The Journal of School Nursing
© The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1059840520940379
journals.sagepub.com/home/jsn
2022, Vol. 38(4) 358–367
Early alcohol use places youth at risk for adverse health, academic, and legal consequences. We examined the content of the total array of self-cognitions in urban youth to determine whether specific self-concept profiles were associated with early drinking, drinking-related self-cognitions, and conduct problems. We conducted a secondary analysis of data from a cross-sectional study with 9- to 12-year-old predominantly Black and Hispanic youth (N = 79) who attended urban school and summer youth programs. Measures included an open-ended self-description task and questionnaires to measure presence/absence of a drinking-related self-cognition, alcohol use, and conduct problems. We content analyzed 677 self-descriptors; cluster analysis revealed six unique self-concept profile groups. In a cluster group distinguished by negative self-content, 37% drank alcohol and 42% had a drinking-related self-cognition. Youth in this group also had conduct problems. School nurses are in prime positions to identify and intervene with youth who have at-risk self-concept profiles.
Keywords
self-concept, youth, alcohol use, drinker possible self, conduct disorder, maladaptive behaviors
Early alcohol use is associated with immediate and future health, academic, and legal consequences as well as delinquent behavior (Jackson et al., 2015; Komro et al., 2010; Zernicke et al., 2010). Moreover, drinking onset prior to the age of 14 is a powerful predictor for the later development of alcohol problems (Colder et al., 2018) and alcohol dependence (Spear, 2015). Population-based data show that by eighth grade, one in four youth have begun drinking and 1 in 10 report getting drunk (Miech et al., 2019). Although many studies have shown that non-Hispanic White youth begin drinking earlier than Black and Hispanic youth (Alvanzo et al., 2011), early alcohol use in Black and Hispanic youth is also problematic. For example, a longitudinal study to examine initiation and progression of alcohol use in a large sample of low-income Black and Hispanic youth in Grades 5–8 in Chicago showed that those who used alcohol were more likely than those who did not to engage in delinquent behaviors in eighth grade (Komro et al., 2010). These results suggest that any use of alcohol in early adolescence is associated with other high-risk behaviors. This underscores the importance of identifying modifiable factors that lead to early initiation and progression of alcohol use among low-income Black and Hispanic youth as a key factor to prevent not only alcohol problems but also future delinquent behavior.
Self-cognitions may be important modifiable factors that lead to early initiation and progression of alcohol use. Self-cognitions are mental representations about the self in specific content domains. Studies have shown that having a self-cognition related to drinking predicts alcohol use and/or alcohol problems in youth (Corte & Szalacha, 2010), adolescents (Lee, Corte, Stein, Finnegan, et al., 2015), college students (Lee, Corte, & Stein 2018; Lindgren et al., 2016; Ramirez et al., 2017), and young adults (Corte & Stein, 2007). Studies have tended to focus on the influence of a single domain-specific self-cognition, for example, drinker (Lee, Corte, Stein, Finnegan, et al., 2015), popular (Stein et al., 1998), and physically active (Robbins et al., 2004). Each individual has a unique collection of domain-specific self-cognitions that function together as a system—this collection is referred to as a self-concept profile. To our knowledge, no studies to date have examined associations between self-concept profiles and risk behavior. In this study, we examine self-concept profiles of youth and associations between these profiles and a drinking-related self-cognition (a known risk factor for alcohol problems) and conduct problems.
We use Markus’s model of the self-concept (Markus & Nurius, 1986; Markus & Wurf, 1987) as the theoretical framework guiding our study. According to this model, the self-concept is comprised of mental representations about the self called self-schemas. Self-schemas are functional cognitive structures that serve as lenses through which social information is interpreted (Markus, 1977). Studies have shown that people pay more attention to and process information more quickly in domains that are self-relevant, that is, domains in which they have a self-schema. For example, using a response latency task, Stein and Corte (2007) found that women who had a fat self-schema processed “fat†related words (such as chubby, heavy, overweight) more quickly than women who did not have a fat self-schema (Stein & Corte, 2007). Moreover, Lee, Stein, and Corte (2018) found that a drinker self-schema not only facilitates the processing of drinking-related stimuli but also smoking-related stimuli. In addition to facilitating information processing, self-schemas include behavioral strategies and routines that facilitate behavior in the domain. For example, in a sample of youth, Robbins et al., (2004) found that having a physically active self-schema was positively associated with the level of physical activity. Self-schemas have also been shown to predict behavior longitudinally. For example, in young women, a fat self-schema predicted disordered eating behavior 1 year later (Stein & Corte, 2008) and having a drinker self-schema predicted increases in alcohol consumption and alcohol problems over 2 years (Lindgren et al., 2016).
