The Spanish Journal of Psychology (2018), 21, e34, 1–10.

© Universidad Complutense de Madrid and Colegio Oficial de Psicólogos de Madrid

doi:10.1017/sjp.2018.35

Psychometric Properties of the Spanish Short

Depressive Rumination Scale in a Nonclinical Sample

Jorge Javier Ricarte1, Eva Aizpurúa2, Laura Ros1, José Miguel Latorre1 and Filip Raes3
1 Universidad de Castilla-La Mancha (Spain)
2 Trinity College Dublin (Ireland)
3 University of Leuven (Belgium)

Abstract. Repetitive negative thinking (RNT) is considered a transdiagnostic variable underlying common symptoms
(e.g., depressed mood) across various mood disorders. Depressive rumination is one typical and frequent manifestation
of RNT and is a well-known vulnerability factor of depression onset, maintenance and recurrence. Due to the time-
related constraints in assessment settings and the association of rumination on sadness with diagnosis of depression, the
rapid identification of individuals scoring high on this construct may become a useful screening tool in non-clinical
samples. The main aim of the present study was to test the psychometric properties of the Spanish version of the Short
Depressive Rumination Scale (SDRS) in a large sample (N = 649). Confirmatory Factor Analysis supported a one-factor
model accounting for 75% of variance with an excellent internal consistency (α = .93) in spite of the reduced number of
items (4 items). Criterion validity results based on associations with other well-established rumination (sub)scales, age,
scores in depression and gender differences, were congruent. The results suggest that the SDRS, the shortest existing
scale assessing depressive rumination to our knowledge, can be a useful instrument for a rapid assessment of depressive
rumination in community samples.

Received 11 May 2017; Revised 2 July 2018; Accepted 16 July 2018

Keywords: depression, negative repetitive thinking, Short Rumiation Scale.

Repetitive Negative Thinking (RNT) is assumed to (Nolen-Hoeksema, 1991) where rumination is con-
be a transdiagnostic variable that may account for sidered a possible consequence of mood answer to
the high levels of comorbidity observed between emo- distress focused on passive thinking about one´s
tional disorders especially depression and anxiety symptoms of depression. The tendency to engage in
(Kertz, Koran, Stevens, & Björgvinsson, 2015; Watkins, a ruminative process when distressed is a stable indi-
2015). Rumination significantly mediates the associ- vidual difference (Nolen-Hoeksema & Davis, 1999).
ation between negative affect and depressive symp- According to this theory, this form of rumination
toms (e.g., Verstraeten, Bijttebier, Vasey, & Raes, 2011). increases the likelihood that the symptoms of depres-
The general description of rumination as repetitive sion become chronic evolving into episodes of major
thinking on a theme related to personal goals and depression. The mechanisms by which rumination
concerns (Martin & Tesser, 1996) implies both con- may prolong depression are: Activation of negative
structive and unconstructive consequences (Watkins, thoughts, decreases effective problem solving, inter-
2008). However, in the literature on consequences of feres with instrumental behavior and loss of social
rumination for depressive symptoms, researchers support.
typically measure negative depressive rumination. Other theories have linked rumination to self-
Depressive rumination is defined as “behavior and regulation within a multilevel approach. The Self-
thoughts that focus one’s attention on one’s depressive Regulatory Executive Function (S-REF, Wells &
symptoms and on the implications of these symptoms” Matthews, 1996), emphasizes the role of metacognitive
(Nolen-Hoeksema, 1991, p. 569). Depressive rumina- beliefs in perpetuating ruminative thought. In this
tion is a core variable in the onset and maintenance model, rumination is considered a generic dysfunctional
of clinical diagnosis of depression (Williams et al., process not only present in depression. Rumination is
2007). also considered as the engagement in mental activities
Explanatory theories of depressive rumination were
initially developed under the response styles theory
How to cite this article:
Ricarte, J. J., Aizpurúa, E., Ros, L., Latorre, J. M., & Raes, F.
Correspondence concerning this article should be addressed to
(2018). Psychometric properties of the Spanish Short Depressive
Laura Ros Segura. Departamento de Psicología de la Facultad de
Rumination Scale in a nonclinical sample. The Spanish Journal of
Medicina de la Universidad de Castilla-La Mancha. Albacete (Spain).
Psychology, 21. e34. Doi:10.1017/sjp.2018.35
E-mail: [email protected]

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 2  J. J. Ricarte et al.

