WHat are important quotes from the text? Introduction The ecological theory of human…
- User
WHat are important quotes from the text? Introduction The ecological theory of human development has mixed findings which suggest adolescent work has both negative and positive impacts on adolescent development (Zimmer-Gembeck & Mortimer, 2006). The mixed findings may partially be due to the research ignoring an important source of work context heterogeneity which has theoretical reasons for being quite different—the family firm. Working in a family firm at a young age, as opposed to a nonfamily firm, likely influences adolescent development differently. This is a critical theoretical gap in the literature as the prevalence of family firms around the globe, including those of developed countries (Arregle, Hitt, Sirmon, & Very, 2007; La Porta, Lopez-de-Silanes, & Shleifer, 1999), indicates that a large number of parents have the option of hiring their own children, especially their own adolescents in their own business. One important decision these parents make is whether to involve the next generation early on in the family business. Family legacy has a significant influence on the decision-making process of family firms (Hammond, Pearson, & Holt, 2016) such that the choosing and developing of a successor, while considering parent– child relationship quality, is one of the most important decisions that family business owners make (Cater & Justis, 2009). Care for the next generation prompts many family owners to hire their adolescents into the business. In developed countries, many adolescents end up working in their family firms at a young age. For example, according to nationally representative data gathered from the Youth in Transition Survey (Statistics Canada, 2012), 45% of Canadian adolescents who were 15 years old as of December 1999 reported having worked in their family business at one point. An important question thus emerges about the consequences of working in the family business on adolescents.With such a high percentage of adolescents in society having worked in a family firm, and given adolescence is a critical stage in developing one’s identity (Mortimer, 2003; Mortimer, Lam, & Lee, 2014), we must better understand the implications of adolescents working in their family business on the adolescent’s psychological well-being and relationship with parents. There has been a debate in the family studies literature regarding the impact of general work on adolescent development and relationships with parents. While some argue this influence is harmful (Greenberger & Steinberg, 1986) noting that this kind of work may lead to stress and a lower level of family commitment (Steinberg, Grennberger, Ruggiero, Garduque, & Vaux, 1982), others posit positive outcomes, such as a greater sense of independence (Mortimer & Shanahan, 1994). Recently, the literature has started to recognize the heterogeneity of outcomes among adolescents on the transition into the working world. Vuolo, Mortimer, and Staff (2014) recently started to disentangle these trajectories in terms of work identity outcomes categorizing the outcomes into “successful” and “floundering” based on a variety of characteristics including work experiences during transitional years. Mortimer and her colleagues (Mortimer, 2003; Mortimer, Kim, Staff, & Vuolo, 2016) start to introduce psychological outcomes such as self-efficacy, finding that “high-quality” employment is linked to increased self-efficacy during these transitional years. The impact on broader psychological well-being and the relationship with parents still needs further disentanglement. Additionally, these studies have been limited to general business work placements, and the dynamics are likely to differ when working with family members due to the inherent nature of the business family overlap. Working with family members is likely to affect the quality of the work experience dramatically, quite possibly providing higher quality employment experiences overall for adolescents. The psychological and relationship impacts of adolescent work become more complex when the work context is the family firm, where familial relationships are particularly affected due to the overlap of family and work settings. Empirical evidence in general settings suggests work has a negative influence on parent–adolescent relationships. One key explanation highlighted in the family studies literature is that working reduces the amount of time adolescents spend with their families (Ashbourne & Daly, 2010). However, in the context of the family firm, adolescents are able to spend more time with their parents at work and be exposed to a unique work environment. Since the underlying theoretical time overlap assumptions are different in the family business context, we expect that in the family business context, the impact of adolescent work may be different than suggested in the general work context typically used in the family studies literature. In other words, working in a family business may be an important source of heterogeneity in the quality of the work experience. This study is the first of its kind to explore this important relationship and make a direct comparison between adolescents who work for nonfamily firms with those who work in their