Contribution of Social Bonds to the Development of Scientific Research Identity in Adolescent Participants of a Science Enrichment Program

Abstract

An abundance of research on self-determination theory has shown that satisfaction of basic needs for autonomy, competence, and relatedness among adolescents promotes the experience of optimal career-related outcomes such as academic motivation and engagement. However, few studies have explored how satisfaction of these needs promotes identification with particular careers, and less attention has been paid to the role of relatedness satisfaction in these developmental processes compared to autonomy and competence. We addressed these issues in the current study by examining the relationship between relatedness and identification as a research scientist. Five latent classes of scientific research identity development were identified using growth mixture modeling: (a) strong positive growth, (b) moderate positive growth, (c) weak positive growth, (d) no growth, and (e) strong negative growth. Results of logistic regression analyses indicated that need for relatedness was a significant positive predictor of membership in the strong positive growth class relative to the no growth class. Implications for identity development as a research scientist and the social and motivational influences undergirding this process are discussed.

Keywords

scientific research identity vocational identity need for relatedness self-determination theory adolescent development

Introduction

The formation of a personal identity involves a process of categorizing oneself in terms of roles and internalizing the meanings and expectations of those roles into the self (Stets & Burke, 2000). These roles become increasingly rooted in relationships during the teenage years as adolescents begin to form more intimate bonds with their peers, become members of various social groups, and generally ascribe greater importance to identifying themselves with others. The distinction between personal and social identity development is often much less pronounced during this period because although adolescents may strive to maintain their uniqueness, they are likely to view the world and behave in ways that are consistent with the perspective of the groups to which they belong. It can be argued that the fundamental need to form enduring social bonds (Baumeister & Leary, 1995) is perhaps the most influential determinant of identity development during adolescence. The roles that adolescents assume become increasingly differentiated as a result of maturation in cognitive, emotional, and physical functioning as well. Growth in these areas allows for greater exploration and mastery of one’s environment, thus affording adolescents the ability to engage in career-related activities that inform vocational identity development (Chávez, 2016).

Vocational psychology researchers who have examined adolescent identity issues have often applied global theories of identity development to the domain of work in efforts to better understand general career exploration and decision-making processes (Luyckx et al., 2010; Negru-Subtirica et al., 2015). Much of this research has largely drawn from Marcia’s (1966, 1993) identity status theory, which posits four identity types: (a) diffusion, (b) foreclosure, (c) moratorium, and (d) achievement. Individuals are theorized to begin in the diffusion stage, which is characterized by low exploration and commitment, and ideally proceed to the most adaptive stage of achievement where exploration and commitment are high. Studies involving adolescents have generally supported this sequence (e.g., Klimstra et al., 2010; Porfeli et al., 2011; Skorikov & Vondracek, 1998), and linked vocational constructs (e.g., career indecision) to the statuses in ways that accord with theoretical expectation (e.g., Luyckx et al., 2010; Vondracek et al., 1995). The domains that influence and are influenced by a well-established vocational identity also extend beyond the realm of work. For example, Hirschi (2012) reported a positive association between adolescent vocational identity and general well-being, while Negru-Subtirica and Pop (2018) documented evidence of a bidirectional relationship between adolescents’ vocational and academic identity development. Thus, vocational identity tends to be correlated with other salient identities and influences overall psychological functioning in beneficial ways (Porfeli et al., 2011; Skorikov & Vondracek, 1998). Indeed, striving to achieve a clear vocational identity during adolescence is a development task that is associated with the formation of a positive overall self-concept (Skorikov & Vondracek, 2007).

