Interdependent Self-Construal, Anger Suppression, And Depression In Middle-Aged And Older Adults

Interdependent self-construal, anger suppression, and depression

How a person defines, oneself can play a major role in the way emotions are experienced, regulated, and expressed. In their theory of the self, Markus and Kitayama (1991) suggested that interdependent self-construal relate to the self-being defined through social relationships, which highlights connectedness with people. As such, individuals with higher levels of interdependent self-construal tend to suppress angry feelings for the sake of maintaining social harmony (Markus & Kitayama, 1991), which can result in greater susceptibility to psychological distress.

Research on anger expression has well-documented associations between anger and different interpersonal and relational outcomes (Cheung & Park, 2010; Park & Zane, 2004). Anger suppression, which relates to how often a person experiences angry feelings but does not express them (i.e., anger expression-in) (Spielberger, 1999), has been found to be linked with adverse emotional outcomes, such as depression (Cheung & Park, 2010; Kitamura & Hasui, 2006; Park, Kim, Cheung & Kim, 2010).

Research on interdependent self-construal has been relatively inconsistent in its relation to anger suppression and depression. While some studies have found a positive association between interdependent self-construal and anger suppression (Cheung & Park, 2010; Kitamura & Hasui, 2006), others have found no such relation (Akutsu, Yamaguchi, Kim & Oshio, 2016). Similarly, research on interdependent self-construal and depression has also had somewhat different findings, with some studies finding the two to be positively associated in Asian American adolescents (Liu & Goto, 2007; Norasakkunkit & Kalick, 2002), while others have reported no such significant association (Park et al., 2010).

However, in their study with a sample of 166 adolescent Korean Americans, Park and colleagues (2010) found that interdependent self-construal was negatively linked with externalising anger, which was to be expected given that individuals with interdependent self-construal tend to avoid conflict, and value maintaining relational harmony. This result is consistent with the findings of a more recent study (Akutsu et al., 2016) that used a subset sample of Americans from the Midlife Development in the United States (MIDUS) and Japanese from the Midlife Development in Japan (MIDJA). Overall, the results of this study indicate that an individual’s definition of ones’ self, has a limited effect on anger suppression in Japan. Moreover, Park and colleagues (2010) found that anger suppression was a significant predictor of depressive symptoms. However, the generalisability of these findings is somewhat limited given that the study’s sample explicitly comprised of Korean American adolescents who were recruited through ethnic churches.  

As Cheung & Park, (2010) suggested, the relationship between anger suppression and depression must be considered within a cultural context. In their study, Cheung and Park (2010) examined the role of anger suppression as a mediator between temperament (i.e., trait anger) and family processes (i.e., family cohesion and conflict) and their effects on depressive symptoms in a sample of 365 Asian and European American college students. Also, they also examined the moderating role of interdependent self-construal in the link between anger suppression and depression. As hypothesised, they found anger suppression to play a significant role in mediating the influence of trait anger and family processes on depressive symptoms in this sample of college students.

Positive Effects of Anger Suppression

Furthermore, they also found that interdependent self-construal moderated the link between anger suppression and depression in such a way that it weakened this link. In other words, individuals high in interdependent self-construal experienced less adverse effects when they suppressed their anger, which could be explained through their more top focus on relational harmony and possible positive results this entails for these individuals. These findings are consistent with those of Kitamura and Hasui (2006).

Contrastingly, in a study with a sample of 73 Canadian undergraduates (Le & Impett, 2013), which examined the daily experience of people who are dating, it was found that suppression of negative emotions plays a vital role in relationship quality and individual well-being, in the context of sacrifice. However, this was seen to be the case only in individuals with a higher level of interdependent self-construal – those who value putting their relationship above their self-interest. These findings suggest that there are instances where suppression of negative emotions can be interpersonally and individually beneficial. Similarly, Akutsu and colleagues (2016) found that interdependent self-construal was positively linked with life satisfaction in both the American and the Japanese sample.

Research in this area appears to be somewhat inconsistent regarding the role that interdependent self-construal plays in the link between anger suppression and depression. Indeed, Akutsu and colleagues (2016) found this self-construal to have no relation to anger suppression. Moreover, it has been positively linked to life satisfaction, relationship quality, and personal wellbeing (Akitsu et al., 2016; Le & Impett, 2013). It appears that in populations that have more Asian cultural values higher levels of interdependent self-construal weaken the adverse psychosocial effects associated with anger suppression.