family business. The specific research question we explore in this article centers on the consequences of adolescents working in their family business. We focus our attention on how working in this context influences adolescent development and the parent–child relationship differently. Finally, for those children who are likely to find themselves working in their family-owned firm, we empirically test the best strategy for improving psychological and family outcomes for adolescents. We construct the article as follows. First, drawing on ecological theories of human development, we briefly theorize about the consequences of an adolescent working along with their parents on two important outcomes: parent–adolescent relationships and adolescent wellbeing. We then use the National Longitudinal Survey of Children and Youth (NLSCY), a unique Canadian longitudinal data set well suited to shedding light on this question. To ensure the robustness of our findings, we use different operationalizations of our primary variables of interest and, due to the mixed definitions of family business in the literature, define a variety of business family risk sets. Finally, we discuss our findings and the managerial implications for family firm managers. The Consequences of Adolescents Working in Their Family Business The ecological theory of human development (Bronfenbrenner, 1979) helps understand the spillover between work and family settings and how experience in one setting influences the experience in another setting. The theory, specifically, delineates different types of systems that shape individual development. The lowest level is the micro system in which individuals function in different settings such as school, family, and work. The micro system is concerned with how ties developed in these settings shape the individual and ultimately his or her experience. The next system is the meso system in which the settings found in the micro system are assumed to interact with each other. Their interactions make it possible for relationships formed in one setting to influence the individual’s experience in other settings (Bronfenbrenner, 1979). These settings become more interconnected within the context of family firms. Family businesses have unique characteristics that entail both family and business life and require more nuanced theorizing to study (Chua, Chrisman, & Steier, 2003; Gersick, Lansberg, Desjardins, & Dunn, 1999; Sirmon & Hitt, 2003). They possess distinct traits from nonfamily firms such as the influence of family relations on simultaneous roles and shared identity (Tagiuri & Davis, 1996). This overlap of both roles and identity creates a need to modify broader theory when considering such contexts. The decision-making process of parents as the owners and managers of family firms to a significant extent is shaped by their intention and preference in perpetuation of their family dynasty (e.g., Jaffe & Lane, 2004; Lambrecht, 2005) and keeping family legacy (Hammond et al., 2016). Families have a varying degree of preference for different types of legacy—biological, material, and social. All these types of family legacy could serve as a source of motivation for involving the next generation in the business. However, given those families with a higher biological legacy orientation have a preference for passing the business through the family’s bloodline, they are more likely to consider employing their adolescents in the business. Prior studies suggest that concerns for keeping biological legacy influence how leaders of family firms make strategic decisions (e.g., Gersick, Davis, Hampton, & Lansberg, 1997; Hammond et al., 2016). We, therefore, deduce that the business decision to hire familial adolescents implies parents and adolescents spending more time with each other as well as adolescents working in a unique environment. In the family business context, parents as managers and owners pay attention to providing adolescents with a different type of working experience than their peers would receive from working for strangers. As Hammond et al. (2016) noted, “parental altruism, which represents a trait linking the parents’ welfare to their children’s welfare, is grounded in the notion that parents view their children as an extension of themselves” (p. 1216). Under such context, as family business owners, while spending more time with their children, parents also have incentive to provide their adolescents a unique working environment that in turn affects the adolescent’s relationship with them as well as the adolescent’s psychological well-being. Parent–Adolescent Relationship Adolescence is a critical stage in one’s life (Hall, 1904) and the work experience during this phase conveys a developmental importance both in the short and long run (Mortimer, 2010). An important consequence of work is the mark it leaves on the type of relationship one develops with her parents. Family studies researchers have studied this relationships by making an underlying assumption that adolescent work takes time away from family time (e.g., Ashbourne & Daly, 2010; Steinberg & Dornbusch, 1991). In fact, many of the studies in the literature highlight the importance of the role time plays in the parent–child relationship. Time is a tool families use to craft and