Although these findings are important to be sure, researchers have tended to focus on general vocational identity development (e.g., Gushue, Clarke, et al., 2006; Gushue, Scanlan, et al., 2006; Jantzer et al., 2009; Li et al., 2019) rather than studying these processes within specific career domains. This may be due to the fact that many adolescents have not had sufficient exposure to particular careers to internalize a stable image of themselves performing in a vocational role. In other words, adolescents may have little to identify with by this point in their development because their learning experiences and interactions with role models are typically limited to family members and teachers in formal academic settings. To the extent that adolescents can learn from adults performing in particular career roles (e.g., work-based learning; Alfeld et al., 2013; Kenny et al., 2010; Lapan et al., 2003), and share a collective identity with others in line to pursue those same careers, they can begin to narrow (Gottfredson, 2002) or expand their career options and make more informed decisions. For students developing preliminary interests in science careers, learning from scientists and peers who are conducting genuine research can help them form a concept of themselves as functioning in a similar capacity. In the present study we address the dearth of research on domain-specific vocational identity development by focusing our attention on the development of adolescents’ identities as research scientists in the context of an authentic research experience marked by collaboration, social support, and a shared sense of community. Scientific research identity is operationally defined in the current study as a self-concept characterized by an assimilation of the values, roles, and behaviors associated with conducting scientific research. Authentic research represents scientific inquiry that is hypothesis-driven, thus it offers a sharp contrast to more structured, course-based research in which outcomes are often predetermined (Spell et al., 2014). Such an experience of engaging in authentic research also reflects an approximation of work-based learning insofar as the activities entailed within resemble those which a scientist would carry out in an actual research setting.

Social Motives and Scientific Identity Development

The career development of scientists has a rich focal history in the vocational psychology literature, with descriptions of the characteristics of scientists dating back to the seminal work of Anne Roe (1953, 1961). It was during this period that lay conceptions of the scientist as having a solitary nature were first empirically documented (Mead & Metraux, 1957), and though such notions no longer reflect the contemporary collaborative nature of science (e.g., Allen et al., 2018; Hall et al., 2018), the belief that scientists tend to be “loners” continues to persist as an intractable stereotype held by both children (e.g., Sharkawy, 2009) and adults (e.g., Bruun et al., 2018; Cheryan et al., 2013). Such stereotypical beliefs represent one source driving the proliferation of research on the social-motivational factors underlying students’ decisions to pursue careers in science. Some researchers have examined science career choice behaviors by framing their investigations from a goal congruity perspective (Diekman et al., 2011) – that is, the degree to which people pursue goals that are concordant with one’s interests and values (Sheldon & Elliot, 1999), such as communion with others (i.e., communal goals) and self-direction (i.e., agentic goals).

Despite the explanatory power of the goal congruity perspective, studying these issues from diverse theoretical viewpoints can help scholars converge on a fuller understanding of the interpersonal motivations underlying research engagement. Self-determination theory (SDT; Deci & Ryan, 1985; Deci & Ryan, 2000; Ryan & Deci, 2000) may be a useful framework for broadening such analyses. According to SDT, individuals are theorized to experience a higher quality of motivation when they perceive themselves to be the cause of their own behavior as opposed to feeling controlled by others or contingencies in the environment. To this end, satisfaction of innate needs for autonomy (internal locus of causality), competence (feeling of efficacy), and relatedness (feeling connected to others) should manifest in the most self-determined form of self-regulation – intrinsic motivation – whereas more controlling forms of regulation (i.e., extrinsic motivation and amotivation) are presumed to result when these needs are not satisfied. Satisfaction of these needs has been linked not only to more adaptive motivation, but other important outcomes such as classroom engagement (e.g., Jang et al., 2012) and psychological well-being as well (e.g., Reis et al., 2000).

SDT research within the academic/career development realm has tended to focus on the relations of needs and motivation types to academic outcomes such as performance, engagement, persistence intentions, and self-efficacy (Howard et al., 2021). Identity has emerged as an equally important construct of interest in this area of research because it shares in common with SDT a focus on the internalization of adaptive attitudes that promote psychosocial functioning (Ryan & Deci, 2003). Identity development involves the assimilation of roles and practices in a way that mirrors the internalization of values that occurs within the more autonomous SDT form of external regulation referred to as identification (Ryan & Deci, 2000). Research has shown that both general need satisfaction (e.g., Luyckx et al., 2009) and need satisfaction in the context of family relationships (e.g., Erçelik & Dost-Gözkan, 2022) facilitate general identity development. Luyckx et al. also found that relatedness not only influences identity development, but identity development has a positive reciprocal effect on relatedness need satisfaction as well, thus lending compelling support to the notion that identity is firmly rooted in both personal and social conceptions of the self (Albarello et al., 2018). These findings illustrate the importance of social cohesion in shaping identity development. Need for relatedness, which is operationally defined for the purpose of the present study as a need to feel connected to and supported by program participants and faculty, is particularly important to science career development because research is increasingly conducted collaboratively and in the context of cross-disciplinary teams (Hall et al., 2018). However, there are two issues in the literature that warrant attention. First, relatedness is considered perhaps the most essential determinant of identity formation (Ryan & Deci, 2003) and may even be a more potent predictor of science identity than autonomy need satisfaction (Chiu, 2023), yet researchers have tended to highlight the latter form of need satisfaction as the predominant motivational process in student learning (e.g., León et al., 2015). Indeed, there is some evidence to suggest that need for relatedness is more strongly associated with science identity than needs for either competence or autonomy (Skinner et al., 2017) but otherwise there has been surprisingly little research linking basic psychological needs to science identity. The paucity of research on the unique role of relatedness in shaping identity thus suggests this is an issue in need of further investigation. Second, research has mainly centered on the relationship between need satisfaction and general identity development, with fewer studies examining whether these basic needs serve as a foundation for domain-specific identity development. Some studies have linked SDT motives to science identity among both college students (e.g., Smith et al., 2014) and adolescents (Deemer et al., 2022), but scholars have yet to determine whether psychological needs themselves influence the growth of science identity over time.