to get consistent feedback about the progress toward Watkins, 2016). Regarding age, older adults ruminate
our goals in the control theory (Martin & Tesser, 1996). less than young counterparts (Ricarte, Ros, Serrano,
More recently, Watkins & Nolen-Hoeksema (2014) con- Martínez-Lorca, & Latorre, 2016). In fact, the reduction
ceptualized rumination as a mental habit to explain the of rumination in old age has been suggested as a pos-
relationship between habits and goals. sible mechanism of the increased positivity affect which
Rumination can be distinguished from negative increases with age (Erskine, Kvavilashvili, Conway, &
automatic thoughts or private self-consciousness Myers, 2007) and may be used with therapeutic pur-
because rumination is a process of turning attention poses in older adults (Ekkers et al., 2011).
to one´s symptoms and analysis of the causes and Depressive rumination is typically assessed with the
consequences of those symptoms. Brooding, a fre- 22-item Ruminative Response Scale (RRS) subscale
quently used term in rumination research, reflects a from the Response Styles Questionnaire (RSQ; Nolen-
passive comparison of one´s current situation with Hoeksema & Morrow, 1991). The RSQ is a 71-item, self-
some unachieved standard (e.g., ‘‘Why do I have prob- report instrument designed to assess an individual’s
lems other people don’t have?’’) (Treynor, González, & characteristic tendency to engage in ruminative, distract-
Nolen-Hoeksema, 2003). Thus, brooding is more ing, problem-solving, or dangerous coping behavior in
similar to a self-blame construct (Roberts, Gilboa, & the presence of a dysphoric or stressful situation.
Gotlib, 1998) than depressive rumination. In fact, when Considerable evidence indicates that the RRS subscale
an individual presents extreme difficulty in disen- consists of items confounded with the construct of
gage attention from that perceived discrepancy, this depression as well as a heterogeneous factor structure
results in self-regulatory perseveration (Pyszczynski & (Armey et al., 2009; Bagby & Parker, 2001). In response
Greenberg, 1987). Thus, although similar maladap- to these shortcomings, Treynor et al. (2003) elimi-
tive repetitive thinking constructs can activate self- nated items related to the symptoms of depression
negative comparisons and self-schemas; in depressive and developed a non-affectively-confounded, two-
rumination, the contents of repetitive thinking are factor (brooding and reflection) solution. Brooding,
focused on depression symptoms. defined as negative self-reflection as well as a focus on
Ruminative thought occurs in both normal individ- obstacles to problems (e.g., Nolen-Hoeksema, Wisco, &
uals and clinical patients and is also a feature of anx- Lyubomirsky, 2008), represents a component of depres-
iety disorders. Although rumination can be initially sive rumination (Haigh, Fresco, & Armey, 2004) and
used as a cognitive strategy to control distress, the shows more psychopathological consequences than
continued use of rumination may led to intrusions, reflection (Watkins, 2008). In fact, the reflection com-
impaired problem-solving and a perpetuation of ponent of rumination has not emerged consistently
mood disturbance (Matthews & Wells, 2004). Worry in psychometric factor analyses studies (e.g. Griffith &
and rumination may overlap between a 16–21% in Raes, 2015).
both non-clinical and clinical samples (Segerstrom, To avoid confusion between components of rumina-
Tsao, Alden, & Craske, 2000) and rumination may tion, Conway, Csank, Holm, and Blake (2000) devel-
increases both depression and anxiety (Fresco, Frankel, oped the Rumination on Sadness Scale (RSS). This
Mennin, Turk & Heimberg, 2002). However, although 13-item self-report questionnaire that includes items
rumination and worry have several overlapping features, such as “I repeatedly analyze and keep thinking about
they also represent distinct cognitive processes that the reasons for my sadness”, was originally developed
makes rumination more closely related to depression and to assess only the construct of rumination on current
past events while worry is more closely related to anxiety feelings of distress. It excludes social sharing on sad-
and future thinking, respectively (Papageorgiou & ness as part of rumination. Although initial research
Wells, 1999). confirmed the one-factor structure, Raes, Hermans,
Gender differences in rumination mediates the Williams, Bijttebier, and Eelen (2008), found a three-
gender difference in depression (Butler & Nolen- factor solution after confirmatory factor analyses.
Hoeksema, 1994). In other words, the fact that women Regarding its convergent validity, the RSS correlated
are more likely to suffer from major depression than positively with depression severity and shared more
men could be, among other causes, grounded on the unique variance with depression than the RRS.
presence of higher levels of rumination in women than Subsequently, Raes et al. (2008) revised and extended
men. This greater tendency to ruminate may keep the RSS in the Leuven Adaptation of the Rumination
women stuck in cycles of passivity. Females consistently on Sadness Scale (LARSS) in a 17-item self-report
report higher trait rumination than males (Lopez, questionnaire. The LARSS consists of three subscales:
Driscoll, & Kistner, 2009). In fact, rumination signifi- Causal analysis (causes of sadness), understanding
cantly mediates the association between gender and (meaning of one’s sadness), and uncontrollability (lack
depression scores (Ricarte, Navarro, Latorre, Ros & of control over sad thoughts). The uncontrollability

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 The Spanish Short Depressive Rumination Scale  3

