Identity Development and Its Relationship to Family History Knowledge among Late Adolescents

Abstract

Identity development among late adolescent university students and its relationship to family history knowledge was examined in this study. Identity development was examined using Marcia’s individual developmental framework (1988) of exploration and commitment and Stutman and Lich’s family systems framework (1984) of autonomy and relatedness. It was proposed that late adolescents’ personal exploration of and commitment to roles and values may be influenced by knowledge of parent and grandparent histories. It was also proposed that late adolescents’ achievement of personal autonomy and positive family relatedness may be influenced by knowledge of parent and grandparent histories. The sample consisted of 239 university students. The Parental Relationship Inventory (PRI) and the Ego Identity Process Questionnaire (EIPQ) were used to measure identity development. The Do You Know? (DYK) scale measured family history knowledge. Multiple regression analyses indicated a significant positive relationship between commitment and family history knowledge and relatedness and family history knowledge, a negative relationship between autonomy and family history knowledge, and a weak correlation between exploration and family history knowledge. Findings indicated that family history knowledge may influence components of identity development. This has implications for those working to enhance adolescent development.

1. Introduction

In 2002, observing the vast global reservoir of family history collections, and noting the growing number of young people in western countries struggling to find well-being and flourish, an Australian-based family history charity, the National Heritage Foundation (NHF), proposed that family history stories were a potentially powerful, untapped resource that could have a significant influence on positive youth development. This proposition was founded on the premise that families are central to personal well-being; and that if ancestors are included in our common reckoning of ‘family’, we acquire a substantial learning resource full of constructive lessons arising from many generations of experience and a sense of connectedness that goes beyond the immediate family where relationships may be strained. NHF conducted several experimental youth programs to test this proposition with family history story exploration and reflection as the main components. Programs included family history research, ancestor interviews, historical re-enactments, ancestor tribute performances and artwork, journal writing, and various other means to collect, explore, and reflect on family history stories. Two key themes that participants repeatedly referred to during and after these programs were personal and family identity. Thus, the study of youth identity development and its relationship to family history knowledge was identified as a key subject for additional study and research.

Although the processes involved in the development of personal and family identities are complex and include psychological, sociological, cultural, and historical factors (Schwartz et al. 2006), this study is limited to an analysis of psychological and sociological components within the context of individual and family development. Within these contexts, the individual developmental and family systems frameworks are two perspectives researchers have used to explain the adolescent identity development process (Adams 1998; Galliher et al. 2017; Grotevant and Cooper 1985; Lippold et al. 2017). These frameworks provide both psychological and sociological views of identity development. Although some argue that these two perspectives are not compatible (Slife and Williams 1995), others contend that despite being theoretically and methodologically distinguishable, they share common theoretical roots and can be integrated (Allen et al. 1994; Davies and Cicchetti 2004; Galliher et al. 2017; Perosa et al. 2002; Sabatelli and Mazor 1985).

In general, these models emphasize the need for individuals to explore and make commitments to roles and ideals while negotiating the balance between autonomy and relatedness within the family, peer, and other social relationships (Adams et al. 2006; Erikson 1959, 1968; Galliher et al. 2017; Mathies and Adams 2004; Marcia 2002; Perosa et al. 2002).

Current Study

Several researchers have suggested that knowing family history stories and participating in family history-related rituals may contribute to adolescent identity development by providing opportunities for adolescents to examine roles and ideals (Fivush et al. 2008; Gagalis-Hoffman 2004; Hammond 2001; Merrill and Fivush 2016; Merrill et al. 2018; Pratt and Fiese 2004) and strengthen family relatedness (Fivush et al. 2008; Gagalis-Hoffman 2004; Hammond 2001; Merrill and Fivush 2016; Merrill et al. 2018; Pratt and Fiese 2004). In addition, as indicated above, limited research with small sample sizes conducted by NHF suggested that family history knowledge may contribute to positive adolescent identity development. Hence, the purpose of this study was to further explore the relationship between psychological and sociological components of adolescent identity development and family history knowledge, with a particular emphasis on late adolescent university students, by testing the following hypothesis:

H₁. 

There is a positive relationship between identity development in late adolescent university students and family history knowledge.

H₂. 

There is a positive relationship (p < 0.05) between commitment (commitment subscale of EIPQ questionnaire) to personally selected life roles and ideological values and beliefs in late adolescent university students and family history knowledge.

H₃. 

There is a positive relationship (p < 0.05) between exploration (exploration subscale of EIPQ questionnaire) of life roles and ideological values and beliefs in late adolescent university students and family history knowledge.

H₄. 

There is a positive relationship (p < 0.05) between maintaining autonomy (autonomy subscale of PRI questionnaire) from the parental family in late adolescent university students and family history knowledge.

H₅. 

There is a positive relationship (p < 0.05) between relatedness (relatedness subscale of PRI questionnaire) with parents in late adolescent university students and family history knowledge.

2. Review of Literature

2.1. Identity Development

Erikson’s eight-stage psychosocial model of human development is a widely recognized theory of identity (Adams 1998; Erikson 1968; Marcia 1993; Marcia and Josselson 2013). According to Erikson (1959, 1968, 1985), the life cycle is divided into eight key stages. During each stage, there is a psychosocial crisis, a consequence of contradictory personal characteristics. Identity-Identity Diffusion is the fifth stage, taking place during the critical transition from adolescence to adulthood (Erikson 1959, 1968, 1985; Marcia 1980, 1993, 2002). In this model, positive identity development is seen as a major component in the healthy psychological development of late adolescence (Adams 1998; Erikson 1968; Marcia 2002).

