The Relationship Among Purpose Classification, Purpose Engagement, and Purpose Commitment in Low Socioeconomic Status and Ethnically Diverse Adolescents

Martin Seligman, who is commonly referred to as the father of positive psychology, asks the roomful of teachers two simple and important questions: “In two words or fewer, what do you most want for your children in life?” Several audience members chime in: “happiness;” “love;” “purpose.” Seligman then asks, “In two words or fewer, what do schools teach?” A few more teachers rattle off: “compliance;” “facts;” “numeracy;” “literacy.” And then, Seligman delivers the punch line: “Notice, there is no overlap between the two lists” (Character Lab, 2016). Many adults would agree that while they want their children to have a purpose in life, our schools generally do a poor job of cultivating this virtue. Given the gap between the desire to educate for purpose and the present reality, it is worth examining what is meant by purpose, what makes a purpose in life worth having, what youth purpose currently looks like, what allows for a strong commitment to purpose, and how purpose might be assessed.

Bronk’s (2014) review of the purpose literature concludes that the majority of definitions of purpose consist of three “irrefutable components” (p. 6): commitment, goal-directedness, and personal meaningfulness. For instance, McKnight and Kashdan (2009) conceptualize purpose as, “A central, self-organizing life aim that organizes and stimulates goals, manages behaviors, and provides a sense of meaning” (p. 242). Damon, Mennon, and Bronk (2003) define purpose as, “A stable and generalized intention to accomplish something that is at once meaningful to the self and of consequence to the world beyond the self” (p. 121). Taken together, these definitions and Bronk’s (2014) review describe purpose as a personally meaningful aim that provides a sense of direction in life beyond “low-level goals such as ‘to get to the movie on time’” (Damon et al., 2003, p. 121). Given this helpful starting point for thinking about purpose, the present study defines purpose as a personally meaningful and enduring aim that provides an individual with a nonharmful direction in life. While this definition can include a beyond-the-self dimension, as alluded to in the Damon et al. (2003) definition of purpose, this definition can also include self-oriented purposes that are not intentionally harmful to other people, such as a purpose to become a professional basketball player.

The Holocaust survivor and psychiatrist, Viktor Frankl, made a compelling case for the importance of purpose in life with the publication of his influential book, “Man’s Search for Meaning” (1959/2006). Frankl believed that a purpose could propel someone to overcome dire circumstances, even as extreme as the Holocaust. He wrote:

Frankl’s personal and powerful account of the benefits of purpose helped pave the way for much of the empirical research on purpose that has ensued (Damon et al., 2003).

The research on purpose demonstrates positive associations with mental health, including hope and life satisfaction (Bronk, Hill, Lapsley, Talib, & Finch, 2009; Burrow, O’Dell, & Hill, 2010), as well as serving as a protective factor against depression in adolescence (Brassai, Piko, & Steger, 2011). Purpose is also associated with better physical wellbeing, including longevity and reduced rates of mortality (Boyle, Barnes, Buchman, & Bennett, 2009) and lower rates of Alzheimer’s disease (Boyle, Buchman, Barnes, & Bennett, 2010). Studies with adolescents have demonstrated associations between purpose and academic and social outcomes. Specifically, an intervention designed to build students’ purpose contributed to academic achievement, including increases in grade point averages (Pizzolato, Brown, & Kanny, 2011). Purpose has also been associated with lower rates of self-reported violence (Durant, Cadenhead, Pendergast, Slavens, & Linder, 1994) and maladaptive coping behaviors like substance abuse (Minehan, Newcomb, & Galaif, 2000). Recent studies with college-aged students have also found positive associations between purpose and the psychologically adaptive constructs of grit (Hill, Burrow, & Bronk, 2016) and self-efficacy (DeWitz, Woolsey, & Walsh, 2009), both of which are associated with persistence.