Present Study

The purpose of the present study was twofold. First, we set out to determine whether students participating in the Summer Science Program (SSP) increasingly identify as scientific researchers over time. SSP strives to create a community of practice (Wenger, 1998) characterized by collaboration, intellectual curiosity, and mutual support. These elements of the SSP learning environment are thought to engender in students an esprit de corps that facilitates satisfaction of their needs for social connection and shapes their identification with the scientific community. Exposure to such role models affords students an opportunity to engage in the important developmental process of seeing themselves functioning in an impactful fashion in the future, and reconciling any discrepancies that may exist between their “possible selves” (Markus & Nurius, 1986) and current self-conceptions. We assumed an exploratory perspective by attempting to establish whether different scientific identity growth trajectories could be identified for various subgroups of students. Thus, our second aim was to examine whether perceptions of relatedness in the program were predictive of scientific identity growth both within the overall sample, and within any subgroups emerging from our analyses. We hypothesized that (a) scientific identity would exhibit unconditional growth over the course of the program, and (b) perceptions of relatedness at time 1 (T1) would be a significant predictor of latent class membership in at least one of the emergent subgroups demonstrating a positive identity growth trajectory. Given that women have been historically underrepresented in the physical sciences (Eccles, 2007), we also examined latent class frequencies by gender and project type as an exploratory analysis.

Method

Participants

Eight hundred eighty-eight high school students participated in the study. A total of 102 participants provided demographic information prior to entering the program but provided no responses to the substantive measures. These cases were removed from the data set, resulting in a final N of 786. A slight majority of participants identified as male (49.9%), 49.6% identified as female, and .5% reported a nonbinary gender identity. With regard to race/ethnicity, most participants identified as East Asian (34.9%), followed by White/European American (23.1%), Indian (13.9%), multiracial (7.7%), Hispanic/Latino/Latina (7.1%), Black/African American (3.2%), Southeast Asian (3.1%), Arabic/Middle Eastern (1.4%), Native American (.1%), and Pacific Islander (.1%). A small percentage of participants (.6%) identified as “other,” and 4.8% preferred not to disclose their race/ethnicity.

Measures

Relatedness Need Satisfaction

The eight-item Relatedness subscale of the Basic Need Satisfaction at Work Scale (BNSWS; Deci et al., 2001) was used to measure participants’ perceptions of the degree to which the program satisfied students’ needs for social connection. The BNSWS was originally designed to assess need satisfaction at work but we adapted the relatedness items such that they tapped perceptions of social connectedness within the SSP program. An original item states “I get along with people at work” while the adapted item states “I get along with people in this program.” Items are rated on a Likert scale ranging from 1 (not at all true) to 7 (very true). Need for relatedness scores at T1 demonstrated good internal consistency reliability (α = .83).

Scientific Research Identity

We used the six-item Scientific Identity Scale (SIS; Chemers et al., 2011) to measure participants’ identification with scientific research. We adapted the items such that the term “scientist” was replaced with the term “researcher.” An original item example states “Being a scientist is an important reflection of who I am” while the adapted version states “Being a researcher is an important reflection of who I am.” Participants rate SIS items on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s alpha coefficients were .86, .87, and .90 for T1, T2, and T3, respectively.