subscale showed the highest association with depres- (UCLM) website and in the local newspaper, giving a
sion symptomatology. A 4-item short version of the telephone number and an email address to contact the
LARSS was tested by Raes, Hermans, Williams, and researchers. Exclusion criteria were chronic illness, cog-
Eelen (2007), showing promising results in terms of nitive impairment or a current episode of psychiatric
internal consistency and validity. illness. All participants were assessed using the Mini
The first main aim of the current research is to fur- International Neuropsychiatric Interview (Sheehan
ther examine the psychometric properties of the one- et al., 1998) to rule out psychiatric disorders.
factor, 4-item reduced version of the LARSS which we
have named the Short Depressive Rumination Scale Variables and Measure
(SDRS). The SDRS was initially tested in a non-clinical
Short Depressive Rumination Scale (SDRS)
Belgian sample (Raes, et al., 2007). In that study,
Cronbach’s alpha for the four items was .84. Total Participants were requested to rate on a scale of 0
scores on the SDRS correlate highly with scores on (never) to 10 (very often) how often they experienced
widely used rumination scales as the Rumination on the following situations when feeling sad, down or
Sadness Scale (RSS; Conway et al., 2000) and the depressed: (a) “I have difficulty getting myself to stop
Ruminative Response Scale (RRS; Nolen-Hoeksema & thinking about how sad I am”; (b) “I get absorbed in
Morrow, 1991), respective rs being .82 and .77 (Raes, thinking about why I am sad and find it difficult to
Hermans, de Decker, Eelen, & Williams, 2003). In a think about other things”; (c) “I repeatedly try to figure
Spanish sample of non-clinical young and older adults, out, by doing a lot of thinking, what might be the
the SDRS was positively correlated with depression causes of my sadness”; and (d) “I keep thinking about
and with functional avoidance (Ros, Latorre, Serrano, & how I feel, to understand myself and my sad feelings
Ricarte, 2017). Second, due to the limited presence of better”. These items were taken from the Leuven
validated measures of rumination in Spanish speaker Adaptation of the Rumination on Sadness Scale (LARSS;
samples, the current research aims to provide an efficient Raes et al., 2008), which has a three-factor structure. The
tool for assessing rumination in those samples. There is a first two items of the SDRS stem from the LARSS-
need to test whether the structure of rumination is con- Uncontrollability factor, while the other two stem from
sistent across languages and communities. Finally, due the Causal Analysis and Understanding factors. The sum
to the time-related constraints in assessment settings of the scores for each item gave a total score ranging
and the association of rumination on sadness with a between 0 and 40. Raes et al. (2007) reported a .84
poor treatment prognosis, the rapid identification of Cronbach’s alpha for this 4-item scale. The items were
individuals scoring high on this construct may become translated from Spanish to English by the first author of
a useful screening tool in non-clinical samples. this study. A native English speaker then translated them
Based on the functioning of this construct in pre- back into English. The translation matched with the ini-
vious research, in addition to an association with tial English version in the article by Raes et al. (2007). See
depression, we expect a higher presence of depressive Appendix for instructions and items in Spanish. Direct
rumination in women than in men (e.g. Ros et al., 2014) addition of all items is necessary to obtain the total score.
and a negative association of rumination with age
(Ricarte, Navarro et al., 2016). We expect our data Ruminative Response Scale (RRS; Treynor et al., 2003)
analysis confirm a single factor structure as well as a
This self-report questionnaire asks participants to
positive association between the total score of the
rate the frequency with which they think or do cer-
scale and both depression and rumination scores as
tain things when they feel sad, down or depressed. It
measured by other instruments.
comprises two subscales assessing two subtypes of
rumination: brooding and reflection. Statements are
Method
rated on a four-point Likert-type scale ranging from
Participants almost always to almost never. The Spanish validated
version was used in the current study (Hervás, 2008).
The study sample comprised 649 volunteer partici-
The complete scale obtained a Cronbach’s alpha value
pants, 399 women and 250 men. Ages ranged between
of .81. Cronbach’s alpha in this study was .74 for the
18 and 81 years. The mean age of males was 29.12
Brooding subscale and .73 for the Reflection subscale.
(SD = 17.73) and 31.45 (SD = 19.49) for females. A total of
68.2% had university studies, 21.3% secondary studies,
The Response Styles Questionnaire to Depression (RSQD;
and 10.5%, primary studies. Inclusion criteria were
Lima, Fernández-Berrocal, Extremera & Queirós, 2006)
aged between 18 and 85, with at least primary studies.
Participants were recruited through advertisements The RSQD was based on the RRS (Nolen-Hoeksema &
placed on the Universidad de Castilla-La Mancha Morrow, 1991). It has 25 items and was validated for a

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 4  J. J. Ricarte et al.

Spanish population (Extremera & Fernández-Berrocal, trained in the test administration. The SDRS scale was
2006). The RSQD assesses three facets: rumination administered first to avoid cognitive contamination
with 10 items (e.g. I think about how sad I feel), prob- from other scales. Following this, the BDI-II was
lem solving with 4 items (e.g. I make a plan of action administered, which collects information on mood
and follow it), and distraction with 11 items (e.g. I help state over the past two weeks. Finally, the other two
someone else with something so I don’t think about scales including rumination variables, the RSQD and
my own problem). All items are rated on a Likert-type the RRS, were applied.
scale from 1 (never) to 4 (always). Higher scores repre-
sent greater use of rumination, distraction and prob- Data analysis
lem solving when feeling sad. A Cronbach’s alpha of
The factor validity of the scale was verified using con-
.79 was obtained for the total scale in the current sam-
firmatory factor analysis. Since the asymmetry results
ple with Cronbach’s alpha values for rumination,
(β1,p = 2.57, p < .001) and kurtosis results (β2,p = 34.22,
problem solving and distraction being .82, .67, and .79,
p < . 001) indicated that multivariate normality was
respectively.
not satisfied, the parameters were estimated using the
Beck Depression Inventory (BDI–II; Beck & Steer, 1984) maximum likelihood mean adjusted (MLM) procedure
in Mplus 7.4 (Muthén & Muthén, 2012). In accordance
The BDI–II is a widely used and well-validated, with the recommendations in the literature, the good-
21-item self-report questionnaire for depressive symp- ness of fit was assessed using absolute and relative
toms. Participants are asked to appraise their feelings indicators. The methods used were the chi-squared
over the last two weeks regarding each item. Each item test, the comparative fit index (CFI), the Tucker-Lewis
is rated on a four-point scale. The Spanish validated Index (TLI), the standardized root-mean-square
version of this instrument was used (Sanz, Perdigón & residual (SRMR) and the root-mean square error of
Vázquez, 2003). Tested in adults from general popula- approximation (RMSEA). We considered indicative
tion, this adaptation obtained a unique depression of a good fit to the model values higher than .90 for
dimension formed by two highly related factors CFI and TLI and values lower than .08 for SRMR and
(cognitive-affective and somatic-motivational) and a lower than .06 for RMSEA (Hu & Bentler, 1999).
Cronbach´s alpha of .87. The Cronbach’s alpha in the Comparison between nested models was conducted
present study was .80. using the Satorra-Bentler scaled chi-square differ-
ence test (Satorra & Bentler, 2001).
Procedure Internal consistency of the scale was measured using
Two psychology doctoral students answered the phone Cronbach’s alpha while the relationships of the scale to
calls and emails from possible candidates and pro- other instruments was analyzed using Pearson’s corre-
vided information on the general characteristics of the lation coefficient. We compared the scores obtained on
study (objective, test types, place and length of test the SDRS by males and females using the t-student test
application). During these conversations, a short soci- for independent samples. The results were comple-
odemographic questionnaire was conducted to check mented by measuring the effect size using Cohen’s d.
inclusion and exclusion criteria. If respondents were Finally, an Analyses of Covariance (ANCOVA) was
university students, they were offered the options of conducted to test whether SDRS scores varied by age
completing the questionnaire with an experimenter at group.
their own school or at the School of Psychology. The
tests were administered in small groups with a min- Results
imum of 5 and a maximum of 20 participants. A period
Factor Analysis Validity
of two years was required to complete the final study
sample. On arrival at the School of Psychology, partic- A total of 110 multivariate outliers were identified using
ipants were provided with a summary of the informa- the robust Mahalanobis distance (RMD) (Rousseeuw &
tion previously offered by phone or email. It was Leroy, 1987). The factor analysis was then conducted
stressed that they could end their participation at any on the complete sample (N = 649) and the sample
point and that participation was voluntary and minus outliers (n = 539). In both cases (Models 1a
unremunerated. Once it was clear that participants and 1b), the parameters were statistically significant
had understood the instructions and the conditions, (t > 1.96), the factor loadings were high (over .60) and
informed consent was obtained. The tests were the factor explained between 66% and 91% of the vari-
administered by 7 psychologists (all Ph. D) forming ance of each indicator (Table 1).
part of the academic and research staff of the UCLM Table 2 shows the indices of goodness of fit for the
Department of Psychology who had previously been two models proposed. All the measures of fit were

