Original

Leticia Camarillo1, 2
Francisco Ferre1,3 Partner’s Emotional Dependency
Scale: Psychometrics
Enrique Echeburúa4
Pedro J. Amor5

1
Instituto de Psiquiatría y Salud Mental del Hospital Gregorio Marañón. Madrid, España
2
Facultad de Psicología. Universidad Complutense de Madrid, España
3
Facultad de Medicina. Universidad Complutense de Madrid
4
Facultad de Psicología. Universidad del País Vasco (UPV-EHU), San Sebastián, España
5
Facultad de Psicología. Universidad Nacional de Educación a Distancia (UNED), Madrid, España.

Introduction. Emotional dependency in a partner implies Escala de Dependencia Emocional de la Pareja:

an excessive permanent affectional bonding to the other in- propiedades psicométricas
dividual that is dysfunctional, associated with low self-esteem
and conceals a lack of affection. Emotional dependency gen- Introducción. La dependencia emocional de la pareja
erates a series of negative emotional consequences: symp- implica una vinculación afectiva permanente excesiva de
toms of anxiety and depression, obsessive thoughts, sleep la otra persona, que resulta disfuncional, que se asocia con
disorders and abandonment of social relations and leisure. In una baja autoestima y que encubre carencias afectivas. La
recent decades several tools have been developed to measure dependencia emocional genera una serie de consecuencias
emotional dependency, but they are more focused on mea- emocionales negativas: síntomas ansioso-depresivos, pen-
suring dependent personality disorder, centered on an exclu- samientos obsesivos, alteraciones del sueño y abandono de
sively young population or are not validated in the Spanish relaciones sociales y de ocio. En las últimas décadas se han
population. The aim of this research is to design and validate desarrollado diferentes instrumentos para medir la depen-
a new scale to overcome these limitations. dencia emocional, pero están más centrados en medir el
trastorno de personalidad por dependencia, se focalizan en
Method. The sample population included 166 adults (53 una población exclusivamente joven o no están validados en
men and 113 women) from the general population, to whom población española. El objetivo del estudio es diseñar y va-
a new scale was applied (partner’s emotional dependency lidar una nueva escala que haga frente a estas limitaciones.
scale, SED), as well as the CDE (Emotional Dependency Ques-
Método. La muestra ha contado con 166 personas adul-
tionnaire).
tas (53 hombres y 113 mujeres) de la población general, a
Results. The scale has a unidimensional structure, show- las que se ha aplicado la nueva escala (Partner’s Emotional
ing good internal consistency (Cronbach’s alpha = .90) and a Dependency Scale, SED) y el CDE (Cuestionario de Depen-
high convergent validity with the CDE (r = .86). There is no dencia Emocional).
difference between men and women with regarding emo- Resultados. La escala presenta una estructura unidi-
tional dependency. Some cut-off points have been estab- mensional, muestra una buena consistencia interna (alfa
lished based on the degree of emotional dependency (low, de Cronbach=0,90) y una alta validez convergente con el
moderate, high and extreme). CDE (r=0,86). No hay diferencias entre hombres y mujeres
respecto a la dependencia emocional. Se establecen unos
Conclusions. The SED is a brief assessment tool, simple to
points de corte en función del grado de dependencia emo-
conduct and allows the degree of emotional dependency to cional (bajo, moderado, alto y extremo).
be assessed unidimensionally with regard to a stable intimate
relationship (current or past) in a normative population. Conclusiones. La SED es un instrumento de evaluación
Key words: Emotional Dependency, Intimate Relationship, Partner’s Emotional Dependency
breve, sencillo de cumplimentar y que permite evaluar uni-
Scale, Assessment Scale, Psychometrics dimensionalmente el grado de dependencia emocional con
respecto a la relación de pareja estable (actual o pasada) en
Actas Esp Psiquiatr 2020;48(4):145-53
una población normativa.
Palabras Clave: Dependencia Emocional, Relación Íntima, Escala de Dependencia
Correspondence:
Emocional de la Pareja, Escala de Evaluación, Psicometría
Francisco Ferre
C/ Doctor Esquerdo, 46
28007 – Madrid, Spain
E-mail: francisco.ferre@salud.madrid.org