The self-concept also includes future-oriented cognitions about the self called possible selves—the selves one hopes to become, expects to become, and fears becoming (Oyserman et al., 2004). In addition to vivid images of the self in the future state (e.g., popular, physically fit), possible selves include strategies to achieve (or avoid) the possible self (Oyserman et al., 2004). Although not as well formed as self-schemas, studies have shown that possible selves powerfully predict behavior. For example, in youth, having a possible self related to academics has been shown to predict grade-point average and the number of hours spent studying (Oyserman et al., 1995); having an expected possible self as a drinker has been shown to predict alcohol problems 1 year later (Lee, Corte, Stein, Finnegan, et al., 2015).
Self-cognitions develop as a function of interaction with the social environment in domains considered important to the individual (Markus & Wurf, 1987). Development of self-cognitions is influenced by the child’s past experiences, the available opportunities in their social context, and support from parents and other key adults (Oyserman & Fryberg, 2006; Zhu et al., 2014). Given individual differences in experiences, values, attitudes, and interests, each person has a unique collection of self-cognitions, that is, a unique self-concept profile. While many studies have shown that the content of individual self-cognitions predicts behavior in the related domain (Corte & Szalacha, 2010; Kendzierski & Whitaker, 1997; Markus, 1977; Robbins et al., 2004; Stein & Corte, 2008), very few (Lee, Corte, Stein, Park, et al., 2015) have examined the content of the total array of self-cognitions, and none have determined whether specific self-concept profiles are associated with maladaptive behavior in youth.
In this study, we addressed the following two research questions: (1) Are there subgroups of youth characterized by unique self-concept profiles? (2) Are self-concept profiles differentially associated with alcohol use, conduct problems, and presence of a drinker possible self (a known risk factor for alcohol use in youth)?
This was a secondary analysis of data from a cross-sectional mixed-methods study to examine the influence of parental alcohol problems on the self-concept and alcohol use in youth (Corte & Szalacha, 2010). The study was approved by the Institutional Review Board (IRB) at the University of Illinois at Chicago. Youth age 9- to 12-year-old (n = 79) were recruited from urban schools and two summer youth programs located in a lower income area of a metropolitan Midwest city. Data were collected in individual sessions.
Content of self-cognitions. An open-ended self-description task and methodology developed by Markus (1977) was used to reveal the total array of youth’s self-cognitions. Participants were asked to describe themselves as completely as possible use a set of blank index cards. More specifically, they were asked to write one thing on each card to describe themselves and that they could include anything that was important to them. Youth were also told that these things may tell how they are like other people or how they are different from them and that they could include things they like about themselves and things they don’t like about themselves. Finally, youth were told that they could use as many or as few cards as necessary to completely describe themselves. Open-ended self-description tasks have been used in many previous studies (Chung & Pennebaker, 2008; Stein et al., 2013; Stein & Corte, 2007, 2008), including with adolescent samples (Montemayor & Eisen, 1977; Stein et al., 1998).
Drinker possible self. A single item from a widely used closed-ended Possible Selves Questionnaire (Stein et al., 1998) was used to measure the presence/absence of a drinker possible self. This questionnaire is comprised of 31 questions addressing the likelihood of specific self-descriptions in the Next Year, for example, likelihood of being a good student NEXT YEAR, being popular NEXT YEAR, using drugs NEXT YEAR, with responses on a 5-point Likert-type scale ranging from 0 = not at all to 4 = very much. An affirmative response (1 = a little to 4 = very much) in response to the drinker possible self item, “drink too much alcohol is likely to describe me NEXT YEAR (0 = not at all, 4 = very much),†suggests cognitions about a “drinker possible self,†whereas a negative response (0 = not at all) suggests no cognitions about a drinker possible self. Validity of this single item regarding the drinker possible self in youth has been shown (Corte & Szalacha, 2010; Lee, Corte, Stein, Finnegan, et al., 2015). Construct validity has been demonstrated in a longitudinal study of youth that showed that availability of a drinker possible self in the eighth grade predicted alcohol misuse in the ninth grade, even controlling for eighth-grade alcohol misuse (Lee, Corte, Stein, Finnegan, et al., 2015). One-year test–retest reliability was Spearman’s Ï = .39 (Lee, 2013). In addition, likelihood ratings have been successfully used to measure possible selves in a variety of domains, for example, popularity, conventionality, deviancy (Stein et al., 1998), and academics (Kemmelmeier & Oyserman, 2001).
Alcohol use. A single item, “Have you ever had a drink of alcohol (even a few sips)?†on an Activities Questionnaire was used to determine whether youth had begun to drink alcohol. Single questions about ever using alcohol are widely used to determine drinking onset in youth and has been shown to be reliable and valid (National Institute on Alcohol Abuse and Alcoholism, 2011). Responses were coded yes/no (0 = no, 1 = yes).