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 The Spanish Short Depressive Rumination Scale  5

Table 1. Results of Testing for Differences between Nested Models the reflection one (rxy = .48, p < .001), they did not differ
of the SDRS Scale significantly (z = .88, p = .19).
The correlations between the SDRS and the RSQD
Comparison Statistics total score and its Rumination subscale were posi-
tive and large (rxy = .38, p < .001 and rxy = .73, p < .001,
More constrained Less constrained ΔS-Bχ2 Δdf p respectively). In contrast, the correlation between the
SDRD and the Distraction subscale was negative and
Model 1a Model 1c 28.22 1 < .001
in the small-to-moderate range (rxy = –.13, p < .05). As
Model 1b Model 1d 24.07 1 < .001
shown in Table 4, no correlation was found for the
Note: Model 1a = complete sample; Model 1b = Outliers Problem Solving subscale (rxy = .02, p = .680).
excluded; Model 1c = complete sample, correlated errors As expected, there was a moderate-to-strong corre-
permitted; Model 1d = Outliers excluded, correlated errors lation between the Beck Depression Inventory (BDI–II)
permitted. and the SDRS (rxy = .29, p < .001).

higher than the commonly accepted cutoff values Analysis of the items
(CFI > .90; TLI > .90; SRMR <. 08), with the exception Table 5 shows the descriptive statistics of the
of the RMSEA, with values below the .06 threshold. indicators and the corrected item-total correlation,
Considering items SDRS1 and SDRS2 were taken from Cronbach’s alpha if the item is eliminated from the
the same uncontrollability dimension of the LARSS, test and the total Cronbach’s alpha for the scale. As
the model was respecified to allow the correlation can be seen, the highest mean corresponds to Item 4
between its error terms (Models 1c and 1d). After (“I keep thinking about how I feel, to understand
respecification, the indicators showed an adequate fit myself and my sad feelings better”), while the lowest
to the model in both the complete sample (Model 1c), is in Item 3 “I repeatedly try to figure out, by doing
and in the sample minus outliers (Model 1d). The a lot of thinking, what might be the causes of my
Satorra-Bentler scaled chi-square difference test sadness”). The indicators have a positive asymmetry
(Satorra & Bentler, 2001) between nested models was and negative kurtosis. Hence, the hypothesis of
statistically significant (p < .001), suggesting that the normality is rejected in all cases. The correlations
models allowing correlation between error terms between each item and the scale are high (rit > .70) and
show a better fit to the data (see Table 3). As can be the internal consistency of the total scale is excellent
seen in Table 3 in models 1c and 1d, all the indices (α = .93).
are within the accepted limits of error, the factor load-
ings are equal to, or higher than, .8, and explained var- Differences by gender and age
iance is high (R2 > .70).
Finally, we analyzed the differences by gender in the
total score and the score for each of the indicators.
Concurrent validity
The results show that women scored higher (M =
To verify the concurrent validity, we calculated the 17.02; SD = 10.94) than men (M = 11.53; SD = 9.89)
correlations between the SDRS and other theoreti- on the global scale. The difference in the means,
cally related instruments (see Table 4). As expected, 5.49, 95% CI [3.86, 7.12], was statistically significant
the relationship with the Ruminative Response (t = 6.46, p < .001) and the effect size was medium
(RRS) scale and its subscales were positive and large. (Cohen’s d = 0.52). Nevertheless, the size effect was
Although the coefficient was slightly higher for the greater for the items representing the dimension of
brooding subcomponent (rxy = .53, p < .001) than for Uncontrollability.

