For instance, to measure the relative strength of automatic associations between the BII bipolar dimensions of cultural harmony and cultural blendedness with self-related words, participants were asked to complete the following tasks. During one block of trials, participants had to categorize “self” words with words that depict the “Harmony” concept pole of the BII dimension on one side and “other” words with words that represent the “Conflict” concept pole of the BII dimension on the other side. In a second block of trials, “self” words shared the same response as the “Conflict” pole of the BII dimension and “other” words shared the same response option as the “Harmony” pole of the BII dimension. The second IAT followed the same logic in that participants had to discriminate between words that represent the “Blended vs. Distance” dimension of BII, paired with words that differentiate between “self vs. other” concept pair. In total, participants completed seven blocks of trials. Each block included 20 practice trials and 40 test trials. The order of blocks was randomized across participants. For each bicultural participant, the software randomly determined the order in which the seven blocks were completed. This procedure limits the influence of order effects on the obtained results. Explicit measures. After completing the two IATs,square plastic planter participants filled out the battery of acculturation and outcome measures described in Study 1. The implicit measures were administered first. The opposite order might have produced more noise in the implicit data due to fatigue. Please see Study 1 for a description of the explicit measures .
Data screening of test trials showed an overall error rate of 5.6% and a mean response latency of 901 ms for the IAT that assessed the conflict versus harmony dimension of BII. Furthermore, there was an overall error rate of 8.1% and a mean response latency of 919 ms for the IAT that assessed the blendedness versus distance dimension of BII. For the experimental condition, data screening of test trials showed an error rate of 5.5% and a mean response latency of 934 ms for the IAT that assessed the conflict versus harmony dimension of BII. There was an overall error rate of 7.2% and a mean response latency of 954 ms for the IAT that assessed the blendedness versus distance dimension of BII. For the control condition, there was an overall error rate of 5.7% and a mean response latency of 866 ms for the IAT that assessed the conflict versus harmony dimension of BII. Finally, there was an overall error rate of 9.1% and a mean response latency of 881 ms for the IAT that assessed the blendedness versus distance dimension of BII. Taken together, these results suggest that participants had little trouble completing the IATs. Similarly to Study 1, any outliers of extremely fast or slow responses were noted and excluded during data screening. Based on the criterion developed by Greenwald et al. , there were a total of 6 participants eliminated from the study. See Study 1 for how the IAT D effect is calculated. In Study 2, for the Harmony vs. Conflict IAT, a positive score indicated that the Self + Harmony association was stronger than the Self+ Conflict association. A negative score indicated that the Self + Conflict association was stronger than the Self+ Harmony association. Forthe Blendedness vs. Distance IAT, a positive score indicated that the Self + Blendeness association was stronger than the Self+ Distance association. A negative score indicated that the Self + Distance association was stronger than the Self+ Blendedness association.
Two one-sample t-tests were performed in order to examine if the two IAT D means differ from 0 . For the Harmony versus Conflict IAT, results indicated that the sample mean of .396 was significantly greater than 0, t=9.60, p<.001. This result suggested that overall participants implicitly identified more strongly with being a harmonious bicultural relative to being a conflicted bicultural. For the Blendedness versus Distance IAT, results indicated that the sample mean of .270 was significantly greater than 0, t=7.95, p<.001. This result suggested that overall participants implicitly identified strongly with being a blended bicultural compared to being a distant bicultural. Finally, a paired sample t-test was performed to examine the difference between these two groups and results suggested there to be a significant difference between the harmony vs. conflict IAT and blendedness vs. distance IAT, t=2.20, p=.024. This result suggested that Mexican American participants implicitly showed a stronger self-attachment to harmony relative to self + blendedness association. An independent sample t-test was performed to examine if there was a difference in IAT D effect between the experimental and control conditions for both dimensions ofBII. Results indicated that for the harmony versus conflict dimension of BII, there was no significant difference in the IAT D effect between the control condition and experimental conditions, t=.645, p=.520. For the blendedness versus distance dimension of BII, there was no significant difference in the IAT D effect between the control and experimental conditions, t=1.38, p=.171. A second independent sample t-test was performed to examine if there was a difference in the explicit BII scores between the experimental and control conditions for both dimensions of BII. Similarly to the implicit measures, results indicated that for the harmony versus conflict dimension of BII, there was no significant difference in the explicit scores between the control condition and experimental conditions, t=.113, p=.910.