Conduct problems. Conduct problems were measured with the Antisocial Behavior Checklist for Youth (ASB) developed by Zucker and Fitzgerald (1996). This 63-item questionnaire measures youth’s participation in a variety of aggressive and antisocial activities. We omitted 13 items that were age-inappropriate for our sample (e.g., lied to your spouse, changed jobs more than 3 times in 1 year). The items on the ASB Questionnaire include overt forms of aggression such as “cursed at your parents to their face†and more covert forms of aggression like “lying to your teacher,†which may better reflect conduct problems in girls. Responses were coded yes/no (0 = no, 1 = yes). The score is a sum of the yes responses. Expected negative correlations with positive behavioral conduct were found in adolescent participants of the longitudinal family study of risk for alcoholism both in early adolescence, r(264) = –.31, and three years later in mid-adolescence, r(264) = –.42 providing evidence of construct validity (Zucker et al., 2000). In the same sample of adolescents, 3-year stability of measure was r(264) = .62. Cronbach’s a in this sample was = .89.
Demographics. A demographic questionnaire was used to measure age in years, gender (boy, girl), and race/ethnicity (White, Black, Hispanic, Asian, Other).
After obtaining approval from the IRB at the University of Illinois at Chicago, parental permission, and participant assent, youth completed the open-ended self-descriptor measure first to avoid priming them regarding content related to the self. Youth then completed the Possible Selves Questionnaire (drinker possible self), Activities Questionnaire (alcohol use), ASB (conduct problems), and demographic questions. Youth were compensated for their time with a US$10 gift card.
Content analysis of self-descriptors. We qualitatively content coded the spontaneously generated self-descriptors. This approach was chosen rather than using a pre-existing coding scheme in order to avoid introducing bias or placing constraints around the data. Authors read and reread the self-descriptors and identified codes to reflect the content of the self-descriptors. Disagreement between the authors was resolved through discussion. Codes were then entered into a codebook.
Cluster analysis of self-descriptors. We conducted hierarchical agglomerative cluster analysis of the coded self-descriptors in order to identify subgroups of youth who had similar self-concept profiles. Hierarchical agglomerative cluster analysis maximizes the differences between the groups and minimizes the differences within the groups (Eshghi et al., 2011). We chose Minkowski methods and nearest neighbor (Bratchell, 1989; Fraley & Raftery, 1998; Gower, 1967; Milligan & Cooper, 1987) to ensure that coded self-descriptors would be grouped by similarity, not by an average or centroid used in other probabilistic cluster analysis methods such as Latent Class Analysis (Hsu et al., 2007). Minkowski methods (Euclidean) and nearest neighbor were also useful in exploring the organization of patterns when the number of clusters and structure was unknown (Halkidi et al., 2002) and helped maintain the natural structure of how items cluster (Fraley & Rafter, 1998; Milligan & Cooper, 1987). The authors examined the x-axis for naturally occurring breaks and the cophenetic coefficients on the y-axis for branches in the dendrogram to identify clusters of similarity (Beckstead, 2002).
Demographic information, outcome variables, and cluster differences. Descriptive statistics were performed on demographic information and outcome variables for each cluster group. Chi-square and analysis of variance (ANOVA) were used to compare the clusters.
The sample consisted of 79 youth who were on average 10.7 (SD 1.1) years old. As shown in Table 1, slightly more than half of the sample were Black and nearly 40% were (Hispanic/Latino). Slightly more than half the sample were girls.
Across all 79 youth, a total of 667 self-descriptors were generated. Using content analysis, 27 codes were derived from the self-descriptors. Codes appear in the left column and examples of the codes appear in the right column (see Table 2).