Table 2. Indicators of Goodness of Fit for the SDRS with Four Indicators

Model S-Bχ2 χ2(df) CFI TLI SRMR RMSEA [95% CI]

1a. With outliers 18.90 34.69(2)** .983 .950 .023 .159 [.115, .207]
1b. Without outliers 13.48 28.38(2)** .983 .950 .025 .156 [.109, .210]
1c. With outliers, correlated errors permitted 1.66 2.58(1) .999 .995 .005 .049 [.000, .128]
1d. Without outliers, correlated errors permitted .51 .91(1) 1.00 1.00 .003 .000 [.000, .112]

Note: SB-χ2: Satorra-Bentler scaled chi-square difference test; χ2 = Chi Square; CFI = Comparative Fit Index; TLI = Tucker-
Lewis Index; SRMR = Standardized Root-Mean-Square Residual; RMSEA = Root-Mean Square Error of Approximation.
**p < .01

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 6  J. J. Ricarte et al.

Table 3. Factor Loadings and Coefficients of Determination robust and significant even when outliers were included
in the analysis. Consequently, this scale can be adminis-
Models Item R R2 Models Item R R2 tered in the presence of outliers and the high explained
variance is maintained. Internal consistency was excel-
Model 1a SDRS1 .88 .77 Model 1c SDRS1 .80 .64 lent (α = .93), despite the low number of items. The
SDRS2 .96 .92 SDRS2 .90 .81 internal consistency was, in fact, higher than that
SDRS3 .85 .72 SDRS3 .90 .81 obtained for the original version of the scale (α = .84)
SDRS4 .82 .67 SDRS4 .86 .73
proposed by Raes et al. (2007) and higher than that of
Model 1b SDRS1 .90 .80 Model 1d SDRS1 .82 .67 the Brooding subscale from the RRS (α = .74).
SDRS2 .96 .92 SDRS2 .90 .80 Regarding the individual performance of each item,
SDRS3 .85 .73 SDRS3 .91 .82 the results were also excellent, yielding internal reli-
SDRS4 .81 .66 SDRS4 .86 .73 ability values of .80 or higher for all items. However,
Note: Model 1a = complete sample; Model 1b = Outliers
the results yielded little statistical difference between
excluded; Model 1c = complete sample, correlated errors scores for Items 1 and 2 and the total score. This might
permitted; Model 1d = Outliers excluded, correlated errors be due to the somewhat similar nature of the items
permitted. (“I have difficulty getting myself to stop thinking
about how sad I am”; “I get absorbed in thinking about
Regarding age, scores decreased as age increased why I am sad and find it difficult to think about other
(rxy = –.33, p < .001). To further explore the effect of age things”). However, exclusion of one of these items
on rumination scores, respondents were classified into impedes the correct functioning of the scale as the
three groups: 20 and younger (48.15% of participants), model became not statistically significant when one
21 to 30 years old (27.47%), and 31 and older (24.38%). of those items was not included in the analyses.
To assess whether rumination varied by age group, an Furthermore, the uncontrollability of sad thoughts,
Analysis of Covariance (ANCOVA) was performed from which these two items were taken, is the sub-
using the SDRS total score as the outcome variable scale with the most robust association with depression
while controlling for the respondents’ sex. There was a scores (Raes et al., 2008). Hence, we advise against
significant age group effect after controlling for the these two items being excluded on both statistical and
effect of sex, F(2, 644) = 39.95, p < .001, partial n2 = .11. theoretical grounds. However, because models allow-
Planned contrasts revealed that the oldest group had ing measurement errors to correlate tend to be more
lower scores than the youngest group (t = 8.53, p < .001, unstable, further validation of this scale with other
partial n2 = .10) and the middle group (t = 7.22, p < .001, samples will be beneficial.
partial n2 = .08). However, those in the youngest and The results for the convergent validity were as
middle groups did not differ significantly in their expected. Especially noteworthy was the high correla-
SDRS scores (p = .662). tion with the RRS Brooding subscale (Treynor et al.,
2003). Both scales were designed with the intention of
excluding specific depressive symptoms for the mea-
Discussion
surement. In fact, the correlations between these scales
The results of this study confirm that the one-factor and depression are similar and lower than the correla-
structure of this depressive rumination scale functions tion between depression and the RSQD rumination
well. Furthermore, the one-factor structure remained subscale. Hence, the SDRS has a high correlation with

Table 4. Correlations between the SDRS, RRS and RSQD

1 2 3 4 5 6 7 8 9

1. SDRS 1.00
2. RSS total .58** 1.00
3. RRS brooding .53** .87** 1.00
4. RSS reflection .48** .87** .50** 1.00
5. RSQD total .38** .59** .45** .59** 1.00
6. RSQD rumination .74** .68** .68** .50** .55** 1.00
7. RSQD distraction –.13* .16** –.00 .29** .72** –.13* 1.00
8. RSQD problem solving .02 .21** .07 .30 .63** –.00 .51** 1.00
9. BDI–II .29** .22** .28** .10 .10 .39** –.19** –.14* 1.00

**p < .01. *p < .05

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 The Spanish Short Depressive Rumination Scale  7