shape family ties (Gillis, 1996). In families where both parents and children work, family members may be forced to sacrifice spending time with each other to meet other demands arising from work obligations, which may give rise to work–family conflict for both parents and adolescents. The value of time becomes even greater for adolescents, compared with younger children, as they have less access to family time due to their external activities such as staying out late with friends and working at jobs (Larson, Richards, Moneta, Holmbeck, & Duckett, 1996). While these findings are not conclusive, some studies suggest that reducing the amount of time family members spend with one another may weaken family ties (e.g., Mortimer & Shanahan, 1994; Steinberg & Dornbusch, 1991). The family–work literature also posits that family and work domains influence each other. While empirical evidence supports the notion that the number of work hours is negatively related to work– family balance (Valcour, 2007), it is driven by the tradeoff between working and spending time at home that causes distress to individuals rather than the absolute number of working hours (Barnett & Gareis, 2000). Furthermore, to the extent to which work interferes with family, the family provides less emotional support (Adams, King, & King, 1996)The reduction of time adolescents spend with their families, not surprisingly, is highlighted as one explanation for more negative outcomes associated with adolescents working (Ashbourne & Daly, 2010). These studies, however, do not take into account the nature of the relationship of the employer with the adolescents. While many of these studies highlight the importance of the role time plays in the parent–child relationship, they do not differentiate between who the adolescents are spending time with at work—a stranger or a relative. In family firms, spending time at work or with family is not mutually exclusive. We therefore expect that in the family firm context, the impact of working on adolescent behaviors may be different than currently suggested in the family studies literature. Within the context of family firms, the involvement of adolescents in the business carries its own weight. Family businesses are viewed by many family members as a form of collective wealth that provides utility to the members and while contributing to it is usually encouraged, disengagement from it may be frowned on for both financial and emotional reasons (Eckrich & Loughead, 1996). Because family members view working together and spending time together as advantageous to running a family firm (Cole, 2000), they may be disappointed when some members choose not to engage fully in the firm. Moreover, family firms count on the human capital represented by their members; and given that the size of families is shrinking, these resources have become more valuable (Aldrich & Cliff, 2003). Those family members who do not facilitate running the business may damage their relationships with prominent family members, in this case their father and mother. Part of this unique work experience is that the nature of work contribution also differs for adolescents who work in their family firms than those who work for other people. Those who work for their family firms directly contribute to their family business and hence their relationship with their parents may be more directly and positively influenced. The work setting also allows adolescents to establish new types of ties, in addition to familial ones. Not only can they learn more about their family firm and gain appreciation for the work their family puts into the business they might be able to better express themselves in new dimensions (Ward, 2004). Finally, adolescents who work in their family firms are exposed to unique work experiences that shape them differently than their counterparts. For example, one study showed that those who work in their family business are more likely in the future to enjoy richer career networking and select themselves into organizations with more fitting characteristics than other adolescents who work for a nonfamily member (Houshmand, Seidel, & Ma, 2014). The experience in the work role could spill over to the family role through both the affective and instrumental paths which makes the family role more positive or negative depending on the experience (Greenhaus & Powell, 2006). Consequently, we expect a richer work experience and a positive working relationship between parents and adolescents at work to favorably influence general parent–adolescent relationships. Spending time with family members coupled with the other reasons discussed previously such as establishing new types of connections, finding new avenues to express themselves, and contributing to the family business all suggest that adolescents develop better relationships with their parents when they work consistently in their family firms. Hypothesis 1: Adolescents who have year-round involvement in their family firms report a better relationship with their parents than all other adolescents. Psychological Well-Being Work that is dissatisfying or stressful can impair the psychological well-being of adolescents. Stress and depression in the workplace is often the result of feeling responsible for outcomes beyond one’s