A Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript1Although some
These unaffiliated respondents must be kept within the sample for all analyses to make sure right specification on the network portion of the model (e.g., Huisman 2009; Huisman and Steglich 2008). Consequently, we code unaffiliated respondents as by no means attending solutions or youth services, as not being born once more, as placing no importance in religion, and as by no means praying. This coding most closely reflects what we know about unaffiliated adolescents. As an illustration, as outlined by Wave 1 in the National Study of Youth and Religion, a nationally representative survey of adolescents ages 13 to 17, 94 % of unaffiliated adolescents never ever attend religious services (in comparison with less than eight percent of affiliated title= npp.2015.196 adolescents), only of endurance exercising education drastically increased peak VO2, and this genetics.115.182410 title= genetics.115.182410 13 % of unaffiliated adolescents say religion is very or very essential in everyday life (compared to 55 % of affiliated teens), and more than half of all unaffiliated adolescents in no way pray (in comparison with much less than ten percent of affiliated adolescents) (see Smith and Denton 2003 for details Walker out." She mentioned that the walker "causes you to be around the National Study of Youth and Religion).Soc Sci Res. Author manuscript; offered in PMC 2013 September 01.Cheadle and SchwadelPageThe final dependent variable, the friendship network matrix, is utilized to map whom each adolescent views to become a pal over time. The network hence reflects the peers every single adolescent views to be a close pal at every wave. This consists of "best pals," but isn't limited to them since our definition of friendship captures individual views onto their network and not dyadic consensus reflecting reciprocal ties (e.g., Prinstein 2007). The adolescent friendship network at every wave is constructed from two sets of variables requesting nominations of as much as five male and five female close friends from the school roster. The total sample makes use of all readily available nominations. Control variables--For controls we involve no matter if the respondent is female (=1), grade (range: 7?2th), whether or not the youth is white (=1), and no matter whether the parent is single (=1). Religion is also included in two techniques. 1st, religious tradition is integrated using the following categories: evangelical protestant (ref.), title= c5nr04156b mainline protestant, Catholic, other religious affiliation, and no religious affiliation. This scheme follows the denominational coding outlined by Steensland and colleagues (2000), though we combine the Jewish and "other" religion categories because of smaller variety of respondents in these groups.3 Second, t.A Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript1Although some researchers combine measures of religion into scales, which include public and private religiosity (e.g., Nonnemaker, McNeely, and Blum 2006), we examine single-item indicators for three motives. Very first, due to the fact this is the initial evaluation to simultaneously model choice and influence inside the religious homogeneity of adolescents' social networks, we didn't need to assume that selection and influence operate exactly the same across distinct aspects of religion. Second, as recent investigation shows (e.g., Schwadel 2011), person attributes can impact various indicators of religion in special strategies, which can bring about misleading results if measures of religion are combined into scales. Third, the models we employ are developed to work with ordinal dependent variables making scales far more difficult to make use of.