In Re iew
Valida ion o he Spanish e sion o he S uden Adap a ion
o College Ques ionnai e (SACQ-50) wi h Pe u ian S uden s
Jou nal:
Jou nal o Ame ican College Heal h
Manusc ip ID
JACH-2021-07-0532.R2
Manusc ip Type:
Resea ch A icle
Keywo ds:
Adjus men o college, Fac o analysis, College s uden s, SACQ-5O
Abs ac :
Objec i e: o e alua e he psychome ic p ope ies o he sho e sion
o he Spanish S uden Adap a ion o College Ques ionnai e (SACQ-50,
Spanish e sion). Pa icipan s: 1513 s uden s om 14 uni e si ies in
Pe u, mainly emales (61.5%), aged be ween 18 and 30 yea s. Me hod:
C oss-sec ional s udy wi h he ques ionnai e adminis e ed in pe son.
Con i ma o y ac o ial analysis was conduc ed o con i m he scale
alidi y. Resul s: adequa e i s we e ob ained o he mul idimensional
s uc u e and o he second o de ac o o he es . Alpha and omega
coe icien s indica ed adequa e es eliabili y. Conclusions: The Spanish
e sion o he SACQ-50 is a mul idimensional scale displaying adequa e
eliabili y and alidi y. The scale may be use ul o esea che s and o he
p o essionals wo king in he uni e si y con ex .
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
In Re iew
Valida ion o he Spanish e sion o he S uden Adap a ion o College
Ques ionnai e (SACQ-50) wi h Pe u ian S uden s
Abs ac
Objec i e: o e alua e he psychome ic p ope ies o he sho e sion o he Spanish S uden
Adap a ion o College Ques ionnai e (SACQ-50, Spanish e sion). Pa icipan s: 1513
s uden s om 14 uni e si ies in Pe u, mainly emales (61.5%), aged be ween 18 and 30 yea s.
Me hod: C oss-sec ional s udy wi h he ques ionnai e adminis e ed in pe son. Con i ma o y
ac o ial analysis was conduc ed o con i m he scale alidi y. Resul s: adequa e i s we e
ob ained o he mul idimensional s uc u e and o he second o de ac o o he es . Alpha
and omega coe icien s indica ed adequa e es eliabili y. Conclusions: The Spanish e sion
o he SACQ-50 is a mul idimensional scale displaying adequa e eliabili y and alidi y. The
scale may be use ul o esea che s and o he p o essionals wo king in he uni e si y con ex .
Page 1 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
In oduc ion
Du ing uni e si y li e, young people ace a se ies o obs acles and challenges ha o m pa
o hei de elopmen and ha in ol e coping wi h new, o en un amilia , si ua ions. Du ing his li e
s age, s uden s expe ience di e se demands om amily and om social and in e pe sonal
en i onmen s, and hey also encoun e he cha ac e is ic challenges o his de elopmen al s age, such
as de eloping iden i y, becoming independen , o ming iendships, choosing a ca ee and
es ablishing u u e goals.1,2 A his ime o li e, he uni e si y en i onmen ep esen s an impo an
aining g ound, whe e young people ace a se ies o challenges o academic, social and ins i u ional
na u e.3 S uden s who ind i di icul o adap o hese demands will gene ally be mo e suscep ible o
de eloping men al heal h p oblems.4
In esponse o he impo ance o s udying he p ocess o imme sion in uni e si y li e, Bake
and Si yk5 de eloped a sel - epo ins umen denomina ed he
S uden Adap a ion o College
Ques ionnai e
(SACQ). The SACQ aims o e alua e he p ocess o adap ing o he mul iple challenges
ha may a ise du ing academic aining wi hin he uni e si y con ex . The au ho s o he scale
p oposed ha he demands o uni e si y li e can be g ouped in o ou speci ic dimensions: academic
adjus men , social adjus men , pe sonal-emo ional adjus men and ins i u ional a achmen .