The few studies that have documented the types of purposes and meanings people maintain, based on free response (Ebersole & DeVogler, 1981; Hill, Burrow, O’Dell, & Thornton, 2010) and semistructured interview formats (Damon, 2008), have demonstrated that it is common for teens and early adults to have multiple purposes. Hill et al. (2010) observed that among a sample of adolescents (N = 229) from one Catholic high school and one suburban public high school, the majority endorsed more than one purpose. Several of the purpose categories in this study included “happiness,” “religion,” and “occupational and financial.” Specifically, 43% of the sample endorsed two categories, 21% endorsed three categories, 4% endorsed four categories, and only 30% endorsed one category. When Ebersole and DeVogler (1981) asked college students (N = 112) to elaborate on and categorize their strongest meaning in life, 21% categorized their meanings as “miscellaneous” because they tried to “jam several meanings into their single essay” (p. 293). Notably, these studies, and the purpose literature in general, do not tend to focus on marginalized populations, including students from low-income, urban, and ethnically diverse households (Sumner, Burrow, & Hill, 2018), which suggests that these observations about purpose are somewhat limited.

Damon (2008) also speaks to this point of multiple purposes, observed during his long and storied career interviewing and studying adolescents. But he goes a step further to distinguish between “parallel” (p. 69) purposes, “purposes across several … fronts” (p. 69) (which will be referred to as “integrated” purposes from this point forward), and “one burning purpose” (p. 69) (which will be referred to as “one purpose” from this point forward). To clarify, people who maintain parallel purposes have multiple long-term intentions that do not reinforce each other and can lead to conflict between purposes, whereas people with integrated purposes have long-term intentions that intersect and reinforce each other. Bronk (2014) essentially describes an integrated purpose and support for categorizing purposes in this way when she writes the following:

An integrated purpose is clearly a phenomenon that has been observed in the field, yet its frequency and clinical implications have not been clearly documented.

Purpose engagement, defined as a “behavioral manifestation of the purpose” (Moran,2009, p. 145), appears to be fairly common and seems to increase with age. Among a geographically diverse sample of American youth (N = 270), 50% expressed engagement with either a self-oriented or beyond-the-self purpose (Moran, 2009). In this cross-sectional dataset, engagement increased with age in that 25% of the sixth graders, 34% of the ninth graders, 56% of the 12th graders, and 84% of the college students expressed engagement with either a self-oriented or beyond-the-self purpose (Moran, 2009). A follow-up study that consisted of interviewing 146 students from the original study 2 years later (Quinn, 2016), generally showed a similar upward trend in engagement. The exceptions included the 11th graders, who remained stagnant between the two time points, such that 37.5% reported engagement with a self- or other-oriented purpose in ninth grade as well as in 11th grade; and the college students, who dropped several percentage points in level of engagement, from 83.3% to 81%. Approximately one fifth (21.7%) of sixth graders engaged with their purpose, whereas 2 years later as eighth graders, over one third (34.8%) engaged with their purpose. Slightly over one half (55.6%) of 12th graders engaged with their purpose, whereas 2 years later, the vast majority (88.9%) indicated engagement with their purpose. The fairly consistent trend supports the notion that engagement with purpose generally seems to increase with age, at least into early adulthood. In a sample (Mariano, Going, Schrock, & Sweeting, 2011) of sixth-, seventh-, and eighth-grade adolescent girls across six middle schools in an urban area in the southeastern United States (N = 46), 24% expressed engagement with either a self-oriented or beyond-the-self purpose, which is fairly consistent with Moran’s (2009) study. These numbers seem to indicate that roughly one quarter of middle school students in the United States would state that they act on their purpose.

The number of students who express any kind of purpose, regardless of level of engagement or classification of purpose, has been documented by Quinn (2016) in her sample of 146 students in sixth grade, ninth grade, 12th grade, and college who were interviewed at two time points over the course of 2 years. She looked at the number of students with an “intention,” which she defines as, “what the individual desires to accomplish through his or her life purpose” (p. 2). Similar to level of engagement, the upward trend in purpose tends to increase across cohorts. Most pertinent to the present study is Quinn’s (2016) finding that 41.3% of sixth graders expressed an intention, and this increased 2 years later to 52.2% when the students were eighth graders (Quinn, 2016). Mariano’s (2011) study of middle school girls (N = 46) again lends support to the numbers by demonstrating that 39.2% of her sample had an intention, which roughly resembles Quinn’s (2016) findings. Unfortunately, Mariano (2011) does not distinguish intention rates between grade levels and so a more granular comparison cannot be made.