control (Shanahan, Finch, Mortimer, & Ryu, 1991). Work that interests and challenges the worker can lead to positive psychological outcomes. For example, boys reported a greater sense of well-being when assigned interesting and complex work; girls reported greater self-esteem and well-being when they perceived school and work settings to be mutually supportive (Mortimer, Finch, Shanahan, & Ryu, 1992). Parents are altruistic toward their children and family members of a family business also display altruistic behavior toward each other (Schulze, Lubatkin, & Dino, 2003). Families, therefore, have incentives to provide positive and developmental working conditions for their own children. To keep family and family firm legacy, parents groom the next generation as their business successors (e.g., Hammond et al., 2016). We thus argue that in family firms, adolescents are more likely to have access to developmental work assignments than in situations where they work for strangers. They are also more likely to receive support from other settings such as school and family for the work they are doing at their family firms. A study of children growing up in the Great Depression concluded that those who helped their financially troubled families by either working within or outside the home had a higher level of self-confidence than those who did not have the same opportunity (Elder & Rockwell, 1979). In the same vein, we predict that due to the better working conditions and the ability to financially help their family, those adolescents who have year-round involvement in their family firms develop greater psychological well-being. Taking into account the role time plays, compared with their younger siblings, adolescents have fewer opportunities to spend time with their family and may also face work–family conflict due to their scarcity of time; which becomes divided among school, friends, work, and family (Larson et al., 1996). Higher level of work–family conflict is positively associated with psychological distress and depression for adults (Frone, 2000; Hammer, Neal, Newsom, Brockwood, & Colton, 2005). Similarly, adolescents facing time constraints and higher levels of work–family conflict may experience lower psychological well-being. Working adolescents in family firms, however, have more opportunities to spend time with their families than their counterparts. They, additionally, have more spare time to allocate to other settings such as social and school. Therefore, these adolescents are better able to handle environmental demands and experience less distress. When parents hire their adolescents into the business, they provide a unique work environment to adolescents. The unique work environment stems from the familial ties these young workers have with their employers (Houshmand et al., 2014) who are more likely to encourage the youngsters to engage in more interesting and complex tasks than other potential employers at odd jobs elsewhere. This positive experience in the work role could spill over to other settings (Greenhaus & Powell, 2006). Moreover, while all adolescents are pulled by many external forces in different directions, those who work in their family firms are more likely able to manage their family time more efficiently by having the opportunity to spend time with family at work. Based on this logic, we argue that adolescents who have year-round involvement, defined as both during the school year and summer, in their family firms have higher psychological well-being than their peers. Hypothesis 2: Adolescents who have year-round involvement in their family firms report higher psychological well-being than all other adolescents. Method Data and Sample We identified a longitudinal data set well suited to exploring our questions called the NLSCY. The NLSCY was collected by Statistics Canada and Human Resources and Skills Development Canada. We used the longitudinal weights provided by Statistics Canada to account for attrition between cycles, which ensures the models are representative of the targeted population. As recommended by Statistics Canada, we conducted Stata BRR (balanced repeated replication) bootstrapping for all of our models. Employing this method helps ensure that our sample represents the target population and allows us to draw more generalizable conclusions from the results. The NLSCY focuses on child growth development from an ecological perspective. According to NLSCY user guides (e.g., Statistics Canada, 2010), survey content including subject areas, priorities, and survey questions were determined by (a) the NLSCY expert advisory group, which consists of researchers in the area of child development and the social sciences, (b) federal departments, and (c) representatives responsible for child development programs in the provinces and territories. As a result, several prevalidated scales were collected; and we predict two in our current study (Marsh & O’Neill, 1984; Radloff, 1977). For the purpose of this study, we used NLSCY data spanning 8 years, from when our subset of adolescents, representing over 570,000 Canadians who were 8 to 9 years old in 1994-1995 to when they were 16 to 17 years old in 2002-2003. In this study, we tested our theory using our primary independent variables of interest when adolescents