e al., 2014), bu mos o hese sco es came close o .50, conside ed by He nández Mendo, Mo ales Sánchez and T igue o Molido (2013) as a eliable cons uc indices. The AVE in mos cases was highe han he .50, abo e he ecommended alue (Hai e al., 2014), indica ing he cons uc explained mo e han hal o he a iance o all o i s own indica o s. In ela ion o disc iminan alidi y, he esul s indica ed he cons uc s we e su icien ly independen among each o he , gi en ha in each ac o he squa e o he ac o s co ela ions was below he AVE o he same (Hai e al., 2014). Ou esul s subs an ia ed a h ee- ac o model, simila o he model o he o iginal s udy (Hausenblas e . al., 1999), and an imp o ed adjus ed wo- ac o model gi en ha he h ee- ac o model should be imp o ed. A possible explana ion o he disc epancies be ween he esul s o model in he p esen s udy and he model o he s udy o iginal may by he di e ence in he sample popula ions unde s udy. Hansenblas e al.’s (1999) s udy analysed wo samples, 89 % and 97.6%, espec i ely, o women ae obic p ac i ione s, in compa ison o he p esen s udy wi h 77% male unde g adua es o he Bachelo o Physical Ac i i y and Spo s Sciences deg ee who egula ly p ac ised an a ay o spo s. Hall (1995) asse ed men al image y was ela ed o he ype o exe cise, and may e en be ela ed o excessi e exe cise beha iou such as exe cise dependence. Mo eo e , expe ience ( he leng h o ime one has been exe cising) may also impac he unc ions o image y (Mun oe-Chandle & Gammage, 2005). The sco es o he Spanish e sion o he EIQ scales we e simila o hose p e iously ob ained by Gammage e al. (2004), who ound signi ican co ela ions (p < .01) among he h ee EIQ scales: echnique image y .52 wi h appea ance image y and .56 wi h ene gy image y; and ene gy image y .48 wi h appea ance image y. Besides, Gammage e al., (2004) ob ained signi ican co ela ions be ween he EIQ scales and o he scales ela ed o exe cise. Technique image y co ela ed .21 (p < .01) wi h he Sel -P esen a ion in Exe cise Ques ionnai e (SPEQ,