Studies seem to be reasonably consistent in regards to the positive relation between anger suppression and depression; however, most of the research in this area has been limited to student samples, which are not very representative of the older population, thus limiting the generalizability of these findings to adults. Therefore, this study aimed to examine the role that interdependent self-construal and anger suppression have in depression in middle-aged and older adults. More specifically, the study sought to find significant predictors of depression in this adult sample.

First, it was hypothesised that anger expression-in would be a significant mediator in the relationship between interdependent self-construal and depression in this sample of American middle-aged and older adults. Second, it was hypothesised that interdependent self-construal would be a significant moderator in the relationship between anger expression-in and depression.  Consider the hypotheses below.

The study consisted of a voluntary sample of 1255 midlife adults living in the United States. The participants were 542 males and 713 females, with ages ranging from 35 to 86 years, and a mean age of 57.32 years (SD = 11.5). Inclusion criteria included middle-aged to older adults only.

1. Interdependent self-construal:

Interdependent self-construal was measured using the 12-item subscale of the 24-item Self-Construal Scale (SCS; Singelis, 1994), which uses a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). One example of an item from the subscale is: “I will sacrifice my self-interest for the benefit of the group I am in”. The internal consistency of this scale was questionable, with a Cronbach’s alpha of 0.69.

1. Anger Expression-In:

Research Methods

Inward anger expression or anger suppression was measured using the eight-item Anger Expression-In subscale of the State-Trait Anger Expression Inventory (STAXI-2; Spiel Berger, 1999). The measure uses a 4-point Likert scale measuring the frequency of angry feelings being suppressed, ranging from (1) almost never to (4) almost always. A sample item is: “In general when I feel angry or furious, I boil inside but don’t show it”. The internal consistency of anger expression-in scale was right (α = 0.82).

1. CES-D Depression Scale:

Depressive symptoms were assessed using the 20-item self-report depression scale (CES-D Depression Scale; Radloff, 1977). The measure uses a 4-point Likert scale measuring how often an individual experienced each of the symptoms, ranging from (1) rarely or none of the time to (4) most or all of the time. One example of a scale item is: “During the past week, I felt depressed”. Cronbach’s alpha of the CES-D depression scale was 0.90. Items were recoded to a 0-3 level (i.e., 1=0, 2=1, 3=2, 4=3), and reverse-question items were then reverse-coded so that high scores represent higher scores in depressive symptoms. Following this, scale scores were summed for all items, except for those where there was missing data.

Data were derived from a subset of the Midlife Development in the United States (MIDUS) survey (i.e., MIDUS-II Project 4 Biomarker data) which was conducted in 2004 as a follow-up from the original MIDUS, which was done in 1995 and 1996. The study was approved by the National Institute on Aging (NIOA) at the University of Wisconsin-Madison, and informed consent was obtained from all participants before commencing the survey administration.

For preliminary analyses, descriptive statistics for age and gender will be computed, assumptions of linearity, independence, homoscedasticity, multicollinearity and normality will be checked, as well as any outliers and missing data. Means, standard deviations, median and interquartile range, and effect sizes will be presented in Table 2.

Correlations between variables will be presented in a correlation matrix. All the outliers will be removed and replaced using their respective variances. Three standard deviations cut-off will be used as it corrects the value to a lesser extent and therefore will not likely affect the outcome of the analysis. As a primary analysis, using an SPSS program, a regression analysis of mediation and moderation of interdependent self-construal on the relationship between anger suppression and depression will be performed (α = .05 threshold).

Frequency analysis is applied in this study to measure the descriptive statistics. The distribution of both the variable and non-variables was showed to test the normality of the data before the analysis. The pie chart and histogram below is illustrate the distribution of gender as one of the demographic variables in the study. The table 1 and 2 below displays the descriptive statistics of three variables used in the study before and after data cleaning.