were 14 to 15 years old to predict outcomes in later years, that is, when they were 16 to 17 years old. We also used various control variables from as early as ages 8 to 9, as detailed below. Measures Dependent Variables. Our questions lead to two sets of dependent variables for this study. We theorized that adolescents who have year-round involvement in their family firms have a better relationship with their parents and have higher psychological well-being than all other adolescents. For robustness, we operationalized these constructs in multiple ways as follows. General relationship with parent(s). We used a number of measures to capture the parent–adolescent relationship construct from different perspectives: the adolescents’ perception of how much parental affection they receive, how much the adolescents feel their parent(s) understand them, and the overall adolescent impression of their relationship with their parent(s). The questions pertaining to fathers and mothers were asked separately. Three measures captured adolescents’ perceptions of their relationship with each applicable parent. The questions were phrased as follows: (a) Affection— How much affection do you receive from your (mother/ father)? (b) Understanding—How well do you feel that your (mother/father) understands you? (c) Overall Relationship—Overall, how would you describe your relationship with your (mother/father)? The affection, understanding, and overall relationship variables were each coded with three ordered outcomes: 1 for very little/not at all, 2 for some and 3 for a great deal. We used a latent construct approach to measure parent–adolescent relationship. First, we created a general measure for relationship with father and a general measure for relationship with mother. Cronbach’s alpha for the father measures are .85 and .84 when adolescents are 14 to 15 years old and 16 to 17 years old, respectively. For the mother measures, they are .80 and .81. Although these measures originate from the NLSCY, the internal consistency of the scale is high. The measures were determined by domain experts and continue to be adopted in research regarding relations with parents (e.g., Myers, 2005) as well as caregivers (e.g., Flynn, Robitaille, & Ghazal, 2006). Furthermore, these studies exhibit similarly high levels of scale reliability. Our approach to this variable is similar to Bushnik’s (2005) article examining youth depressive symptoms and changes in relationships with parents and peers in that we use the same three parent measures.1 However, we then combined the measures as a single broader construct of overall relationship with parents by averaging all valid parent–adolescent relationship components (i.e., three components for single-parent families and six components otherwise). As a robustness check, we found that results using the broader construct were consistent with the separate measures for father and mother,2 and as such present the results using this broad latent construct for our sample measured both when adolescents are at ages 14 to 15 (T1) and again at the ages 16 to 17 cycle (T2). Psychological well-being. To capture adolescent psychological well-being, we used two measures, one of depression, the other of self-esteem. We measured depression with a version of the depression scale developed by Radloff (1977) with a Cronbach alpha of .82 based on T2 data (when adolescents were 16 to 17 years old). The scale sums 12 items, each ranging from 0 (infrequent prevalence) to 3 (frequent prevalence), to capture occurrence of depression symptoms as indicated by respondents within a 1-week span. The original depression scale, which comprised 20 questions, is also known as the Center for Epidemiologic Studies Depression Scale and continues to be widely used to measure symptoms of depression. The NLSCY’s reduced 12-item scale ranges between 0 and 36, with a higher score representing greater presence of depression symptoms. The full schedule of items is in Appendix A. Statistics Canada collected the depression scale in T2 (when our subset of adolescents were 16 to 17 years old), but not in T1 (when our subset of adolescents were 14 to 15 years old). Our second measure of psychological well-being is an index of self-reported self-esteem. Four self-esteem items were taken from the Marsh Self-Description Questionnaire (SDQ; Marsh & O’Neill, 1984) with a Cronbach alpha of .73 based on T1 data. The full-schedule Marsh SDQ measures multiple dimensions pertaining to an individual’s self-concept. We operationalized self-esteem using items from the SDQ’s General-Self Scale. The Marsh SDQ, including its underlying dimensions and (re)iterations (e.g., SDQ-I, SDQ-II, SDQ-III, and reduced versions thereof), are well established and have been applied broadly. The scale utilizes the sum of four items, each ranging on a Likert-type scale from 0 (mostly false) to 4 (mostly true), to denote general selfesteem. The scale ranges from 0 to 16, with a higher score indicating higher self-esteem. The full schedule of items is in Appendix B. The scale was collected in both T1 and T2 for our subset of adolescents. Independent Variables. When adolescents were 14 to 15 years old, they were asked what type of job they had during the previous summer, and since the beginning of the school