The most common coded self-descriptor was likes with 67.1% of the sample (n = 53) describing themselves in terms of things they like. The next most common coded self-descriptor was think/know with 45.6% of the sample (n = 36) describing themselves in terms of being smart, curious, and so on. The next most common coded self-descriptor was Positive Personal Qualities with 35.5% of the sample (n = 28) describing themselves in terms of being kind, thoughtful, and so on. Nearly 32% of the sample (n = 25) described themselves in terms of their family, for example, brother/sister, and 25.3% (n = 20) described themselves in terms of personality, for example, cool, funny, and so on. Nearly 22% (n = 17) of the youth described themselves in terms of personal characteristics (e.g., height, weight, gender), positive self-views (e.g., like myself), and being friendly (e.g., lots of friends). Just over 20% of the sample (n = 16) described themselves in terms of sports (e.g., “I play football,†most valuable player, cheerleader). Nineteen percentage of the sample described themselves in terms of social acceptance (e.g., being good, popular), dislikes (e.g., “I don’t like to do homework,†“don’t like people who are racialâ€), and ways they were distinguished from others (e.g., different, tomboy). The next most common coded self-descriptors were play (17.7%, n = 13), physical appearance (16.7%, n = 13), level of activity (15.2%, n = 12), behavior not engaged in (13.9%, n = 11), indicators of SES (13.9%, n = 11), negative dimensions of self (12.7%, n = 10), favorite things (10.1%, n = 7), and passive observation such as video games or TV (10.1%, n = 7). Less than 10% of the sample described themselves in terms of having internal fortitude (6.3%, n = 5), being future focused (5.1%, n = 4), a state of equanimity (5.1%, n = 4), positive affect (3.8%, n = 3), having cultural pride (3.8%, n = 3), and religion (2.5%, n = 2). Figure 1 shows a bar graph that depicts the percentage of youth who generated at least one self-descriptor in each of the content domains.
We found six distinct clusters (see Figure 2). Cluster 1 included 10 youth who all described themselves primarily according to their likes. Cluster 2 included 10 youth who described themselves primarily by their positive self-views (e.g., like myself, love myself) as well as their likes. Cluster 3 included 10 youth who primarily generated descriptors identifying personal characteristics (e.g., height, weight, name) and personality (e.g., funny, crazy, silly). Cluster 4 included 18 youth who described themselves by their personality traits and by how they were distinguished from others (e.g., different). This was the only cluster in which almost no youth generated any self-descriptors related to their likes. Cluster 5 included 12 youth who described themselves according to their likes, favorites (e.g., color, hobby), being friendly, and behavior they did not engage in (e.g., don’t drink, don’t do drugs, don’t eat a lot of chocolate). Cluster 6 included 19 youth who described themselves by their likes, being friendly, personality traits, having negative dimensions about themselves (e.g., mean, bad, get in trouble), social acceptance, positive self-views, indicators of socioeconomic status (e.g., wealthy, not rich like others, wish I had more money), and personal characteristics such as height/weight.
The demographic composition of the youth in each of the cluster groups is shown in Table 3. There were no significant differences between the cluster groups in age, F(5, 73) = 1.9, p = .10; sex (χ2 = 3.4, p = .64); or race/ethnicity (χ2 = 8.5, p = .13).
Although there were no significant differences between cluster groups in the number of youth who had a drinker possible self (χ2 = 9.8, p = .08) or the number of youth who reported ever drinking (χ2 = 8.7, p = .12), over half (n = 8) of the 15 youth who had a drinker possible self were in Cluster 6, diversified with negative dimensions. The remainder of youth who had drinker possible self were distributed across the other clusters. Another 15 youth reported ever drinking. Nearly half of these youth (n = 7) were also in Cluster 6, diversified with negative dimensions. The other ever drinkers were dispersed across the other five clusters. The overall ANOVA was significant for differences in conduct problems, F(5, 73) = 3.4, p = .008. Post hoc pairwise comparisons using least significant difference showed that youth in Cluster 6, diversified with negative dimensions, reported significantly more conduct problems than youth in Cluster 1, predominantly likes (mean difference = –5.84, p = .02); Cluster 2, likes and positive self-views (mean difference = –5.74, p = .03); and Cluster 5, prosocial with restraint (mean difference = –5.15, p = .04).
The purpose of this study was to identify subgroups of youth who were characterized by similar within-group self-concept profiles and to determine whether subgroup-specific self-concept profiles were differentially associated with maladaptive outcomes, for example, presence of a “drinker†possible self, alcohol use, and conduct problems. Youth generated a wide variety of self-descriptive content. We found six distinct subgroups, each characterized by a unique self-concept profile. The subgroups differed according to their level of risk as indicated by associations with drinker possible self (a known predictor of alcohol problems), ever drinking, and conduct problems.
Although the cluster sizes were small, it appears that the most at-risk group was Cluster 6. Approximately one half of youth in this cluster group had a drinker possible self and reported ever drinking. Youth in Cluster 6 also had the highest level of conduct problems. While the self-concept profiles of youth in this cluster were the most diversified in terms of content, they were the only group who generated negative self-descriptors. Other studies have similarly shown that negativity in the self is associated with alcohol use (Corte & Zucker, 2008) and conduct problems (Moretti et al., 2001). Because our study is cross-sectional, it is not clear whether negative dimensions of the self contributed to or were a consequence of drinking and conduct problems. Yet there is some evidence from longitudinal studies that negativity in the self antedates drinking. In a study of rural youth, Long and Boik (1993) found that having negativity in the self in third and fourth grade predicted greater likelihood of drinking alcohol in sixth and seventh grade. In a longitudinal study, Corte and Zucker (2008) found that among those who reported drinking by mid-adolescence (ages 15–17), having many negative self-cognitions in early adolescence (ages 12–14) distinguished earlier drinking onset from later age of onset. This held true even when controlling for the number of positive self-cognitions, conduct problems, and parental alcoholism. Finally, in terms of conduct problems, there is evidence that negativity in the self of youth is associated with conduct problems (Barry et al., 2003; Muratori et al., 2018) and prospectively predicts conduct problems (Gerard & Buehler, 2004).