Actas Esp Psiquiatr 2020;48(4):145-53 145

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

INTRODUCTION Emotional dependency is linked to a series of negative

emotional consequences: symptoms of anxiety and depres-
Emotional sensitivity and emotion regulation have been sion, obsessive thoughts, difficulty concentrating, sleep dis-
found to be predictive of most personality dimensions and orders, abandonment of social relationships, neglect of lei-
symptoms of most personality disorders1,2. More specifically, sure activities and professional projects, etc. If the
emotional dependency in a partner implies an excessive per- relationship has ended, there is a compulsive need to resume
manent affectional bonding to the other individual that is contact with the partner (despite a history of suffering and
dysfunctional, associated with low self-esteem and conceals humiliation). It is no longer about I love you, but rather I
a lack of affection. This dependency is accompanied by a need you (toxic relationship)3. There is also a close relation-
biased perception of reality, an intolerance of loneliness and ship between emotional dependency and partner violence,
an inner emptiness. Specifically, an intimate relationship, either as a victim or as an aggressor9,11.
regardless of what type it is, takes priority over any other
activity or value in the life of the affected person3. In recent decades, different tools have been developed
to measure emotional dependency. Among those published
Functionally speaking, a person may be considered de- in the English-speaking world are the Interpersonal Depen-
pendent when they perceive that the current and past as- dency Inventory (IDI)12, the Spouse-Specific Dependency
sessment of their stable intimate relationship is negative Scale (SSDS)13 and the Relationship Profile Test (RPT)14. All
and they consider to break off that relationship, but feel of the above tools focus more on measuring dependent
unable to do so, without financial dependency or threats personality disorder than emotional dependency on a part-
that explain the permanence of that relationship4. ner as such. The SSDS was only applied to university stu-
dents with an average age of 19.9 years, which makes it
From an attachment perspective, dependency is a dis-
play of pathological, anxiety/ambivalent attachment be- difficult to generalise across other adult age ranges. The
haviours in interpersonal relationships that express unsatis- IDI, meanwhile, is more focused on dependency on other
fied emotional needs and prevent the relationship from significant persons in an excessively general way. Finally,
ending despite dissatisfaction5,6. the RPT was validated on psychology students (average
age: 18 years).
Unlike dependent personality disorder, people who are
emotionally dependent on their partners can be indepen- In the Spanish-speaking world, two further tools have
dent in other spheres (social or employment, for example). In been created: the Emotional Dependency Questionnaire
some ways emotional dependency on a partner is more sim- (CDE)15 and the Emotional Dependency Scale in Partner Re-
ilar to an addictive disorder: partner submission and ideali- lationships in Young People and Adolescents (DEN)16. The
sation, with false expectations of change in a partner; un- CDE, prepared following Beck’s cognitive model, is a brief
conditional love and controlling behaviour; and in the event tool obtained from a large and representative sample of dif-
of a relationship coming to an end, desperate attempts to ferent age groups, but it has not been validated in the Span-
get back together due to suffering a sort of emotional absti- ish population and measures, as well as the central factors,
nence syndrome7. Low self-esteem and the presence of ob- six sub-scales, which are excessive in a 23-item tool, espe-
sessive personality traits, as well as a history of a lack of af- cially as some of them only contain 2 or 3 items (for exam-
fection or previous traumatic intimate relationships, may ple, fear of loneliness, expression of limits or attention-seek-
lead to emotional dependency8. In these cases, individuals ing), which under-represent the corresponding constructs.
prefer to continue suffering instead of facing the new real- The DEN, meanwhile, is an interesting tool validated among
ity of breaking up with their partner and being forced to the Spanish population, but referring to partner relation-
cross the unknown abyss. At a clinical level, emotional de- ships in young people and adolescents, with an average age
pendency may share traits in common with a dependent of 20.46 years.
personality and an addictive disorder9.
For the above reasons, this research aims to construct
New technologies may lead to the development of emo- and validate a brief partner’s emotional dependency scale in
tional dependency on a partner, contribute to prolonging the adult Spanish population in relationships with a mini-
and perpetuating a toxic behavioural pattern for the depen- mum duration of six months, determine if there are differ-
dent person and may also make it difficult to overcome this ences between men and women in this variable and calcu-
condition due to the enormous possibility of permanent late some percentiles that allow a person’s degree of
contact with and control of the person on whom the indi- dependency on their partner to be identified (emotional
vidual depends10. dependency should not be considered as a dichotomous