There is reason to believe that having one purpose would be associated with a stronger commitment to purpose than having parallel purposes. Theoretical perspectives on purpose preference one purpose over parallel purposes, as the latter can lead to potential conflicts (Duckworth & Gross, 2014) and decreased effort (McKnight & Kashdan, 2009). In support of this argument, research on the related goal-oriented constructs of goal-differentiation and goal-integration (Sheldon & Emmons, 1995), as well as grit (defined as “perseverance and passion for long-term goals”) (Duckworth, Peterson, Matthews, & Kelly, 2007, p. 1087), suggest that purposes or goals that are connected with lower level goals and not hindered by conflicting or unrelated goals, help to explain greater goal commitment and success (Duckworth & Gross, 2014; Sheldon & Emmons, 1995).

Beyond the distinction between one purpose and parallel purposes, integrated purposes that touch on and reinforce multiple life areas and domains of purpose might allow for a stronger sense of identity and in turn, a stronger commitment to purpose than a singular purpose that is narrowly focused. In a longitudinal and in-depth multiple case study of adolescent purpose exemplars (N = 8), Bronk (2011) found that “purpose led to identity development, and identity development reinforced commitments to purpose” (p. 38). So, a key question is, what characteristics of purpose are critical for a strong sense of identity that will then reinforce purpose commitment? Bronk (2014) suggests an answer to this question in her review of the purpose literature when she hints that a purpose that intersects with multiple life areas (or domains of purpose) may be more indicative of a purpose that is central to identity, which can then reinforce one’s commitment to his or her purpose. She writes that purpose is “diffuse and influences many aspects of life, likely including career, hobby, and familial choices. In fact, a purpose can be so broad that it composes a central component of one’s sense of identity” (p. 109). In other words, the broader the purpose, the more it can influence identity, and the more it influences identity, the more it can increase commitment to purpose.

Commitment to purpose is also likely to be influenced by engagement with purpose. That is, if a student indicates present engagement with their stated purpose, it would be reasonable to predict that the student would have a stronger commitment to his or her purpose than a student who did not indicate present engagement. A study on purpose exemplars (N = 9) suggests that the exemplars “would not have discovered noble purposes in the areas they did had they not been involved in those areas early on, often as children…. Without exposure, noble purpose seems doubtful” (Bronk, 2012, p. 105). Similarly, a large-scale and representative sample of American youth living in suburban, rural, and urban areas, and ranging in age from 12 to 22 (N = 270) demonstrated that “lack of engagement [with their purpose] in the present is associated with vague statements for most of our sample; young people don’t talk as well about things that haven’t happened and may need exposure to concrete events” (Moran, 2009, p. 155). In short, engagement seems to be an essential step toward commitment. The aforementioned observations suggest that students who engage with a purpose that integrates multiple life areas (or domains of purpose) will likely indicate a stronger commitment to purpose than students who either do not act on their purpose or engage with a narrowly focused purpose or multiple and potentially conflicting purposes.

A variety of measures have been developed to assess for purpose. Damon et al. (2003) and Bronk (2014) provide helpful reviews of a range of measures, which include diary entries, self-report surveys, and semistructured interviews. Hill et al. (2010) have also published results from a free-response purpose essay, and Malin, Liauw, and Damon (2017) have designed a purpose scale with predefined purposes that students can select. There are advantages and disadvantages to using each of these measures.

Although a semistructured interview allows for students to elaborate on their purposes and level of engagement, interviews are also very resource intensive, requiring either many skilled interviewers or a few interviewers spending a significant amount of time interviewing.

Self-report purpose surveys are more efficiently administered and analyzed than semistructured interviews, but typically assess for sense of purpose without capturing the types of purposes that students have.

One advantage of a purpose essay is that it prompts students to write about purpose rather than waiting for the subject to spontaneously arise, as would be the case with diary entries. Another advantage of a purpose essay is that it provides room for elaboration about purpose and engagement, without confining students to describe particular categories of purpose or a limited number of purposes, as might be the case with a purpose scale. Additionally, an essay can be efficiently administered to an unlimited number of students simultaneously. Conversely, essays do not allow for follow-up and clarifying questions and they can be difficult to reliably code.

Malin et al. (2017) have provided a model for reconciling some of these concerns by supplementing a purpose scale that captures a predefined set of purpose categories, with a semistructured interview process. Yet time and resource constraints could still make this approach unfeasible for many research teams. It is clear that every approach has both its advantages and disadvantages.