Table 1: Descriptive statistics before data cleaning

	CESD: Center for Epidemiologic Studies Depression Scale	Spielberger Anger Expression: Anger/In	Relational Interdependent self-construal scale
N	Valid	1246	1250	1249
Missing	9	5	6
Mean	8.730	14.647	4.979
Median	7.000	14.000	5.100
Std. Deviation	8.1914	4.1603	.9487
Variance	67.099	17.308	.900
Skewness	1.596	.744	-.423
Std. Error of Skewness	.069	.069	.069
Minimum	.0	8.0	1.8
Maximum	54.0	31.0	7.0

Table 2: Descriptive statistics table of all the variables used in the study

	N	Mean	Std. Deviation	Skewness	Kurtosis
				Std. Error		Std. Error
Depression (B4QCESD)	1255	8.7305	8.16195	1.602	.069	3.130	.138
Anger Expression-in (B4QAE_AI)	1255	14.6474	4.15204	.746	.069	.507	.138
Interdependent Self-construal (B4QRISC)	1255	4.9785	.94640	-.424	.069	-.169	.138
Valid N (listwise)	1255

1. Correlation Analysis

Pearson’s correlation analysis was conducted to test the relationship between each independent variable and the dependent variable. There was a significant correlation between all the three variables with a p-value of < .001 for the relationship between each independent variable (anger expression-in and interdependent self-construal) and the dependent variable (depression), and a p-value of .004 for the relationship between the two independent variables.

Pearson’s correlation shows a weak positive correlation between anger expression-in and level of depression (r = .44). On the other hand, there is a weak but significant negative correlation between interdependent self-construal and level of depression (r = -.17). Similarly, there is a weak but significant negative correlation between the two independent variables (r = -.08). The table 3 below illustrates correlation output.

Coefficientsa
Model	Unstandardized Coefficients	Standardized Coefficients	t	Sig.	95.0% Confidence Interval for B
B	Std. Error	Beta	Lower Bound	Upper Bound
1	(Constant)	2.524	1.254		2.013	.044	.064	4.984
SMEAN(B4QAE_AI)	.776	.045	.429	17.073	.000	.687	.866
SMEAN(B4QRISC)	-1.097	.200	-.138	-5.497	.000	-1.488	-.705
a. Dependent Variable: SMEAN(B4QCESD)

1. Regression Analysis

 After confirming that the relationship existed between the three variables, a regression analysis of mediation and moderation was conducted to evaluate the relationship that each variable had on the other, mainly, how the two independent variables affect the dependent variable. The regression analysis had an adjusted R-Squared of 0.213, which means that the model summary is predicting about 21% of the variance.

Model Summaryb
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate	Durbin-Watson
1	.462a	.213	.212	6.66454	2.033
a. Predictors: (Constant), SMEAN(B4QRISC), SMEAN(B4QAE_AI)
b. Dependent Variable: SMEAN(B4QCESD)

1. Hypothesis testing

Analysis of variance (ANOVA) test was conducted to determine if there existed any significance difference between the dependent and the independent variable. The null and alternative hypotheses were formulated as follows;

Ho: There is no significant difference in the mediating and moderating roles of Interdependent Self-construal and anger suppression on depression in an American adult sample.

H_A: There is significant difference in the mediating and moderating roles of Interdependent Self-construal and anger suppression on depression in an American adult sample.

To test the hypothesis, it was necessary to check for the normality of the data and the box plots below confirms the normality of the variables under the study.

The ANOVA showed that the model is significant, F (2, 1252) = 169.54, p < .001). The coefficient of the regressions presented in the Table 3 below shows that both anger expression-in (B4QAE_AI) and interdependent self-construal (B4QRISC) are predictors of depression. The table below illustrate the output of the result from SPSS.

ANOVAa
Model	Sum of Squares	df	Mean Square	F	Sig.
1	Regression	15060.589	2	7530.295	169.540	.000b
Residual	55608.921	1252	44.416
Total	70669.510	1254
a. Dependent Variable: SMEAN(B4QCESD)
b. Predictors: (Constant), SMEAN(B4QRISC), SMEAN(B4QAE_AI)

When a mediator was used, which is a variable created by getting the product of anger expression-in and interdependent self-construal, both the old model and the mediator model became significant with F(2, 1252) = 169.54, p < .001, and F(2, 1252)  = 113.18, p < .001, respectively. However, the overall regression model changed when the mediator was included. For example, interdependent self-construal became less significant at F (2, 1252) = -.89, p = .38. Therefore, it can be concluded that the relationship between the level of depression and interdependent self-construal was fully mediated by levels of anger expression-in. The Mediation table is presented in the Table 4

 Before moderation analysis was performed, two additional centered variables were created, namely Centered Anger Expression-in and Centered Interdependent Self-Construal.

 The interaction between the two variables was used to test the effect of mediation of interdependent self-construal and anger expression-in on depression in an American adult population. The mediation had a significant increase in the percentage coverage of variance shown by R-Square change of .213, p < .001. The value of F also changed significantly (F = .448, p < .001). This result is a preliminary indicator that mediation has an effect on the level of depression of American adult population.