Two of the cluster groups had an intermediate level of risk (Clusters 3 and 4) based on conduct problems and ever drinking. Conduct problems scores were 2–2.5 times higher in Clusters 3 and 4 than in Clusters 1, 2, and 5. Moreover, 20% of those who reported ever drinking were in Cluster 3. As such, this group bears monitoring for escalation of alcohol use and conduct problems. Taken together, there were three “at-risk†groups (Clusters 3, 4, and 6) and three “lowrisk†groups (Clusters 1, 2, and 5).
One factor that seemed to distinguish the three at-risk cluster groups (Clusters 3, 4, and 6) from the three low-risk cluster groups (Clusters 1, 2, and 5) was self-descriptors related to personality traits. While youth in all three of the at-risk groups described themselves in part by personality traits, none of the youth in the low-risk groups described themselves in terms of personality traits. It is interesting to note that the personality trait self-descriptors were primarily extroversion traits (e.g., loud, funny, talkative, outgoing). Although we do not have any evidence that personality traits caused maladaptive behavior in this study, there is considerable evidence that personality traits related to extroversion predict alcohol use in youth (Adan et al., 2017; Dick et al., 2013), college students (Martsch & Miller, 1997), and adults (Adan et al., 2017; Hakulinen et al., 2015). There is also evidence that extroversion traits are associated with conduct problems (Khan et al., 2005; Lopez-Romero et al., 2015).
Although we were examining self-concept profiles in relation to risk behaviors, we could not rule out that youth in the at-risk groups were simply older and thus more likely to drink alcohol. Because of small cell sizes, we combined the three at-risk groups (Clusters 3, 4, and 6) and the three low-risk groups (Clusters 1, 2, and 5) to determine whether age may be a distinguishing factor. Youth in the at-risk group were significantly older (11.0 years) than youth in the low-risk groups (10.4 years, t = 2.4, p = .02). Based on our data, however, it is not possible to know whether onset of drinking and conduct problems were a function of age, the content of the self-descriptors, or both.
Finally, the most common self-descriptor generated was “likes†with 70% or more of the youth in five of the six cluster groups describing themselves in terms of “likes.†In a study of 262 children and adolescents, Montemayor and Eisen (1977) found a curvilinear relationship between age and likes such that the use of “likes†to describe the self generally increased from ages 10 to 14, and then subsequently decreased. Given that our sample was comprised of 9- to 12-year-old youth, generating self-descriptors related to likes appears to be a developmental phenomenon. Likes did not, however, distinguish the “at-risk†from the low-risk cluster groups.
Our findings have important implications for school nurses. Based on our findings, youth who have negative self-beliefs and personality traits related to extroversion may be at risk for early drinking onset and conduct problems. School nurses are in an ideal position to assess children’s beliefs about themselves. Assessing children’s beliefs about themselves can be as simple as asking the child to describe themselves (either verbally or on paper), for example, things they like about themselves, things they don’t like about themselves, how they are like other kids or different from them, and so on. Alternatively, a closed-ended self-description questionnaire could be used where children simply check those characteristics that describe themselves. This would enable the school nurse to identify those youth who hold negative self-beliefs. And while personality traits related to extroversion are not in and of themselves problematic, given their associations with alcohol use and conduct problems, it may be useful for screening purposes. Considering the powerful role of self-cognitions in motivating behavior in youth (Lee, Corte, Stein, Park, et al., 2015) together with the fact that the self is malleable in youth (Harter, 1990; Oyserman & Markus, 1990; Oyserman et al., 2002), assessing the content of the self as part of a routine assessment may be a valuable way to identify youth at risk.
Rather than a checklist with predefined categories, we used an open-ended task to determine the content of self-cognitions. This approach along with allowing domains/categories emerge from the data to describe self-profiles, give youth voice. Moreover, we used a person-centered approach to analyze the data to identify clusters of youth distinguished by their self-concept profiles versus a variable-centered approach in which the average effect of each self-content domain on outcomes was examined.