Despite the presence of integrated purposes among adolescents (Bronk, 2014; Damon, 2008) and research indicating that integrated higher order goals are associated with greater goal commitment and grit than differentiated or conflicting aims (Duckworth & Gross, 2014; Sheldon & Emmons, 1995), the literature does not seem to provide insight into whether adolescents—particularly those from ethnically diverse and low socioeconomics status (SES) families—with integrated purposes are also more committed to their purposes than adolescents with one purpose or adolescents with parallel purposes. Furthermore, the literature does not indicate if these purpose classifications interact with level of purpose engagement to influence purpose commitment. As such, this study attempted to answer the following research questions: (1) Developmentally, how did eighth grade students compare to seventh grade students in terms of purpose classification, purpose engagement, and purpose commitment?; (2). Did strength of purpose commitment differ across purpose classifications (i.e., one purpose, integrated purpose, parallel purpose, no purpose)?; and (3) Was there an interaction effect between purpose engagement and purpose classification on strength of purpose commitment?

It was hypothesized that while the majority of seventh and eighth grade students would have some form of purpose classification, a plurality would have one purpose. Given that extant literature does not offer clear guidance on a 1-year grade level difference, this study explored whether eighth graders would be more likely than seventh graders to have integrated purposes, as well as more likely to have one purpose; whether seventh graders would be more likely than eighth graders to have parallel purposes, as well as more likely to have no purpose; and whether eighth graders would also be more likely than seventh graders to be engaging with their purpose and to be more committed to their purpose. Additionally, it was hypothesized that students with an integrated purpose would have the greatest commitment to purpose on average, followed by students with one purpose, which would be followed by students with parallel purposes, and lastly followed by students with no purpose. Finally, it was hypothesized that there would be an interaction between purpose engagement and the integrated purpose classification, as well as with the one purpose classification, whereas there would be an additive effect between purpose engagement and the parallel purpose classification.

The sample included 124 students in a public urban middle school in New Jersey. The sample was nearly evenly split between females (49.2%) and males (50.8%), as well as between seventh (49.2%) and eighth (50.8%) grade students. The majority (79%) of students qualified for free or reduced price lunch, an indicator of low SES. A plurality of the students was Hispanic (44.4%), although White (24.2%), Black (21%), and Asian (10.5%) students were also represented. The sample was drawn from one of the six schools in the study that fulfilled several criteria—a majority of the seventh and eighth graders completed a purpose essay assignment; the school was ethnically diverse; and the assignment was reliably administered and not altered. All three of these criteria were not met in the other schools.

Throughout the 2015–2016 school year, students were exposed to a socioemotional and character development curriculum that also included content on purpose in life. The students’ advisory teachers taught the curriculum for 15 minutes per day. Students in this sample completed a purpose essay prompt during the school day in March of 2016. The essay assignment was administered by the students’ language arts teachers and disconnected from the socioemotional and character development intervention. Several months later, toward the end of the school year, the students were administered self-report surveys, measuring various social and emotional skills as well as character strengths and purpose.

The purpose essay prompt that was administered to this study’s sample asked students to define purpose, to describe their own purpose and motivation for their purpose if they feel they have one, and to discuss how they might be engaging with their purpose. The prompt encouraged students to write their essays in a five-paragraph format, with an introduction, three body paragraphs, and a conclusion. This essay format was decided on in collaboration with the district middle schools’ language arts instructors, who reported that students would be expected to write essays in this format as part of their usual Language Arts curriculum, and that asking students to write such an essay could be requested as a typical assignment and not associated with the intervention program the students received.

A set of codes was developed for the purpose essays based on purpose category, purpose classification, and purpose engagement. The author, who was the designated master coder, initially trained four undergraduate research assistants to code for these areas. When reliability proved insufficient, the author made revisions to the codes and retrained one of the original research assistants. Subsequently, the author coded all 124 essays and aimed to reach 80% agreement with the newly trained research assistant, on 20% of the essays (n = 25) that were randomly selected.

To determine how each essay should be classified, the coders first determined the purpose category (or categories) that was being described in each essay. The purpose categories are listed in Table 1 and are based primarily on categories developed by DeVogler and Ebersole (1980, 1981, 1983) in their studies with college students, adults, and adolescents respectively; Bundick et al. (2006) in their work with adolescents; as well as on an initial review of approximately 15 students’ purpose essays. Several of the categories include family-oriented, religiously- or spiritually-oriented, and activity-oriented.