146 Actas Esp Psiquiatr 2020;48(4):145-53

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

variable), which allows for early detection of the problem borderline expression (3 items, .62 [.50]) and attention
and the establishment of specific intervention programmes. seeking (4 items, .78 [.75]). The internal consistency
(Cronbach’s alpha) of the total score was .927 (α=.938
in this study).
METHOD
Procedure
Participants
The sample of this instrumental study was obtained be-
The total sample was made up of 166 individuals, 53 tween January 2015 and June 2017 in a general population
men (31.9%) and 113 women (68.1%), with an average age belonging to people in active employment in the Basque
of 39.17 years (SD=11.60) and ranging from 21 to 63 years, Country. The assessment tools were administered by psy-
taken from the general population. 80.5% were in active chology graduates who had received prior training to carry
employment and the majority (76.9%) had university quali- out this task. The participants responded voluntarily to the
fications, 18.2% secondary education and 4.8% primary ed- SED and the CDE after having signed an informed consent
ucation. form in which the main aspects of this research were ex-
plained, and the confidentiality of the data obtained was
In terms of civil status, 74.7% of the sample was mar- assured.
ried or in a relationship, 21.1% single and 4.2% separated or
divorced. The relationship length was between 1 and 40
years (M=14.44; SD=10.95). 56% of the sample had children Data analysis
(range: 1-5, Mdn=1, SD=1.16).
The internal consistency of the SED was calculated us-
The inclusion criteria to be part of the study were as ing Cronbach’s alpha and the item discrimination index
follows: a) 18 or over; b) in a relationship currently (or have based on the corrected item-total correlation. The dimen-
been in one recently) lasting at least six months; and c) re- sionality of the SED was analysed using an Exploratory Fac-
spond to all the items in the Partner’s Emotional Depen- tor Analysis with the total sample. The following was cal-
dency Scale and the Emotional Dependency Questionnaire. culated: a) dispersion matrix: Polychoric correlations; b)
factor extraction: Unweighted Least Squares (ULS); c) fac-
Instruments tor rotation: Promin17; and d) determination of the number
of dimensions: Optimal implementation of Parallel Analysis
- Partner’s Emotional Dependency Scale (SED) (Appen- (PA)18, Minimum Average Partial Test (MAP)19, and Root
dix). A questionnaire containing 22 items aimed at as- Mean Square Error of Approximation (RMSEA). The possi-
sessing emotional dependency on a partner and mea- ble unidimensionality of the questionnaire was also exam-
sured using a Likert scale ranging between 0 and 4 ined using overall and item-level assessment20, calculating
(range: 0-88 points). The higher the score, the higher Unidimensional Congruence (UniCo), Explained Common
the emotional dependency (all items point in this direc- Variance (ECV) and Mean of Item Residual Absolute Load-
tion, only number 1 is inverted). This tool aims to assess ings (MIREAL). The data were treated as essentially unidi-
a partner’s emotional dependency through various mensional if values UniCo>.95, ECV>.85 and MIREAL<.30.
questions taking into account the heterogeneity of this To calculate concurrent validity, the Pearson correlation
construct. This scale is a self-report that can be used coefficient was used between the SED and the CDE. Spear-
both in a group and in an individual format. man’s correlation coefficient was also calculated between
- Emotional Dependency Questionnaire (Cuestionario de SED and CDE items (total score and sub-scales). The cor-
Dependencia Emocional, CDE)15. Adapted to the Colom- relation effect size (ES) was considered: |r|=.1, .3 and .5
bian population, this tool has 23 items (range: 23-138 correspond to small, medium, and large ES, respectively.
points) measured using a Likert scale that goes from 1 The differences between mean values in men and women
(not at all like me) to 6 (exactly like me). This test mea- were compared using the Student’s t-test and Hedges’ g
sures six dimensions of emotional dependency (number (|g| =.2, .5 and .8 correspond to small, medium, and large
of items, α of the original test [α obtained in this re- ES, respectively). Finally, in order to get a classifying crite-
search]): separation anxiety (7 items, .87 [.87]), couple’s rion, the percentiles of the SED were obtained in the total
affective expression (4 items, .83 [.89]), change of plans sample. The analyses were performed using the programs
(4 items, .75 [.72]), fear of loneliness (3 items, .80 [.83]), SPSS 24.0 and FACTOR 10.7.0121.

Actas Esp Psiquiatr 2020;48(4):145-53 147

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

RESULTS items). The mean discrimination was .53. Eliminating any of the
22 items did not increase the reliability of the scale (Table 1).

