Over the past few decades, the concept of mindfulness gained great attention in the Western world. Jon Kabat-Zinn (2003) described mindfulness as “the awareness that emerges through paying attention on purpose, in the present moment, and nonjudgmentally to the unfolding of experience moment by moment” (p. 145). Originating from the Buddhist tradition, the concept of mindfulness has received growing interest also in the Western world. Mindfulness-based group programs were developed, most prominently mindfulness-based stress reduction (MBSR), initially described in 1990 (Kabat-Zinn), and mindfulness-based cognitive therapy (MBCT; Segal et al., 2002). MBCT was originally designed as a relapse prevention for major depression disorder and combines components of MBSR as well as elements of cognitive behavioral therapy (Teasdale et al., 2000).
Most studies on mindfulness are based on adult samples and support beneficial effects of mindfulness in a variety of outcomes: Regarding psychological symptoms, current research shows the most promising results of mindfulness-based interventions (MBIs) in reducing symptoms of depression and anxiety (Basso et al., 2019; Blanck et al., 2018; Goldberg et al., 2018; Hilton et al., 2017; Hofmann & Gomez, 2017; Khoury et al., 2013; Wielgosz et al., 2019). This was further supported by a recent review of meta-analyses by Goldberg et al. (2022). Furthermore, MBIs may ameliorate cognitive functions (Basso et al., 2019), schizophrenia spectrum disorder (Jansen et al., 2020), quality of life, and pain management (Hilton et al., 2017). The application of MBIs has also been researched in different settings, reaching from group programs (e.g., MBCT; Segal et al., 2002) to individual therapy (Mander et al., 2019), as well as stand-alone interventions (Blanck et al., 2018).
There is also growing literature on the application and effects of mindfulness on children and adolescents. One of the most prominent adaptations for youths is mindfulness-based cognitive therapy for children (MBCT-C; Semple et al., 2011). It originates from the MBCT program (Segal et al., 2002) and was adjusted for children between 9 and 12 years to help them cope with anxiety. MBCT-C contains 12 sessions for group or individual therapy. Children learn about the concept of mindfulness, practice mindfulness using their senses (e.g., mindful eating, mindful hearing), and transfer the concept of mindfulness into their everyday lives. The MBSR program (Kabat-Zinn, 1990) was also adapted for younger people. Biegel et al. (2009), for instance, adjusted it for 14- to 18-year-olds by shortening several elements (e.g., shorter home practice, no day-long retreat) and using stress-related topics that are especially relevant to teenagers (e.g., self-image, self-harming behaviors).
A review of randomized controlled trials by Dunning et al. (2019) found positive effects of MBIs in children and adolescents on executive functioning, attention, depression, anxiety/stress, and negative behaviors when compared to non-active control groups. When only compared to active control groups, effects were still significant for depression and anxiety/stress. The effect of MBIs on anxiety in children and adolescents was also supported by a review of Borquist-Conlon et al. (2019) who found a moderate effect of MBIs on anxiety in youths with anxiety disorders. Reangsing et al. (2021) reviewed 29 studies and found a small effect of MBIs on depressive symptoms in children and adolescents. Based on previous promising findings, further investigation on mindfulness in the so far rather underresearched group of children and adolescents is needed (Kalmar et al., 2022). Basic requirement for that is a valid and reliable instrument assessing mindfulness skills in children and adolescents. At the present time, two measures are most commonly used: The Mindful Attention Awareness Scale for Adolescents (MAAS-A; Brown et al., 2011) and the Child and Adolescent Mindfulness Measure (CAMM; Greco et al., 2011).