Table 5. Psychometric Properties of the Items Comprising the Scale Ros et al., 2016). Although there are few studies on the
subject and the explanatory mechanisms of this associ-
Items M SD Assymetry Kurtosis K-S rit α-i Α ation are unknown, a number of studies suggest that
the decrease in rumination is one explanation for the
SDRS1 3.79 2.91 .31 –1.09 .13** .82 .91 .93 higher levels of psychological well-being in older
SDRS2 3.86 3.05 .23 –1.30 .15** .89 .89 adults (Erskine et al., 2007).
SDRS3 3.21 2.92 .57 –0.91 .17** .83 .91
In addition to a lack of an appropriate measure of
SDRS4 4.06 3.08 .21 –1.21 .13** .80 .80
divergent validity, the main limitation of this study is
Note: M = mean; SD = Standard deviation; K-S = Kolmogorov the absence of a clinical group. Although considered as
Smirnov with Lilliefors correction; rit = corrected item-total a transdiagnostic variable, the functioning of depres-
correlation; α-I = α if item is eliminated. sive rumination through different diagnosis and clin-
**p < .01 ical status is still scarce. The correlational nature of our
data is also a shortcoming; thus, more experimental
and longitudinal assessments are in need to test the
the construct of brooding and a low overlap with diag-
real impact of depressive rumination on emotional
nostic assessment of depression. However, compared
state. Finally, the SDRS items may appear complex
to the RRS brooding subscale, the SDRS items repre-
asking about two things at once. Future versions of this
sent uncontrollability, causal analysis and understanding
questionnaire should advise participants about it to
factors. Furthermore, repetitive thinking, as measured
ensure a clear response.
by the SDRS, acts as a cognitive strategy for coping
In short, the SDRS appears to be a highly reliable
with negative feelings which differs from distraction
instrument for measuring depressive rumination.
(negatively correlated) and is unrelated to coping strat-
Administration is especially quick and simple.
egies focused on problem solving (not significantly
Consequently, it could be used to screen for repetitive
correlated). Though our results provide evidence for
negative thinking in a Spanish-speaking population in
the validity of the scale, one limitation of this study is
non-clinical settings (e.g. educational settings). Given
that we only assessed convergent validity. A sub-
the demonstrated association between depressive
sample of 71 participants completed the State-Trait
rumination and depression, the potential for preven-
Anxiety Inventory (Spielberger, Gorsuch, & Lushene,
tion and prognosis of depression justifies the use of
2011), and although the correlation with the SDRS
this short scale.
was significant (rxy = .38, p < .001), this correlation was
smaller than the correlation with other rumination
subscales. This result is consistent with expected References
results regarding divergent validity (depressive rumi- Armey M. F., Fresco D. M., Moore M. T., Mennin D. S.,
nation and anxiety may be related but not so strongly Turk C. L., Heimberg R. G., … Alloy L. B. (2009).
as with other rumination scales). Future studies should Brooding and pondering: Isolating the active
advance in this direction, assessing its discriminant ingredients of depressive rumination with exploratory
validity more consistently. factor analysis and structural equation modeling.
Assessment, 16, 315–327. https://doi.org/10.1177/
The results of this study are also useful to identify the
1073191109340388
normative function of depressive rumination according
Bagby R. M., & Parker J. D. A. (2001). Relation of
to sociodemographic variables. Several studies have rumination and distraction with neuroticism and
shown the onset of gender differences in the frequency extraversion in a sample of patients with major
of rumination initiating at the age of 13 (Jose & Brown, depression. Cognitive Therapy and Research, 25, 91–102.
2008). In our sample, with a mean age of around https://doi.org/10.1023/A:1026430900363
30 years, women scored 5.9 points higher than males Beck A. T., & Steer R. A. (1984). Internal consistencies of
(d = .52). Moreover, these differences were even greater the original and revised Beck Depression Inventory.
for the items from the Uncontrollability dimension. Journal of Clinical Psychology, 40, 1365–1367. https://doi.
These results suggest the importance of prevention of org/10.1002/1097-4679(198411)40:6%3C1365::AID-
negative rumination in women with intervention spe- JCLP2270400615%3E3.0.CO;2-D
Butler L. D., & Nolen-Hoeksema S. (1994). Gender
cially focused on lack of control of negative thoughts
differences in responses to depressed mood in a college
which is, in turn, the factor which most highly corre-
sample. Sex Roles, 30, 331–346. https://doi.org/10.1007/
lates with depressive symptoms. In fact, depression BF01420597
is more prevalent in women than in men (Nolen- Conway M., Csank P. A. R., Holm S. L., & Blake C. K.
Hoeksema, 1990). Regarding age and use of rumina- (2000). On assessing individual differences in rumination
tion, our data support previous studies showing a on sadness. Journal of Personality Assessment, 75, 404–425.
decrease in the use of rumination with age (Ricarte, https://doi.org/10.1207/S15327752JPA7503_04

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 8  J. J. Ricarte et al.