Reliability of the Partner’s Emotional Dependency

Validity of the Partner’s Emotional Dependency
Scale
Scale
Internal consistency. The Cronbach’s alpha was .90 (.91
Construct validity (factor structure). The adequacy of
based on standardised elements) for the SED total score (22 the correlation matrix was analysed using the Kaiser-Mey-
er-Olkin (KMO) test=.885 and Bartlett’s test (231)=1420.9
(p<.001). Table 2 shows factor loads, communality values and
Table 1 Descriptive statistics and reliability other statistics that serve as criteria to analyse the possible
in the Partner’s Emotional unidimensionality of the scale. The factor loads of the items
Dependency Scale (SED) were between .44 and .82, while communalities were be-
If item deleted Correlations tween .19 and .67. Items 1, 2, 14 and 22 showed less commu-
nality than the others. In turn, the different procedures for
Items Mean SD α rcItem-total R2 determining the number of dimensions suggest a single fac-
tor that would explain the 45.33% of the variance:
Item 1a 17.17 12.15 0.900 0.38 0.32
MAP=.02734, RMSEA=.034, Bias-Corrected Bootstrap 95%
Item 2 16.11 11.96 0.900 0.40 0.32 Confidence Interval=[.0148, .0483].
Item 3 16.39 11.65 0.897 0.54 0.44
On the other hand, both the overall assessment (Uni-
Item 4 16.63 11.80 0.899 0.47 0.35
Co=.977, ECV=.871, and MIREAL=.222) and the item-level
Item 5 16.03 11.86 0.898 0.49 0.39 assessment (18 items have I-UniCo values>.95, 16 items
Item 6 17.04 11.80 0.895 0.57 0.50 I-ECV values>.85 and 17 items I-REAL values<.30) suggest
that the data can be treated as essentially unidimensional.
Item 7 17.60 11.97 0.895 0.61 0.55

Item 8 16.78 11.92 0.896 0.53 0.39 After analysing the reliability and dimensionality of the
SED, therefore, the items are observed to assess a dimension
Item 9 17.33 11.70 0.893 0.65 0.58
related to the construct of emotional dependence in the re-
Item 10 17.72 12.14 0.898 0.47 0.35 lationship.
Item 11 17.69 12.09 0.897 0.53 0.50
Convergent validity. The total SED score was statistically
Item 12 17.58 12.03 0.896 0.58 0.59 related to the total CDE score (r=.86, p<.01) and its different
Item 13 17.07 11.94 0.899 0.44 0.40 sub-scales: separation anxiety, couple’s affective expression,
Item 14 17.01 11.97 0.899 0.44 0.29 change of plans, fear of loneliness, borderline expression and
attention seeking, with values ranging between .61 and .79
Item 15 17.60 12.00 0.897 0.52 0.57
(see Table 3).
Item 16 17.29 11.77 0.893 0.67 0.61
Although all items on the SED were statistically related
Item 17 17.72 12.13 0.898 0.50 0.36
to the different CDE sub-scales (except item 1, which was not
Item 18 17.67 12.07 0.897 0.56 0.50 statistically related to the attention-seeking dimension), the
Item 19 17.27 11.92 0.896 0.53 0.41 different magnitudes of association ranged from small to
large. Overall, items 3, 5, 12, 19, 21 and 22 of the SED were
Item 20 17.72 12.05 0.896 0.58 0.59
those most related with the CDE total score (effect size large,
Item 21 16.81 11.66 0.892 0.71 0.54 r>.50). Specifically, the SED items most related to the differ-
Item 22 16.88 11.85 0.899 0.46 0.32 ent CDE sub-scales (effect size large) were the following:
a
=Reverse score; α=Cronbach’s alpha; rcitem-total=Corrected item-total items 3, 12, 19, 21 and 22 (separation anxiety), items 5, 12
correlation; R2=Squared multiple correlation; Mean discrimination and 21 (couple’s affective expression), item 14 (change of
(Range) of the SED=.529 (.381-.707) plans), item 19 (fear of loneliness), item 22 (borderline ex-
pression), and item 21 (attention seeking).

148 Actas Esp Psiquiatr 2020;48(4):145-53

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

Table 2 Factorial Matrix and Closeness to Unidimensionality Assessment of the Partner’s Emotional
Dependency Scale (SED) in the Total Sample (N=166)