The MAAS-A has a one-dimensional factor structure with 14 items. It was originally adapted for 14- to 18-year-olds from the Mindful Attention Awareness Scale (MAAS; Brown & Ryan, 2003) by eliminating one inappropriate item. It was validated in a Dutch (de Bruin et al., 2011), a Persian (Mohsenabadi et al., 2019), and a Spanish sample (Calvete et al., 2014). An additional version, the MAAS-C for fourth to seventh grade children, was evaluated by Lawlor et al. (2014). The CAMM is based on the Kentucky Inventory of Mindfulness Skills (KIMS; Baer et al., 2004), which contains 39 items and assesses four facets of mindfulness: observing (noticing of external and internal phenomena), acting with awareness (undivided attention on the present activity), accepting without judgment (being nonjudgmental about the present-moment experience), and describing (nonjudgmental description of the present moment).
The CAMM was developed in an elaborate process of adapting items of the KIMS in order to make them suitable for children and adolescents. During initial consideration, all items of the describing facet were eliminated due to their complexity, which might not be suitable for younger users. Of the remaining three factors, 25 items were preselected by the authors. In a next step, item comprehensibility was evaluated and modified by expert raters. Further modifications were applied after presenting and gaining feedback on these items from a sample of 10- to 17-year-olds. During subsequent analyses, all nine items originating from the factor observing were excluded from the 25 selected items due to inconsistent correlations of this factor to other measures. An exploratory factor analysis and further considerations based on the remaining 16 items indicated a ten-item version with one single factor. The proposed one-factor structure was further examined by a confirmatory factor analysis (CFA) which reached good fit indices. In the original study by Greco et al. (2011), significant correlations to measures of somatization symptoms, quality of life, psychological inflexibility, thought suppression, and control, as well as social skills, were found.
The CAMM was validated in several different languages, e.g., in Dutch (de Bruin et al., 2014), in Catalan (Vinas et al., 2015), in Italian (Chiesi et al., 2017; Saggino et al., 2017), in French-Canadian (Dion et al., 2018), in French (Roux et al., 2019), in Spanish (García-Rubio et al., 2019; Guerra et al., 2019), in Portuguese (Cunha et al., 2013), in Greek (Theofanous et al., 2020), and in Persian (Mohsenabadi et al., 2020). Moreover, the CAMM was used in a number of further validation or mindfulness studies, e.g., the validation of the CAMM in a sample of minority adolescents from low-income environments (Prenoveau et al., 2018), analyzing the role of mindfulness and emotional intelligence on emotional states (Mestre et al., 2019), investigating mindfulness in relation to gambling behavior (Pace et al., 2021), the relation of mindfulness and psychiatric symptoms (Pepping et al., 2016), as well as mindfulness and correlations with chronic pain (Waldron et al., 2018). In summary, when compared to the MAAS-A, the CAMM has fewer items, relies on a more elaborate adapting process to ensure applicability on youths, and is better integrated into the literature.
Previous validation studies of the CAMM reported inconsistent findings regarding gender differences. No significant gender differences were found by de Bruin et al. (2014), Chiesi et al. (2017), Cunha et al. (2013), Kuby et al. (2015), Saggino et al. (2017), Vinas et al. (2015), and in the original study by Greco et al. (2011). On the contrary, several studies on the CAMM reported gender differences indicating lower mindfulness skills in females (Dion et al., 2018; Roux et al., 2019; Theofanous et al., 2020). Interestingly, Theofanous et al. (2020) and Chiesi et al. (2017) found the CAMM to be gender invariant, meaning that it measures the same construct in boys and girls. These findings suggest that, if differences are found between girls and boys, they are not based on different understandings of the content of the CAMM.
The original version of the CAMM by Greco et al. (2011) was based on participants between the ages of 10 and 17 years. While several following validation studies included participants under 17 years (de Bruin et al., 2014, Guerra et al., 2019; Kuby et al., 2015; Roux et al., 2019; Vinas et al., 2015), other studies validated the CAMM on participants up to 18 or 19 years (Chiesi et al., 2017, Cunha et al., 2013, Dion et al., 2018, García-Rubio et al., 2019, Mohsenabadi et al., 2020, Saggino et al., 2017) and even up to a mean of 22.44 years (Theofanous et al., 2020).