Positive identity development involves gaining a strong sense of self (Adams 1998; Marcia 2002). The process is one in which individuals must explore roles and values and make independent decisions and commitments regarding occupation; religious, political, and social beliefs; and interpersonal and sexual values (Kroger and Marcia 2011; Marcia 2002; Marcia and Josselson 2013). Utilizing these dimensions of exploration and commitment, Marcia outlines four statuses of identity: (a) identity achievement; (b) foreclosure; (c) moratorium; and (d) identity diffusion (Kroger and Marcia 2011; Marcia 1980, 2002). Each status represents a level of exploration and commitment (see Figure 1).

Figure 1. Identity statuses (Marcia 2009, p. 672).

Identity achievement is reached after one has undertaken a process of exploration, has made decisions and is now pursuing self-directed occupational and ideological goals. Foreclosure is a state in which one has committed to a set of values and beliefs and is pursuing an identified occupation. However, there has been no individual exploration, and this commitment has been based on parental views and values. Moratorium describes a state of active exploration where no commitment has yet been made. This is a time when adolescents can work toward developing their own set of guiding values and beliefs. Finally, identity diffusion is a state where individuals have made no commitments and are not seeking to explore the available alternatives (Kroger and Marcia 2011; Marcia 1980, 2002).

Connected with determining one’s own ideological and occupational identity is the need to psychologically separate self from parents and family and the ability to see oneself as a separate and distinct individual (Adams 1998; Anderson and Sabatelli 1990). This process of psychological separation has been labeled individuation (Adams 1998; Marcia 1980, 1993). From a psychoanalytical perspective, the individuation process is completed when fusion with others ceases, and autonomy from the family of origin has been achieved (Anderson and Sabatelli 1990; Perosa et al. 2002). Fusion is defined as a state of embeddedness where there are no clear boundaries in relationships with others and emotional dependence on others is high (Perosa et al. 2002; Sabatelli and Mazor 1985).

Obtaining a sense of self, however, is dialectic and also requires the attainment of a sense of belonging. This is acquired through relatedness with and acceptance and recognition from family and peers (Adams 1998; Mathies and Adams 2004; Scabini and Manzi 2011). Adolescents “are sometimes morbidly, often curiously, preoccupied with what they appear to be in the eyes of others as compared with what they feel they are” (Erikson 1968, p. 128). Thus, the peer group, in particular, has a strong influence on the personal role and ideological and relationship choices (Erikson 1959; Verhoeven et al. 2018). According to Moore and Boldero (1991), the peer group provides feedback about how individuals are seen by others. This feedback contributes to self-discovery and self-concept through reciprocity. They also suggest the peer group provides models to copy, an empathetic support base while adolescents are establishing autonomy from parents, and assistance in the development of intimacy through compromise and the sharing of confidences.

Although a strong influence on individual identity development, the peer group does not completely remove the influence of the family (Mathies and Adams 2004; Perosa et al. 2002). In fact, it is argued that “secure family relations are the basis for entry into the peer system and success within it” (Hartup 1983, p. 172). In this regard, family system theorists have emphasized the role of the family in the process of individuation and autonomy achievement. (Lippold et al. 2017; Mathies and Adams 2004).

2.2. Family Systems Framework

In seeking to explain the individuation process, family system theorists emphasize the influence of family relationships on individual development and healthy autonomy achievement (Lippold et al. 2017; Mathies and Adams 2004; Perosa et al. 2002; Scabini and Manzi 2011). Family systems theory states that families can be likened to a system (Broderick 1993). Families, like systems, must be looked at as a whole. The individual actions of one member of the system affect all other individuals and vice versa. (Broderick 1993; White and Klein 2008).

As such, the family system impacts identity development. Family system theorists argue that individual development and autonomy achievement occur within healthy functioning family systems characterized by age-appropriate levels of autonomy and relatedness. In such families, individual autonomy is encouraged within the context of warm and supportive relationships (Lippold et al. 2017; Mathies and Adams 2004; Perosa et al. 2002; Stutman and Lich 1984). Stutman and Lich define this condition as Healthy Differentiation.

Differentiation describes the degree to which an individual has developed autonomy from the family of origin and the degree to which the family system allows for such autonomy to be developed while maintaining close and supportive relations (Mathies and Adams 2004; Perosa et al. 2002; Stutman and Lich 1984). According to Stutman and Lich, Healthy Differentiation is indicated by a combination of high autonomy and high relatedness (see Figure 2).

Figure 2. Stutman and Lich (1984) Family Differentiation Categorization System.

Stutman and Lich also identify three maladaptive forms of differentiation: Overinvolvement—Consonant type, Overinvolvement—Dissonant type, and Underinvolvement. Overinvolvement—Consonant type describes individuals who are overly involved with their family of origin and who are unable to achieve autonomy. They are unable to make independent decisions or commit to personally selected roles and values. In addition, these individuals are comfortable with their dependent state. Overinvolvement—Dissonant type also describes individuals who are overinvolved with their family of origin and who are unable to achieve autonomy. In this case, however, these individuals are aware of their regressive and childlike tendencies, leading to feelings of resentment. Underinvolvement describes individuals who have a high degree of emotional and physical separateness from their family of origin. This form of differentiation is characterized by a significant lack of connection, intimacy, or commitment to the family of origin. The condition is described as false autonomy, as the appropriate balance between independence from and relatedness to the family has not been achieved. According to the family systems model, optimal identity development occurs when healthy differentiation exists (Mathies and Adams 2004; Perosa and Perosa 1993; Perosa et al. 2002; Stutman and Lich 1984).