Ekkers W., Korrelboom K., Huijbrechts I., Smits N., and associations with depressive symptoms. Journal of
Cuijpers P., & van der Gaag M. (2011). Competitive Clinical Child & Adolescent Psychology, 38(1), 27–35.
Memory Training for treating depression and rumination https://doi.org/10.1080/15374410802575412
in depressed older adults: A randomized controlled trial. Martin L. L., & Tesser A. (1996). Some ruminative thoughts.
Behavior Research and Therapy, 49, 588–596. https://doi. In R. S. Wyer (Ed.), Ruminative thoughts: Advances in social
org/10.1016/j.brat.2011.05.010 cognition (Vol. 9, pp. 1–47). Hillsdale, NJ: Erlbaum.
Erskine J. A. K., Kvavilashvili L., Conway M. A., & Matthews G. & Wells A., (2004). Rumination, depression
Myers L. (2007). The effects of age on psychopathology, and metacognition: The S-REF Model. In C. Papageorgiou &
well-being and repressive coping. Aging & Mental Health, A. Wells (Eds.), Depressive rumination. Nature, theory
11, 394–404. https://doi.org/10.1080/13607860600963737 and treatment (pp. 125–151). Chichester, UK: John
Extremera N., & Fernández-Berrocal P. (2006). Validity Wiley & Sons.
and reliability of Spanish versions of the Ruminative Muthén L. K., & Muthén B. O. (1998–2012). Mplus user’s
Responses Scale-Short Form and the Distraction guide. Seventh edition. Los Angeles, CA: Author.
Responses Scale in a sample of Spanish high school and Nolen-Hoeksema S. (1990). Sex differences in depression.
college students. Psychological Reports, 98, 141–150. Stanford, CA: Stanford University Press.
https://doi.org/10.2466/pr0.98.1.141-150 Nolen-Hoeksema S. (1991). Responses to depression and
Fresco D. M., Frankel A. N., Mennin D. S., Turk C. L., & their effects on the duration of depressive episodes. Journal
Heimberg R. G. (2002). Distinct and overlapping features of Abnormal Psychology, 100, 569–582. https://doi.
of rumination and worry: The relationship of cognitive org/10.1037/0021-843X.100.4.569
production to negative affective states. Cognitive Therapy Nolen-Hoeksema S., & Davis C. G. (1999). ‘‘Thanks for
and Research, 26, 179–188. https://doi.org/10.1023/ sharing that’’: Ruminators and their social support
A:1014517718949 networks. Journal of Personality and Social Psychology, 77,
Griffith J. W., & Raes F. (2015). Factor structure of the 801–814. https://doi.org/10.1037/0022-3514.77.4.801
Ruminative Responses Scale: A community-sample study. Nolen-Hoeksema S., & Morrow J. (1991). A prospective
European Journal of Psychological Assessment, 31, 247–253. study of depression and posttraumatic stress symptoms
https://doi.org/10.1027/1015-5759/a000231 after a natural disaster: The 1989 Loma Prieta earthquake.
Haigh E. A. P., Fresco D. M., & Armey M. (2004, November). Journal of Personality and Social Psychology, 61, 115–121.
Brooding and pondering: Isolating the active ingredients https://doi.org/10.1037/0022-3514.61.1.115
of depressive rumination with confirmatory factor analysis Nolen-Hoeksema S., Wisco B. E., & Lyubomirsky S.
in an adult clinical sample. A poster session presented at the (2008). Rethinking rumination. Perspectives on
38th annual convention of the Association for Advancement of Psychological Science, 3, 400–424. https://doi.org/
Behavior Therapy. New Orleans, LA. 10.1111/j.1745-6924.2008.00088.x
Hervás G. (2008). Adaptación al castellano de un Papageorgiou C., & Wells A. (1999). Process and
instrumento para evaluar el estilo rumiativo: La escala metacognitive dimensions of depressive and anxious
de respuestas rumiativas [Spanish adaptation of an thoughts and relationships with emotional intensity.
instrument to evaluate rumination style: The scale of Clinical Psychology and Psychotherapy, 6, 156–162. https://
ruminative responses]. Revista de Psicopatología y doi.org/10.1002/(SICI)1099-0879(199905)6:2<156::AID-
Psicología Clínica, 13, 111–121. CPP196>3.0.CO;2-A
Hu L.-T., & Bentler P. M. (1999). Cutoff criteria for Pyszczynski T., & Greenberg J. (1987). Self-regulatory
fit indexes in covariance structure analysis: perseveration and the depressive self-focusing style:
Conventional criteria versus new alternatives. Structural A self-awareness theory of reactive depression.
Equation Modeling, 6, 1–55. https://doi.org/10.1080/ Psychological Bulletin, 102, 122–138. https://doi.org/
10705519909540118 10.1037/0033-2909.102.1.122
Jose P. E., & Brown I. (2008). When does the gender Raes F., Hermans D., Williams J. M. G., & Eelen P. (2007).
difference in rumination begin? Gender and age A sentence completion procedure as an alternative to the
differences in the use of rumination by adolescents. Autobiographical Memory Test for assessing overgeneral
Journal of Youth and Adolescence, 37, 180–192. https://doi. memory in non-clinical populations. Memory, 15, 495–507.
org/10.1007/s10964-006-9166-y https://doi.org/10.1080/09658210701390982
Kertz S. J., Koran J., Stevens K. T., & Björgvinsson T. Raes F., Hermans D., de Decker A., Eelen P., & Williams
(2015). Repetitive negative thinking predicts depression J. M. G. (2003). Autobiographical memory specificity and
and anxiety symptom improvement during brief cognitive affect regulation: An experimental approach. Emotion, 3,
behavioral therapy. Behaviour Research and Therapy, 68, 201–206. https://doi.org/10.1037/1528-3542.3.2.201
54–63. https://doi.org/10.1016/j.brat.2015.03.006 Raes F., Hermans D., Williams J. M. G., Bijttebier P., &
Lima M. M., Fernández-Berrocal P., Extremera N., & Eelen P. (2008). A “Triple W”- model of rumination on
Queirós P. S. (2006). Physical activity in the elderly: sadness: Why am I feeling sad, what’s the meaning of my
Analysis of relations to perceived emotional intelligence sadness, and wish I could stop thinking about my sadness
and styles of response to depression. Ansiedad y Estrés, 12, (but I can’t!). Cognitive Therapy and Research, 32, 526–541.
293–303. https://doi.org/10.1007/s10608-007-9137-y
Lopez C. M., Driscoll K. A., & Kistner J. A. (2009). Sex Ricarte J. J., Navarro B., Latorre J. M., Ros L., & Watkins E.
differences and response styles: Subtypes of rumination (2016a). Age and gender differences in emotion regulation