For the blendedness versus distance dimension of BII, there was no significant difference in the explicit scores between the control condition and experimental conditions, t=1.16, p=.248. Taken together, these results suggested that the manipulation did not work. First, correlation analyses were performed to examine the overall pattern among the two IATs . Results indicated that there was no significant relationship between the Harmony vs. Conflict IAT and the Blendedness vs. Distance IAT , r=-.07, p=.46. Next correlational analyses were performed to examine the relation between explicit BII measures and both IATs. Overall, results showed that there was only one significant correlation between the implicit harmony versus conflict BII dimension and explicit BII dimension of blendedness versus distance . This result suggested that the more Mexican American participants explicitly selfidentified as being a blended bicultural, the less they implicit self-identified as being a harmonious bicultural. The rest of the results indicated that there were no other significant relationships among the other three pairings: implicit and explicit conflict versus harmony dimension of BII , implicit and explicit blendedness versus harmony dimension of BII , and the implicit blendedness versus distance BII dimension with the explicit BII dimension of harmony versus conflict . Even though the experimental manipulation did not work, the aforementioned relationships were examined separately for the experimental and control conditions. Reason being is that these correlations will help to explain the discrepancy in the results in the general discussion section. Results indicated that there was no significant relationship between the harmony vs. conflict IAT and the blendedness vs. distance IAT for both the experimental and control groups . Next correlational analyses were performed to examine the relation between explicit BII measures and both IATs. Similarly to the overall pattern of results, for the control group, there was a significant correlation between the implicit harmony versus conflict BII dimension and explicit BII dimension of blendedness versus distance . This result was consistent with the overall correlation; however, the association was stronger. In regards to the experimental group,square plastic plant pot there was no significant correlation between implicit harmony versus conflict BII dimension and explicit BII dimension of blendedness versus distance . The rest of the results indicated that there were no other significant relationships among the other three pairings for both the experimental and control groups . To further explore the validity of the implicit BII measure, additional correlations were run to assess the overall relationship between the two IATs and a battery of explicit acculturation and outcome measures. Interesting enough, there were no significant correlations between the implicit BII measures and the battery of explicit acculturation and outcomes measures with the exception of two correlations. Results indicated a marginal correlation between implicit BII dimension of blendedness versus distance and the explicit acculturation strategy measure of separation and explicit U.S. Identification . These correlations suggested that the more Mexican American participants implicitly self-identified as being a blended bicultural, the less they explicitly endoresd the separation strategy and identified with U.S. culture. The same correlations were also run separately for both the experimental and control groups. Results indicated the same pattern of correlations found in the overall correlation patterns, however differed by either the experimental or control groups . Overall, these correlational analyses indicated that there is little to no relationship between the implicit BII measure and the battery of acculturation and outcome measures including the explicit BII measure. Participants were categorized into two groups based on their explicit BII scores. First, participants received two BII scores for blendedness and harmony. A median split was performed on the two BII scores in order to determine the high versus low groups for each BII dimension. For both BII dimensions, the median score was 4.0. Any participant that scored below the median was categorized as being a conflicted or distant bicultural, whereas any participant that scored at the median or higher was categorized as being a harmonious or blended bicultural.
Based upon these categorization criteria, results from the explicit data indicated that 51% of participants were categorized as high blendedness and 57% were categorized as high harmony, while 49% were categorized as low blendedness and 43% were categorized as low harmony. Participants were then categorized into two groups based on their implicit IAT D scores for comparison purposes. First, participants received two implicit IAT D scores for blendedness and harmony. A median split was performed on the two implicit IAT D scores in order to determine the high versus low groups for each BII dimension. For the harmony vs. conflict dimension of BII, the median score was .459. For the blendedness vs. distance dimension of BII, the median score was .263. Any participant that scored below the median scores of each BII dimension was categorized as being a conflicted or distant bicultural, whereas any participant that scored at the median or higher of each BII dimension was categorized as being a harmonious or blended bicultural. Based upon these categorization criteria, results from the implicit data indicated that 52% of participants were categorized as high blendedness and 51% were categorized as high harmony, while 48% were categorized as low blendedness and 49% were categorized as low harmony. Upon comparing the categorizations based upon the explicit and implicit data, results indicated that the categorizations for high versus low blendedness were roughly identical. However, categorization results based on the explicit and implicit data showed some discrepancy between the high versus low harmony groups. These results indicated that the Mexican American participants tend to assess their blended bicultural identities at both levels of awareness with little variation; however tend to assess their harmonious bicultural identities across two levels of awareness with a little more variation. To further examine the validity of the implicit BII measures, four independent sample tests were performed to examine the overall differences between explicit high and low blendedness and harmony groups on the two IAT measures . Results indicated that there was a significant difference between the explicit high and low blendedness groups on the harmony versus conflict IAT measure, t=-2.23, p=.027. This result suggested that Mexican American participants who explicitly selfidentified as being high in blendedness, implicitly self-identified weaker with harmony compared to those explicitly low in blendedness. Results further indicated that there was no significant difference between the explicit high and low blendedness groups on the blendedness versus distance IAT measure, t=1.31, p=.194. Results for the high versus low explicit harmony groups showed that there was no significant difference between the high and low harmony groups on the harmony versus conflict IAT measure, t=-.526, p=.600. Results further indicated that there was no significant difference between the explicit high and low harmony groups on the blendedness versus distance IAT measure, t=-1.75, p=.083. For the experimental condition, results showed that that there was no significant difference between explicit high and low blendedness and harmony groups on the two IAT measures .