Limitations of this study include cross-sectional design, small sample size, and self-report of alcohol use and conduct problems. The cross-sectional design prevents us from drawing any conclusions about causal direction. The small sample size precluded use of latent class analysis and may have prevented us from detecting some significant findings. Moreover, it is not clear whether a larger sample would have yielded the same subgroups. As such, the self-concept profile cluster groups identified in this study must be considered preliminary. Because the sample was comprised of Black and Hispanic urban youth, we cannot generalize the findings to other groups of youth such as those who are White Non-Hispanic or those who live in rural areas. Finally, because alcohol use and conduct problems were based on self-report data, it is possible that youth underreported these behaviors.
In conclusion, the constellation of self-cognitions that comprise a youth’s self-concept profile may help to identify youth who are at risk for maladaptive behaviors. Although the content of the self may be a viable prevention and early intervention target, this study requires replication with a larger sample to validate the findings. Replication with other samples will also enable us to better characterize specific cultural groups.
This article is based on data published in a podium presentation at the 2018 Conference for the Advancement of Nursing Science.
Both authors contributed equally to the conception of the manuscript, data acquisition as well as analysis, drafting, and critical revisions and gave approval on the final draft. They agreed to be accountable for all aspects of work ensuring integrity and accuracy.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Colleen Corte received funding from the UIC College of Nursing Dean’s Fund.
Celeste M. Schultz, PhD, RN-BC, CPNP-PC https://orcid.org/0000-0001-7270-6377
Adan, A., Forero, D. A., & Navarro, J. F. (2017). Personality traits related to binge drinking: A systematic review. Frontiers in Psychiatry, 8, 134. https://doi.org/10.3389/fpsyt.2017.00134
Alvanzo, A. A., Storr, C. L., La Flair, L., Green, K. M., Wagner, F. A., & Crum, R. M. (2011). Race/ethnicity and sex differences in progression from drinking initiation to the development of alcohol dependence. Drug and Alcohol Dependence, 118, 375–382. https://doi.org/10.1016/j.drugalcdep.2011.04024
Barry, C. T., Frick, P. J., & Killian, A. L. (2003). The relation of narcissism and self-esteem to conduct problems in children: A preliminary investigation. Journal of Clinical Child and Adolescent Psychology, 32(1), 139–152.
Beckstead, J. W. (2002). Using hierarchical cluster analysis in nursing research. Western Journal of Nursing Research, 24(3), 307–319.
Bratchell, N. (1989). Cluster analysis. Chemometrics and Intelligent Laboratory Systems, 6, 105–125.
Chung, C. K., & Pennebaker, J. W. (2008). Revealing dimensions of thinking in open-ended self-descriptions: An automated meaning extraction method for natural language. Journal of Research in Personality, 42(1), 96–132.
Colder, C. R., Shyhalla, K., & Frndak, S. E., (2018). Early alcohol use with parental permission: Psychosocial characteristics and drinking in late adolescence. Addictive Behaviors, 75, 82–87. https://doi.org/10.1016/j.addbeh.2017.07.030
Corte, C., & Stein, K. F. (2007). Self-Cognitions in antisocial alcohol dependence and recovery. Western Journal of Nursing Research, 29(1), 80–99.
Corte, C., & Szalacha, L. (2010). Self-cognitions, risk factors for alcohol problems, and drinking in preadolescent urban youths. Journal of Child & Adolescent Substance Abuse, 19(5), 406–423. https://doi:10.l080/1067828X.2010.515882
Corte, C., & Zucker, R. A. (2008). Self-concept disturbances: Cognitive vulnerability for early drinking and early drunkenness in adolescents at high risk for alcohol problems. Addictive Behaviors, 33, 1282–1290.
Dick, D. M, Aliev, F., Latendresse, S. J., Hickman, M., Heron, J., Macleod, J., Joinson, C., Maughan, B., Lewis, G., & Kendler, K. S. (2013). Adolescent alcohol use is predicted by childhood temperament factors before age 5, with mediation through personality and peers. Alcoholism: Clinical and Experimental Research, 37(12), 2108–2117.
Eshghi, A., Haughton, D., Legrand, P., Skaletsky, M., & Woolford, S. (2011). Identifying groups: A comparison of methodologies. Journal of Data Science, 9, 271–291.
Fraley, C., & Raftery, A. E. (1998). How man clusters? Which clustering method? Answers via model-based cluster analysis. The Computer Journal, 41(8), 578–588.
Gerard, J. M., & Buehler, C. (2004). Cumulative environmental risk and youth maladjustment: The role of youth attributes. Child Development, 75(6), 1832–1849.
Gower, J. C. (1967). A comparison of some methods of cluster analysis. Biometrics, 23(4), 623–637.