To summarize, identity development is a critical component of the maturing process. Successful identity development requires an individual to establish both psychological autonomy and familial intimacy. This is achieved through a process of exploring and making commitments to personally determined roles and values and developing age-appropriate autonomy while maintaining intimate ties with parents and the family of origin. These two facets of identity are braided and integrated together. The ideal state is neither independence nor dependence; identity achievement and healthy differentiation are conditions of interdependence. Thus, personal exploration, making decisions and commitments, developing autonomy, and maintaining a healthy relationship with one’s family all contribute to the identity development process.

2.3. Family History and Identity

One factor that may contribute to identity development through its influence on both psychological autonomy and strengthening family relatedness is family history knowledge (Fivush et al. 2008; Gagalis-Hoffman 2004; Hammond 2001; Merrill et al. 2018). The individual developmental perspective suggests the need for personal exploration of, and commitment to, roles and ideological values. Research suggests that such personal exploration and commitment may be facilitated through knowing and reflecting on family history stories. Gagalis-Hoffman (2004) reported that “kinship with story characters appeared to increase desire in both parents and their…children to emulate the traits and characteristics ascribed to their ancestors in family stories” (p. 41). Hammond (2001) highlighted the use of family history knowledge as a facilitator of values transmission and reflection. His results suggested that exploring the family’s historical traditions and values was important in building the next generation’s identity (pp. 80–82). Merrill et al. (2018) investigated the accessibility and functions of intergenerational narratives that adolescents and emerging adults know of their parents and “captured instances when the participant derived information about the self from understanding parent’s characteristics, traits, or values” (p. 17). Similarly, Pratt and Fiese (2004) argue that adolescents “are seen as drawing on the cultural reservoir of stories to provide elements from which the…sense of self is constructed…” and that “…family stories serve as an important way of communicating, negotiating, and re-negotiating that identity” (p. 17).

A young high school dropout attending an applied ancestry program provides an example of how family history stories may be a medium individuals use to construct a sense of self by exploring roles and values. The individual concerned discovered that a deceased uncle was a marketing professional and the creator of a famous advertising slogan. Previously without a vision of what occupational role to pursue, this youth now had a new sense of who he could be. He subsequently returned home and reenrolled in school with the intent to pursue tertiary studies in marketing. In another example, a young man who was reintroduced to stories about his grandfather’s experiences as a soldier in World War II chose to pursue a career in the armed forces. These examples illustrate how family history stories can facilitate occupational role exploration (Rancie, Peter. 2003. National Heritage Foundation, Barjarg, VIC, Aust. Personal communication). In addition, however, the literature suggests family history knowledge may also contribute to the development of family relatedness through the family system (Fivush et al. 2008; Gagalis-Hoffman 2004; Merrill et al. 2018; Pratt and Fiese 2004).

Family stories link or connect generations, creating a sense of connectedness, belonging, and relatedness contributing to positive family relationships (Fivush et al. 2008; Gagalis-Hoffman 2004; Merrill et al. 2018; Pratt and Fiese 2004). For example, Gagalis-Hoffman (2004) found that parents and children who knew family history stories felt they “belonged to a group, which in turn gave them a feeling of…family identity” (p. 24). McGoldrick (1995) suggests that families “communicate their connectedness through rituals [including family stories] and patterns passed from generation to generation” (p. 100). Similarly, Fivush et al. (2008), citing Pratt and Fiese (2004), claim that “family stories are the way in which we connect across generations to create family history and family identity. Through the telling and sharing of family history stories children develop a sense of self as connected to previous generations” (p. 135). Further, stories of parent’s and grandparent’s lives, and other stories from previous generations, “create meaning beyond the individual, to include a sense of self through historical time and in relation to family members” (Fivush et al. 2008, p. 134). Reese et al. (2010), reported that, “adolescents made personal connections to their parents by drawing comparisons between their parents’ life circumstances and their own” (in Merrill et al. 2018, p. 3). In these ways, family history stories may contribute to increased relatedness between members of the family system.

This body of evidence suggests that family history knowledge may have a positive influence on identity development. In addition, the literature suggests that when considering the influences of family history knowledge on identity development, a sense of self as a unique individual (identity achievement) and as connected to family (healthy differentiation) should be considered (Fivush et al. 2008; Merrill et al. 2018). That is, consideration should be given to the process of exploration and commitment and relatedness and autonomy. Further, the significance of late adolescence in the identity formation process and the findings of Fivush et al. (2008), indicating family history knowledge contributed to preadolescents’ sense of self, provoked interest in whether results could be replicated for late adolescent university students.

3. Methods

3.1. Study Design

The purpose of this study was to further explore the relationship between psychological and sociological components of adolescent identity development and family history knowledge, with a particular emphasis on late adolescent university students, by testing the five research hypotheses.

Data were collected from late adolescent university students. The convenience sample (N = 239) consisted of 186 (77.8%) females and 53 (22.2%) males aged between 18 and 20 years old, drawn from seven United States universities: Western Kentucky (31.6%); Michigan State (28.3%); Clemson (16.9%); Brigham Young (13.1%); Indiana (5.5%); Texas A&M (4.2%); and, Western Washington (0.4%). Each participant completed two identity questionnaires and a family history knowledge questionnaire. The majority of participants were religious (73.4%) and had parents who were married (77.2%). In addition, the overwhelming majority grew up with their biological family (97.5%). No racial/ethnic data were collected. A convenience sample was chosen due to limited resources and time constraints. The sample size was determined by a power analysis. Data collection was via an online questionnaire.