Acco ding o hese au ho s, academic adjus men in ol es he deg ee o success in acing academic
challenges and in academic achie emen ; social adjus men e e s o handling o social links,
in e pe sonal ela ionships and in eg a ion in social sphe es wi hin he uni e si y; pe sonal/emo ional
adjus men e alua es he emo ional s a e associa ed wi h uni e si y li e and in ol es esponses o
s ess and anxie y o physical eac ions (e.g. sleep dis up ion) o academic demands; and, inally,
ins i u ional a achmen e lec s he deg ee o which he s uden iden i ies wi h and eels commi men
owa ds he uni e si y o which hey belong.5,6
The SACQ has been widely used o s udy he a iables a ec ing uni e si y adap a ion and
has enabled iden i ica ion o s uden s who expe ience di icul ies in adjus men .3,6–10 The ins umen
has also iden i ied a iables ela ed o academic success and academic pe sis ence,11–13 academic
Page 2 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
bu nou 14, uni e si y s uden e en ion, social suppo and whe he o no he s uden has ecei ed
counselling 3. Likewise, he SACQ has e ealed links be ween a iables associa ed wi h men al
heal h, such as he p esence o dep essi e symp oms, s ess, coping and he use o psychological
counselling se ices.3,14,15
The SACQ has been used in di e se con ex s,3,9,16–19 and di e en s udies ha e demons a ed
he adequa e eliabili y o he ins umen .8,14,17 In addi ion, alpha coe icien s highe han .90 ha e
been epo ed o all subscales o he SACQ in a me a-analysis o 237 independen samples.3 Indeed,
since publica ion o he scale, a ious s udies ha e epo ed a es eliabili y consis en wi h ha
epo ed in he manual.10
Howe e , a ious c i icisms ha e been made in ega d o ce ain i egula i ies in he
de elopmen o he scale, which a e expec ed o a ec i s capaci y o ep oduce he p oposed ac o ial
s uc u e.20 In his espec , he scale has only been es ed in eshmen yea s uden s, despi e ha ing
been p oposed o use in he gene al uni e si y popula ion. This calls in o ques ion he heo e ical
basis used by he au ho s o o ganize he dimensions. In addi ion, he e is poo dis inc ion be ween
he es i ems, which hus end o be sa u a ed by mo e han one ac o .3,10,20,21 Finally, he use o
p incipal componen s analysis wi h he in e co ela ion ma ix o he SACQ subscales in he o iginal
s udy has been c i icised,6 as his may p e en adequa e explo a ion o he ac o ial s uc u e o he
i ems and whe he hey co espond o he domains o which hey supposedly belong.20
Rod íguez e al22 conduc ed a new e ision o he SACQ which imp o ed he p e ious e sion
o he ins umen while e aining i s s ong poin s. This new ins umen denomina ed he SACQ-50,
Spanish e sion,
e ained he mos ep esen a i e i ems in each domain and ep oduced he o iginal
s uc u e o he es . The esea che s p oposed elimina ing hose i ems yielding a s anda dized
coe icien below .30, hose wi h seconda y loadings in o he ac o s and hose ha did no e lec he
ac ual si ua ions in La in-Ame ican and Ibe o-Ame ican uni e si ies.