While deciding on the purpose categories, the coders classified the essays as having one purpose, no purpose, parallel purposes, or an integrated purpose. Students who wrote about one purpose category would be classified as one purpose. Students who wrote that they did not have a purpose would be classified as no purpose. Students who wrote about multiple categories of purpose (e.g., family-oriented and activity-oriented) or multiple types of purpose within a single category (e.g., doctor and professional basketball player) and drew no link between the categories or types of purposes would be classified as parallel purposes. Lastly, students who included multiple purpose categories and described the categories as overlapping and reinforcing would be classified as integrated purpose (see Appendix for examples of these classifications).

The coders also coded the purpose essays for purpose engagement. Engagement was based on a rating of students’ behavioral involvement with their purpose. Purpose engagement is a binary variable that gauges the presence (“Student reports engaging in activities aligned with their purpose”) and absence (“No indication of any action taken related to their purpose”) of engagement. Initially, semidirect engagement was a third category of purpose engagement, but semidirect engagement was merged with the presence of purpose engagement and a binary variable was created. A combination of factors led to this decision: (1) There were conceptual similarities between semidirect engagement and presence of purpose engagement; (2) Too few students had semidirect engagement; and (3) The binary variable allowed for greater reliability between coders. Taken together, these factors suggested that opting for a binary variable was a sensible choice.

An example of the presence of purpose engagement is:

An example of the absence of purpose engagement is:

Items from two adolescent purpose scales were combined and used to assess for strength of purpose commitment in the present study’s sample. A primary rationale for reducing the length of the commitment measure was to reduce the likelihood of testing fatigue among students. Furthermore, modifications made to the scales were based on theoretically and psychometrically informed decisions. The first of the two scales is widely used in research (Lippman et al., 2014) and consists of three items (1) My life has no meaning; (2) My life will make a difference in the world; and (3) I am doing things now that will help me to achieve my purpose in life) responded to via a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” In the present study, the first item was removed from this measure due to school-level concerns about the grim nature of the question. The scale was supplemented by a subset of items (1. My life has a clear sense of purpose; 2. I am always working toward accomplishing my most important goals in life; and 3. I have a purpose in my life that says a lot about who I am) from the Revised Youth Purpose Survey’s (Bundick et al., 2006) purpose identification subscale, which is closely aligned with commitment to purpose and has in fact been labeled as “commitment” in a previous study (Burrow et al., 2010, p. 1268). The complete and finalized purpose measure used in this study that will be referred to as the Modified Purpose Scale, included the five remaining aforementioned items from the Lippman et al. (2014) and Bundick et al. (2006) purpose scales, and a 5-point Likert scale (“disagree a lot,” “disagree a little,” “neither agree nor disagree,” “agree a little,” and “agree a lot”). The interitem Pearson product-moment correlations ranged from 0.54 to 0.67, and the scale had a Cronbach’s alpha of 0.88.

A modified subscale of the Short Grit Scale (Grit-S) (Duckworth & Quinn, 2009), which is an 8-item 5-point Likert scale ranging from “not at all like me” to “very much like me” and measures perseverance and passion for long-term goals, was used as a supplement to the purpose scale to assess strength of purpose commitment. The Grit-S has two subscales focused on consistency of interest and perseverance of effort. A modified version of the perseverance of effort subscale was used for the present study. The subscale consists of four items, has reliability alphas ranging from 0.60 to 0.78 across four samples, and better predicted GPA among seventh to 11th grade students over a 1-year period than the consistency of interest subscale and the full Grit-S measure (Duckworth & Quinn, 2009). The perseverance of effort subscale was used in the present study because of its demonstrable psychometric strengths (Credé, Tynan, & Harms, 2016), particularly among a seventh- to 11th-grade sample (Duckworth & Quinn, 2009) as well as with a sample of seventh and eighth grade students who were predominantly low income and Latino (Hatchimonji, 2016) and more closely resembled the current study’s population. This latter study (Hatchimonji, 2016), which also revealed that the consistency of interest subscale had a low reliability alpha of 0.52 as well as showing that one of the perseverance of effort items (“setbacks don’t discourage me”) loaded on the wrong factor, influenced the decision to remove the consistency of interest subscale as well as the reverse-worded “setbacks don’t discourage me” item. With these modifications, the present study’s finalized perseverance of effort subscale, which will be referred to as the Modified Short Grit Scale, consists of three items (1. I finish whatever I begin; 2. I am a hard worker; and 3. I am diligent. Diligent means I am careful and responsible in the things I do) on a 5-point Likert scale (“not at all like me,” “a little like me,” “half the time like me,” “usually like me,” and “always like me”). A definition of “diligent” was provided for students based on their confusion around its meaning, in prior studies. The present study’s Modified Short Grit Scale demonstrated interitem Pearson product-moment correlations ranging from 0.55 to 0.60, and the scale had a Cronbach’s Alpha of 0.81, which exceeds the reliability alphas of five previous studies (Duckworth & Quinn, 2009; Hatchimonji, 2016).