3.2. Measures

3.2.1. Parental Relationship Inventory (PRI)

The PRI (Lich 1985; Stutman 1984; Stutman and Lich 1984) was used to measure the identity development components of autonomy and relatedness. The PRI consists of 25 items (14 autonomy items and 11 relatedness items). A sample item on the Autonomy scale reads, “Many times when something happens to my parents, I feel like it’s happening to me.” A sample item on the Relatedness scale reads, “It is fun to be with my parents.” Items are answered on a 4-point Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree). Autonomy and relatedness scores are summed separately, providing individual total scores. Participants can also be divided into high and low groups utilizing a median split, allowing for categorization into family differentiation categories (Stutman and Lich 1984). For this sample, Cronbach’s alpha coefficients were 0.79 for autonomy and 0.87 for relatedness.

3.2.2. Ego Identity Process Questionnaire (EIPQ)

The EIPQ (Balistreri et al. 1995) was used to measure identity development components of exploration and commitment. The EIPQ is a 32-item questionnaire designed to assess exploration and commitment (16 exploration items and 16 commitment items) within four ideological domains (politics, religion, occupation, and values) and within four interpersonal domains (friendships, dating, sex roles, and family). A sample item on the Exploration scale reads, “I have consistently re-examined many different values in order to find the ones which are best for me.” A sample item on the Commitment scale reads, “I have firmly held views concerning my role in my family.” Each item is answered on a six-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Item scores are summed separately, providing individual total scores for both exploration and commitment. Participants can also be divided into high and low groups utilizing a median split, allowing for categorization into identity statuses. (Balistreri et al. 1995). For this sample, Cronbach’s alpha coefficients were 0.697 for exploration and 0.78 for commitment.

3.2.3. Do You Know (DYK)

The DYK scale was used to measure family history knowledge (Duke et al. 2008). The DYK is a 20-item yes or no questionnaire designed to measure family history knowledge. A sample item on the DYK scale reads, “Do you know some of the lessons that your parents learned from good or bad experiences?” The scale tests respondents’ knowledge of major events, places lived, occupations, and family anecdotes from the lives of parents and grandparents. Each item on the questionnaire is worth 1 point, and the higher the score, the higher the knowledge of family history (Duke et al. 2008).

3.2.4. Socio-Demographic Variables

Four socio-demographic variables were used as independent variables: parent’s marital status (married/divorced); family of origin status (natural/step or adopted); gender (male/female); and religiousness (religious/non-religious).

3.3. Analysis

Descriptive statistics (SPSS 17.0) were performed. Pearson Product Moment zero-order correlations were calculated to check for multicollinearity and significant bivariate relationships among variables. With the use of a p < 0.01 criterion, the Mahalanobis distance was used to examine the normality, skewness, linearity, and homoscedasticity of residuals. Hypotheses were tested using multiple regression analysis with identity measures (autonomy, relatedness, exploration, and commitment) as separate dependent variables and knowledge of family history and socio-demographic variables as independent variables. In addition, an analysis of variance was also conducted to compare family history knowledge means with Marcia’s identity statuses (Balistreri et al. 1995) and Stutman and Lich’s (1984) family differentiation categories. A Tukey’s Post Hoc test was performed to examine the relationship between and within the identity statuses and family differentiation categories.

4. Results

With the use of a p < 0.01 criterion for the Mahalanobis distance, four outliers among the cases were identified. Due to having multiple variables lying outside the normal distribution, two of these were eliminated to reduce skewness and improve the normality, linearity, and homoscedasticity of residuals. As the remaining outliers lay just outside the specified parameters and their inclusion was deemed significant to the overall analysis of the data, they were included. No cases had missing data, and no suppressor variables were found.

The mean scores for the identity constructs were autonomy 33.29 (SD = 5.27; Range = 21–50), relatedness 35.89 (SD = 5.38; Range = 14–44), commitment 65.01 (SD = 9.98; Range = 33–88), and exploration 64.91 (SD = 8.70; Range = 41–90). The family history knowledge mean was 16.22 (SD = 3.08). Zero-order correlations were used to examine bivariate relationships, and meaningful correlations were found among the variables.

When considering Stutman and Lich’s differentiation categories (1984) of the sample, 75 (31.6%) were underinvolved, 57 (24.1%) were overinvolvement-dissonant, 73 (30.8%) were overinvolvement-consonant, and 32 (13.5%) were healthily differentiated. For Marcia’s identity development statuses (1980) of the sample, 49 (20.9%) were diffused, 72 (30.6%) were foreclosed, 71 (30.2%) were in a moratorium, and 43 (18.3%) were identity achieved. The sample aligned with similar late adolescent samples (Adams 1998).

4.1. Hypothesis Testing

The following analysis shows that hypotheses H₁, H₂, and H₅ were supported in the multiple regression models. H₃ and H₄ were not supported in this data set.

H₁. 

There is a positive relationship between identity development in late adolescent university students and family history knowledge. Results from H₂ and H₅ generally support this hypothesis.

H₂. 

There is a statistically significant positive relationship between commitment (commitment subscale of EIPQ questionnaire) to personally selected life roles and ideological values and beliefs in late adolescent university students and family history knowledge (β = 0.254, p < 0.01).

H₃. 

There is not a statistically significant relationship between exploration (exploration subscale of EIPQ questionnaire) of life roles and ideological values and beliefs in late adolescent university students and family history knowledge (β = 0.029, p < 0.654).

H₄. 

There is not a statistically significant positive relationship between maintaining autonomy (autonomy subscale of PRI questionnaire) from the parental family in late adolescent university students and family history knowledge (β = −0.267, p < 0.01).

H₅. 

There is a statistically significant positive relationship between relatedness (relatedness subscale of PRI questionnaire) with parents in late adolescent university students and family history knowledge (β = 0.402, p < 0.05).