Procedure

Data were collected from six cohorts of students participating in SSP (https://summerscience.org) between 2017 and 2022. SSP is an annual residential science enrichment program in which high school students engage in authentic research in astrophysics, biochemistry, or genomics over a period of 39 days. Prospective students are admitted to the program through a formal application process and they may choose only one research project to engage in. Students in the astrophysics project use mathematical and astronomical techniques to determine the orbit of an asteroid whereas biochemistry students use bioinformatic tools and wet lab biochemical methods to characterize an enzyme from a fungal crop pathogen with the end goal of developing a fungicide. The genomics project, which was added to the SSP curriculum in 2021, involves conducting research on the evolution of antibiotic resistance. The curriculum for each project remained the same through the entirety of the study. Prominent individuals from academic and corporate settings also provide guest lectures on scientific, ethical, and career development topics. Institutional review board (IRB) approval was obtained from the first author’s institution prior to initiating data collection. Parents of students under the age of 18, as well as students who were 18 years or older, provided their written consent to participate in the study. Data were collected within each offering of the program at 11-day intervals across three measurement occasions using an online survey. We measured identity at 11-day intervals because these measurement occasions were thought not to interfere with particular program activities (e.g., field trips). This interval length was thought to be sufficient for identity change to occur because students are exposed to robust socializing agents (i.e., faculty and fellow students) via concentrated instruction and close collaboration in a very specific career domain. Students were delivered a link to a webpage at each measurement occasion which summarized their rights and responsibilities as participants and afforded them the opportunity to rescind their assent to continuing in the study. Those who provided their assent were directed to a portal which contained the survey.

Results

Latent growth curve modeling operates on the assumption that all individuals in a sample belong to a single homogeneous population, however, this may be an untenable assumption because there are often subpopulations of individuals that have particular growth trajectories.

We used growth mixture modeling (GMM; Muthén, 2004) in the present study to account for the possibility that there are different patterns of growth in scientific identity over time. All analyses were conducted using Mplus 7.4 statistical software (Muthén & Muthén, 2016). The following fit statistics were used to assess model fit and determine the optimal number of latent classes: (a) Akaike Information Criterion (AIC), (b) Bayesian Information Criterion (BIC), (c) Vuong-Lo-Mendell-Rubin (VLMR) likelihood ratio test (Lo et al., 2001), (d) bootstrap likelihood ratio test (BLRT), and (e) entropy statistic. Smaller values of AIC and BIC indicate better model fit. Likelihood ratio tests involve assessing the degree of improvement in model fit when comparing models with different numbers of latent classes. Entropy values range from 0 to 1, with higher values indicating greater accuracy in assignment of class membership. Entropy values of .80 and higher indicate acceptable classification accuracy (Clark, 2010), therefore we used this value as a criterion for model retention. The analyses were conducted using robust maximum likelihood estimation, 500 initial stage random starts, 50 final stage optimizations, and 50 iterations in the first step of the optimization process. Because multivariate outliers can exert an undue influence on latent class formation (Peel & McLachlan, 2000), we identified 16 cases with significant Mahalanobis distance values (p < .001) and removed them from the data set.

We established a baseline model and tested our first hypothesis by specifying a single-class latent growth model with no covariates using maximum likelihood as the estimation method. The fit of this model to the data was evaluated using the following fit indexes: (a) model chi-square test, (b) comparative fit index (CFI), (c) root mean square error of approximation (RMSEA), and (d) standardized root mean square residual (SRMR). CFI values greater than .90 indicate good fit (Hu & Bentler, 1999). RMSEA values of less than .07 (Steiger, 2007) and SRMR values of less than .08 (Hu & Bentler, 1999) indicate acceptable fit. Results indicated the model provided an excellent fit to the data, χ² (1) = .02, p = .88, CFI = 1.00, RMSEA = .000 (90% CI: .000, .047), SRMR = .001, as the mean intercept (estimate = 21.35, p < .001) and slope (estimate = 1.10, p < .001) growth factors were statistically significant. Our hypothesis that research identity would exhibit unconditional growth was therefore supported (Table 1).

Table 1.

Descriptive Statistics for the Study Variables.

