# Answered the questionnaire in Study 1. To avoid analyzing the identical students

This statistical analysis was performed with SPSS 22.0.(5 items) didn't significantly differ amongst groups (Table 3).Outcome from analysis 3: correlations between the subscales in the AMI-MeT and SCSResultsResults from analysis 1: multi-group structural equation modelingDescriptive statistics and correlation matrixes on the Al group was higher among those using a university degree (see AMIMeT items are reported in detail in Further file 1. Soon after testing the configural invariance model, we tested the other invariance models in line with a number of sets of constraints. The chi-square test (2 df) along with the chi-square distinction test (2) showed that the weak invariance model was the most effective match for the information (Table two), even though the items did not follow a regular distribution (Additional file 1). The fit indices also indicated that the weak invariance model was the very best match for the data (Table 2). Accordingly, we regarded the weak invariance model as the very best model, and have displayed the standardized estimations from the model in Fig. 1. In the weak invariance model, the obtained for AMO and AMS ranged from .6775 and .7279, respectively, indicating higher internal consistency. The results with the SEM along with the high internal consistency reliability coefficients indicate that the two-factor model from the AMI-MeT was trustworthy and that the aspect loadings from the AMO and AMS had been prevalent across the three groups.Result from analysis 2: one-way evaluation of variance of AMI-MeT scores amongst the three groupsBefore examining the correlations, we calculated the Cronbach's alphas on the subscales of each the AMIMeT as well as the SCS, as follows: AMS and AMO, = .79 and .75, respectively; IndSC and InterSC, = .69 and .74, respectively. The correlation evaluation revealed that the AMS scores had been far more associated using the IndSC scores (r =.Answered the questionnaire in Study 1. To prevent analyzing the same students twice, we created sure to treat them as separate groups while handing out the questionnaire. Of the 518 students given the questionnaire, 335 filled it in entirely (200 females and 135 men; mean age = 19.6; SD = two.9); we excluded any answer sheets that contained missing data. Motivation for health-related treatment was measured working with the AMI-MeT. To validate the dualistic ideas from the AMI-MeT subscales, we administered the Self-Construal Scale (SCS) [40], that is based around the notion of selfconstrual [23]. The SCS was translated into Japanese by Takahashi et al. [41]. Working with a 7-point Likert-type scale (1 = strongly disagree to 7 = strongly agree), participants indicated their amount of agreement with 12 things assessing IndSC (e.g., "I favor to be direct and forthright when coping with people I've just met," "I delight in becoming exceptional and unique from other folks in a lot of respects") and 12 statements assessing InterSC (e.g., "It is important for me to keep harmony within my group," "It is significant to me to respect choices created by the group").Analysis three: correlations amongst the subscales of your AMI-MeT and SCSTo demonstrate the convergent and divergent validity from the subscales of your AMI-MeT together with the dualisticHatta et al. BMC Healthcare Informatics and Choice Generating (2016) 16:Page five ofconcepts with the self, we calculated the Pearson's correlation coefficients among the AMI-MeT subscales and SCS subscales.