4.2. Multiple Regression Analyses

Four multiple regression models using sequential regression were computed to examine the relationship between the identity variables and family history knowledge among late adolescent university students beyond the bivariate level. Autonomy, relatedness, commitment, and exploration were assigned as dependent variables. In the first block for each model, demographic variables (parent’s marital status, family of origin status, gender, and religiousness) were assigned as independent variables. In the second block for each model, family history knowledge was added as an independent variable. Table 1 displays the correlations between the variables, the unstandardized regression coefficients (B) and the intercept, the standard errors (SE B), the standardized regression coefficients (β), R2, and adjusted R2 for each model.

Table 1. Summary of Multiple Regression Equations.

Variables	B	SE B	β	p
Identity Autonomy (N = 237)
Block 1 R2 = 0.018 (p = 0.389)
Constant	34.11	0.890		0.001 **
Divorced	1.174	0.842	0.094	0.165
Step/Adopted a	0.296	2.254	0.009	0.896
Female	−0.463	0.837	−0.036	0.581
Religious b	−0.998	0.789	−0.084	0.207
Block 2 ∆R2 = 0.067 (p < 0.01)
Constant	41.458	1.961		0.001 **
Divorced	0.648	0.823	0.052	0.432
Step/Adopted	0.183	2.179	0.005	0.933
Female	−0.183	0.811	−0.014	0.822
Religious	−1.046	0.762	−0.088	0.171
Family History Knowledge	−0.457	0.110	−0.267	0.001 **
Identity Relatedness (N = 237)
Block 1 R2 = 0.037 (p < 0.05)
Constant	34.130	0.896		0.001 **
Divorced	−1.296	0.825	−0.101	0.117
Female	1.125	0.843	0.087	0.184
Religious	1.610	0.791	0.132	0.043
Block 2 ∆R2 = 0.179 (p < 0.001)
Constant	22.816	1.876		0.001 **
Divorced	−0.471	0.766	−0.037	0.540
Female	0.696	0.776	0.054	0.371
Religious	1.677	0.726	0.138	0.022 *
Family History Knowledge	0.704	0.105	0.402	0.001 **
Identity Commitment (N = 237)
Block 1 R2 = 0.048 (p < 0.05)
Constant	60.781	1.658		0.001 **
Divorced	−0.210	1.569	−0.009	0.894
Step/Adopted	3.078	4.200	0.049	0.464
Female	0.956	1.559	0.040	0.540
Religious	4.707	1.469	0.209	0.002 **
Block 2 ∆R2 = 0.091 (p < 0.001)
Constant	47.508	3.662		0.001 **
Divorced	0.740	1.538	0.031	0.631
Step/Adopted	3.283	4.069	0.052	0.421
Female	0.449	1.515	0.019	0.767
Religious	4.794	1.423	0.213	0.001 **
Family History Knowledge	0.825	0.205	0.254	0.001 **
Identity Exploration (N = 237)
Block 1 R2 = 0.042 (p < 0.05)
Constant	65.888	1.45		0.000
Divorced	0.917	1.372	0.044	0.504
Step/Adopted	2.143	3.673	0.039	0.560
Female	1.828	1.363	0.087	0.181
Religious	−3.638	1.285	−0.185	0.005 **
Block 2 ∆R2 = 0.022 (p = 0.654)
Constant	64.551	3.312		0.001 **
Divorced	1.013	1.391	0.049	0.467
Step/Adopted	2.164	3.680	0.039	0.557
Female	1.777	1.37	0.085	0.196
Religious	−3.630	1.287	−0.185	0.005 **
Family History Knowledge	0.083	0.185	0.029	0.654

Note. ^a Step/Adopted is coded 0 for respondents with natural parents and 1 for respondents who have step-parents or are adopted. ^b Religious is coded 0 for non-religious and 1 for religious. ** p < 0.01; * p < 0.05.

In the first block predicting autonomy r² was not significantly different from 0 (r² = 0.018, p = 0.389). Adding family history knowledge led to a significant change in the model. Controlling for parents’ marital status, family of origin status, gender, and religiousness, family history knowledge was found to be a significant negative predictor of autonomy (β = −0.267, p < 0.01). In the first block predicting relatedness, the model was significant (r² = 0.044, p < 0.05) However, none of the t-tests for the coefficients were statistically significant. Multicollinearity was suspected. Family of origin status and female were found to be collinear. Family of origin status was removed from the model. In the new first block predicting relatedness, controlling for parents’ marital status and gender, religiousness was found to be a significant predictor of relatedness (β = 0.132, p < 0.05). Adding family history knowledge led to a significant change in the model. Controlling for parents’ marital status, gender, and religiousness, family history knowledge was found to be a significant predictor of relatedness (β = 0.402, p < 0.01). Family history knowledge was the strongest predictor of relatedness even though controlling for marital status, gender, and family history knowledge, being religious was also a significant predictor of relatedness (β = 0.138, p < 0.05).

In the first block predicting commitment, controlling for parents’ marital status, family of origin status, and gender, religiousness was found to be a significant predictor of commitment (β = 0.209, p < 0.01). Adding family history knowledge led to a significant change in the model. Controlling for parents’ marital status, family of origin status, gender, and religiousness, family history knowledge was found to be a significant predictor of commitment (β = 0.254, p < 0.01). Family history knowledge was the strongest predictor of commitment even though controlling for marital status, gender, and family history knowledge, being religious was also a significant predictor of commitment (β = 0.213, p < 0.01). In the first block predicting exploration, controlling for parent’s marital status, family of origin status, and gender, religiousness was found to be a significant negative predictor of exploration (β = −0.185, p < 0.01). Adding family history knowledge did not lead to a significant change in the model.

