Data Analysis
To examine the impacts of acculturation on the development of eating disorders within the first and second-generation Asian-American population through analysis of acquired data, both the use of inferential and descriptive statistical measures was employed. Upon obtaining 130 survey responses over the course of 10 days, researchers assigned numerical values to each of four subscales of responses to the EDE-Q as well as formulated a global score for the test as per guidelines released by the Centre for Integrative Health (Centre for Integrative Health, 2019). Simultaneously, survey responses concerning the SMAS were scored through the addition of the values of participants’ responses. T- and ANOVA tests were run using those scores to produce the results as described below.
Implications
Independent sample t-tests utilizing the SMAS (acculturation) score as a dependent variable of participants’ generation demonstrate a strong positive correlation between generation and SMAS score (Figure 1). With a p-value of .030, these results are eligible to be deemed significant, indicating that participants’ generation identity can serve as a viable measure of their extent of acculturation as defined by the SMAS. Taking this into consideration, researchers can employ generation as a grouping variable representing acculturation for further data visualization testing.
The expected outcome for this study satisfies the hypothesis that there is a significant trend among generations when measuring the level and rate of acculturation in the United States. The p-value was .030, which indicated a significant correlation of acculturation among generations. However, the prevalence of eating disorders was not as statistically significant as expected when it was measured within each generation -- t-test and regressions run using the values for those respective variables produced a p-value of .159, which exceeded the .05 significance threshold. Thus, the conclusion that acculturation does not have as significant of an impact on eating disorder prevalence as originally thought to be, contradicting researchers’ hypothesis predicting a positive correlation between the extent of acculturation – as defined by both participants’ generation and SMAS score – and prevalence of eating disorders, as measured in accordance with EDE-Q score.
Nonetheless, the production of descriptive visuals, including that of a box and a bar graph (Figure 2) detailing the correlation between generation and EDE-Q score, demonstrates a consistent negative correlation between generation and each of the 4 subscales of the EDE-Q, suggesting an underlying relationship between the two despite values despite formal T- and ANOVA tests deeming those values insignificant. This implies the prospect for the need for further study in the topic to investigate whether or not the significance of such a relationship fluctuates when different parameters -- such as the usage of larger sample size or consideration of a third or fourth generation -- are taken into account.
Although the relationships observed in surveyed data are statistically insignificant, they nonetheless produce a trend consistent with that described in previous studies concerning populations of different racial backgrounds, including that which took into account the extent to which acculturation impacted Mexican Americans’ tendencies towards eating disorder-associated behaviors (Cachelin et. al. 2006). Those studies, as our plots vaguely suggest, demonstrate a positive relationship between values of acculturation and eating disorder risk, enforcing the hypothesis that assimilation into American culture tends to produce detrimental effects when concerning acculturated populations’ likelihood of acquiring eating disorders. Our data contrasts those claims as applied to an Asian American population, as the plots generally display a lessened EDE-Q value as in correspondence with participants’ generation (which has proved in Fig. 1 to be a reliable measure of acculturation). The difference between Asian Americans’ results in this study as compared to that of Mexican Americans in Cachelin et. al.’s previous work introduces the predicament that consideration of native cultures’ tendencies towards unhealthy eating habits and body image influences the nature of which assimilation into American culture affects later generations’ tendencies towards those same habits. Because those identifying with Asian, particularly East Asian, cultures such as those in Japan, Hong Kong, Singapore, Taiwan, and South Korea tend to globally be among the most at risk for eating disorder contraction (Kathleen 2015), integration into American society reduces their risk relative to their native cultures. On the contrary, demographics who originate from cultures without a significant susceptibility to eating disorders tend to demonstrate a higher risk for eating disorders after acculturation, as evidenced in Cachelin et. al. 2006.
To further investigate other demographic variables that might have affected the outcome of statistical tests, researchers divided participants in accordance with their ethnicity, a variable acquired from demographic questions in the survey. The two largest populations, Indian (43.9%) and Chinese (33.3%) Americans, demonstrated a fluctuation in their results.
As exhibited in nearly all 4 of the EDE-Q subscale results of the two respective populations, Indian American participants tended to produce a slightly more statistically significant (in congruence with p-value) correlation between EDE-Q score as affected by the extent of their acculturation into American culture as compared to the Chinese population. This suggests that ethnicity may have played a role in scattering the spread of data and thus contributed to the lack of overall statistical significance in initially run general t-tests. Such phenomena imply the need for future studies investigating ethnicity-specific groups in relation to the topic at hand to explore the different nuances between Asian American populations, which should not be regarded as a single monolith.