Factor structure Item-level assessment

Items BCa 95% CI Communality Comunalidad I-UniCo I-ECV I-REAL

Item 1ª 0.45 [0.41 - 0.59] 0.21 0.99 0.90 0.16

Item 2 0.44 [0.44 - 0.53] 0.19 0.86 0.63 0.35

Item 3 0.64 [0.64 - 0.66] 0.41 0.98 0.84 0.29

Item 4 0.54 [0.55- 0. 63] 0.29 0.99 0.89 0.20

Item 5 0.55 [0.55 - 0.64] 0.30 0.99 0.90 0.19

Item 6 0.63 [0.62 - 0.70] 0.40 0.88 0.64 0.49

Item 7 0.73 [0.67 - 0.82] 0.54 0.94 0.73 0.46

Item 8 0.60 [0.58 - 0.73] 0.36 1.00 0.96 0.12

Item 9 0.79 [0.73 - 0.90] 0.62 1.00 0.95 0.18

Item 10 0.74 [0.64 - 10.0] 0.54 1.00 0.98 0.12

Item 11 0.69 [0.64 - 0.82] 0.47 0.99 0.87 0.27

Item 12 0.76 [0.75 - 0.85] 0.58 0.97 0.80 0.39

Item 13 0.54 [0.50 - 0.62] 0.29 1.00 0.97 0.10

Item 14 0.50 [0.52 - 0.70] 0.25 1.00 0.93 0.14

Item 15 0.68 [0.64 - 0.89] 0.47 0.94 0.73 0.43

Item 16 0.73 [0.68 - 0.82] 0.53 0.99 0.89 0.26

Item 17 0.71 [0.65 - 0.83] 0.51 1.00 1.00 0.01

Item 18 0.76 [0.70 - 0.90] 0.58 1.00 0.99 0.09

Item 19 0.62 [0.60 - 0.68] 0.38 0.99 0.87 0.25

Item 20 0.82 [0.68 - 10.0] 0.67 1.00 0.94 0.22

Item 21 0.78 [0.78 - 0.84] 0.61 1.00 1.00 0.02

Item 22 0.51 [0.51 - 0.66] 0.26 1.00 0.91 0.17

a
=Reverse score; BCa 95% CI=Bias-Corrected and accelerated bootstrap 95% Confidence Intervals for loading values; I-UniCo=Item Unidimensional
Congruence (larger than .95 suggests that data can be treated as essentially unidimensional); I-ECV=Item Explained Common Variance (larger than
.85 suggests that data can be treated as essentially unidimensional); I-REAL=Item REsidual Absolute Loadings (lower than .30 suggests that data can
be treated as essentially unidimensional).

Gender differences in total scores on the CDE were not statistically significant, t(164)=1.21. p = .227.
Partner’s Emotional Dependency Scale and the Hedges’ g=.20, 95% Confidence Interval [-.13, .53].
Emotional Dependency Questionnaire

The mean differences among men (n=53, M=19.45, Calculation of percentiles for the Partner’s
SD=12.89) and women (n=113, M=17.26, SD=12.25) on the Emotional Dependency Scale
SED were not statistically significant, t(164)=1.05, p=.291,
Hedges’ g=.17, 95% Confidence Interval [-.15, .50]. Similarly, The percentiles for the total score in the SED were cal-
the mean differences among men (n=53, M=45.98, culated to create a classification criterion for the normative
SD=18.37) and women (n=113, M=42.48, SD=16.82) on the group. Those individuals scoring less than 10 points (≤25th

Actas Esp Psiquiatr 2020;48(4):145-53 149

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

Table 3 Pearson and Spearman Correlations between Partner’s Emotional Dependency Scale (SED) and
Emotional Dependency Questionnaire (CDE) and Descriptive Statistics of CDE