 This research study made three significant findings. First, this study confirmed that both interdependent self-construal and anger suppression are factors that influence the level of depression among the adult population of the United States. This result was confirmed in two ways. First, the Bivariate Pearson’s correlation showed that all the three variables (i.e. interdependent self-construal, anger suppression, and depression) are significantly correlated to each other.

This finding is consistent with the results of Park et al., (2010). Park and his colleagues found that there is a positive relationship between the levels of anger suppression and depression. According to the study, those people who demonstrated depressive symptoms were associated with high levels of anger, lower levels of family cohesion and high level of family conflicts.

Secondly, the first regression analysis indicated that both interdependent self-construal and anger suppression could predict the level of depression amongst adult American population. From the regression equation, the level of depression increased with the increase in the level of anger suppression, evidenced by positive regression coefficient of 0.776. This finding is in tandem with the research findings of Jackson et al., (2011) that concluded that anger suppression has a direct role in the level of depression among adolescents. On the other hand, the level of depression would decrease with the decrease in the level of interdependent self-construal, evidenced by the negative regression coefficient of -1.097.

The second significant findings that this study made was that the relational interdependent self-construal moderates the relationship between anger suppression and level of depression on the adult American population. In other words, this study revealed that the level at which anger suppression affects the level of depression can be influenced by an individual’s level of interdependent self-construal. Therefore, the second hypothesis that interdependent self-construal would be a significant moderator in the relationship between anger suppression and depression amongst the adult population of the United States was supported.

 The finding was consistent with that of Park and colleagues (2010) that was conducted using 166 American teenagers of Korean origin. The study found out that there is a negative correlation between interdependent self-construal and externalising anger. Generally, it is expected that people with interdependent self-construal often avoid conflicts and have an incredible understanding.

From the table of regression coefficient below, both variables were statistically significant (p < .001) before mediation. However, even after mediation, both remained statistically significant (p < .001). Moreover, the interaction between the centered variable was not significant, F = -.759. p = .448. This interaction implies that mediation does not affect the interaction of the variables.

Also, Akutsu and colleagues (2016), in their study found that individuals in Japan with interdependent self-control are slow to anger. The study was conducted using the subset of the Midlife Development in the United States (MIDUS) survey and the corresponding Midlife Development in Japan (MIDJA). However, Akutsu and his colleagues did not find a significant association between anger control and interdependent self-construal.

According to research (Kitayama & Markus 1994; Kitayama et al., 2004), self-construal my effect how a person controls his or her emotions. Emotions are also considered to be dependent on culture. According to Mesquita and Boiger, (2014), culture can influence someone’s emotion to the extent that it serves the central purpose of that culture. For instance, the researchers argue that anger depends on one’s ability to accept the conditions of a situation, willingness to accommodate divergent ideas, as well as ability to accept that other people have more control over other people in a situation. According to Boiger et al., (2013), anger is a controlled emotion that defines one’s limit and in most Western countries like North America, whose culture is characterised by competitive individualism, anger becomes a beneficial value of achieving personal goals.  

 The last significant finding that this study made is that interdependent self-construal does not have a mediating effect on the relationship between anger suppression and the level of depression among the older population of the United States. Therefore, the first hypothesis that anger suppression would be a significant mediator in the relationship between interdependent self-construal and depression in this sample of American middle-aged and older adults was not supported. This result differs from the findings of Cheung & Park, (2010), which concluded that anger suppression was a significant mediator in the relationship between anger suppression and depression among Asian Americans.

While separate studies conducted by Akitsu and colleagues (2016), and Le and Impett, (2013) found that interdependent self-construal has a positive correlation with life satisfaction, personal wellbeing and relationship quality, Akutsu and colleagues (2016) found out that interdependent self-construal has no mediating effect on the relationship between anger and level of depression on the middle-aged American population.

Conclusion

Finally, this study concludes that, first, both anger suppression and interdependent self-construal affect the level of depression on the middle-aged American population. Secondly, the effect of anger suppression on the level of depression on middle-aged American people can be moderated by the level of interdependent self-construal but cannot be mediated by the same variable. However, this study would recommend further studies on the mediation effect of interdependent self-construal on the relationship between anger suppression and depression across all age groups in the United States.