Page 3 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Wi h his aim, he esea che s e alua ed a sample o 300 i s -yea s uden s and inally
concluded ha 50 o he es i ems wo ked bes as hey we e e icien ly dis ibu ed in he ou
dimensions, ollowing he model p oposed by Bake and Si yk.6 Howe e , al hough he esul s
demons a ed adequa e psychome ic unc ioning o his s uc u e, he possible exis ence o a global
dimension o adap a ion was no explo ed,6 despi e he ac ha he esea che s p oposed ha h ee
i ems o he SACQ-50 would con ibu e o his gene al measu e.23 Analysis o his second o de
dimension has been lacking in alida ion s udies o has only been applied o he i ems con o ming
he ou heo e ical dimensions, bu no o he o he es i ems.24
The new, sho e sion mus be suppo ed by psychome ic da a, pa icula ly when ob ained
om di e se samples, as his would demons a e he anscul u al alidi y o he scale. The e o e, he
objec i e o he p esen s udy was o e alua e he psychome ic p ope ies o he SACQ-50, Spanish
e sion. A u he aim was o es he possibili y o es ablishing a second o de scale ha includes he
4 dimensions and all o he es i ems. In addi ion, gi en ha he dimensionali y o he SACQ and i s
a ian s ha e been called in o ques ion, unidimensional models will be compa ed in o de o e alua e
he iabili y o he mul idimensional aspec o he heo y.3,6,24
Me hod
Pa icipan s
The pa icipan s comp ised 1513 s uden s om ou een uni e si ies (6 p i a e and 8 public) in eigh
ci ies in Pe u. The uni e si ies all belong o he
Red Pe uana de Uni e sidades
(Pe u ian Uni e si y
Ne wo k) (RPU), which is o med by 26 o he uni e si ies in he coun y. Mos o he pa icipan s
we e emale (61.5%) and hey anged in age be ween 18 and 30 yea s (
M =
20.82,
SD
= 2.38). A high
p opo ion o he pa icipan s (41.8%) we e unde g adua es (i.e. in yea s 1 o 4 o hei uni e si y
cou ses, and mos (61.37%) we e a ending a public uni e si y. The pa icipan s we e en olled in he
ollowing acul ies: 45.1%, A and Social Sciences; 18.9%, Sciences and Enginee ing; 18.4%,
Biological o Heal h Sciences; 16.8%, Managemen and Accoun ing; and 0 8%, o he acul ies.
Page 4 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Ins umen s
The ins umen e alua ed in he p esen s udy was he Spanish, 50-i em e sion o he SACQ
(Spanish SACQ-50).23
This e sion is based on he ins umen de eloped by Bake and Sy ik6
including 4 dimensions: academic adjus men , social adjus men , pe sonal-emo ional adjus men and
ins i u ional a achmen . These dimensions we e conside ed oge he wi h an addi ional 3 i ems o
indica e he o e all adjus men . The ins umen is based on a Like scale wi h 9 esponse op ions
anging om 1 (“ o ally disag ee”) o 9 (“ o ally ag ee”).
Rega ding he eliabili y o he scale, alpha and omega coe icien s highe han .7 we e
epo ed o he ou dimensions: academic adjus men (⍺ = .85, ⍵ = .86), social adjus men (⍺ = .85,
⍵ = .84), pe sonal-emo ional adjus men (⍺ = .86, ⍵ = .84) and ins i u ional a achmen (⍺ = .90, ⍵ =
.92).23
Page 5 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
P ocedu e
The p ojec was app o ed by an acc edi ed e hics commi ee. Ins i u ions belonging o he
RPU we e hen in i ed o pa icipa e in he s udy. They we e in o med abou he s udy objec i es and
pe mission was eques ed o he esea che s o ex ac he equi ed da a.
In he uni e si ies ha ag eed o pa icipa e in he s udy, in o ma ion was collec ed in ace- o-
ace in e iews in he class ooms. In o med consen was ob ained om pa icipa ing s uden s, who
signed a o m ou lining he olun a y, con iden ial and anonymous na u e o he s udy. The s uden s
comple ed he pape ques ionnai es in he class ooms, supe ised by he esea che s.
Da a analysis
The da a analysis was conduc ed wi h R s a is ical so wa e, e sion 4.0.4.25 Following he
ecommenda ions o Bake and Si yk,6 i ems wi h h ee o mo e missing esponses we e excluded
om he da a se . Likewise, in cases wi h one o wo missing esponses, he mean alue was calcula ed
o he i em.