Variable	M	SD	Observed Range
T1 need for relatedness	47.65	6.01	18–56
T1 scientific research identity	21.33	4.23	6–30
T2 scientific research identity	22.49	4.01	10–30
T3 scientific research identity	23.57	4.26	6–30

We used the three-step latent class modeling procedure (Asparouhov & Muthén, 2014; Nylund-Gibson et al., 2014) in which we (a) estimated a growth mixture model without covariates, (b) assigned participants to classes based on posterior probabilities of class membership, and (c) introduced the relatedness predictor into the model to evaluate its relationship with the latent class variable. Within this process we sought to identify a model with the optimal number of latent classes by incrementally testing models with k + 1 classes until they no longer provided a conceptual or statistically significant improvement over previous models. To determine the number of latent trajectory groups we examined models with one to six classes, with the 1-class model serving as a baseline upon which subsequent models could be compared. As Table 2 indicates, the 2-class and 5-class models fit the data significantly better than their k – 1 class counterparts, and both models produced entropy values of greater than .80. The 5-class model offered greater population heterogeneity than the 2-class model and yielded the lowest BIC value (i.e., 10010.06) of any of the six models, therefore we retained the 5-class model for further analysis.

Table 2.

Fit Statistics for Six Growth Mixture Models of Scientific Research Identity Development.

Model	Log Likelihood	AIC	BIC	VLMR LRT p value	BLRT p value	Entropy
1-class model	−6706.98	13433.95	13480.41	--	--	--
2-class model	−4974.34	9970.68	10021.79	<.001	<.001	.81
3-class model	−4970.98	9969.95	10035.00	.34	.43	.63
4-class model	−4957.99	9949.98	10028.97	.16	<.001	.75
5-class model	−4938.57	9917.13	10010.06	< .001	< .001	.82
6-class model	−4931.20	9908.39	10015.26	.03	<. 001	.79

Note. AIC, Akaike information criterion; BIC, Bayesian information criterion; BLRT, bootstrap likelihood ratio test; VLMR LRT, Vuong-Lo-Mendell-Rubin likelihood ratio test. Bolded print indicates the chosen model for analysis.

Classification probabilities for participants’ most likely latent class membership exceeded .80 in all five classes, with class 5 yielding the highest probability at .916 (see Table 3). A growth plot of the sample means for each class is presented in Figure 1. Class 1 comprised 20.00% of the sample (n = 154) and had the highest initial level of scientific research identity of any of the classes (mean intercept = 24.92, p < .001) as well as the strongest positive growth rate (mean slope = 1.96, p < .001). We therefore labeled this group strong positive growth. Participants in class 2 (6.88%, n = 53) had the second lowest initial status at T1 (mean intercept = 16.30, p < .001) and exhibited no growth over the course of the program (mean slope = −.001, p = .998), therefore this group was labeled no growth. Class 3 was the smallest group identified in the analysis, as it comprised only 1.82% of the sample (n = 14). Participants in this group endorsed the lowest initial level of scientific research identity (mean intercept = 15.94, p < .001) and their perceived identities as scientific researchers declined significantly over time (mean slope = −2.16, p < .001), therefore this class was labeled strong negative growth. Class 4 was the second largest group that emerged from the analysis, consisting of 23.25% of the sample (n = 179). Participants in this group had a comparatively low initial level of scientific identity (mean intercept = 18.97, p < .001), but a significant and positive mean slope growth factor (mean slope = .86, p < .001). Because this group yielded a much weaker rate of growth in identity development than class 1, this group was labeled weak positive growth. Finally, class 5 was the largest group to emerge from the analysis at 48.05% of the sample (n = 370). This group produced the second highest initial level of research identity development (mean intercept = 22.18, p < .001) and a mean slope growth factor (mean slope = 1.18, p < .001) that was weaker than class 1 but stronger than class 4. We therefore labeled this the moderate positive growth group. Because the strong positive growth and the weak positive growth groups were similar in size, we conducted two separate chi-square tests of independence to assess the associations between these latent classes and both gender and research project type.

Table 3.

Classification Probabilities for Most Likely Latent Class Membership.

Latent class	1	2	3	4	5
1. Strong positive growth	.878	.000	.000	.001	.120
2. No growth	.000	.826	.004	.160	.010
3. Strong negative growth	.000	.095	.856	.041	.007
4. Weak positive growth	.000	.026	.000	.833	.141
5. Moderate positive growth	.033	.000	.000	.051	.916

Figure 1.

Growth trajectories for five latent classes of scientific research identity development.