The notable number of validation studies and its use in diverse research questions underlines the applicability of the CAMM and the demand for a reliable and valid measurement in current research on mindfulness in children and adolescents. To our knowledge, there is no validation study of a German translation of the CAMM available. Therefore, the aim of this study was to explore the psychometric properties of the German version of the CAMM in a community sample of children and adolescents based on the following hypotheses: (1) the one-factor structure of the original CAMM can be replicated in a CFA; (2) internal consistency is comparable to the original by Greco et al. (2011); (3) meaningful correlations between the German CAMM and measures of self-compassion, internalizing and externalizing symptoms, and quality of life indicate convergent validity of the German CAMM. Furthermore, (4) potential differences in CAMM mean scores between female and male participants and between the age groups of 10 to 17 and 10 to 19 years were explored.
Methods
Participants
Initially, 2000 children and adolescents were invited to participate in the survey. Inclusion criteria were an age above 9 and under 20 years. Of the 285 respondents, 32 canceled the survey after the first few questions, four were excluded due to implausible data (unrealistically short time to fill in the questionnaires), and one person filled in the questionnaire twice. After exclusion, data of 248 participants were used for further analyses (response rate 12.40%). Participants were between 10 and 19 years old. More females than males participated in the study and a clear majority of the students attended Gymnasium (type of school with 12 to 13 years of education). Additional information is provided in Table 1.
Table 1
Sample description (n = 248)
M | SD | Range | |
|---|---|---|---|
Age (years) | 14.85 | 2.55 | 10–19 |
n | % | ||
Gender | |||
Female | 146 | 58.87 | |
Male | 102 | 41.13 | |
School type | |||
Hauptschule | 1 | .40 | |
Realschule | 17 | 6.88 | |
Gymnasium | 169 | 68.42 | |
University | 18 | 7.28 | |
Other | 42 | 17.00 | |
Procedures
Contact data of the 2000 children and adolescents were provided by the municipal office. These 2000 persons received an invitation by mail with the participation link and information about the background and aims of the survey. Incentive for participating was a draw of three 10 € coupons for a local toy- and drugstore. The study information and declaration of consent were provided online for the participating children and adolescents and their parents. Before participating in the survey, informed consent was obtained. Complete anonymity was ensured and email addresses (required only for taking part in the draw) were saved separately. The survey was implemented on the platform SoSci Survey (www.soscisurvey.de; Leiner, 2019). It took participants on average 18.37 min (SD = 7.38) to complete the survey.
Measures
The CAMM (Greco et al., 2011) assesses mindfulness skills in children and adolescents. It consists of 10 negatively worded items on a 5-point Likert scale ranging from 0 (never true) to 4 (always true). After reverse-scoring of the items, higher scores indicate higher levels of mindfulness. Internal consistency is good with Cronbach’s α = 0.81 (Greco et al., 2011). After obtaining authorization from the author, the translation of the CAMM into German was performed by independent forward–backward translations. Two research associates each formulated a German version of the CAMM and discussed any differences leading to a joint first version. A native-speaking associate retranslated this version, which led to further adaptation and eventually to the final version.
Self-compassion was assessed by the Self-Compassion Scale (SCS) (Neff, 2003) adapted for children and adolescents (Stolow et al., 2016). The original version of the SCS (Neff, 2003) consists of 26 items with six subscales (self-kindness, self-judgment, common humanity, isolation, mindfulness, over-identification). Stolow et al. (2016) validated a version of the SCS for children and adolescents in which the language was made more suitable for younger persons (e.g., “not good enough” instead of “flaws and inadequacies”). They identified one “positive” factor (SCS-POS) containing 12 items of the subscales self-kindness, common humanity, and mindfulness, and one “negative” factor (SCS-NEG) with 13 items of the subscales self-judgment, isolation, and over-identification from the original SCS by Neff (2003). Items are rated on a 5-point Likert scale ranging from 1 (almost never) to 5 (almost always). Internal consistency is good with α = 0.87 for SCS-POS and 0.92 for SCS-NEG (Stolow et al., 2016). Analogous to the CAMM, the SCS for children and adolescents was translated into German for the purpose of this study. Internal consistency in our sample was good with α = 0.89 and ω = 0.91.
