text new page (beta)
English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
This study examines which particular aspects of self-construal may be associated with positive and negative affect within the contemporary Mexican cultural context. In order to accomplish this, we first examine evidence as to the nature of self-construal in two Mexican regions, using recently developed measures. We then examine the reasons why associations between self-construal and affect might be expected to vary between different cultural contexts.
Self-construal
Cross-cultural psychologists (e.g., Smith, Vignoles, Fischer & Bond, 2013) have typically considered the way in which individuals perceive themselves as dependent upon the type of cultural context within which they are located. Following the pioneering study of Markus and Kitayama (1991), a distinction between independent self-construal and interdependent self-construal has been widely used. Those who perceive themselves as relatively independent of others are more frequently found within more individualistic nations, while those who see themselves as more strongly interdependent with others are more frequently found within more collectivistic nations. A measure of independent and interdependent self-construal developed by Singelis (1994) has been used in many studies (Oyserman, Coon & Kemmelmeier, 2002), but doubts have been raised as to its measurement validity. Reliability for this scale varies between samples, and the items do not control for the tendency for some respondents to respond positively to all items. Some authors have also suggested that the items fail to distinguish self-construal from other aspects of cultural difference such as attitudes, values and beliefs. There has been a need for newer and more carefully designed measures that can differentiate the elements of Singelis’ more global measure.
Within Mexico, Diaz-Loving and his colleagues (La Rosa & Diaz-Loving, 1991; Díaz-Loving & Draguns, 1999; Diaz-Loving, 2005, 2006, 2015) have developed a comprehensive set of scales measuring self-construal, tapping physical, social, emotional, moral and occupational aspects. Within these domains they noted the strong emphasis on interpersonal relationships that is characteristic of collectivist cultures, exemplified by a willingness to preserve harmony by adapting to the wishes of others, obeying one’s parents and presenting oneself as happy and cheerful. These ways of describing oneself are clearly characteristic of more interdependent self-construals, but we need more recent information, given the substantial social changes that have occurred with Mexico over the past several decades.
Aside from the social changes that have occurred in all areas of Mexico, it is likely that the values and personality of those living in close proximity to the US border will have been distinctively affected by the threats to their identity that this entails. These stigmatizing threats include unfavorable comparisons of oneself with those north of the border, the presence of constant migratory flow, and images of local violence (Campos & Ortiz, 2012). Given the contrast between the relatively tight structure of Mexican culture and the distinctively loose culture of neighboring California (Gelfand et al., 2011; Harrington & Gelfand, 2014), a more strongly individualistic self-construal can be predicted.
In a recent study, Vignoles et al. (2016) distinguished between seven different aspects of self-construal. Each of these aspects was seen as expressing an element of independence versus interdependence, but doing so in a way that is not necessarily correlated with other elements. The new measure was first developed using samples of students from 16 nations and then fully tested within 55 samples of adults drawn from 33 nations. The items comprising these measures improved on earlier measures in three ways. Firstly, the contrasting poles defining each dimension were identified. For instance, the first dimension was defined in terms of self-reliance versus dependence on others, rather than in terms of high versus low self-reliance. Secondly, some of the items defining this dimension were phrased in terms of self-reliance and others were phrased in terms of dependence on others. This makes it possible to estimate and control for acquiescent responding. Thirdly, the response scales for each item were keyed in terms of ‘how well does this describe you?’, which provides a more direct focus on oneself in contrast to the ‘agree/disagree’ scales used in earlier measures. The seven dimensions of self-construal are shown in Table 1, along with a sample illustrative item for each.