year. Qualifier questions ascertained whether they had held any jobs during those two periods. Later, questions were designed to narrow down the specific type of employment. The questions were structured to elicit yes or no responses. For example, “Last summer, did you do any work: at your family’s farm or business (with or without pay)?” or “Since September 1, have you done any work: at your family’s farm or business (with or without pay)?” We categorized those respondents who answered “yes” to both the above questions as year-round family firm workers and distinguished them from other adolescents. We used this measure for all of our analyses to compare those adolescents who work in their family firm on a year-round basis with all other adolescents. The year-round engagement in the business, we believe, allows for the development of better relationships with parents as well as developing better psychological well-being. Control Variables. We controlled for general individual attributes such as gender, age, and race. Female is a dichotomous measure of gender, with male respondents coded as 0 and female respondents coded as 1. Age is a continuous measure of the respondents’ actual age at the time of their T1 interview and can differ from their survey age. That is, although the survey age at T1 is 14 to 15 years, the actual range for this variable spanned from 13 to 16 years. Though most respondents were either 14 or 15 years old at T1, interviews were not conducted on the same day corresponding to the day respondents were surveyed in the first cycle of the NLSCY (i.e., when they were 8 or 9 years old). Because respondents are not surveyed on the same day each time, a few will be 1 year younger or older from their formal survey age. Our third general individual attribute variable is race. White is a dichotomous measure of race, with White respondents coded as 1 and non-White respondents coded as 0. Number of siblings has been associated with perceived parent–adolescent relationships (Kidwell, 1981) and research shows that socioeconomic status (SES) influences children’s health, cognitive, and socioemotional outcomes (Bradley & Corwyn, 2002). As such, we also included four family-level variables to control for differences across families—number of siblings, single-parent family, parent(s) average age, and SES. All independent and control variables were identified when adolescents were 14 to 15 years old except for the SES and single-parent family variables, which were identified when adolescents were 12 to 13 years old. The latest cross-sectional SES captured in the survey for our age group was based on the SES of up to both parents from when the adolescents were 12 to 13 years old (Statistics Canada, 2010). Number of siblings was coded as a count of the number of siblings belonging to a respondent. The single-parent family variable was also dummy coded as 1 for families with only one parent, and 0 for two-parent families. The household SES variable was derived using up to five variables: household income (in $1000s), years of education (for up to two parents), and occupational prestige (for up to two parents) using a modified version of the scale developed by Pineo, Porter, and McRoberts (1977). The SES variable was only coded for two-parent families that answered at least 3 of the 5 questions, and for single-parent families that answered at least two of the three questions posed to them. Statistics Canada, in their coding of the SES variable, makes a final transformation by standardizing all components of the SES variable, and taking the mean of all available components (Statistics Canada, 2002). Two versions of the SES variable were derived in the NLSCY: longitudinal SES and cross-sectional SES. The differences between the two are minor; however, the former is recommended for examining within-household SES changes, while the latter is suited for between-household differences in SES (Statistics Canada, 2002). Therefore, we use the cross-sectional version of the SES variable during the time that respondents were 12 to 13 years old. As single-parent households naturally score lower on the SES scale, Statistics Canada recommends that researchers control for single-parent households when using the scale; and we followed this recommendation (Statistics Canada, 2010). In the same vein, we control for parent(s) average age because older parents are more likely to have higher SES (Statistics Canada, 2002). Additionally, we included a dummy variable for working in general in our models to tease out the difference in outcomes for those who have work experience. Those who worked at the ages of 14 to 15 years either during the summer or/and school year received a score of 1 on this variable, while those who did not were coded as 0. All respondents who indicated working for their family firm for any part of the year were therefore also coded as 1 along with those who had general work experience. Including this control variable allowed us to differentiate the effect of working in family firm from the effect of general work. In other words, because we have variables for family firm work as well as general work, any family firm work effects can be interpreted as occurring above and beyond that of general work effects. Indeed, the motivation for including this variable is central to our argument that year-round family firm involvement has benefits beyond those associated with working in general (e.g., Houshmand et al., 2014). While many family farms are family businesses, they possess unique characteristics that might be different than nonfarm family businesses (Astrachan & Shanker, 2003; Dumas, Dupuis, Richer, & St.