Hakulinen, C., Elovainio, M., Pulkki-Raback, L., Virtanen, M., Kivimaki, M., & Jokela, M. (2015). Personality and depressive symptoms: Individual participant meta-analysis of 10 cohort studies. Depression and Anxiety, 32(7), 461–470.
Halkidi, M., Batistakis, Y., & Vazirgiannis, M. (2002). Cluster validity methods: Part I. SIGMOD Record, 31(2), 40–45.
Harter, S. (1990). Developmental differences in the nature of self-representations: Implications for the understanding, assessment and treatment of maladaptive behavior. Cognitive Therapy and Research, 14(2), 113–142.
Hsu, C. C., Chen, C. L., & Su, Y. W. (2007). Hierarchical clustering of mixed data based on distance hierarchy. Information Science, 177, 4474–4492.
Jackson, K. M., Colby, S. M., Barnett, N. P., & Abar, C. C. (2015). Prevalence and correlates of sipping alcohol in a prospective middle school sample. Psychology of Addictive Behaviors, 29, 766–778. https://doi.org/10.1037/adb0000072
Kemmelmeier, M., & Oyserman, D. (2001). Gendered influence of downward social comparisons on current and possible selves. Journal of Social Issues, 57(1), 129–148.
Kendzierski, D., & Whitaker, D. J. (1997). The role of self-schema in linking intentions with behavior. Personality and Social Psychology Bulletin, 23(2), 139–147.
Khan, A. A., Jacobson, K. D., Gardner, C. O., Prescott, C. A., & Kendler, K. S. (2005). Personality and comorbidity of common psychiatric disorders. British Journal of Psychiatry, 186, 190–196.
Komro, K. A., Tobler, A. L., Maldonado-Molina, M. M., & Perry, C. L. (2010). Effects of alcohol use initiation patterns on highrisk behaviors among urban, low-income, young adolescents. Prevention Science, 11(1), 14–23.
Lee, C. K. (2013). One-year test-retest reliability of the drinker identity in 8th and 9th grade adolescents. Unpublished raw data.
Lee, C. K, Corte, C., & Stein, K. F. (2018). Drinker identity: Key risk factor for adolescent alcohol use. Journal of School Health, 88(3), 253–260.
Lee, C. K., Corte, C., Stein, K. F., Finnegan, L. F., McCreary, L., & Park, C. G. (2015). Expected problem drinker possible self: Predictor of alcohol problems and tobacco use in adolescents. Substance Abuse, 36, 434–439. https://doi.org/10.1080/08897077.2014.988323
Lee, C. K., Corte, C., Stein, K. F., Park, C. G., Finnegan, L., & McCreary, L. L. (2015). Prospective effects of possible selves on alcohol consumption in adolescents. Research in Nursing & Health, 38, 71–81.
Lee, C. K., Stein, K. F., & Corte, C. (2018). Effects of drinker self-schema on drinking- and smoking-related information processing and behaviors. Substance Abuse, 39(1), 32–38.
Lindgren, K. P., Ramirez, J. J., Olin, C. C., & Neighbors, C. (2016). Not the same old thing: Establishing the unique contribution of drinking identify as a predictor of alcohol consumption and problems over time. Psychology of Addictive Behaviors, 30(6), 659–671.
Long, K. A., & Boik, R. J. (1993). Predicting alcohol use in rural children: A longitudinal study. Nursing Research, 42(2), 79–86. https://doi.org/10.1097/00006199-1993030000-00004
Lopez-Romero, L., Romero, E., & Andershed, H. (2015). Conduct problems in childhood and adolescence: Developmental trajectories, predictors and outcomes in a six-year follow up. Child Psychiatry Human Development, 46, 762–773.
Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35(2), 63–78.
Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41(9), 954–969.
Markus, H., & Wurf, E. (1987). The dynamic self-concept: A social psychological perspective. Annual Review of Psychology, 38(1), 299–337.
Martsch, C. T., & Miller, W. R. (1997). Extraversion predicts heavy drinking in college students. Personality and Individual Differences, 23(1), 153–155.
Miech, R. A., Schulenberg, J. E., Johnston, L. D., Bachman, J. G., O’Malley, P. M., & Patrick, M. E. (2019, December 19). National adolescent drug trends in 2019: Findings released. Ann Arbor, MI: Monitoring the Future. Retrieved February 29, 2020, from http://www.monitoringthefuture.org
Milligan, G. W., & Cooper, M. C. (1987). Methodology review: Clustering methods. Applied Psychological Measurement, 11(4), 329–354.
Montemayor, R., & Eisen, M. (1977). The development of self-conceptions from childhood to adolescence. Developmental Psychology, 13(4), 314–319.
Moretti, M. M., Holland, R., & McKay, S. (2001). Self-other representations and relational and overt aggression in adolescent girls and boys. Behavioral Sciences and the Law, 19, 109–126.