Internalizing and externalizing symptoms were measured by the Youth Self-Report (YSR) (Achenbach, 1991) in a short form, derived from a Portuguese version of the YSR (Cruz et al., 2014). Originally consisting of 112 items, the Portuguese short form contains only 33 items. For the purpose of this study, these 33 items were paralleled to the German 112-item version resulting in a German short form of the YSR. Items are rated on a three-step scale from 0 (not true) to 2 (frequently true). Cruz et al. (2014) found four underlying dimensions: internalization-depression, internalization-anxiety, externalization-destructiveness, and externalization-exhibitionism (boundary violations). Furthermore, a total score can be calculated with higher scores indicating higher problem severity. Internal consistency is acceptable to good with α ranging between 0.70 and 0.85 in the respective dimensions of the Portuguese version (Cruz et al., 2014). In our sample, internal consistency was good with α = 0.88 and ω = 0.90.
Quality of life was assessed by the Inventory for Measuring Quality of Life in Children and Adolescents (Inventar zur Untersuchung der Lebensqualität von Kindern und Jugendlichen; ILK) (Mattejat & Remschmidt, 2006). It contains nine questions about areas of quality of life (school, family, social relations to peers, interests and leisure time, physical health, psychological health, overall quality of life) that are rated on a 5-step scale from 1 (very good) to 5 (very bad). The first seven items constitute the quality of life score, which can be further transformed into a percent range from 0 to 100 with higher scores indicating higher quality of life. Internal consistency is poor to acceptable with Cronbach’s α ranging across different samples between 0.55 and 0.76 (Mattejat & Remschmidt, 2006). In our sample, internal consistency was good with α = 0.81 and ω = 0.84.
Data Analyses
Item discrimination was calculated by correlating the respective item with the sum of the remaining items. In order to test the factor structure of the CAMM, a CFA was performed using the maximum-likelihood method. The interpretation of resulting fit indices was based on guidelines by Hu and Bentler (1999), who recommend a comparative fit index (CFI) ≥ 0.95, the root mean square error of approximation (RMSEA) ≤ 0.06, and the standardized root mean square residual (SRMR) ≤ 0.08. Due to unsatisfactory results, items with low factor loadings were excluded and another CFA was conducted with a reduced version of the CAMM. Both CFAs are based on non-correlated error co-variances. Internal consistency was estimated by Cronbach’s alpha (Cronbach, 1951) and McDonald’s omega (McDonald, 1999). Convergent validity was assessed by correlations of the CAMM with the SCS for children and adolescents, YSR short form, and ILK using Pearson’s correlation coefficient r. To test for possible gender differences in CAMM scores, an unpaired two-sample t-test was conducted, with Cohen’s d (Cohen, 1988) as a measure of effect size. To address potential age differences, we ran the same analyses a second time after excluding participants older than 17 years. All calculations were performed with R, version 4.2.0 (R Core Team, 2022), and RStudio, version 2021.09.0 (Rstudio Team, 2021). The CFA was conducted using the R package lavaan (Rosseel, 2012).
Results
Comparing our sample to the original population of 2000 contacted persons, no significant differences were found regarding age (M = 14.85 years in our sample, M = 15.17 in the population, t(2246) = − 1.75, p = 0.080) and gender distribution (58.87% female in our sample, 49.90% in the population, χ2(1) = 0.71, p = 0.399). The mean total score of the German CAMM is 3.38 (SD = 0.76) and mean item scores ranged between 2.84 and 3.76 (SD between 1.06 and 1.27). Item discrimination ranged between 0.38 and 0.62. Further descriptive analyses are shown in Supplement 1. Inter-item correlations were lowest between item 1 and item 5 (r = 0.13, p = 0.038) and between item 6 and item 10 (r = 0.18, p = 0.064). The highest inter-item correlation was found between item 8 and item 9 (r = 0.61, p < 0.001).