The Mahalanobis dis ance was used o de ec a ypical cases, and i ems wi h ex eme alues
(p < .001) we e excluded. In addi ion, Ma dia’s es o mul i a ia e no mali y, ca ied ou wi h he
MVN 5.8 package,26 indica ed ha he da a we e no no mally dis ibu ed (Ma dia’s coe icien o
mul i a ia e ku osis o 85.73 and a coe icien o skewness o 47767.13).
Owing o he non-no mal dis ibu ion o he da a and ollowing p e ious ecommenda ions,3
a con i ma o y ac o ial analysis (CFA) was conduc ed wi h he Diagonally Weigh ed Leas Squa es
(DWLS) es ima ion me hod, o e alua e he in e nal s uc u e o he scale. In addi ion, he
unidimensional i was de e mined o each dimension o he sho , Spanish e sion o he SACQ-50,
wi h he aim o iden i ying dimensions wi h poo i s.
Two models we e subsequen ly es ed: (1) a ou - ac o model composed o 47 i ems included
in he Spanish SACQ-50 scale, as p oposed by Rod íguez e al23, and (2) a second-o de model
including a gene al ac o composed o he ou dimensions o he SACQ and he h ee i ems
sugges ed by Rod íguez e al23 as pa o he o e all adap a ion sco e. Bo h models we e compa ed
wi h wo unidimensional models including espec i ely 47 and 50 i ems, wi h he aim o es ing he
obus ness o he mul idimensional scale.
Page 6 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Among he a ious goodness o i indices a ailable, he ollowing inc emen al i indices
we e conside ed: he Compa a i e Fi Index (CFI) and he Tucke Lewis Index (TLI) and he absolu e
i indices Roo Mean Squa e E o o App oxima ion (RMSEA) and S anda dized Roo Mean Squa e
Residual (SRMR). Values o CFI and TLI > .90, and alues o RMSEA and SRMR < .08 we e
conside ed o indica e adequa e i s.27 The in e nal s uc u e and eliabili y o he scale we e e alua ed
using he la aan 0.6-8 package,28 and he inal model was plo ed using he semPlo 1.1.2 package.29
Wi h he aim o acili a ing in e p e a ion o he da a and compa ison wi h he indings o p e ious
s udies, he i ems a e p esen ed using he o iginal nume a ion p oposed by Bake and Si yk6. Finally,
he es eliabili y was measu ed wi h he alpha and omega coe icien s o each dimension and he
o al scale, measu ed using he semTools 0.5-4 and semPlo 1.1.2 packages30.
Resul s
E alua ion o he unidimensional s uc u e o each dimension o he Spanish SACQ-50
e ealed an adequa e i o each o he models (see Table 1). The Pe sonal-emo ional dimension
yielded an almos pe ec i o he da a, unlike he Academic and Ins i u ional dimensions, which
yielded ma ginal alues o he co esponding indica o s. Thus, he Academic dimension yielded
alues a he uppe limi o he 90% con idence in e al o he RMSEA abo e .08, and he Ins i u ional
dimension yielded an SRMR alue o .082. A he local le el, only one i em yielded a alue below .3
(i em 44 = .24). None o he i ems we e emo ed om he indi idual dimensions due o he o e all
i , wi h he aim o main aining he o iginal s uc u e o he scale. In addi ion, McDonald’s alpha and
omega coe icien s o eliabili y indica ed adequa e eliabili y o all o he models (see Table 1).
Secondly, a one- ac o solu ion was compa ed wi h he ou - ac o model p oposed by
Rod íguez e al23. The esul o his analysis showed ha he uni ac o ial model did no p o ide a
good da a i (see Table 1). In addi ion, compa ison o he models wi h a Chi squa ed di e ence es
showed ha he mul i ac o ial model, based on he heo y p oposed by Bake and Si yk6, p o ided a
signi ican ly be e da a i han he model wi h a single o e all dimension (Δχ2 = 4053.69,
d
= 6,
p
< .001). A he local le el, only i em 44 ( om he 69-i em SACQ scale) yielded a ac o ial loading
Page 7 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
below .3 in he unidimensional model (i em 44 = .26). Likewise, excep o he ela ionship be ween
he Ins i u ional and Pe sonal-Emo ional dimensions (
= .44), he in e co ela ions be ween he
dimensions o he Spanish SACQ-50 anged om .54 o .86. The McDonald’s alpha and omega
coe icien s o eliabili y o his model anged om .83 o .88 and om .83 o .84 espec i ely o
he dimensions o he Sp SACQ-50, indica ing adequa e eliabili y o his solu ion.