4.3. Analysis of Variance

Analysis of variance was also conducted to compare family history knowledge means with Marcia’s identity statuses: identity achieved; moratorium; foreclosure; and diffused (Balistreri et al. 1995), and Stutman and Lich’s (1984) four family differentiation categories (Healthy Differentiation, Overinvolvement—Consonant type, Overinvolvement—Dissonant type, and Underinvolvement).

ANOVA results for Marcia’s identity statuses (Table 2 and Table 3) suggest that there was a significant difference between family history knowledge and identity development stages (F (3, 231) = 7.121, p < 0.000). The family history mean identity achieved was the highest (µ = 17.6279), and diffused was the lowest family history knowledge mean (µ = 15.3878). Tukey’s Post Hoc test (Table 4) indicates there was no difference in family history knowledge between identity stages diffused and moratorium (p = 0.997), as well as between stages foreclosed and identity achieved (p = 0.407). There were, however, significant differences between the stages of diffused and moratorium when compared to foreclosed and identity achieved.

Table 2. Descriptives: Family History Knowledge and Marcia’s Identity Stages (1980).

Category	N	M	SD	df	F	Sig.
1. Diffusion	49	15.63	2.62	3 (between)	7.12	0.001 **
2. Foreclosed	72	16.76	2.90	231 (within)
3. Moratorium	71	15.49	3.42
4. Identity Achieved	43	17.63	2.06

Note. Group sizes are unequal; however, as noted by Tabachnick and Fidell (2001), unequal n often reflects true differences in the nature of the population, and efforts to equalize them artificially may distort the differences. Given the large F result, it was determined that the unequal group sizes would not lead to a Type 1 error. ** p < 0.001.

Table 3. Analysis of Variance Family History Knowledge and Marcia’s Identity Stages (1980).

	SS	df	MS	F	Sig.
Between Groups	177.78	3	59.26	7.12	0.000 **
Within Groups	1922.41	231	8.32
Total	2100.19	234

Note. ** p < 0.001.

Table 4. Post Hoc Test: Tukey Multiple Comparisons (Family History Knowledge and Marcia’s Identity Stages (1980).

(I) Differ Differentiation	(J) Differ Differentiation	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Lower Bound	Upper Bound
1.00 Identity Achieved	2.00	0.86402	0.55599	0.407	−0.5748	2.3028
	3.00	2.13495 *	0.5745	0.001	−0.6924	3.5775
	4.00	2.24015 *	0.60281	0.001	0.6802	3.8001
2.00 Foreclosed	1.00	−0.86402	0.55599	0.407	−2.3028	0.5748
	3.00	1.27093 *	0.48249	0.044	−0.0223	2.5195
	4.00	21.37613	0.53425	0.052	−0.0064	2.7587
3.00 Moratorium	1.00	−2.13495 *	0.55745	0.001	−3.5775	−0.6924
	2.00	−1.257093 *	0.48249	0.044	−2.5195	−0.0223
	4.00	0.10520	0.53577	0.997	−1.2813	1.4917
4.00 Diffused	1.00	2.24015 *	0.60281	0.001	−3.8001	−0.6802
	2.00	−1.37613	0.83425	0.052	−2.7587	0.0064
	3.00	−0.10520	0.53577	0.997	−1.4917	1.2813

Note. * The mean difference is significant at the 0.05 level.

When considering Stutman and Lich’s four family differentiation categories (healthy differentiation, overinvolvement—consonant type, overinvolvement—dissonant type, and underinvolvement), ANOVA results (Table 5 and Table 6) indicated a significant difference between underinvolvement and the other three differentiation categories (F (3, 233) = 14.74, p < 0.001). The family history knowledge mean (µ = 17.06) was high for those who were classified as healthily differentiated. As expected, categories with lower levels of autonomy (overinvolvement-consonant and overinvolvement—dissonant) also had high family history knowledge means. Underinvolvement had the lowest mean (µ = 14.55). Tukey’s Post Hoc (Table 7) indicated that Healthy Differentiation, Overinvolvement-Consonant, and Overinvolvement—Dissonant all differed significantly from the Underinvolvement category, characterized by higher autonomy and lower family history knowledge scores.

Table 5. Descriptives: Family History Knowledge and Stutman and Lich’s (1984) Differentiation Categories.

Category	N	M	SD	df	F	Sig.
1. Healthy Differentiation	32	17.06	2.31	3 (between)	14.74	0.001 **
2. Overinvolvement-Consonant	73	17.52	1.97	233 (within)
3. Overinvolvement-Dissonant	57	16.26	2.92
4. Underinvolvement	75	14.55	3.59

Note. Group sizes are unequal; however, as noted by Tabachnick and Fidell (2001), unequal n often reflects true differences in the nature of the population, and efforts to equalize them artificially may distort the differences. Given the large F result, it was determined that the unequal group sizes would not lead to a Type 1 error. ** p < 0.001.

Table 6. Analysis of Variance Family History Knowledge and Stutman and Lich’s (1984) Differentiation Categories.

	SS	df	MS	F	Sig.
Between Groups	356.292	3	118.764	14.737	0.001 **
Within Groups	1877.733	233	8.059
Total	2234.025	236

Note. ** p < 0.001.

Table 7. Post Hoc Test: Tukey Multiple Comparisons (Family History Knowledge and Stutman and Lich’s (1984) Differentiation Categories).