CDE Total CDE-SA CDE-CAE CDE-CP CDE-FL CDE-BE CDE-AS

SED Total 0.86** 0.79** 0.71** 0.72** 0.67** 0.61** 0.64**

Item 1 -0.25** -0.22** -0.20** -0.19* -0.27** -0.25** -0.13

Item 2 0.36** 0.30** 0.29** 0.48** 0.15* 0.22** 0.30**

Item 3 0.53** 0.55** 0.43** 0.42** 0.35** 0.25** 0.44**

Item 4 0.37** 0.32** 0.29** 0.40** 0.24** 0.18* 0.30**

Item 5 0.59** 0.46** 0.63** 0.44** 0.35** 0.31** 0.47**

Item 6 0.41** 0.36** 0.38** 0.36** 0.26** 0.25** 0.37**

Item 7 0.38** 0.35** 0.30** 0.31** 0.32** 0.32** 0.32**

Item 8 0.43** 0.43** 0.31** 0.43** 0.33** 0.28** 0.29**

Item 9 0.49** 0.45** 0.38** 0.35** 0.40** 0.41** 0.38**

Item 10 0.38** 0.38** 0.26** 0.32** 0.25** 0.31** 0.24**

Item 11 0.46** 0.44** 0.37** 0.33** 0.35** 0.36** 0.36**

Item 12 0.60** 0.56** 0.54** 0.37** 0.44** 0.42** 0.47**

Item 13 0.38** 0.34** 0.31** 0.33** 0.23** 0.30** 0.27**

Item 14 0.43** 0.39** 0.31** 0.51** 0.28** 0.32** 0.19**

Item 15 0.44** 0.47** 0.34** 0.30** 0.38** 0.34** 0.29**

Item 16 0.47** 0.45** 0.36** 0.47** 0.33** 0.29** 0.29**

Item 17 0.34** 0.34** 0.29** 0.22** 0.24** 0.25** 0.28**

Item 18 0.43** 0.39** 0.37** 0.35** 0.32** 0.38** 0.40**

Item 19 0.57** 0.54** 0.48** 0.37** 0.66** 0.39** 0.45**

Item 20 0.45** 0.47** 0.39** 0.24** 0.36** 0.39** 0.34**

Item 21 0.68** 0.60** 0.62** 0.47** 0.45** 0.46** 0.56**

Item 22 0.58** 0.50** 0.47** 0.49** 0.40** 0.49** 0.37**

No. of items 23 7 4 4 3 3 2

Range 23-138 7-42 4-24 4-24 3-18 3-18 2-12

M 43.60 13.54 9.52 7.57 4.90 37.0 4.36

DT 17.36 6.54 4.75 3.31 2.68 1.33 2.31

α 0.938 0.874 0.892 0.718 0.830 0.503 0.754

CDE-SA=Separation Anxiety; CDE-CAE=couple’s affective expression; CDE-CP=change of plans; CDE-FL=Fear to Loneliness; CDE-BE=borderline
expression; CDE-AS= ttention seeking; *p≤.05; **p≤.01. α=Cronbach’s alpha

150 Actas Esp Psiquiatr 2020;48(4):145-53

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

percentile) were classed as having low levels of emotional

Table 4 Percentile scores of the Partner’s dependency; between 10 and 21 points moderate depen-
Emotional Dependency Scale (SED) in dency; between 22 and 36 points high dependency; and
the Total Sample (N=166)
more than 37 points extreme dependency (≥90th percentile)
Percentiles SED Total Classification (Table 4).
1 0.00 Low (0-9 points)

2 0.00 CONCLUSIONS
3 2.00 In this research, the psychometrics of a new brief tool
4 2.68 (the SED) were studied to assess emotional dependency in
intimate relationships using an adult Spanish population
5 3.35
currently in or which has been in a stable relationship. It is
10 5.00 the first specific scale in this field.
15 7.00
The SED is a unidimensional 22-item tool that assesses
20 8.00 the overall degree of emotional dependency in the general
25 9.00 population. The psychometrics of the scale are satisfactory.
Specifically, the internal consistency of the scale was high
30 11.00 Moderate (10-21
points) (alpha=.90), as was convergent validity with the total CDE
35 12.00 score (r=.86). As for the validity of the construct, the explor-
40 13.00 atory factor analysis and the calculation of unidimensional
congruence lead us to confidently conclude the existence of
45 14.00
a single factor that allows us to explain the 45.3% variance
50 15.00 in the variable examined (partner’s emotional dependency).
55 16.00 This result differs from that of other studies, which have
obtained different dimensions, probably because they used
60 17.00 statistical different procedures and the factors obtained are
65 18.55 saturated by very few items.15,16
70 21.00 Unlike the results of other studies15, there are no gender
75 22.00 High (22-36 points) differences in this research between men and women in
terms of emotional dependency. It is likely that the be-
80 24.60
haviour among one gender or another differs, but we have
85 29.00 not studied this factor here, leaving it to future research.
87 33.00
Based on the fact that the differences between a cou-
88 34.92 ple’s healthy dependence (healthy emotional attachment)
89 36.00 and pathological dependence are a matter of degree, as a
classification criterion, this study has calculated the percen-
90 37.30 Extreme (37-88 points)
tiles of the total SED score (range: 0-88). Specifically, it has
92 39.28 been determined that high emotional dependence is estab-
94 42.98 lished from a score of 22 (75th percentile) and extreme from
a score of 37 (90th percentile). This is an interesting point
95 45.65 that does not feature in all the scales analysed. It allows the
96 46.64 problem to be detected early on and intervention strategies
to be formed for the individuals affected by pathological
97 48.99
dependency22.
98 53.64
If the degree of emotional dependence of a person on
99 63.65
their partner can be assessed with a certain degree of preci-
Descriptive statistics of the SED (Range: 0-88): Mean=17.96, sion, the greater or lesser vulnerability of a victim to male
SD=12.47, Skewness=1.376 (p =.188), Kurtosis=2.103 (p=.375)
violence can be determined, as well as the difficulties of get-

Actas Esp Psiquiatr 2020;48(4):145-53 151

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

ting out of it, or the greater or lesser propensity of certain CONFLICT OF INTERESTS
persons to become abusers based on their dependence on
The authors have no conflict of interests to declare.
the victim23,24. This topic was not the subject of this work
(because the sample was limited to the general population),
but could be studied in future research. REFERENCES