[Table 1 he e]
In esponse o he high le els o in e co ela ion be ween he dimensions o he Spanish
SACQ-50, he exis ence o a second-o de ac o was e alua ed by compa ison o he model wi h a
single ac o model composed o all 50 i ems, ollowing he ecommenda ions o Rod íguez e al23.
The esul s o his analysis showed ha he uni ac o ial model did no p o ide a good da a i (see
Table 1). Likewise, compa ison o he models using a Chi squa ed di e ence es showed ha he
second-o de model yielded a signi ican ly be e da a i han he unidimensional model (Δχ2 = 3856,
d
= 3,
p
< .001). A he local le el, i em 44 yielded a s anda dized coe icien o less han .3 in he
second o de model (i em 44 = .26).
[Figu e1 he e]
Rega ding he eliabili y o his solu ion, he alpha and omega alues o espec i ely .939 and
.924 indica ed adequa e in e nal consis ency o he o e all scale. These coe icien s we e also
calcula ed o he subscales o he Spanish SACQ-50, ob aining α = .83 and ω = .82 o he Academic
dimension, α = .83 y ω = .80 o he Social dimension, α = .84 y ω = .80 o he Pe sonal-Emo ional
Page 8 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
2012;48(2):1-36.
29. Epskamp S. SemPlo . Pa h diag ams and isual analysis o a ious SEM packages Ou pu . R
package e sion 1.1.2. Published online 2019. h ps://c an. -p ojec .o g/package=semPlo
30. Jo gensen TD, Po np ase mani S, Schoemann AM, e al. Use ul ools o s uc u al
equa ion modeling. Published online 2021. h ps://c an. -
p ojec .o g/web/packages/semTools/index.h ml
31. Ame ican Educa ional Resea ch Associa ion, Ame ican Psychological Associa ion, Na ional
Council on Measu emen Educa ion.
S anda ds o Educa ional and Psychological Tes ing
.;
2014.
32. Cook DA, Beckman TJ. Cu en concep s in alidi y and eliabili y o psychome ic
ins umen s: Theo y and applica ion.
Am J Med
. 2006;119(2):166.e7-166.e16.
doi:10.1016/j.amjmed.2005.10.036
33. Rocha M, Ma os PM. Adap ação do S uden Adap a ion o College Ques ionnai e (SACQ) a
uma amos a de adolescen es de escolas egula es, p o issionais e pólos de ap endizagem.
Psicol Educ e Cul
. 2008;12(1):171-196.
34. Dominguez-La a S, P ada-Chapoñan R, G a ini-Donado M. In e nal s uc u e o S uden
Adap a ion o College Ques ionnai e (SACQ) in i s -yea uni e si y s uden s.
Educ Medica
.
2021;22(5):335-345. doi:10.1016/j.edumed.2019.06.004
35. Cha loque Céspedes R, Va a-Ho na A, Lopez-Oda D, San i-Hua anca I, Diaz-Rosillo A,
Asencios-Gonzalez Z. Ausen ismo, p esen ismo y endimien o académico en es udian es de
uni e sidades pe uanas.
P opósi os y Rep esen
. 2018;6(1):83-107.
doi:10.20511/py 2018. 6n1.177
36. O’Donnell MB, Shi ley LA, Pa k SS, Nolen JP, Gibbons AM, Rosén LA. The college
adjus men ques ionnai e: A measu e o s uden s’ educa ional, ela ional, and psychological
adjus men o he college en i onmen .