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 The Spanish Short Depressive Rumination Scale  9

strategies: Autobiographical memory, rumination, Therapy and Research, 24, 671–688. https://doi.org/
problem solving and distraction. The Spanish Journal of 10.1023/A:1005587311498
Psychology, 19, E43. https://doi.org/10.1017/sjp.2016.46 Sheehan D. V., Lecrubier Y., Sheehan H., Amorin P.,
Ricarte J. J., Ros L., Serrano J. P., Martínez-Lorca M., & Janaus J., Weiller E., … Dunbar G. C. (1998). The
Latorre J. M. (2016b). Age differences in rumination Mini-International Neuropsychiatric Interview (M.I.N.I.):
and autobiographical retrieval. Aging and Mental Health, The development and validation of a structured Diagnostic
20, 1063–1069. https://doi.org/10.1080/13607863.2015. Psychiatric Interview for DSM–IV and ICD–10. Journal of
1060944y Clinical Psychiatry, 59, 22–33.
Roberts J. E., Gilboa E., & Gotlib I. H. (1998). Ruminative Spielberger C. D., Gorsuch R. L., & Lushene R. E.
response style and vulnerability to episodes of dysphoria: (2011). STAI. Cuestionario de Ansiedad Estado-Rasgo
Gender, neuroticism, and episode duration. Cognitive [STAI. State-Trait Anxiety Inventory]. Madrid, Spain:
Therapy and Research, 22, 401–423. https://doi.org/ Ediciones TEA.
10.1023/A:1018713313894 Treynor W., González R., & Nolen-Hoeksema S. (2003).
Ros L., Latorre J. M., Serrano J. P., & Ricarte J. J. (2017). Rumination reconsidered: A psychometric analysis.
Overgeneral autobiographical memory in healthy young Cognitive Therapy and Research, 27, 247–259. https://doi.
and older adults: Differential age effects on components of org/10.1023/A:1023910315561
the capture and rumination, functional avoidance, and Verstraeten K., Bijttebier P., Vasey M. W., & Raes F. (2011).
impaired executive control (CaRFAX) model. Psychology Specificity of worry and rumination in the development
and Aging, 32, 447–459. https://doi.org/10.1037/ of anxiety and depressive symptoms in children. British
pag0000175 Journal of Clinical Psychology, 50, 364–378. https://doi.
Ros L., Ricarte J. J. Serrano J. P., Nieto M., Aguilar M. J., & org/10.1348/014466510X532715
Latorre J. M. (2014). Overgeneral autobiographical Watkins E. R. (2008). Constructive and unconstructive
memories: Gender differences in depression. Applied repetitive thought. Psychological Bulletin, 134, 163–206.
Cognitive Psychology, 28, 472–480. https://doi.org/ https://doi.org/10.1037/0033-2909.134.2.163
10.1002/acp.3013 Watkins E. (2015). An alternative transdiagnostic
Rousseeuw P. J., & Leroy A. M. (1987). Robust regression and mechanistic approach to affective disorders illustrated
outlier detection. New York, NY: John Wiley & Sons, Inc. with research from clinical psychology. Emotion Review, 7,
Sanz J., Perdigón A. L., & Vázquez C. (2003). Adaptación 250–255. https://doi.org/10.1177/1754073915575400
española del Inventario para la Depresión de Beck–II Watkins E. R., & Nolen-Hoeksema S. (2014). A habit-goal
(BDI–II): 2. Propiedades psicométricas en población framework of depressive rumination. Journal of Abnormal
general [The Spanish adaptation of Beck’s Depression Psychology, 123, 24–34. https://doi.org/10.1037/a0035540
Inventory-ll (BDI-II): 2. Psychometric properties in the Wells A., & Matthews G. (1996). Modelling cognition in
general population]. Clínica y Salud, 14, 249–280. emotional disorders: The S-REF model. Behaviour Research
Satorra A., & Bentler P. M. (2001). A scaled difference and Therapy, 34, 881–888. https://doi.org/10.1016/
chi-square test statistic for moment structure analysis. S0005-7967(96)00050-2
Psychometrika, 66, 507–514. https://doi.org/10.1007/ Williams J. M. G., Barnhofer T., Crane C., Herman D.,
BF02296192 Raes F., Watkins E., & Dalgleish T. (2007).
Segerstrom S. C., Tsao J. C. I., Alden L. E., & Craske M. G. Autobiographical memory specificity and emotional
(2000). Worry and rumination: Repetitive thought as a disorder. Psychological Bulletin, 133, 122–148. https://
concomitant and predictor of negative mood. Cognitive doi.org/10.1037/0033-2909.133.1.122

Appendix

SDRS

A continuación, debe indicar en una escala de 0 a 10, donde 0 indica nunca y 10 muy a menudo, cuánto hace lo que
se describe en las siguientes frases cuando se siente triste, desanimado o deprimido:

1- Me cuesta conseguir parar de pensar sobre lo triste que estoy:

Nunca Muy a menudo
0 1 2 3 4 5 6 7 8 9 10

2- Me quedo pensando sobre por qué estoy triste y me cuesta pensar en otras cosas:
Nunca Muy a menudo
0 1 2 3 4 5 6 7 8 9 10

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35
 10  J. J. Ricarte et al.

3- Paso bastante tiempo intentando encontrar cuáles podrían ser las causas de mi tristeza:
Nunca Muy a menudo
0 1 2 3 4 5 6 7 8 9 10

4- Me quedo pensando sobre cómo me siento, para entenderme a mí y a mis sentimientos de tristeza mejor:
Nunca Muy a menudo
0 1 2 3 4 5 6 7 8 9 10

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 14 Aug 2018 at 23:17:33, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/sjp.2018.35