The first step toward analyzing this study’s relevant data involved coding the students’ purpose essays for purpose classification and purpose engagement. Reliability was based on the percent agreement of the coding between the author of this paper, who was the designated master coder, and a research assistant who had been trained on the coding system. The master coder coded all 124 essays, and the research assistant coded 20% of the essays (n = 25) that were randomly selected. The percent agreement between the two coders on purpose engagement reached 84% while the percent agreement on purpose classification was 60%. The master coder reconciled any coding discrepancies by defaulting to his rating.

The Modified Purpose Scale and the Modified Short Grit Scale demonstrated a correlation of r = .62, suggesting these scales shared meaningful variance that could be labeled as purpose commitment but also contained distinct elements that supported their separate analysis. In order to see more clearly the nature of the convergence and divergence of purpose commitment as assessed by these two scales, the scales were divided into low, medium, and high groups. Approximately one third of the sample population was accounted for in each group. All analyses also were conducted with these scales being cut into a low and high group, where the medium group was combined with the high group to increase the cell count for chi-square analyses, as well as a low and high group that excluded the medium group, in order to accentuate the group differences. However, all analyses yielded nonsignificant results. As a result, the data are presented using low, medium, and high grouping for heuristic clarity.

Demographic (gender, grade level, and ethnicity) differences on purpose engagement, purpose classification, and purpose commitment (as measured by the Modified Short Grit Scale and the Modified Purpose Scale), were analyzed using parametric and nonparametric tests. The cross tabs of purpose engagement and purpose classification by all three demographic variables appear in Table 2. The means and standard deviations on the Modified Short Grit Scale and the Modified Purpose Scale by all three demographic variables appear in Table 3.

Regarding gender, male and female participants did not demonstrate a significant difference on purpose engagement, purpose classification, or purpose commitment. a chi-square analysis was conducted for purpose engagement, a binary variable. Similarly, a chi-square analysis was conducted for purpose classification, which is a categorical variable. Lastly, a t test was run for purpose commitment, on both the Modified Purpose Scale and the Modified Short Grit Scale.

In terms of grade level, seventh and eighth graders did not show a significant difference on purpose engagement, purpose classification, or purpose commitment. Chi-square analyses were used for purpose engagement as well as for purpose classification. T tests were run for both purpose commitment measures.

Lastly, the four ethnicities in the sample did not show a significant difference on purpose engagement, purpose classification, or purpose commitment. Chi-square analyses were conducted for purpose engagement as well as for purpose classification. Analysis of variance (ANOVA) was run for each of the two measures of purpose commitment.

These analyses revealed no statistically significant differences in purpose engagement, purpose classification, or purpose commitment between genders, grade levels, or ethnicities. Therefore, these demographic variables were not controlled for in later analyses.

In order to explore the relationship of purpose classification and purpose commitment, two sets of analyses were done. A series of chi-square tests depict whether students with integrated purposes tended to have higher commitment scores than students with one purpose, whether students with one purpose tended to have higher commitment scores than students with parallel purposes, and whether students with parallel purposes tended to have higher commitment scores than students with no purpose. After performing the chi-square tests, no relationship was found between purpose classification and purpose commitment, the latter variable being grouped by low, medium, and high Modified Short Grit Scale Scores, as well as by low, medium, and high Modified Purpose Scale Scores. ANOVAs were also run with purpose commitment as a continuous variable in order to provide additional support for the presence or absence of a significant finding. ANOVA’s confirmed that purpose classification was not significantly associated with purpose commitment on the Modified Purpose Scale or the Modified Short Grit Scale.