Table 1 Dimensions of individual-level self-construal
| Independence | Interdependence | Sample item |
|---|---|---|
| Self-Reliance | Dependence on Others | You prefer to ask other people for help rather than relying only on yourself |
| Self-Containment | Connection to Others | Your happiness is unrelated to the happiness of your family |
| Difference from Others | Similarity to Others | Being different from others makes you feel uncomfortable |
| Self-Interest | Commitment to Others | You value personal achievements more than good relations with those around you |
| Consistency | Variability | You behave differently when you are with different groups of people |
| Self-Direction | Receptiveness to Influence | You prefer to do what you want without letting your family influence you |
| Self-Expression | Harmony with Others | You try to adapt to those around you even if it means hiding your inner feelings |
Vignoles et al. (2016) showed that the endorsement of these seven aspects of independence/interdependence across their 55 samples did not match exactly with the contrast between cultures thought to be individualistic and those thought to be collectivistic. Looking at the different scores can provide a more precise indication of the distinctive ways in which culture members perceive themselves. The survey by Vignoles et al. included nine samples from South America, but none from meso-America. The South American samples scored distinctively high on self-expression and on difference from others. The first task of the present paper will be to examine which aspects are emphasized within a contemporary Mexican sample.
Self-construal as a predictor of affect
Positive and negative affects are known to be associated with both personality and a wide variety of short-and long-term life experiences (e.g., Carver & Connor-Smith, 2010; Kim et al., 2016). In addition to these more immediate causal agents, there are further factors that may enhance or mitigate their effect. One of these is the cultural context within which an individual is located. If a person construes him or herself in a relatively similar way to those with whom they associate, a negative mood may be mitigated and a positive mood may be enhanced. In contrast, where self-construals of a person and those around them differ, these effects may be reversed. In more specific terms, we can predict that independent self-construals will be a stronger protective factor against depression and in favour of self-esteem in cultural groups where independent self-construals are prevalent. Conversely, interdependent self-construal should be a stronger protective factor against depression and in favour of self-esteem where interdependent self-construals are prevalent.
An initial study relevant to these predictions was reported by Chen, Chan, Bond, & Stewart (2006). Comparing students from Hong Kong and the US, these authors found support for their prediction that self-efficacy would be a stronger protective factor against depression in the US than in Hong Kong. However, a prediction that relationship harmony would be a stronger predictor in Hong Kong was not supported. Smith et al. (2016) extended this study by surveying students from ten nations. Self-efficacy and a measure of relationship harmony were found to be protective factors against depression in almost all samples. However, as predicted, self-efficacy was found to be a stronger protective factor against depression in more individualistic nations. Furthermore, the measure of relationship harmony was found to be a stronger protective factor in nations where some (but not all) of the specific aspects of interdependent self-construal identified by Vignoles et al. (2016) were prevalent. In the most recent study of this type, Maass et al. (2019) found that independent self-construal was a significant protective factor against depressive mood among a sample of Dutch nationals. However, among a sample of immigrants from more collectivist cultures who were living in the Netherlands there was no such relationship, and there was a trend toward a link between interdependent self-construal and low depression. Thus, there is strong evidence that independent self-construal is a protective factor against depression, but mixed evidence relating to interdependent self-construal. We need more evidence from cultural groups in collectivist contexts.
Self-esteem in 53 nations has been surveyed by Schmitt and Allik (2005). Mean scores proved unrelated to levels of individualism-collectivism. The reason for this may be provided by the results of Becker et al. (2014) who examined the bases self-esteem among adolescents across 20 different cultural groups. Their results showed that levels of self-esteem in different groups varied in relation to cultural values. For instance, in more individualistic groups self-esteem was predicted by feelings of being in control one’s life. In contrast in more collectivistic groups self-esteem was predicted by feelings that one was doing one’s duty. Thus, more finely specified measures of self-construal may show which aspects of independence and interdependence relate to self-esteem in a given cultural context.
The prior studies linking affect to culture form the basis for the present investigation. We explore links between affect and self-construal within a collectivist culture. The previous studies of La Rosa and Diaz-Loving (1991) and Diaz-Loving (2006) provided evidence confirming the conclusions of Hofstede (1980) and others that Mexico is a collectivist culture. However, as in other nations, there will be individual-level variability in how individuals construe themselves, with some favouring independent self-construal, and rather more favouring interdependent self-construal. We can therefore test whether interdependent self-construal would be a protective factor against depression, and would favour self-esteem. The availability of measures for the seven dimensions of self-construal devised by Vignoles et al. (2016) means that this hypothesis can be tested seven times, to give maximum clarity as to which dimensions are most relevant to positive and negative affect.
