Body awareness is usually defined as intentionally focusing on and being aware of one’s body sensations [1]. In recent decades there has been a growing research interest in the role of body awareness in relation to health [2,3]. When defined as an ability to recognize subtle body cues, studies have shown that body awareness can be beneficial in managing health conditions, such as different types of long-term pain [4] or congestive heart failure [5], as well as mental health issues [6]. Enhancing body awareness can thus be an important part of interventions promoting well-being [3] and improving sexuality [7]. Drawing on these results, more research needs to be conducted to better understand how body awareness relates to well-being in different population groups and in different cultures. An important aspect of conducting this research is ensuring that there is a valid measurement instrument for body awareness available in multiple languages.

In a review of self-report instruments for body awareness, Mehling et al. [2] concluded that the Body Awareness Questionnaire (BAQ) [8] was the best performing measure. BAQ was developed to measure a person’s sensitivity to non-emotive and normal bodily processes and consists of 18 items. These items were selected to capture “sensitivity to body cycles and rhythms, ability to detect small changes in normal functioning, and ability to anticipate bodily reactions” [8] (p. 802). Even though the BAQ is intended to function as a unidimensional measure, exploratory factor analysis indicated that the items fall into four factors relating to ‘noting responses or changes in body processes’, ‘predicting body reactions’, ‘sleep–wake cycle’, and ‘prediction of onset of illness’ [8]. Various studies have supported the reliability and discriminant as well as convergent validity of the BAQ [2]. For instance, the BAQ discriminates between body awareness and general neuroticism [8] and is positively correlated with measures of self-consciousness [9]. The instrument has been translated into several different languages, such as Swedish [10], Turkish [9] and German [11], among others. However, a Spanish version of the BAQ has not been developed and evaluated.

The current study addresses this lacuna by conducting a translation, psychometric evaluation, and validation of the BAQ in Spanish. To evaluate the discriminant and convergent validity of the translated BAQ, it is important to explore its relationships with associated but distinct constructs. The choice of other variables has been informed by the model of Farb et al. [12], who theorize that interoceptive awareness, e.g., body awareness, forms a foundational layer of interoception, which can influence higher-order constructs. They suggest, for example, that problematic interoceptive processes can lead to a vast array of health problems. For example, awareness of visceral sensations can be beneficial or harmful, depending on how this is interpreted. While contemplative practices may enhance well-being, misinterpretation of interoceptive signals can contribute to panic disorders. Understanding how to skillfully engage with these sensations is crucial for advancing interoceptive training. The representation of the momentary sensory self is key to compassion and insight practices, as well as the foundational mindfulness practices that support them. For this reason, in this particular study, we chose to study body appreciation [13], self-compassion [14], and sexual satisfaction [15]. These were selected as they are three very different constructs, but all related to health and theoretically understood as higher-order processes related to body awareness. As such, including them in the study provides, in addition to evaluations of validity, tentative results on the relationships between body awareness and these constructs. As a validated measure on body awareness in Spanish is missing, this is knowledge urgently needed in a Spanish context in relation to a range of different health-related research areas. As such, an additional aim is to provide a basis from which different researchers can develop future research.

Body appreciation is a central aspect of positive body image. It is conceptually linked to body awareness but remains theoretically distinct. Body awareness emphasizes attentiveness to internal bodily signals, whereas body appreciation focuses on respecting, accepting, and valuing the body [13]. Previous research suggests that awareness of the body can be a foundation for body appreciation [16]. The Body Appreciation Scale-2 (BAS-2) is widely used and validated across diverse populations to measure this construct [13]. Including the BAS-2 allows for an examination of whether the BAQ correlates with body appreciation as expected while maintaining distinctiveness, thereby establishing discriminant validity.

Self-compassion is another construct theoretically linked to body awareness. Like body awareness, self-compassion also requires an element of interoceptive ability. However, self-compassion is also an evaluation of one’s state of being, and involves recognizing one’s needs and responding to them with care [14]. As mentioned above, it is not only important to be aware of the sensations of the present moment, but this attention needs to be compassionate so that it can lead to health care and well-being rather than the opposite. The Self-Compassion Scale—Short Form (SCS-S) has demonstrated robust psychometric properties and is frequently employed in studies exploring interoception-related constructs [17]. By examining the relationship between the BAQ and SCS-S scores, the study can assess whether body awareness is related to self-compassion as hypothesized, without complete overlap.