In order to explore whether there was an interaction between purpose engagement and purpose classification, two additional sets of analyses were done. Chi-square tests were performed to more clearly illustrate whether the students who had either integrated or one purpose classifications and were engaged with their purpose tended to have higher purpose commitment scores than students who were not engaged with their purposes or maintained parallel purposes. Notably, the 12 students in the no purpose group were removed from the sample for this set of analyses because students without a purpose would not and did not show engagement with a purpose, indicating that there could not logically be an interaction between the two terms. Removing these students reduced the sample size to 112. The chi-square tests were supplemented by hierarchical linear regressions, which provided additional information regarding the presence or absence of an interaction. Hierarchical linear regressions were run separately for three of the purpose classifications (one purpose, parallel purpose, and integrated purpose). Purpose engagement was multiplied by each of the three purpose classifications to create three interaction terms. Purpose engagement was entered (Step 1), followed by purpose classification (Step 2), followed by the interaction of purpose engagement and purpose classification (Step 3).

These analyses revealed that purpose engagement did not interact with purpose classification. Specifically, results showed that Step 3, which included the interaction term was not significant for each of the three classifications. However, purpose engagement, as a main effect, was significantly associated with purpose commitment, as measured by the Modified Purpose Scale, but not the Modified Short Grit Scale (B = .35, p < .05). Importantly, the model that included the main effects of purpose engagement and the integrated purpose classification was significant, F(2, 109) = 3.09, R² = .05, p < .05. These results are included in Table 4.

In this exploratory study, the sample of 124 seventh and eighth grade low SES and ethnically diverse middle school students did not differ significantly on their purpose engagement, purpose classifications, or purpose commitment by grade, gender, or ethnicity. Additionally, purpose classification was not associated with different levels of Purpose Commitment. Finally, despite there being no interaction between purpose classification and purpose engagement, students who reported engaging with their purpose scored significantly higher on a purpose commitment measure than students who reported no engagement with their purpose.

The finding that purpose engagement was associated with higher purpose commitment scores is reasonable given research showing that a lack of engagement is associated with “vague” descriptions of purpose (Moran, 2009, p. 155). It is hard to imagine that a student who has no experience with and limited knowledge of his or her purpose, which is likely to be vaguely stated, would be strongly committed to it. In contrast, it is quite easy to imagine that a student who has experience with and first-hand knowledge of his or her purpose, which would conceivably be clearly stated, would be strongly committed to it.

Yet the question remains why these results were significant on the Modified Purpose Scale and not on the Modified Short Grit Scale. The content of the questions on the two scales may help explain this outcome. The Modified Short Grit Scale included questions that related to students’ overall perseverance (e.g., “I finish whatever I begin”), whereas the Modified Purpose Scale included questions that spoke to students’ current engagement with and commitment to their purpose (e.g., “I am always working toward accomplishing my most important goals in life”). The Modified Purpose Scale appears to tap directly into purpose engagement whereas the Modified Short Grit Scale does not. In other words, purpose engagement does not necessarily connect to perseverance, but it does connect to a genuine purpose.

The lack of grade level differences in purpose classifications, purpose engagement, and purpose commitment is not overly surprising. The developmental gap between students can vary greatly within and across grades. In a school as socioeconomically, ethnically, and developmentally diverse as the one this sample is drawn from, students who sit in the same classroom can be at vastly different developmental and academic levels, thereby canceling out any grade-level related variance.

The absence of a relationship between purpose classification and purpose commitment could be a function of several factors, including the relatively low reliability among coders as well as the lack of precision of the purpose commitment scales. Self-report purpose commitment scales lend themselves to exaggerated scores, particularly when administered to adolescents, who as Inhelder and Piaget (1958) observed, articulate unrealistic future plans that resemble “fantasies and fabulations” that are “soon abandoned” (p. 344). Accordingly, the purpose commitment scores tended toward the higher end of the scales and made it difficult to distinguish students’ true level of commitment. These findings support the case for longitudinal data that could portray the duration and consistency of student purpose rather than inflated self-report scores. Notably, these reliability and measurement issues might also help explain the lack of an interaction between purpose classification and purpose engagement.

Several limitations of this study have already been alluded to, including low interrater reliability of the purpose classifications, the absence of a longitudinal and behavioral measure of purpose commitment, and the exploratory nature of the study. Challenges with interrater reliability were partly due to the students’ academic and developmental levels. Grammar and clarity in the essays greatly varied, which created room for inference and interpretation on the part of the coders.