Method
Participants
The sample comprised 93 students of varied subjects from the Autonomous National University of Mexico (54 percent male; mean age 19.8), and 130 students of social science majors from El Colegio de la Frontera Norte, Tijuana (n = 14) and the Baja California State University (n= 116) (56 percent male; mean age 22.5). All participants had Mexican nationality. Mean rating on a seven point scale of their rural versus urban origins was 6.8 (highly urban) for Mexico City and 6.5 for Tijuana.
Measures
Each of the seven dimensions of self-construal identified by Vignoles et al. (2016) and shown in Table 1 were surveyed, using six items per dimension. Table 1 shows a sample item for each of the dimensions. Nine-point response scales were used, keyed from ‘Describes me exactly’ to ‘Does not describe me at all’. High scores on each dimension refer to the independence rather than interdependence. Depression was measured with the 20-item version of the Centre for Epidemiological Studies Depression scale (CES-D) (Radloff, 1977). These items have 4-point response scales keyed in terms of frequency of symptom occurrence. A sample item is: ‘I felt that everything that I did was an effort’. Four of the items describe positive symptoms and these are reverse keyed. Self-esteem was measured using Rosenberg’s (1979) self-esteem scale. A sample item is: ‘I feel that I have a number of good qualities’. Responses are on 4-point scales, keyed from ‘strongly disagree’ to ‘strongly agree’. Five of the ten items are reverse keyed. All survey items were translated from English to Spanish and then back-translated independently (van de Vijver & Leung, 1997). Corrections to the translated items were made through discussion.
Procedures
Participants completed paper versions of the survey, which included additional items not analysed in the present paper. 30 of the respondents from Mexico City completed the survey in a classroom and the remainder were approached in public spaces on the university campus. Respondents from Northern Mexico completed the survey in classrooms.
Results
Details of all measures are provided in Table 2. Cron-bach alpha was adequate, except in the case of the self-orientation scale. Fourteen of the 222 respondents had mean scores above 3 on the 4-point scale measuring depression. The mean score for self-esteem is similar to that reported by Schmitt and Allik (2005) for their sample of Mexican students.
Table 2 Details of All Measures
| Mean | Standard Deviation | Cronbach’s alpha | |
|---|---|---|---|
| Self-Reliance versus Dependence on Others | 6.56 | 1.52 | .70 |
| Self-Containment versus Connection to Others | 3.98 | 1.56 | .72 |
| Difference from Others versus Similarity to Others | 6.79 | 1.40 | .67 |
| Self-Interest versus Commitment to Others | 5.46 | 1.39 | .57 |
| Consistency versus Variability | 5.87 | 2.04 | .88 |
| Self-Direction versus Receptiveness to Influence | 6.77 | 1.38 | .73 |
| Self-Expression versus Harmony with Others | 6.15 | 1.57 | .70 |
| Depression | 1.92 | 0.60 | .91 |
| Self-Esteem | 3.12 | 0.60 | .86 |
After controlling for gender, the mean scores for the two samples differed significantly for self-es-teem (MTijuana = 3.20; MMexicoDF = 3.01; p < .05) and for perceiving oneself as different from others (MTijuana=7.12; MMexicoDF = 6.30; p < .001). Six of the nor-thern respondents originated from southern states. After reassigning these cases to permit north-south comparisons, the contrast in perceiving oneself as di-fferent from others remains significant (Mnorth = 7.28; Msouth = 6.38; p < .001). The remaining means did not differ between samples. To test more precisely whether this difference reflects the cultural effect of residence close to the US border, the analysis was repeated, including only those born in Tijuana state and those from Mexico City. Respondents from Tijuana perceive themselves as more different from others (MTijuana = 7.16; MMexicoDF = 6.40; p < .001). They also perceive themselves as even less self-contained tan those in Mexico City (MTijuana = 3.73; MMexicoDF = 4.22; p < .03).
Six of the seven self-construal scales have means in excess of the midpoint, indicating that most respon-dents preferred to see themselves as predominantly independent rather than interdependent. However, Latin American respondents are frequently found to use more extreme points on response scales than tho-se from other cultural groups (Hui & Triandis, 1989; Smith, 2004). It is more instructive to examine which dimensions received the most extreme ratings. Res-pondents reported themselves as distinctively high on self-direction, self-reliance, difference from others and self-expression. They also reported themselves as distinctively low on self-containment. Thus, consistent with the results of Vignoles et al. (2016), some aspects of independence are emphasized, but some aspects of interdependence are also salient.