Finally, the relationship between body awareness and sexual satisfaction is of growing interest in research. Sexual satisfaction, as measured by the New Sexual Satisfaction Scale—Short (NSSS-S) [15], captures an important dimension of well-being [18]. Previous studies have shown that mindfulness practices focusing on body awareness can enhance sexual functioning by reducing psychological barriers and increasing attunement to physical sensations [19]. This suggests a theoretical connection between body awareness and sexual satisfaction, where body awareness may contribute to but remain distinct from sexual satisfaction. Investigating these constructs in tandem allows for a nuanced understanding of their interplay while providing further evidence for the construct validity of the BAQ.

To establish known-groups validity [20], we will further test the ability of the Spanish BAQ to distinguish between two groups expected to differ in body awareness: meditators and non-meditators. These groups were chosen based on previous findings indicating that practicing meditation can increase levels of body awareness through fostering mindfulness [14].

The aim of the present study is to evaluate the psychometric properties of a Spanish version of the BAQ by conducting factor analyses of the structure of the measure. To this objective, a unidimensional factor structure as well as the suggested four-factor structure will be tested through confirmatory factor analyses. The discriminant and convergent validity of the measure will be evaluated through correlations of BAQ scores to other concepts related to self-awareness, specifically body appreciation and self-compassion, while the predictive validity of the translated BAQ will be evaluated through its correlation with sexual satisfaction. Finally, we will estimate the conceptual validity of the translated BAQ through a known-groups comparison: people with and without experience of meditation. As experience of meditation is suggested to increase body awareness [19], this should be reflected in higher BAQ scores among individuals with experience of meditation.

To assess the psychometric structure of the translated BAQ, two CFA models were tested. Model 1 consisted of a one-factor model and was created to test whether the BAQ can best be described as a unidimensional measure, as originally suggested by Shields et al. [8]. Model 2 followed the factor structure identified by Shields et al. [8] and included the four aforementioned factors. Before analysis was initiated, we used the response variance and a QQ-plot of the residuals for the scale mean to search for outliers and insincere respondents. Four respondents showed a suspicious answering pattern with little or no variance in responses, including for the reversed item, and their responses to the BAQ were removed before analysis. A total of 277 responses were included in the factor analyses.

To test the factorability of the scale responses, we conducted Bartlett’s test of sphericity [30] and a Kaiser–Meyer–Olkin test [31], as recommended by Dziuban and Shirkey [32]. The Bartlett test was significant and thus indicated that the correlation matrix was suitable for factor analysis, χ²(153) = 1180.49, p < 0.001. The Kaiser–Meyer–Olkin test showed a ‘meritorious’ sampling adequacy with an MSA of 0.83. However, the MSA for item 10 specifically was only at 0.59, which is defined in Kaiser and Rice [31] as a ‘miserable’ sampling adequacy. For this reason, both factor models were run with and without item 10 to evaluate its impact on the model fit.

The one-factor model showed poor fit, with a CFI of 0.73, an RMSEA of 0.09 (90% CI = 0.08:0.10), and an SRMR of 0.08. Removing item 10 did not improve the model fit; CFI = 0.73, RMSEA = 0.09 (90% CI = 0.08:0.10), SRMR = 0.08. The four-factor model showed a poor fit to the data for CFI (0.80), but acceptable fit for RMSEA and SRMR (RMSEA = 0.08 (90% CI = 0.07:0.09), SRMR = 0.08). Removing item 10 did not improve the model fit: CFI = 0.80, RMSEA = 0.08 (90% CI = 0.07:0.09), SRMR = 0.08. A scaled χ² difference test showed that the four-factor solution had a significantly better fit to the data compared to the one-factor solution, χ² difference(11) = 62.35, p < 0.001. However, as the four-factor model still did not reach all the predetermined cut-offs for model fit, we further investigated the factor structure of the translated measure with an exploratory factor analysis (EFA) using the fa function in the psych package [33].