(I) Differ Differentiation	(J) Differ Differentiation	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Lower Bound	Upper Bound
1.00 Healthy Differentiation	2.00	−0.45805	0.60186	0.872	−2.0155	1.0994
	3.00	0.79934	0.62708	0.580	−0.8233	2.4220
	4.00	2.51583 *	0.59941	0.000	0.9648	4.0669
2.00 Overinvolvement—Consonant	1.00	0.45805	0.60186	0.872	−1.0994	2.0155
	3.00	1.25739	0.50178	0.062	−0.0410	2.5558
	4.00	2.97388 *	0.46674	0.000	1.7661	4.1817
3.00 Overinvolvement—Dissonant	1.00	−0.79934	0.62708	0.580	−2.4220	0.8233
	2.00	−1.25739	0.50178	0.062	−2.5558	0.0410
	4.00	1.71649 *	0.49884	0.004	0.4257	3.0073
4.00 Underinvolvement	1.00	−2.51583 *	0.59941	0.000	−4.0669	−0.9648
	2.00	−2.97388 *	0.46674	0.000	−4.1817	−1.7661
	3.00	−1.71649 *	0.49884	0.004	−3.0073	−0.4257

Note. * The mean difference is significant at the 0.05 level.

5. Discussion

Study results partially support the hypothesis that family history knowledge is related to identity development in adolescents. As Marcia (2002) has explained, positive identity development is dependent upon exploration and commitment. Study results indicated a significant positive relationship between commitment and family history knowledge but not between exploration and family history knowledge. Stutman and Lich (1984) describe adolescent identity development from the perspective of healthy differentiation which is indicated by high levels of relatedness and autonomy. Study results indicated a significant positive relationship between relatedness and family history knowledge but a significant negative relationship between autonomy and family history knowledge. Results, therefore, suggest that family history knowledge is at least partly associated with the definitions of identity achieved status (Marcia 2002) and healthy differentiation (Stutman and Lich 1984). Explanations of why some identity development variables (exploration and autonomy) are not positively associated with family history knowledge are included in this discussion. Researchers note that the results do not claim a cause-and-effect relationship and recognize other limitations in the study. Further research is recommended, including qualitative methods that may tease out the nuances of how family history stories are shared and influence identity development within family and cultural contexts.

5.1. Interpretation of Results

In support of H₂, there was a significant, positive relationship between the identity component of commitment and family history knowledge. This finding is consistent with previous research. Hammond (2001) reported that participating in intergenerational family rituals and storytelling practices contributed to values transmission and reflection, and Gagalis-Hoffman (2004) reported a desire within participants to emulate traits and characteristics of ancestors. Adams (1998) suggests that the level of integration within a life system [family, group, or community] will influence the degree of commitment to social roles and values. Thus, given that a significant relationship was found between relatedness and family history knowledge, it is not surprising that a significant relationship with commitment was also found. It may be that sharing family history knowledge through rituals and storytelling contributes to integration, which in turn influences levels of commitment.

Narrative identity theory provides further insight into how family history knowledge (transferred through family intergenerational rituals and storytelling) may influence commitment to roles and values. McLean and Pasupathi (2012) state that “…narratives are…likely to reflect, and possibly bolster, people’s commitments to important identities (p. 14)”. They cite the example of an adolescent whose reflection on his experience of his parent’s marriage breakdown and divorce (a family history story) influenced his commitment to the value of self-reliance and growth in the face of adversity. They propose that his current commitment is rooted in his past experience and that the experience will continue to be a source for sustained commitment to the identified value. As well as his reflection on his own feelings and actions, his narrative included a reflection on the behavior of his parents:

Obviously, the only people at the time that I could rely on let me down.

(McLean and Pasupathi 2012, p. 18)

Thus, a reflection on his parent’s part in the story, not just his own, contributed to the development of his view on the importance of self-reliance. Such reflection on family history stories may influence adolescents’ motivation to make commitments.

However, findings from this study did not support H₃, that there would be a significant, positive relationship between exploration and family history knowledge. According to Marcia (2002), identity achievement consists of both exploration and commitment. Commitment without exploration has been labeled foreclosure (Adams 1998; Marcia 2002; Hammond 2001; Kleiber 1999; Perosa et al. 2002). Without the exploration component, it could be argued that family history knowledge may just be a means of projecting parents’ values and traditions onto adolescents. However, the ANOVA result indicated that those classified as identity achieved also had the highest family history knowledge mean score. Therefore, at the very least, family history knowledge did not interfere with the process of exploration for those classified as identity achieved.

One possible explanation for the non-significant finding in the relationship between exploration and family history knowledge may be the proposition that it is possible to score high on commitment and low on exploration and not necessarily be foreclosed. Balistreri et al. (1995) suggest “that the commitment dimension is more readily identifiable than the exploration dimension” and question “whether paper-and-pencil measures can adequately tap exploration or whether the probes of interviews are necessary…” (p. 189). Further, Meeus et al. (2005) suggest that exploration is more a measure of the process of identity achievement, whereas commitment is a measure of the strength of identity achievement. Thus, it may be that for some in this study, exploration had occurred, but this was not detected by the measuring instrument.

Another explanation may lie in the significant relationship found with relatedness H₅). It may be that parents who create a nurturing environment, maintain warm and balanced relationships with their children, where individual views are encouraged, fostered, and valued, and who allow adolescents to hold alternative views and opinions may create a natural exploration environment. Hence, some late adolescents who scored high on commitment and family history knowledge may not have felt a ‘crisis’ or a need for significant additional exploration. In this case, parental values might not have been projected or imposed so much as gladly and independently received. As McElhaney et al. (2008) suggest, “During adolescence, the most effective form of parental influence is not limited to control of adolescents’ behavior but rather more broadly encompasses a relationship in which adolescents feel supported and understood enough that they make the choice to follow their parents’ lead (p. 224)”.