-Cyr, 1995; Steier, 2001). Hence, we controlled for whether the family firm is farm family business or nonfarm family business. To differentiate family farm workers from other types, we created two dummy variables, one to capture whether at least one parent worked in agriculture and one to capture whether the family lived in a rural or urban region. We observed that both of these groups were more likely to be employed as year-round family workers than other adolescents. We therefore included the two dummy variables as control variables in our models. Finally, we control for a lagged measure of general relationship with parent(s) from when adolescents were 12 to 13 years old (T0). The coding of this variable is identical to our first dependent variable, with the distinction that it is coded from an earlier time period and used as a control for potential selection bias. That is, the likelihood for adolescents with better parent–adolescent relationships to self-select into the family firm.3 Risk Sets To rule out alternative explanations of the positive relationship between year-round family firm involvement and our dependent variables, we analyzed three samples of data. The first sample examines year-round family firm involvement compared with all other arrangements. However, it remains possible that our predicted yearround family firm involvement effects stem from family firm properties as opposed to actual involvement in the family firm. An alternative explanation for any significant effects can be attributed to the different properties associated with growing up in a family business environment. As a result, we used two subsamples of the first sample. These risk sets help address concerns around these alternative explanations, by examining adolescents that have the opportunity for—or, are at risk of— family firm involvement. In our initial models, we compare adolescents who have year-round work involvement in their family firm to all other adolescents by testing the models on a full population sample representing a population of roughly 500,000 adolescents. In addition to building models of the full population, as a robustness check to rule out selection bias explanations and also to unearth interesting distinctions between children of family firms, we defined two specific additional risk sets to identify those adolescents in our sample that potentially could have had an opportunity to work in their family firms. The first risk set consists of those who reported working for their family business or farm at the minimum of either in the summer or during the school year when they were between 14 and 15 years old. This set represents roughly 115,000 Canadians who were 14 to 15 years old in 2000- 2001. For this set, we compared our independent variable of year-round work involvement which is defined by working for the family firm in both time periods, against partial involvement in either fall or summer. By repeating the same analysis, we conducted using the full population on this particular risk set, we were able to tease out the effect of year-round involvement in family firm from partial involvement. The second risk set was those who had at least one parent who was self-employed. Being self-employed is not equivalent to running a family firm. However, this risk set serves as a proxy for adolescents who have access to working along with their parents when their parents are self-employed. This set represents roughly 120,000 Canadians during the same period. For this group, we compared year-round family firm work involvement with everyone else in the risk set. Results We used ordinary least squares regression in accordance with the provided design weights to examine both parts of our theory (i.e., the effect of year-round family firm involvement on adolescent relationship with parents, and the effect of year-round family firm involvement on adolescent psychological well-being). Ordinary least squares regression is appropriate because our dependent variables are continuous, while utilizing Statistics Canada’s provided design weights allows us to properly account for attrition and generalize our results to our population of interest (i.e., Canadians that were aged 8 to 9 years in 1994-1995). The design of the survey is such that a unique set of weights are provided for each time period. As a result, our modeling choices are limited and we develop separate models to predict our dependent variables for each time period. As part of Statistics Canada’s privacy restrictions, we are not allowed to report the exact sample size or correlations between raw variables. We followed Statistics Canada guidelines for researchers utilizing individual-level data through the secure Research Data Centre program and provide details on the target population weighted appropriately. For our analysis, the target population covered approximately 500,000 people. We evaluated the variance inflation