To determine the suggested factor structure of the data, we used the Very Simple Structure test (VSS) [34] for models ranging from 1 to 10 factors. The VSS function included in the psych package provides several goodness-of-fit statistics, including the VSS, the MAP criteria, the BIC, the sample size adjusted BIC, and the RMSEA (see Table 2).

The highest VSS2 score as well as the lowest MAP criterion and BIC occurred for a solution of two factors, while the sample-size-adjusted BIC suggested an optimal solution of five factors. All solutions using more than one factor resulted in an RMSEA value under 0.06, indicating good fit. See Table 2 for fit statistics for different numbers of factors. To provide insights into the structure of the translated measure, we therefore ran two EFAs with either two or five factors using maximum likelihood with robust errors and oblimin rotations. The fit of the two models was compared using χ² difference testing, which showed that the two-factor model had significantly better fit than the five-factor model, χ² (45) = 142.02, p < 0.001, indicating that despite the five-factor model showing a lower RMSEA, the two-factor model had better data fit once the complexity of the five-factor model was taken into account. Standardized factor loadings are reported in Table 3. Together, the CFAs and the EFAs indicate that the translated BAQ does not display a one-dimensional nor four-dimensional factor structure, but rather a dual factor structure with items relating to ‘monitoring physical energy levels’ loading on one factor (α = 0.80) and items relating to ‘predicting body responses’ loading on another factor (α = 0.72). The factors correlated positively, r = 0.40, The item relating to not noticing the bodily effects of seasonal rhythms (item 10) loaded poorly on either factor and may not be suitable for inclusion in the scale, either because of its reversed answer format or because of its focus on a more external cyclical process.

To examine the relation of the translated BAQ to related measures, correlations between the Spanish BAQ, the BAS-2, the NSSS-S, and the SCS-S (and its subscales) are reported in Table 3. The BAQ showed significant but weak positive correlations with the BAS-2 and the SCS-S but did not correlate significantly with the NSSS-S. Accordingly, the Spanish BAQ showed convergent validity with the BAS-2 and the SCS-S (as well as its subscales) and divergent validity with the NSSS-S. Descriptive data and correlation coefficients for all measures are presented in Table 4.

Firstly, the results are in line with the results found in the English version, showing that the scale presents adequate reliability, with the value of Cronbach’s alpha ranging from 0.80 to 0.88 [5,8]. It is also consistent with the German [11] and Turkish versions [9], as well as the estimated reliability of the Swedish version for a clinical sample [10]. Reliability for the Spanish BAQ was higher than for the Swedish version when used for a non-clinical sample [10].

Regarding the relationship with other scales, the results have confirmed the convergent and discriminant validity of the BAQ, finding a positive and significant correlation with the BAS-2 and SCS scales, but not with the NSSS-S. Because awareness of the body can be thought to be a basis for, but distinct from, appreciation of the body [16], the significant correlation between the BAQ and the BAS-2 supports the convergent and discriminant validity of the translated BAQ. In a similar fashion, both bodily awareness and self-compassion require a degree of introspective ability [14], so the weak but significant correlation between the BAQ and the SCS-S indicates that the translated BAQ is picking up this introspective ability as well as the specific aspects of it related to bodily awareness. Finally, the lack of a significant correlation between the translated BAQ and the NSSS-S supports the discriminant validity of the translated BAQ. These findings converge with previously found correlations between the BAQ and measures such as the Self-Consciousness Inventory and the Body Consciousness Questionnaire, as well as a lack of a correlation with unrelated measures, such as the Rosenberg Self-Esteem Scale [8]. While it is true that the results found are in line with the justification for the selection of these instruments as comparison measures, based on a focus on interoception that can favor health, future studies could include other measures of body connection already validated in Spanish, such as the Scale of Body Connection (SBC) [35]. Although they do not assess exactly the same construct and it has already been pointed out that the BAQ is considered the best scale for assessing body awareness, the inclusion of such scales could increase the correlations found in this study.