Items 2, 5, and 6 presented the lowest communality scores (h2 = 0.13 to h2 = 0.16) and highest uniqueness scores (u2 = 0.84 to u2 = 0.87), indicating lowest associations to the other items. Additionally, factor loadings were lowest in item 6 (λ = 0.37), item 2 (λ = 0.40), and item 5 (λ = 0.40). More information is presented in Table 2. Fit indices of the ten-item version indicated a non-satisfactory fit of the model: χ2(35) = 129.97, p < 0.001, CFI = 0.86, RMSEA = 0.11 (90% CI 0.09, 0.12), SRMR = 0.07. Due to low factor loadings, an alternative seven-item version without items 2, 5 and 6 was evaluated. The seven-item version reaches excellent model fit: χ2(14) = 19.62, p = 0.143, CFI = 0.99, RMSEA = 0.04 (90% CI 0.000, 0.079), SRMR = 0.03. A chi-square difference test supported a significantly better fit of the seven-item version: χ2diff(21) = 110.25, p < 0.001.
Table 2
Translated items and psychometric properties of the German Child and Adolescent Mindfulness Measure (CAMM)
Item | λ | h2 | u2 | α | ω |
|---|---|---|---|---|---|
1. Ich rege mich über mich selbst auf, dass ich Gefühle habe, die keinen Sinn ergeben I get upset with myself for having feelings that don’t make sense | .67 | .44 | .56 | ||
2. In der Schule besuche ich Schulstunde um Schulstunde, ohne richtig zu bemerken, was ich tue At school, I walk from class to class without noticing what I’m doing | .40 | .16 | .84 | ||
3. Ich beschäftige mich ständig mit etwas, damit ich meine Gedanken und Gefühle nicht wahrnehme I keep myself busy so I don’t notice my thoughts or feelings | .62 | .38 | .62 | ||
4. Ich sage mir selbst, dass ich nicht so fühlen sollte, wie ich es tue I tell myself that I shouldn’t feel the way I’m feeling | .66 | .44 | .56 | ||
5. Ich schiebe Gedanken von mir weg, die ich nicht mag I push away thoughts that I don’t like | .40 | .16 | .84 | ||
6. Es fällt mir schwer, mich ausschließlich auf eine Sache zu konzentrieren It’s hard for me to pay attention to only one thing at a time | .37 | .13 | .87 | ||
7. Ich denke über Dinge nach, die in der Vergangenheit passiert sind, anstatt über Dinge nachzudenken, die gerade jetzt passieren I think about things that happened in the past instead of thinking about things that are happening right now | .53 | .28 | .72 | ||
8. Ich rege mich über mich selbst auf, wenn ich bestimmte Gedanken habe I get upset with myself for having certain thoughts | .73 | .53 | .47 | ||
9. Ich denke, dass manche meiner Gefühle schlecht sind und dass ich sie nicht haben sollte I think that some of my feelings are bad and that I shouldn’t have them | .74 | .55 | .45 | ||
10. Ich halte mich davon ab, Gefühle zu haben, die ich nicht mag I stop myself from having feelings that I don’t like | .54 | .29 | .71 | ||
Total ten-item version | .83 | .85 | |||
Total seven-item versiona | .83 | .85 |
Cronbach’s α and McDonald’s ω indicated good internal consistency with α = 0.83 and ω = 0.85 for the ten-item version and α = 0.83 and ω = 0.85 for the seven-item version without items 2, 5, and 6. The seven-item version showed the expected correlations with regard to convergent validity: The seven-item CAMM correlated positively with the total score of the SCS for children and adolescents (r = 0.63, p < 0.001) and the ILK quality of life score (r = 0.61, p < 0.001). Negative correlations were found between the seven-item version of the CAMM and the YSR mean score (r = − 0.63, p < 0.001). Further information on correlations is presented in Table 3.