Partial correlations between self-construal and mood states are shown in Table 3. High self-esteem is significantly associated of five of the seven dimensions of independent self-construal. In contrast, depression is linked with four of the dimensions of interdependent self-construal. In relation to both mood states, the strongest predictor is the dimension of consistency versus variability.
Table 3 Partial Correlations of Self-Construal with Depressive Mood and Self-Esteem, controlling for Gender and Sample
| Self-Construal | Depressive Mood | Self-Esteem |
|---|---|---|
| Self-Reliance versus Dependence on Others | -.07 | .19** |
| Self-Containment versus Connection to Others | -.18** | .10 |
| Difference from Others versus Similarity to Others | -.04 | .13 |
| Self-Interest versus Commitment to Others | -.18** | .26*** |
| Consistency versus Variability | -.25*** | .36*** |
| Self-Direction versus Receptiveness to Influence | -.21** | .24*** |
| Self-Expression versus Harmony with Others | -.08 | .24*** |
Discussion
This study has addressed two issues. Firstly, we considered whether contemporary Mexican students describe themselves in ways that are consistent with earlier findings, and whether there are regional differences in how students now describe themselves. Secondly, we examined the aspects of self-construal most strongly associated with positive and negative mood states. Our use of the scales developed by Vignoles et al. (2016) provides evidence from this sample both for a continuing commitment to collectivism and also for several predominantly individualistic dimensions of self-construal. We cannot judge the extent to which these results provide evidence for change, because the measurement instruments that have been employed in different studies have not been the same. What is clear is that there is substantial overlap between the present results and those reported by Vignoles et al. (2016) for the mean of nine adult samples from South America. In both studies the means for difference from others, for self-reliance and for self-expression were distinctively high. However, Vignoles et al. also found high scores for consistency versus variability, whereas we found an additional high score for self-direction. Further sampling would be required to determine the extent to which these similarities and differences prove replicable. The low score for self-containment, which is equivalent to a high score for connection with others provides the strongest indication of a continuing collectivistic orientation.
The differences found between respondents from Tijuana and those from Mexico City are consistent with the expectation of influence attributable to the proximity of the US border, representing two different styles of life in highly urban Mexican culture. Tijuana respondents most strongly emphasize their difference from others, but also show a weak tendency to feel more connected to others. Interview data would be needed to shed light on this combination of effects.
The finding that positive and negative affect are most closely associated with self-construals for consistency versus variability is of particular interest. We all vary our behaviours between the differing social contexts encountered within a single day. However, this variability appears particularly marked within the collectivist nations of East Asia, where the preference for harmony more strongly mandates the requirement to adjust one’s behaviour to the specific requirements of each context (Tafarodi et al., 2004; English & Chen, 2007). For East Asians but not for European Americans, variability across context has been found to be associated with higher relationship quality (English & Chen, 2011). The results from the present study suggest that respondents interpreted the variability of their behaviour as negative, rather than as being situationally appropriate. It is likely that the self-construal items that we used tapping consistency versus variability are insufficiently precise to distinguish variability that is adaptive rather than disorganised. The significant associations between depressed mood and other aspects of interdependence suggest a loss of autonomy rather than disorganisation. Further studies from other cultural groups are required in order to determine in which contexts variability is experienced positively and in which it is experienced negatively.
In this study, we investigated whether previous results showing independent self-construal to be a protective factor against depression would be replicable in a more collectivist context, or whether interdependent self-construal would instead become a protective factor. The results obtained show that independence remains predictive within the Mexican context. However, our data also showed that contemporary Mexican students construe themselves in ways that have some substantially individualistic components. Even in relation to their endorsement of connection with others (an element of interdependence), there was no association with positive or negative affect. To identify contexts in which interdependence could be protective, it may be necessary to survey samples that are less urban and less cosmopolitan than students from these two major Mexican cities.
