Both Islam and Christianity Invite to Tolerance

The researchers considered Catholic and Protestant adolescents as Christians and excluded the data of other groups of Christian people such as Orthodox and Evangelical Free Church. On the contrary, they considered Muslims as a unified group and they did not exclude any adolescents affiliated to different groups of Muslims. It is important to highlight that in all religions, various subgroups are different in terms of violent behavior (VB) and intergroup conflicts. Moreover, the question of how religions can provoke such oppositions, and whether they do, remains undetermined (Neuberg et al., 2014). Therefore, the artificial grouping of Muslims into one (presumably) homogeneous group does not allow for the differentiation between extremist and mainstream Muslim groups. Although members of both Muslim groups would designate themselves as “Muslims,” it is plainly obvious that members of extremist groups are by definition more likely to be involved in violence.

In the introduction, the author mentions that most Muslims in Germany are immigrants. In addition, based on the population sample, it is assumed that the various Christian groups consisted largely of ethnically native population, that is, Germans, whereas the Muslim group presumably consisted of immigrant groups. Let us, for a moment, borrow a methodological principle from epidemiology, where migrant people are not considered good representatives in surveys because they bear characteristics similar to their country of origin as well as their country of adoption (Chaturvedi & McKeigue, 1994; Haenszel & Kurihara, 1968). Assuming that the same principle applies, by virtue of the fact that Muslims constitute a religious minority in Germany, they should not be considered a representative sample (Chaturvedi & McKeigue, 1994). The author himself acknowledges the anomic situation of Muslims in the last part of his article.

The participants were asked to state which religious affiliation they belonged to. If a participant answered “no religion,” she or he was excluded from the project, and if a participant did not respond to the question, she or he would be classified according to her or his parents’ religion. The frequency of such missing data is not mentioned, even though, in our view, it is a critical point, as nowadays, many people practice religions other than those of their parents’ or they do not practice any religion, even that of their parents. By making the decision to classify a participant according to her or his parents’ religion, the author seems to be implicitly assuming that adherence to a religion is an acculturation and upbringing issue, rather than a personal choice.

More than 50 valid international instruments for the assessment of violence-related behaviors among youths are available (Dahlberg, Toal, Swahn, & Behrens, 2005). Some of them are lengthy general questionnaires containing a number of items evaluating different aspects of VB, whereas others are short, specific questionnaires containing a few items aimed at evaluating a particular aspect of VB. The present research uses a binary coded questionnaire with only 3 items in which a respondent reports her or his history of minor injury or/and severe injury or/and robbery in the past 12 months. In comparison, the Modified Aggression Scale, devised by Orpinas in 1993, is one of the best instruments in this field, consisting of 22 items for four subscales of Fighting, Bullying, Anger, and Cooperative/Caring Behavior (Dahlberg et al., 2005). Baier’s questionnaire rather resembles the Physical Fighting Scale, devised by the Division of Adolescent and School Health of Centers for Disease Control and Prevention (CDC) in 1993, which is a specific instrument consisting of 4 items for measuring the frequency of physical fighting and injuries from fights within the past year (Dahlberg et al., 2005). Accordingly, Baier’s questionnaire does not seem adequate for the thorough study of VB.

More than 10 valid international instruments assessing religiosity are available (Zwingmann, Klein, & Büssing, 2011). As for the previous item, both lengthy, general questionnaires as well as short, specific questionnaires are in use. Baier used a Likert-type scale questionnaire with only 3 items, in which a respondent reported her or his history of praying and visiting a place of worship in the past 12 months along with rating the extent of importance of religion in her or his personal everyday life. In contrast, the System of Belief Inventory, devised by Holland in 1998, is one of the best instruments in this field, consisting of 35 items for four subscales of Deriving Meaning From an Existential Perspective, Frequency of Religious Activities, Relationship to a Higher Power, and Social Support Derived From Religious Community (Cotton, McGrady, & Rosenthal, 2010). Importantly, when an informant is asked to grade religiosity, she or he should precisely consider what religiosity may mean to her or him (Büssing, Pilchowska, Baumann, & Surzykiewicz, 2014). In fact, the rituals that the researcher focused on make up only a small part of religion and, to evaluate religiosity, other important variables should also be studied. As a result, Baier’s questionnaire does not seem to cover all the necessary aspects of religiosity adequately.

Honesty is one of the criteria of religiosity that cannot be assessed by Baier’s instrument. As a result, honest people are more likely to report past misbehavior than liars. Therefore, because honesty correlates with religiosity, according to Baier’s questionnaire of VB, more religious people can score worse than less honest people.

The author does not mention the level of significance in the method part of the article, while he considers this level as .05 in his analytic tables. However, the author considers .06 as significant in one of the most important analyses (Table 4), namely, that of concluding that religiosity in Muslims has a positive effect on VB.

The author controls the alcohol consumption variable by using a multivariate model (Table 5) and in the next step shows that Muslim religiosity increases VB. Likewise, in the conclusion, he states that the Muslim faith absolutely prohibits alcohol and that Muslims consume less alcohol than Christians. In terms of Muslim religiosity, absolute abstinence from alcohol, not moderation in drinking as is the case with Christians, should be considered in analysis.

Figure 1 shows the proportion of adolescents displaying VB, by religiosity, alcohol consumption, and religious group. We know that a Christian might infrequently consume alcohol and still considered as low, medium, and high religiosity. On the contrary, a Muslim cannot consume even a drop of alcohol and be considered as belonging to the high-religiosity group. Consequently, the author’s statement in page 119 “The effect is most pronounced for the infrequent alcohol consumer group: Whereas among low-religiosity Muslim adolescents 27.6% displayed VB at least once, the equivalent figure for high-religiosity Muslim adolescents is 47.6%” is not correct. In other words, there is a discrepancy in the division of low, medium, and high religiosity for Muslims. More importantly, Figure 1 has another important implication. It shows that in all subgroups of Christians and Muslims, those who consume alcohol, even infrequently, are involved in more violence than those who do not consume alcohol.