Table 3
Correlations between the German Child and Adolescent Mindfulness Measure (CAMM) and measures of self-compassion, internalizing and externalizing problems, and quality of life
Seven-item version CAMMa | |
|---|---|
Correlations with the SCS for children and adolescents (Pearson’s correlation coefficient r) | |
SCS total | .63* |
SCS positive subscale | .39* |
SCS negative subscale | − .68* |
SCS self-judgment | − .61* |
SCS over-identification | − .53* |
SCS common humanity | .21* |
SCS isolation | − .61* |
SCS self-kindness | .44* |
SCS mindfulness | .22* |
Correlations with the ILK and YSR | |
YSR internalization-depression | − .59* |
YSR internalization-anxiety | − .57* |
YSR externalization-destructiveness | − .27* |
YSR externalization-exhibitionism (boundary violations) | − .38* |
YSR mean | − .63* |
ILK quality of life score | .61* |
Significant differences were found in total scores of the seven-item CAMM between female and male participants (t(246) = − 2.12, p = 0.035) with females showing lower scores indicating that they consider themselves to have less mindfulness skills than males. However, the effect size of this difference was small: d = 0.28 (95% CI [0.02; 0.58]).
Results were conceptually identical when comparing the younger sample of 10- to 17-year-olds (n = 201) to the sample of 10- to 19-year-olds. Factor loadings of the critical items were slightly lower in the younger age group with item 2: λ = 0.38 (10–17 years) and λ = 0.40 (10–19 years), item 5: λ = 0.35 (10–17 years) and λ = 0.40 (10–19 years), and item 6: λ = 0.29 (10–17 years) and λ = 0.37 (10–19 years). Further results are provided in Supplement 2.
Discussion
This study validated the German version of the CAMM in a community sample. Results did not replicate the one-factorial structure with ten items of Greco et al. (2011) but promoted a single-factor solution with a reduced number of items. Due to comparatively low factor loadings, items 2, 5, and 6 were excluded from the final version. Model fit was unsatisfactory for the original ten-item version and excellent for the seven-item version. This is in line with previous research on psychometric properties of the CAMM which mostly supports a single-factor solution of the CAMM (Chiesi et al., 2017; Dion et al., 2018; García-Rubio et al., 2019; Kuby et al., 2015; Roux et al., 2019; Saggino et al., 2017; Vinas et al., 2015).
Few studies found evidence for a two-factor solution comprising items 3, 5, and 10 (de Bruin et al., 2014; García-Rubio et al., 2019), and items 2 and 10 (Theofanous et al., 2020), on the second factor. They all rejected the two-factor solution due to the small number of items on the second factor and a good fit of the single-factor solution. Solely the Persian version by Mohsenabadi et al. (2020) showed acceptable fit only in the two-factor solution with items 5, 9, and 10 on the second factor.
Several other validation studies support a version of the CAMM with less than ten items due to unsatisfactory item characteristics. Vinas et al. (2015) promoted a nine-item version without item 5 and Saggino et al. (2017) eliminated items 2 and 5, resulting in an eight-item version. García-Rubio et al. (2019) found evidence comparable to ours in their Spanish version of the CAMM: Items 2, 5, and 6 were eliminated due to low factor loadings. The resulting seven-item version showed better fit indices than the original version with ten items in several different samples (Spanish children, Spanish adolescents, Chilean children, Chilean adolescents).
Some other studies found items 2, 5, and 6 to fit worse; for example, Grossman (2011) and Roux et al. (2019) reported the lowest factor loadings in these items. Also, Chiesi et al. (2017) found items 2 and 5 to be less informative. Furthermore, the same pattern was found in the original version by Greco et al. (2011): items 2, 5, and 6 had the lowest factor loadings. In summary, especially items 2, 5, and 6 were found to fit less into the construct measured by the CAMM across different validation studies. Item 10 was found to fit most commonly into a second factor in other validation studies but showed acceptable factor loading (λ = 0.54) in the one-factor solution in the present study. Furthermore, the major part of studies included item 10 with adequate factor loadings to the single-factor solutions.