J Coll S ud De
. 2018;59(1):116-121.
doi:10.1353/csd.2018.0009
Page 15 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Table 1
Con i ma o y ac o ial analysis and eliabili y o he SACQ-50, Spanish e sion
Model
χ2
d
CFI
TLI
RMSEA [90%CI]
SRMR
α
ω
Indi idual dimension
Academic
756.16
77
.931
.919
.076 [.071, .081]
.075
.83
.82
Social
455.45
65
.960
.952
.063 [.058, .069]
.063
.83
.82
Ins i u ional
90.33
20
.975
.965
.048 [.038, .059]
.082
.84
.83
Pe sonal-Emo ional
179.71
54
.991
.989
.039 [.033, .046]
.040
.87
.88
47-i em models
One- ac o model
11812.32
1034
.887
.882
.083 [.082, .084]
.094
.93
.90
Fou - ac o model
7758.63
1028
.930
.926
.066 [.064, .067]
.074
.93a
.93a
50-i em models
One- ac o model
12545.37
1175
.900
.896
.080 [.079, .081]
.090
.94
.91
Second-o de ac o
8689.37
1172
.934
.931
.065 [.064, .066]
.074
.94a
.92a
No e
:
N
= 1515; Es ima o = DWLS; a, alpha and omega alues o he o e all scale.
Page 16 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Figu e 1
Second-o de model
No e.
AA = Academic Adjus men ; SA = Social Adjus men ; IA = Ins i u ional A achmen ; PEA =
Pe sonal-Emo ional Adjus men ; OA = O e all Adjus men ; SACQ = S uden Adap a ion o College
Ques ionnai e, second-o de model.
Page 17 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
Re iew
Answe
1, Table 1 emains di icul o ead. I you wish
o e ain he po ai o ien a ion, hen please
pu he 90% CIs on a line below he RMSEAs
(in he same column). Tha way, i won'
appea ha he SRMR is pa o he CI.
We ha e changed he o ien a ion o he able
in o de o i all he numbe s p ope ly.
2, I'm cu ious as o why you chose he 90% CI
a he han he con en ional 95% CI. Does
ha indica e ha you' e willing o accep a
highe deg ee o unce ain y (i.e., 10% as
opposed o 5%)? I so, please explain. I you
mean o epo 95% CIs ins ead, please
co ec he able.
Acco ding o Klein (2016), a 90% con idence
in e al o RMSEA is mo e common in he
li e a u e because i allows esea che s o
in e he esul s o he hypo hesis es s o
exac , close and non-close i (MacCallum, e
al., 1996).
3, As I unde s and he analyses, you
compa ed a 4- ac o model o a 1- ac o
model wi h 47 i ems and a 4 + 1 (second
o de ) model wi h o a 1- ac o model wi h 50
i ems. Bu you do no seem o ha e
compa ed he 4- ac o model (47 i ems) wi h
he 4 + 1 ac o model (50 i ems), which
would seem like a logical compa ison o
de e mine whe he he second o de ac o
con ibu es o he alue o he scale. F om
he coe icien s epo ed in able 1, i looks as
i he i o he 4 and 4 + 1 models is
equi alen , so adding he second-o de ac o
did no seem o ha e diminished model i .
The 4- ac o model (47 i ems) and he 4 + 1
ac o model (50 i ems) we e no compa ed
because hey a e non-nes ed models. As a
as ou e iew goes, we we e no able o ind
a way o con as hese wo models when
non no mal da a wi h he DWLS es ima o is
used.
4, I 's in e es ing o me ha none o he
unidimensional models showed especially
poo i . In ac , only he Academic dimension
showed ma ginal i . Gi en you commen s
abou in e co ela ions among i ems, I'm
su p ised ha you don' seem o ha e
examined any modi ica ion indices ha
would ha e allowed e o a iances o
co ela e among pai s o heo e ically ela ed
i ems. Tha 's a p o en way o enhance model
i wi hin logical bounds. I sugges you do ha
o all o he models you es ed. You migh
ind ha some models ha don' look good
now a e as ly imp o ed. As long as
modi ica ion indices a e used judiciously,
hey can con ibu e o he cla i y o models.