In further studies this scale could also be applied to clinical 1. Borges LM, Naugle AE. The role of emotion regulation in
predicting personality dimensions. Personal Ment Health. 2017;
samples, such as people affected by anxiety disorders, depres- 11(4):314–34.
sion, chemical and behavioural addictions25 and personality dis- 2. Wall K, Kalpakci A, Hall K, Crist N, Sharp C. An evaluation of
orders, most of all in borderline personality disorder2, 24,26. the construct of emotional sensitivity from the perspective of
emotionally sensitive people. Borderline Personal Disord Emot
This study does, however, have some limitations, such as Dysregul. 2018 Aug;5:14.
the relatively small sample size and the over-representation 3. Castelló, J. [Emotional dependence. Characteristics and
treatment]. Madrid, España: Alianza Editorial; 2005.
of women. In future research, it would be worth analysing
4. Bornstein RF, Hopwood CJ. Evidence-based assessment of
the psychometrics of this tool and considering its use in interpersonal dependency. Prof Psychol Res Pr. 2017;48(4):251-8.
clinical samples, for preventive and therapeutic reasons. An- 5. Alonso-Arbiol I, Shaver PR, Yárnoz S. Insecure attachment,
other limitation is that there are four items (1, 2, 14 and 22) gender roles, and interpersonal dependency in the Basque
on the scale that show less communality with the rest of the Country. Pers Relatsh. 2002;9(4):479-90.
6. Loinaz I, Echeburúa E, Ullate M. [Attachment style, empathy and
test. One possible explanation is the complex and heteroge- self-esteem in partner-violent men]. Terapia Psicológica. 2012;
neous nature of this construct, which makes it difficult to 30(2):61-70.
assess. Item 1, for example, (“In general, are you satisfied 7. Scantamburlo G, Pitchot W, Ansseau M. [Affectice dependency].
with yourself?”) relates to different variables, such as Revue Medicale de Liège. 2013;68(5-6):340-7.
8. Ahmadi, V, Davoudi, I, Ghazaei, M, Mardani, M, Seifi, S. Prevalence
self-esteem and, to a lesser degree, emotional dependency.
of obsessive love and its association with attachment. Procedia -
In any case, we have decided to maintain these four items Soc Behav Sci. 2013;84:696-700.
for theoretical reasons (greater theoretical coherence and 9. Picó-Alfonso M, Echeburúa E, Martinez M. Personality disorder
representativeness of the construct) and practical reasons symptoms in women as a result of chronic intimate male partner
(there is no reduction in the reliability of the scale). violence. J Fam Violence. 2008;23(7):577-88.
10. Fox J, Tokunaga RS. Romantic partner monitoring after breakups:
In conclusion, the SED is a brief assessment tool, simple Attachment, dependence, distress, and post-dissolution online
surveillance via social networking sites. Cyberpsychol Behav Soc
to conduct and allows the degree of emotional dependency Netw. 2015;18(9):491-8.
to be assessed unidimensionally with regard to a stable inti- 11. Bornstein RF. The complex relationship between dependency
mate relationship (current or past) in a normative popula- and domestic violence: Converging psychological factors and
tion. social forces. Am Psychol. 2006;61(6):595-606.
12. Hirschfeld RMA, Klerman GL, Gouch HG, Barrett J, Korchin
SJ, Chodoff P. A measure of interpersonal dependency. J Pers
ETHICAL RESPONSIBILITIES Assess. 1977;41(6):610-8.
13. Rathus JH, O’Leary KD. Spouse-Specific Dependency Scale: Scale
development. J Fam Violence. 1997;12(2):159-68.
Protection of human and animal subjects 14. Bornstein RF, Geiselman KJ, Eisenhart EA, Languirand MA.
Construct validity of the Relationship Profile Test: Links with
the authors declare that the procedures followed are attachment, identity, relatedness, and affect. Assessment.
according to the ethical norms of the responsible human 2002;9(4):373-80.
experimentation committee, the World Medical Association 15. Lemos M, Londoño NH. [Design and validation of the Emotional
Dependence Questionnaire in Colombian population]. Acta
and the Helsinki Declaration. Colombiana de Psicología. 2006;9(2):12740.
16. Urbiola I, Estévez A. [Emotional dependency and early
maladaptive schemas in adolescents and youth dating
Confidentiality of data
relationships]. Psicol Conductual. 2015;23(3):571-87.
17. Lorenzo-Seva U. Promin: A method for oblique factor rotation.
The authors declare that they followed the protocols of
Multivariate Behav Res. 1999;34(3):347-65.
their centre of work regarding the publication of patient data. 18. Timmerman ME, Lorenzo-Seva U. Dimensionality assessment
of ordered polytomous items with parallel analysis. Psychol
Methods. 2011;16(2):209-20.
Right to privacy and informed consent
19. Velicer WF. Determining the number of components from the
matrix of partial correlations. Psychometrika. 1976;41(3):321-27.
The authors have obtained the informed consent of the 20. Ferrando PJ, Lorenzo-Seva U. Assessing the quality and
patients and/or subjects referred to in the paper. This docu- appropriateness of factor solutions and factor score estimates
ment is held by the corresponding author. in exploratory item factor analysis. Educ Psychol Meas. 2018;