A common point of criticism regarding item 2 (“At school, I walk from class to class without noticing what I’m doing”) is that it does not apply to all school systems and is therefore not suitable to every sample. Additionally, some findings suggest that item 2 generally does not fit well into the concept measured by the CAMM. In the sample of the French-Canadian version by Dion et al. (2018), item 2 also shows the lowest factor loading even though the school system applies to the intended meaning since students change their classrooms during the day. After exclusion of participants above the age of 17, we still found item 2 to have low factor loadings.
Item 5 “I push away thoughts that I don’t like” resembles item 10 “I stop myself from having feelings that I don’t like,” so the concept of nonaccepting and pushing away inner states seems not be generally unsuitable to children and adolescents. Perhaps the idea of pushing away feelings is more common in younger people (e.g., watching something funny when feeling sad) than pushing away thoughts.
Item 6 “It’s hard for me to pay attention to only one thing at a time” might rather be associated with attention deficits than with a facet of mindfulness. In conclusion, previous results suggest that an exclusion of several items, mostly 2, 5, and 6, might improve psychometric properties of the CAMM. Internal consistency of Cronbach’s α = 0.83 is in line with the results of the original version by Greco et al. (2011) with α = 0.80 and other recent validations (e.g., Roux et al., 2019; Theofanous et al., 2020).
Convergent validity was indicated by significant correlations with measures of self-compassion, internalizing and externalizing symptoms, and quality of life. Correlations between the seven-item version of the CAMM and the SCS for children and adolescent total score, as well as SCS subscales, were significant. Interestingly, correlations between the CAMM and subscales of the SCS for children and adolescents were higher for subscales representing less self-compassion (self-judgment, over-identification, and isolation). Since the CAMM is also negatively worded (e.g., “I get upset with myself for having feelings that don’t make sense”), it might be plausible that “negative” subscales of the SCS for children and adolescents and the CAMM share more of the same concept than “positive” subscales of the SCS for children and adolescents and the CAMM. This might also explain the relatively small correlation between the CAMM and the mindfulness and common humanity subscales of the SCS for children and adolescents. Accordingly, Neff et al. (2021) found no significant correlations between the CAMM and the mindfulness and common humanity subscales in their validation study of a SCS youth version (r = 0.07, p > 0.05). They explained the absence of significant correlations between the two mindfulness constructs by different conceptualizations of mindfulness (Neff et al., 2021).
We also found significant positive correlations between internalizing and externalizing symptoms measured by the YSR short form and the German CAMM. In the validation study of the original CAMM, Greco et al. (2011) also found correlations to internalizing and externalizing behavioral problems, which is in line with other studies that link mindfulness with anxiety, depression, and negative behaviors in children and adolescents (Dunning et al., 2019; Kallapiran et al., 2015). Highest correlations appeared between the CAMM and the internalization-depression and internalization-anxiety subscales of the YSR, which concurs with the previously reported and well-researched connection between mindfulness and depression and anxiety (e.g., Goldberg et al., 2018). Higher correlations of internalizing behavior and the CAMM compared to externalizing behavior were also reported by Greco et al. (2011).
Correlations between mindfulness and quality of life were also found in the original study, though being clearly higher in our sample (r = 0.25 in Greco et al. (2011) and r = 0.61 in our sample). This might be due to the application of different measurements of quality of life. Accordingly, Hilton et al. (2017) reported in their review on adults with chronic pain that mindfulness meditation improved participants’ quality of life. Kuby et al. (2015) reported MBSR to enhance quality of life scores in a review of studies on healthy subjects.