factor values of all variables in all models and found no significant multicollinearity problems (variance inflation factor < 1.28). Table 1 presents the full results of the regressions on general relationship with parents captured when adolescents are 14 to 15 years old (T1), and again when they are 16 to 17 years old (T1 + 2 years later). The first model includes only the control variables, whereas the following model adds the year-round involvement in the family firm dummy variable. In the subsequent set of models, which look at our dependent variable in T2, we similarly start with a model with only control variables, then add our independent variable of interest. We highlight the year-round involvement in the family firm variable in bold. Tables 2 and 3 present the results for the risk sets. Table 2 illustrates the findings for those adolescents who had indicated that they had some experience working in the family firm during that year, whereas Table 3 depicts those who had at least one self-employed parent. As seen in these tables, when adolescents are 14 to 15 years old, year-round involvement in the family firm generally has a statistically significant positive relationship with adolescents’ general relationship with their parents. Our risk set analyses in Table 2 show those that work for their family firm throughout the year report having a better relationship with parents, compared with those that work for family during only part of the year. Similarly, as shown in Table 3, the effects are positive and significant when looking at the set of respondents who have at least one self-employed parent showing those who have year-round involvement in their family firm experience better relationship with their parents than those who have a self-employed parent but do not have year-round work involvement at the family firm. Two years later, we find a consistent pattern. In the full population, general parental relationships involvement in their family firm. Looking at the size of the coefficients, we do observe an increase from T1 to T2 in all the models of full population and risk sets. By taking this into account, we infer that while working in a family firm appears to be a reliable way to initiate a rich parent–adolescent relationship, the positive relationships with parents continue to strengthen into the future. Together, these findings provide support for Hypothesis 1 that year-round involvement in the family firm leads to better relationship with parents. To gain perspective on the psychological outcome part of our theory, we used two measures: self-esteem and depression, and found empirical support for each. Table 4 presents the full results of the regression on depression. Depression data were collected by Statistics Canada in T2 when adolescents were 16 to 17 years old. As such, we show the analysis for T2 when adolescents were 16 to 17 years old. Self-esteem, on the other hand, was measured in both T1 and T2. Tables 5 to 7 present the results on self-esteem for the full sample and risk sets, respectively. We find consistent statistically significant effects for both measures when working year round in a family firm in T1, though for T2 only. There is only a marginally significant (p < .10) effect for working year round in the family firm in the full population model predicting self-esteem in T1. Both risk sets predicting self-esteem in T1 do not have statistically significant effects. These combined results could suggest that working in a family firm has positive and proximal (i.e., during T1) effects on self-esteem independent of the level of involvement (i.e., year-round or partial); although the evidence for such an effect is only marginal. Working year round in the family firm shows a negative relationship with depression, and a positive relationship with self-esteem, during T2. This pattern of results for the self-esteem models (significant effect in T2, lack of a consistent effect in T1) suggests that there may be a latent effect for working in the family firm year round at the ages of 14 to 15 years that manifests over time. The results of the models predicting general relationship with parents, with their larger effect sizes in T2, also supports the interpretation that year-round involvement of adolescents in the family firm can establish a gradually manifesting path toward desirable outcomes including a better relationship with parents as well as psychological well-being. Together, these findings provide support for Hypothesis 2 that year-round involvement in the family firm yields higher psychological well-being.
- Assistant
- "Working in a family firm at a young age, as opposed to a nonfamily firm, likely influences adolescent development differently."
- "Teenagers who work for their parents may have a better relationship with them compared to those who work for strangers."
- "Adolescents who have year-round involvement in their family firms report better relationships with their parents than all other adolescents."
- "Year-round involvement in the family firm leads to better relationships with parents and higher psychological well-being for adolescents."
- "Working in a family firm appears to be a reliable way to initiate a rich parent-adolescent relationship."
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.