152 Actas Esp Psiquiatr 2020;48(4):145-53

Leticia Camarillo, et al. Partner’s Emotional Dependency Scale: Psychometrics

78(5):762-80. 24. Esbec E, Echeburúa E. Violence and personality disorders: Clinical

21. Lorenzo-Seva U, Ferrando PJ. Factor 92: A comprehensive and forensic implications. Actas Esp Psiquiatr. 2010;38(5):249-
program for fitting exploratory and semiconfirmatory factor 61.
analysis and IRT models. Appl Psychol Meas. 2013;37(6):497-98. 25. Škařupová K, Blinka L. Interpersonal dependency and online
22. Amor PJ, Echeburúa E. [Psychosocial keys for the permanence of gaming addiction. J Behav Addict. 2016 Mar;5(1):108–14.
the victim in an abusive relationship]. Clínica Contemporánea. 26. Carvalho LF, Pianowski G. Differentiating borderline personality
2010;1(2):93-100. disorder and bipolar disorder through emotional dependency,
23. Andrés-Pueyo A, Echeburúa E. [Violence risk assessment: emotional instability, impulsiveness and aggressiveness traits.
Available tools and instructions for use]. Psicothema. 2010; Eur J Psychiat. 2019 Apr;33(2):91–5.
22(3):403-09.

Appendix Partner’s emotional dependency scale (SED)

Instructions: Read the questions below and answer by marking the responses you think best describe how you feel and behave. The response
options are as follows
0 1 2 3 4
Completely untrue Partly untrue Somewhere in between Partly true Completely true
1. In general, are you satisfied with yourself? 0 1 2 3 4

2. Do you often act as though your partner’s well-being is more important than your own? 0 1 2 3 4

3. Does the possibility of your relationship with your partner ending often cause you anguish or sadness? 0 1 2 3 4

4. Do you often apologise to your partner when they are angry, even though you know that you are not responsible for
0 1 2 3 4
them being angry?

5. Do you need your partner to continually show you affection? 0 1 2 3 4

6. Do you normally do things or activities that you don’t like just to please your partner or avoid ending the
0 1 2 3 4
relationship?

7. Do you often justify your partner’s conflicts, criticisms or infidelities as a lesser evil in the relationship? 0 1 2 3 4

8. Do you usually feel guilty about arguments with your partner? 0 1 2 3 4

9. Do you feel incapable of leaving your partner despite the upset and suffering the relationship causes you? 0 1 2 3 4

10. Have you pushed yourself to the limit and done things that you may even recognise as inappropriate (sexual
0 1 2 3 4
behaviour, reckless endangerment, drug use, allowing financial abuse) just to avoid abandonment?

11. Do you think that you are not up to your current partner’s standards and that you may therefore lose them? 0 1 2 3 4

12. Do you feel the need to check-up on your partner (mobile, WhatsApp, etc.) so that you always know where they are
0 1 2 3 4
and who they are with?

13. Do you think that the way you are and act is different since you have been with your current partner? 0 1 2 3 4

14. Do you make all your joint plans according to your partner’s tastes and desires? 0 1 2 3 4

15. Do you constantly suffer from exaggerated jealousy? 0 1 2 3 4

16. Have you given up your duties or stopped caring for family or friends because you are with your partner? 0 1 2 3 4

17. Do you often think that previous relationships were more satisfying for your partner than being with you? 0 1 2 3 4

18. Has your partner expressed how overwhelmed they feel because of your need to constantly be with them or
0 1 2 3 4
continually know their movements?

19. Do you feel helpless and lonely when you have no partner? 0 1 2 3 4

20. Are you convinced or do you constantly suspect that your partner is thinking of ending the relationship? 0 1 2 3 4

21. Do you continually feel the need to please your partner? 0 1 2 3 4

22. Do you think you would be willing to do whatever it took to get your partner back if they left you? 0 1 2 3 4

Actas Esp Psiquiatr 2020;48(4):145-53 153