We found significant gender differences in CAMM total scores with girls reaching lower mean scores indicating lower mindfulness. One possible explanation might be that female adolescents have a higher tendency to ruminate than adolescent boys (Rood et al., 2009). Rumination might be a moderator that complicates mindfulness and promotes depression, which seems to be significantly more frequent in adolescent girls (Salk et al., 2017). Especially young females might benefit from mindfulness trainings, considering the empirical basis on positive effects of MBIs on mental health. Accordingly, Kang et al. (2018) found female sixth grade students to have greater increases than males in positive affect after a meditation training. Also in line with these results, a study found female college students to reach significantly higher mindfulness and self-compassion scores than their male counterparts after participating in a 12-week meditation course. Also, females in this study reported higher decreases in negative affect after the course. Baseline total mindfulness scores were also significantly lower in females (Rojiani et al., 2017).
With regard to different age groups, results were conceptually identical when compared to a reduced sample of 10- to 17-year-olds (as opposed to 10- to 19-year-olds in the whole sample). In the original study by Greco et al. (2011), participants were aged between 10 and 17 years. We decided to also include older participants because of previous validation studies that used samples with a maximum age older than 17 (Chiesi et al., 2017; Cunha et al., 2013; Dion et al., 2018; García-Rubio et al., 2019; Mohsenabadi et al., 2020; Saggino et al., 2017; Theofanous et al., 2020) as well as previous studies that found the CAMM to be invariant of age up to a mean of 22.44 years (Chiesi et al., 2017; Theofanous et al., 2020). Results support the use of the German seven-item CAMM in a broader age range.
Limitations and Future Research
Several limitations of the study need to be considered. First of all, the sample consisted of a community sample and sample size was relatively small when compared to previous validation studies. Second, a vast majority of participants attended a type of school with 12 to 13 years of education (Gymnasium), as compared to fewer years of education. Also, more females than males participated in the study. According to Carlson (2018), this represents the typical sample in mindfulness studies which consists mostly of well-educated females. Additionally, data about prior mindfulness or meditation practice of the participants were not collected. This leads to a limitation in the generalizability of the reported results, which should be addressed in further research.
Third, to be precise, due to its negative wording the CAMM does not measure mindfulness skills but the absence of mindfulness skills. Therefore, all items of the CAMM were reversed. As Greco et al. (2011) state, the reversion of items might not clearly display the opposite concept. This was also discussed in a recent validation study of the Greek version of the CAMM (Theofanous et al., 2020). One example of addressing the critique of negatively worded items was proposed by Höfling et al. (2011), who explored a version of the (also negatively worded) MAAS with additional, positively rephrased “mirror items.” Höfling et al. (2011) stated that mindfulness can be assessed by positively and negatively phrased items, but recommended to apply trait-method models to control for method effects. Further research should be conducted addressing this matter.
Fourth, validity measures should be considered cautiously since the German translations of the respective questionnaires (YSR short form, SCS for children and adolescents) have not been validated. Also, divergent validity was not addressed, for example, by comparing the CAMM to a questionnaire measuring other psychological symptoms.
Finally, van der Eijk and Rose (2015) state that a CFA based on Likert items bears a higher risk of false rejection of a unidimensional structure. This might have led to a (false) rejection of the ten-item unidimensional model. Furthermore, it might be interesting to include parent or teacher perceptions of the child’s or adolescent’s ability to be mindful and to investigate in test–retest reliability of the German CAMM.
Declarations
Ethics Approval
The study protocol was approved by the ethics committee of the Faculty of Behavior and Empirical Cultural Sciences of Heidelberg University, based on the criteria of the Helsinki Declaration (AZ Kal 2018 2/2-A2, 07/16/2020).
Consent to Participate
Informed consent was obtained prior to participation from all individual participants (and their legal guardians if aged under 18) included in the study.
Conflict of Interest
HB, TH, and JM receive royalties from mindfulness books they have authored. TH and JM receive stipends from workshops on mindfulness. The other authors declare that they have no conflict of interest.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.