Muratori, P., Milone, A., Brovedani, P., Levantini, V., Melli, G., Pisano, S., Valente, E., Thomaes, S., & Masi, G. (2018). Narcissistic traits and self-esteem in children: Results from a community and a clinical sample of patients with oppositional defiant disorder. Journal of Affective Disorders, 241, 275–281.
National Institute on Alcohol Abuse and Alcoholism. (2011). Alcohol screening and brief intervention for youth (NIH Publication No. 11-7805). https://www.niaaa.nih.gov/publications/clinicalguides-and-manuals/alcohol-screening-and-brief-interventionyouth
Oyserman, D., Bybee, D., Terry, K., & Hart-Johnson, T. (2004). Possible selves as roadmaps. Journal of Research in Personality, 38, 130–149.
Oyserman, D., & Fryberg, S. (2006). The possible selves of diverse adolescents: Content and function across gender, race, and national origin. Possible Selves: Theory, Research, and Applications, 2(4), 17–39.
Oyserman, D., Grant, L., & Ager, J. (1995). A socially contextualized model of African American identity: Possible selves and school persistence. Journal of Personality and Social Psychology, 69(6), 1216–1232.
Oyserman, D., & Markus, H. R. (1990). Possible selves and delinquency. Journal of Personality and Social Psychology, 59(1), 112–125.
Oyserman, D., Terry, K., & Bybee, D. (2002). A possible selves intervention to enhance school involvement. Journal of Adolescence, 25(3), 313–326.
Ramirez, J. J., Fairlie, A. M., Olin, C. C., & Lindgren, K. P. (2017). Implicit and explicit drinking identify predict latent classes that differ on the basis of college students’ drinking behaviors. Drug and Alcohol Dependence, 178, 579–585.
Robbins, L. B., Pis, M. B., Pender, N. J., & Kazanis, A. S. (2004). Physical activity self-definition among adolescents. Research and Theory for Nursing Practice, 18, 317–330.
Spear, L. P. (2015). Adolescent alcohol exposure: Are there separable vulnerable periods within adolescence? Physiologic Behavior, 148, 122–130. https://doi.org/10.1016/j.physbeh.2015.01.027
Stein, K. F., Chen, D. G. D., Corte, C., Keller, C., & Trabold, N. (2013). Disordered eating behaviors in young adult Mexican American women: Prevalence and associations with health risks. Eating Behaviors, 14(4), 476–483.
Stein, K. F., & Corte, C. (2007). Self-cognitions in antisocial alcohol dependence and recovery. Western Journal of Nursing Research, 29(1), 80–99.
Stein, K. F., & Corte, C. (2008). The identify impairment model: A longitudinal study of self-schemas as predictors of disordered eating behaviors. Nursing Research, 57(3), 182–190.
Stein, K. F., Roeser, R., & Markus, H. R. (1998). Self-schemas and possible selves as predictors and outcomes of risky behaviors in adolescents. Nursing Research, 47(2), 96–106.
Zernicke, K. A., Cantrell, H., Finn, P. R., & Lucas, J. (2010). The association between earlier age of first drink, disinhibited personality, and externalizing psychopathology in young adults. Addictive Behaviors, 35(10), 414–418.
Zhu, S., Tse, S., Cheung, S. H., & Oyserman, D.(2014). Will I get there? Effects of parental support on children’s possible selves. British Journal of Educational Psychology, 84(3), 435–453.
Zucker, R. A., & Fitzgerald, H. E. (1996). Other evidence for at least two alcoholisms II: Life course variation in antisociality and heterogeneity of alcoholic outcome. Development and Psychopathology, 8(4), 831–848.
Zucker, R. A., Fitzgerald, H. E., Refior, S., Putler, L., Pallas, D., & Ellis, D. (2000). The clinical and social ecology of childhood for children of alcoholics: Description of a study and implications for a differentiated social policy. In H. Fitzgerald B. Lester & B. Zuckerman (Eds.), Children of addiction: Research, health, and policy issues (pp.109–141). Routledge.
Celeste M. Schultz, PhD, RN-BC, CPNP-PC, is a Clinical Assistant Professor at College of Nursing, University of Illinois at Chicago, IL.
Colleen Corte, PhD, RN, FAAN, is an Associate Professor and Department Head at College of Nursing, University of Illinois at Chicago, IL.
1 College of Nursing, University of Illinois at Chicago, IL, USA
Corresponding Author:
Celeste M. Schultz, College of Nursing, University of Illinois at Chicago, 845 S. Damen Ave., MC 802, Chicago, IL 60612, USA.Email: cschultz@uic.edu