He e's an in e es ing
discussion: h ps://cen e s a .o g/wha -a e-
modi ica ion-indices-and-should-i-use- hem-
when- i ing-sems- o-my-own-da a/
We e alua ed ou models’ modi ica ion
indices; howe e , we did no ind any
signi ican imp o emen in hei o e all i s. I
is impo an o men ion ha we discuss he
idea o c oss-loadings among i ems only o
he unidimensional models ha es ed he i
o he SACQ subscales and had a ma ginal
RMSEA and SRMR. In i s ea ly de elopmen ,
many i ems we e sha ed among he
subscales o he SACQ (Bake & Si yk, 1989).
We belie ed ha hese common i ems
encou age he p esence o c oss-loadings
among di e en i ems o he SACQ,
unde mining he one-dimensionali y o each
subscale as showed by he Spanish alida ion
o he scale (Taylo & Pas o , 2007; Rod íguez
e al., 2012). Mo eo e , we ound high
in e co ela ions among subscales ha based
ou second o de model o he scale. The
o e all good i o his model is also e idence
ha he i ems a e sepa a e enough and
beha e p ope ly.
Page 18 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Re iew
5, I 's also in e es ing ha he ou
unidimensional models i be e han any o
he mul i ac o ial models. I wonde i his
means ha he SACQ is eally ou sepa a e
es s in a single package a he han a single
es .
Unidimensional models we e used o assess
a eas o local mis i in he Spanish alida ion
o he scale used in his s udy (Rod íguez e
al., 2012). This p ocess educes he numbe
o i ems pe ac o as well as imp o ed he
unidimensional i o each scale. On he o he
hand, issues ega ding a lowe i o he
mul i ac o ial model ha e been discussed
a ound he p esence o i ems belonging o
mul iple ac o s in he o iginal scale
(Rod íguez e al., 2012; Taylo & Pas o ,
2007).
6, Gi en you inding ha i em #44 did no
load >.3 in any o he models, which is
consis en wi h o he 's indings, why no es
46 and 46 + 1 i em models o in es iga e
whe he i em #44 could be d opped wi hou
doing damage o he s uc u e o he es ?
This would be a aluable con ibu ion,
especially gi en you commen s abou he
i ele ance o he i em in he Pe u ian
con ex and pe haps in o he La in Ame ican
con ex s. This commen e lec s my cu iosi y,
no a demand o e ision.
As men ioned in he s udy esul s, we ook
he decision o keep his i em in o de o
p ese e he o iginal s uc u e o he scale
and because o he o e all i o he models
es ed. P e ious alida ions o he SACQ scale
ha e shown di e en in e nal s uc u es wi h
di e en numbe o i ems, making i di icul
o compa e and build a s ong body o
e idence. Fo his eason, we y o keep he
o iginal numbe o i ems in ligh o he
adequa e i o he models.
7. I don' hink you can make in e ences
abou he alidi y o he Spanish SACQ-50
(Abs ac ; page 9, line 37/38 and page 11, line
55/56) based on his s udy because you didn'
assess alidi y. You can a es o he s abili y
o he ac o s uc u e, he eliabili y o he
sub-scales, and he p esence o he second
o de ac o . Please men ion as a limi a ion
he lack o da a on alidi y o he scale,
pe haps wi h a b ie explana ion o why
alidi y was no assessed.
Acco ding o Cook and Beckman (2006) and
AERA, e al. (2014), in e nal/ ac o s uc u e
is a sou ce o alidi y e idence. We add hese
ci a ions a he beginning o ou discussion in
o de o cla i y his posi ion.
Page 19 o 19
URL: h ps://mc.manusc ip cen al.com/jach Email: [email p o ec ed]
JACH-Fo Pee Re iew Only
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