This study relies heavily on the correlation analysis. Therefore, caution should be taken when making conclusions about the causal relationship between anger suppression, interdependent self-construal and level of depression. Moreover, this study focused on the middle age population in the United States. Further research should be conducted that covers all age groups across all cultures to provide a clear picture of the relationship between the three variables.

References

Akutsu, S., Yamaguchi, A., Kim, M., & Oshio, A. (2016). Self-construals, anger regulation, and life satisfaction in the United States and Japan. Frontiers in Psychology, 7(768), 1-12. doi:10.3389/fpsyg.2016.00768

Boiger, M., Mesquita, B., Uchida, Y., & Feldman Barrett, L. (2013). Condoned or condemned: The situational affordance of anger and shame in the United States and Japan. Personality and Social Psychology Bulletin, 39(4), 540-553.

Cheung, R. Y. M., & Park, I. J. K. (2010). Anger suppression, interdependent self-construal, and depression among Asian American and European American college students. Cultural Diversity and Ethnic Minority Psychology, 16(4), 517-525. Doi: 10.1037/a0020655

Cheung, R. Y., & Park, I. J. (2010). Anger suppression, interdependent self-construal, and depression among Asian American and European American college students. Cultural Diversity and Ethnic Minority Psychology, 16(4), 517.

Jackson, J., Kuppens, P., Sheeber, L. B., & Allen, N. B. (2011). Expression of anger in depressed adolescents: The role of the family environment. Journal of abnormal child psychology, 39(3), 463-474.

Kitamura, T., & Hasui, C. (2006). Anger feelings and anger expression as a mediator of the effects of witnessing family violence on anxiety and depression in Japanese adolescents. Journal of Interpersonal Violence, 21(7), 843-855. doi: 10.1177/0886260506288933

Kitayama, S., Snibbe, A. C., Markus, H. R., & Suzuki, T. (2004). Is there any “free” choice? Self and dissonance in two cultures. Psychological Science, 15(8), 527-533.

Le. B. M., & Impett, E. A. (2013). When holding back helps: suppressing negative emotions during sacrifice feels authentic and is beneficial for highly interdependent people. Psychological Science, 24(9), 1809-1815. doi: 10.1177/0956797613475365

Liu, F. F., & Goto, S. H. (2007). Self-construal, mental distress, and family relations: a mediated moderation analysis with Asian American adolescents. Cultural Diversity and Ethnic Minority Psychology, 13(2), 134-142. doi: 10.1037/1099-9809.13.2.134

Markus, H. R., & Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224-253. doi:10.1037/0033-295X.98.2.224

Markus, H. R., & Kitayama, S. (1994). A collective fear of the collective: Implications for selves and theories of selves. Personality and social psychology bulletin, 20(5), 568-579.

Mesquita, B., & Boiger, M. (2014). Emotions in context: A sociodynamic model of emotions. Emotion Review, 6(4), 298-302.

Norasakkunkit, V., & Kalick, S. M. (2002). Culture, ethnicity, and emotional distress measures: the role of self-construal and self-enhancement. Journal of Cross-Cultural Psychology, 33, 56-70. doi:10.1177/0022022102033001004

Park, I. J. K., & Zane, N. W. S. (2004). Ethnic and cultural variations in anger regulation and attachment patterns among Korean American and European American male batterers. Cultural Diversity and Ethnic Minority Psychology, 10(2), 151-168. doi: 10.1037/1099-9809.10.2.151

Park, I. J. K., Kim, P. Y., Cheung, R. Y. M., & Kim, M. (2010). The role of culture, family processes, and anger regulation in Korean American adolescents’ adjustment problems. American Journal of Orthopsychiatry, 80(2), 258-266. doi:10.1111/j.1939-0025.2010.01029.x

Radloff, L.S. (1977). “The CES-D scale: a self-report depression scale for research in the general population.” Applied Psychology Measurement, 1(3), 385-401. Retrieved from https://purl.umn.edu/93227

Ryff, C. D., Seeman, Teresa, & Weinstein, Maxine. Midlife in the United States (MIDUS 2): Biomarker Project, 2004-2009. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2017-11-21. https://doi.org/10.3886/ICPSR29282.v7

Singelis, T. M. (1994). The measurement of independent and interdependent self-construals. Personality and Social Psychology Bulletin, 20, 580-591. Doi: 10.1177/0146167294205014

Spielberger, C. D. (1996). State-trait anger expression inventory: professional manual. Odessa, FL.