Regarding the dearth of statistically significant findings, an exploratory study often risks null findings and, in many ways, this study exemplified that result. In particular, because of the sample size and large number of categories examined in purpose classification, the study may have been underpowered and less equipped to detect differences between the groups of students.

Additional limitations concern internal and external validity. First, causal conclusions cannot be drawn from this study given the lack of both random assignment and the manipulation of the independent variable. In addition, the findings from this study may not generalize to other grade levels or populations that are more affluent, have different ethnicities, or have never been exposed to the topic of purpose.

In planning future studies, the inclusion of longitudinal and behavioral measures of purpose commitment would be most critical for the advancement of the study of purpose. The introduction of such measures would allow future studies to hew more closely to a more accurate and objective conceptualization and operationalization of purpose commitment. Perhaps purpose classifications would in fact demonstrate significant associations with purpose commitment if longitudinal and behavioral measures of commitment were used. These questions could be addressed in stages, following students during each major educational and occupational transition, from high school, to college, as well as to and through career.

To address coding reliability issues, semistructured interviews could be conducted with middle school-aged students who struggle with writing, and computerized text analysis could be used to create systematic and generalizable coding processes. Although semistructured interviews can be resource-intensive, they can also make it easier for students who struggle with writing to communicate their thoughts. Furthermore, interviews can allow for follow-up and clarifying questions. These features of interviews can help improve coding reliability. Computerized text analysis programs such as Linguistic Inquiry and Word Count program (Pennebaker, Booth, Boyd, & Francis, 2015) contain content dictionaries that can be harnessed to analyze texts, such as the purpose essays. Programs like Linguistic Inquiry and Word Count could increase coding reliability and create systems that schools could conceivably use.

There is empirical and philosophical support for the importance of purpose in life. Purpose in life has been associated with many positive outcomes, ranging from mental health to longevity. These observations beg the question of how to cultivate and strengthen purpose. This study demonstrated that current engagement with a purpose was associated with higher purpose commitment scores. Although the directionality of purpose engagement and purpose commitment is unclear (i.e., engagement may lead to stronger commitment; stronger commitment may lead to engagement; or there may be a third variable that causes both of these outcomes), it seems reasonable for educators, parents, and professionals to provide youth with opportunities to explore and engage with potential purposes, particularly in cases where a child is experiencing academic, social, or emotional difficulties.

Parents, researchers, and policymakers are all looking for ways to help young people build resilience and thrive. There is not a simple solution to this great task, and yet research and reasoning suggest that people who have a purpose in life tend to flourish. Although this paper has raised potentially more questions than it has answered, it lends support to the case of encouraging students to explore and engage with their purpose, and has raised further awareness of the purpose construct, as well as urgency for additional research in this area.

My purpose is still unknown to me. I am still deciding what I want to be, what I want to pursue, what my career is going to be and so on. I still do not know why I am here. Numerous things are still unknown to me and my purpose is included in that list. One day I think that I want to be a roboticist, the next I want to be a surgeon. I still have not made up my mind on what I want to be. I don’t know if I want to create things or own a company or even save people. I still don’t know what I love doing, maybe it’s working with technology, maybe it’s robotics, or even just solving problems. Hopefully I will know what I am passionate about and be able to pursue a career in that field.

My purpose in life isn’t really my goal, it’s something that I’ve already achieved and is my intention in life. Helping others (5). Helping people is one of the best feelings, because making someone else happy makes you happy and it make you feel like a very good person…. Giving advice, comforting people, community service … et cetera. Little things like that is a part of my Life that I Just naturally do.

My purpose would be to get good grades because I want to go to a good high school and a good college. I want to excel with As and Bs in school (6). Another purpose would be that I would help people more often (5). I could donate books, clothing, food or find someone a shelter to stay at.... My purpose would be to get good grades. Another purpose would be to help people more often. I could help them carry their groceries to their house.

I think my purpose in this world is to save lives (5). The lives being of people, animals, and whatever else is a living thing (5 and 11). I want to graduate high school and go to college or medical school (6) and study to become a doctor (1). I wish to be able to go work in a hospital or travel all over the world to save sick or injured people’s lives. I also yearn to learn a lot about chemicals and medicine to help create a cure (10) for some of the world’s most deadliest illnesses, such as cancer, HIV, Ebola, asthma, diabetes, polio, AIDS, and lupus. Having some people in my family pass away from some of these diseases makes me want to strive even harder to create cures and prevent any other families from similar experiences.