Although H₄ was not supported, study findings suggest it is possible for healthy differentiation to be achieved when family history knowledge is high. Stutman and Lich’s (1984) differentiation model suggests that healthy differentiation consists of both high autonomy and high relatedness. The ANOVA conducted in this study indicated a high family history knowledge mean for the healthy differentiation category. It appears that for some, high knowledge of family history knowledge did not correlate positively with autonomy, but for others, at the very least, it did not interfere with achieving healthy differentiation. Some evidence suggests it is possible that family history knowledge may have a positive influence on healthy differentiation when it is shared in ways that foster autonomy. Bohanek et al. (2006) found that families who created a safe and comfortable environment for story-telling and who valued and integrated children’s perspectives of stories had children with higher self-esteem. Walker and Taylor (1991), found autonomy was fostered when parents challenged adolescents’ beliefs but only when challenges were issued within the context of warm and respectful relationships. Similarly, Reese et al. (2010) found that adolescents whose mothers asked them elaborative questions in early childhood were more likely to report personal insights in narrating their life events. Thus, within a nurturing environment where relatedness is high and where individual views are encouraged, fostered, and valued, family history knowledge may contribute to the development of both relatedness and autonomy.

5.2. Sociodemographic Variables

Religion was the only sociodemographic variable found to be a predictor of positive identity. Results showed a significant relationship between religion and relatedness and religion and commitment. This is consistent with Pearce and Axinn (1998) and Mahoney and Tarakeshwar (2005), who both reported ties between religion and family cohesiveness and commitment. This is a likely consequence given that most religions promote the importance of family relationships and family values, and obedience to moral laws. Other sociodemographic variables, including the parent’s marital status, family of origin status, and gender, were not found to be significant in these models.

5.3. Limitations and Recommendations for Future Research

There are some important limitations to the current study. First, due to resource and time constraints, a convenience sample was used, and results cannot be generalized beyond the sample group. In addition, the sample was confined to 18–20 years old university students within the United States and is unlikely to represent a broad racial and cultural cross-section. Future research should include late adolescents who do not pursue higher education, younger adolescents, and those from other ethnicities. Given the important role of race and culture in identity development, emphasis should be given to measuring these influences on identity development as they relate to the role of ancestors and the sharing of ancestor stories within the family context. Another limitation is the way in which family history knowledge was measured in the current study. The current instrument does not seek information beyond the lifespan of grandparents and only measures knowledge of family history. In the future, a research instrument designed to detect a greater depth of ancestral knowledge (looking beyond the grandparent generation), emotional reactions, and racial and cultural differences could be adopted. In this regard, qualitative interviews may be the most effective way to draw out the nuances of emotion, race, and culture. Whatever the instrument, measuring emotion, race, and cultural differences will be an important area to study to gain greater insights into the relationship between identity development and family history knowledge. The use of the median split to divide responses into one of the four family differentiation categories is also a limitation of the study, as no previous studies indicating the medians of autonomy and relatedness were available as a comparison. Further, it should be noted that this is a correlational study and will not allow for cause-and-effect conclusions. Finally, the cross-sectional nature of the current study is also a limitation. As such, in the future, it would be important to conduct longitudinal studies.

Further, future research could examine how parental methods for utilizing family history as a tool for personal exploration of roles, ideals, and interpersonal values act as mediators between family history knowledge and commitment, exploration, relatedness, and autonomy. Future research could also examine whether parents can assist adolescents in utilizing family history knowledge to promote a healthy balance between family relatedness and individual autonomy and promote personal exploration of and increased commitment to personally chosen roles, ideals, and interpersonal values. Finally, rather than paper-and-pencil measures, interviews may be more effective in measuring how knowing family history stories influences positive identity development within family and cultural contexts.

5.4. Implications

These results have implications for application within the family context. Parents may find value in utilizing family history stories and rituals to strengthen family relationships and foster independent development. However, it would be important to show parents how to do so in ways optimal for fostering exploration and autonomy. Examples of doing so include showing parents:

• How to present stories in ways that highlight values, roles, and relationships;
• How to ask questions that invite adolescents to respond to stories with personal interpretations and meanings;
• How to listen for understanding;
• How to share their interpretations and values without imposing them upon adolescents.

In addition, it may be that family history knowledge can also be used by youth workers looking for additional ways to foster positive adolescent identity development.

6. Conclusions

Although there are limitations to this study, results offer evidence that family history knowledge may be a potentially powerful, underutilized resource for promoting positive adolescent identity development. Decades of scholarly work have been devoted to understanding identity development (Soenens and Vansteenkiste 2011). Healthy individual identity development includes an important dosage of family identity (Adams 1998; Adams et al. 2006; Erikson 1959, 1968; Galliher et al. 2017; Marcia 2002). Those without a family identity component in their individual identity can lack connection and intimacy, important elements of positive identity and well-being (Soenens and Vansteenkiste 2011; Stutman and Lich 1984). This study provides new insight into how family identity development (facilitated through family history knowledge) might aid individual identity development. However, to have a positive influence, family history knowledge may need to be facilitated in a context of warmth and mutual respect and in a way that promotes personal exploration by encouraging and allowing adolescents to derive personal meaning while encouraging the free expression of their insights, conclusions, and perspectives.

We hope this new insight, part of an ongoing process to discover how family history knowledge can be utilized to strengthen individuals, families, and communities, will inspire others to pursue the subject matter and contribute to further understanding of how this vast genealogical resource can be utilized for positive human development.