With regard to sexuality, although several studies have found changes in body awareness and other variables related to sexual function after Mindfulness-Based Interventions (MBI, e.g., Silverstein et al., [19]), these interventions used scales other than the BAQ, such as the Scale of Body Connection [7,36], or even specific scales of sexual body awareness, such as the Sexual Mindfulness Measure [37]. These scales have been used to compare meditators with non-meditators or to compare the before and after of an MBI applied to sexuality. However, in the present study, only the correlation between the BAQ and the NSSS-S was analyzed directly, but no MBI applied to sexuality was conducted to improve scores on these questionnaires. Correlations between the BAQ and the NSSS-S were also not analyzed with differentiation between meditators and non-meditators. This may have led to the fact that no statistically significant correlations were found between the two scales in the present study. Nevertheless, the BAQ may be appropriate for use in research that aims to improve some aspect of sexuality, e.g., by applying mindfulness meditation (one MBI), taking into account that body awareness may be a mediating or influencing variable in sexuality [38]. In fact, in terms of conceptual validity, statistically significant differences in BAQ scores were obtained between meditators and non-meditators. This result is consistent with those that have found a relationship between meditation and body awareness [39], where body awareness is significantly higher in meditators than in non-meditators [40], as well as in those studies in which an MBI has improved participants’ body awareness [41].

In terms of factor structure, neither the one- or four-dimensional structure identified by Shields et al. [8] showed an optimal fit to the current data. Using exploratory factor analyses, a two-factor structure was instead shown to have the best fit. This is consistent with the Swedish version [10], and concurs with the structure of the German version [11]. This change in the factor structure with respect to the original 1989 English scale, while there was concordance with the factor structure found in later studies, such as the Swedish (from 2013) or the German (from 2018) ones, could point to sociocultural changes. The dissemination among the general population in the West of meditation practices such as MBIs from the 1990s to the present by authors such as Jonh Kabat-Zinn (e.g., Mindfulness-Based Stress Reduction, MBSR) [42] may have had an impact on the way we are aware of our bodily sensations. This could also be the reason for the percentage of participants who meditate within the sample. These socio-cultural changes, with a greater emphasis on mind–body treatments and the fact that the population may have developed a more refined body awareness, may be orienting the factor structure of the scale towards two factors instead of one factor, as was found in the original version several decades ago.

Similar to findings from the English [5] and Swedish versions [10], item 10 showed a weak relationship to the remaining items. This supports the theory that the issue was not with the translation of the item into Spanish, but rather with the content of the item itself. The poorer fit of this item may be due to the fact that it is the only item where the score has to be reversed or because it asks about very general aspects of the influence of seasonal change on the body, which differs considerably from other items. It could therefore be desirable to eliminate item 10 in future versions of the questionnaire.

Regarding the limitations of the study, the sample consisted of a small convenience sample. However, for a first exploratory study of the psychometric properties of this version, it is considered sufficient (e.g., [43]). Nevertheless, the use of the same sample for multiple factor analyses can increase the risk of overfitting and the two-factor structure identified in the current study should therefore be confirmed with additional samples. As it was a convenience sample, the participants were mainly women with a high level of education, which may have influenced the results. However, as there was no significant difference in BAQ score found between women and men, the overrepresentation of women in the sample is unlikely to have substantially influenced the conclusions of the study. The fact that no differences were found between genders may suggest a cultural change in terms of the attention given in our context to the body, where there is increasing attention to health-related factors, such as sports.

On the other hand, the percentage of people who consider themselves bisexual in this sample compared to the general population is striking. This can be considered a strength of this study, as it shows more data from the LGTBIQA+ group than other studies. The timing of the study may also indicate a generational shift in attitudes, labeling, and self-conceptions of sexual orientation. Regarding the representativeness and generalizability of the sample, we are aware that this is only an exploratory study and that these tentative results should be further explored in a Spanish context with larger and more representative samples, to help clarify the reasons for the results found. Although Mehling et al. [2] selected the BAQ as the best scale to assess body awareness, it has not yet been validated in many languages or applied in many studies [44,45]. Therefore, one of the strengths of this study is that it is the first to explore its application in a Spanish sample throughout the Hispanic world. This study can be a starting point for the study of the application of the BAQ in relation to other health variables in clinical samples. For example, it could be applied together with scales such as the Eating Attitudes Test (EAT-40) validated in Spanish (e.g., [46]) to assess changes in people with eating disorders after a psychological intervention (e.g., one MBI). Another strength is the use of expert evaluation of the translation, as well as comparisons between different factor solutions using a diversity of fit measures.