To test our second hypothesis, we conducted a logistic regression analysis in which need for relatedness was specified as a predictor of class membership. Class 2 was chosen as the reference group for this analysis because this group exhibited neither positive nor negative growth. Results indicated that need for relatedness was a significant positive predictor of membership in the strong positive growth group relative to the no growth group (b = .12, p < .001), thus supporting our second hypothesis (see Table 4). This indicates that participants who endorsed a higher need for relatedness were 1.13 times more likely to be classified as members of the strong positive growth group relative to the no growth group. All other results were nonsignificant.

Table 4.

Results of Logistic Regression Analysis with Need for Relatedness Predicting Latent Class Membership.

Group comparison	b	SE	t	p	Odds ratio
Strong positive growth versus no growth	.12	.04	3.18	.001	1.13
Strong negative growth versus no growth	−.07	.04	−1.55	.122	.93
Weak positive growth versus no growth	−.02	.03	−.68	.495	.98
Moderate positive growth versus no growth	.05	.03	1.59	.111	1.05

Finally, for exploratory purposes we examined latent class frequencies by gender and project type (see Table 5). Nonbinary gender and genomics were not included in the analysis due to sparse data in these categories. Results of a 2 (strong positive growth vs. weak positive growth) × 2 (male vs. female) chi-square test was not significant, χ² (1) = 2.72, p = .10. The chi-square test of independence between these latent classes and research project type (astrophysics vs. biochemistry) was also nonsignificant, χ² (1) = .002, p = .97.

Table 5.

Gender and Research Project Frequencies by Latent Class.

Latent class	Gender			Research project
Latent class	Male n (%)	Female n (%)	Nonbinary n (%)	Astrophysics n (%)	Biochemistry n (%)	Genomics n (%)
1. Strong positive growth	87 (60.4)	57 (39.6)	0 (0)	90 (59.6)	58 (38.4)	3 (2)
2. No growth	19 (41.3)	27 (58.7)	0 (0)	34 (64.2)	16 (30.2)	3 (5.7)
3. Strong negative growth	4 (44.4)	5 (55.6)	0 (0)	9 (64.3)	4 (28.6)	1 (7.1)
4. Weak positive growth	80 (50.6)	77 (48.7)	1 (.6)	105 (59.3)	67 (37.9)	5 (2.8)
5. Moderate positive growth	161 (47.9)	173 (51.5)	2 (.6)	194 (53.3)	158 (43.4)	12 (3.3)

Discussion

The study of vocational identity development has historically focused on broad process dimension constructs that reflect decisional states (e.g., exploration, commitment) rather than the assimilation of prospective roles in particular careers. Further, researchers have tended to examine the intrapersonal origins of identity development among adolescents (Albarello et al., 2018) despite the fact that socializing factors also play an equally important role in this process. The current research attempted to address these issues by focusing on the interpersonal bases of adolescents’ identities as research scientists.

The results of this study provided support for our first hypothesis, as the growth trajectory for the overall sample with regards to identification as a researcher was positive and significant across three measurement occasions. The finding that participants increasingly conceived of themselves as functioning in the role of a scientific researcher is consistent with prior research demonstrating growth in more generalized forms of academic (e.g., Klimstra et al., 2010) and vocational (e.g., Hirschi, 2012; Li et al., 2019) identity development among adolescents. We cannot directly attribute this growth to the effects of SSP because that this was not a controlled study, therefore efforts to employ control or comparison groups in future research would be needed if causal inferences are to be made.

The finding of five latent classes via growth mixture modeling permitted a more detailed analysis of which groups of students reported increased identification with scientific research and which groups did not. Supporting our second hypothesis, need for relatedness was a positive predictor of latent class membership as participants who perceived a higher sense of social connection with others were more likely to be categorized as a member of the strong positive growth group compared to the no growth group. Participants who experienced the most rapid development of self-identification thus perceived the program as allowing them to develop satisfying relationships with others. It is also interesting to note that not only did this group experience the strongest rate of growth, they also reported higher initial levels of scientific research identity than any of the other groups. Conversely, over 71% of the sample experienced either moderate or weak research identity development, yet relatedness was not predictive of membership in either of these groups. This suggests that a certain threshold of research identity may first need to be reached before the relationship between social connectedness and one’s self-conception as a research scientist can be strengthened.

Further supporting this interpretation is the finding that participants in the no growth and negative growth classes started with essentially the same level of research identification at T1, thus removing initial research identity as a potentially confounding variable, yet their trajectories diverged over time. Relatedness was not a significant predictor of differential membership in these groups considering that students were only slightly less likely to belong in the negative growth group versus the no growth group as a function of this factor. Relatedness may therefore not be sufficient to determine growth trajectories for students with the corresponding levels of identification observed at T1. Instead, it may be that needs for autonomy and competence better account for the observed divergence of these groups. Perhaps students’ identification with research declined because they did not perceive themselves to be developing their research skills, or because they could not carry out their research in a way that was free of coercion from others. Even when students reported a high level of T1 identification, satisfaction of relatedness needs did not robustly differentiate between membership in the strong positive versus no growth groups despite its statistical significance as a predictor. The influence of relatedness on identity development may be conditioned upon the extent to which students perceive their competence and autonomy needs to have also been met, suggesting an interactive relationship among these variables. Exploring these relations was beyond the scope of this study but it likely would be worthwhile to explore them in future studies.

The threshold effect interpretation described above implies that a strong foundational science identity, established through prior academic experiences, is necessary if one is to use the benefits of social affiliation to build more a more specialized science career identity. It is possible that some participants were predisposed to identifying highly with research prior to entering the program but we operated on the assumption that they had not been exposed to authentic research in their high school science classes, therefore they should presumably have little real-world experience on which to formulate a research identity. Alternatively, the characteristics of the program itself may have facilitated the development of a strong baseline research identity between the time of entering the program and the first measurement occasion 11 days later. In future studies it may be worthwhile to control for basic science identity prior to program entry in order to account for prior science attitudes as a possible source of research identification. Such studies may reveal a threshold effect in which communities of practice require some level of initial interest to achieve a positive outcome for adolescents.

A secondary analysis of the demographic characteristics of the latent classes indicated that a majority of participants in each group were involved in the astrophysics project, and females outnumbered males in three of the five groups. Male participants outnumbered female participants in the strong positive growth and weak positive growth groups whereas females outnumbered males in the no growth, moderate positive growth, and strong negative growth groups. However, results of a chi-square test of independence indicated there were no differences in proportions of participants classified by gender or research project type when comparing the comparably-sized strong positive and weak positive growth groups. Astrophysics participants were just as likely as biochemistry participants, and females were just as likely as males, to experience robust identity development. Given that need for relatedness was positively associated with strong research identity development, these finding offers partial evidence to counter the stereotypical belief that physics is a branch of science best suited for solitary individuals who have no interest in working collaboratively with others (Bruun et al., 2018).

Our results align with related findings indicating that environments in which there is an emphasis on collaboration and helping others are associated with greater science motivation (Brown et al., 2015) and interest in science careers (Steinberg & Diekman, 2017; Thoman et al., 2015) among college students. However, it remains an open question as to whether satisfaction of needs for autonomy and competence play a role in determining the course of research identity development. Individuals who feel coerced into behaving in a particular way are not free to express their interests and values in ways that are consistent with their conceptions of self (Ryan & Deci, 2003). Autonomy may provide an important foundation for the satisfaction of relatedness needs in identity development contexts because people cannot fully engage with others on social and intellectual levels if they feel that certain contingencies in their environments limit their ability to form relationships. Competence need satisfaction may function in a somewhat similar way in that strong self-efficacy beliefs are likely to be essential to generating the domain interest (Lent et al., 1994) needed to facilitate personal identification with scientific research. The importance of career decision-making self-efficacy to the formation of a vocational identity has been well-documented in the literature (Gushue, Clarke, et al., 2006; Gushue, Scanlan, et al., 2006) and socially-based operationalizations of self-efficacy have been shown to influence individual science interest and career intentions (Deemer et al., 2017). Both individual and collective perceptions of competence may influence social identification through motives to enhance self-esteem and feel accepted by others (Luhtanen & Crocker, 1992; Stets & Burke, 2014). Examining more closely the structural relations among autonomy, competence, and relatedness needs relative to scientific identity would be an important avenue for future research. For instance, autonomy may function as a temporal precedent of relatedness need satisfaction while competence may interact with relatedness to produce optimal identification states.

Limitations

There are three limitations in the present study that should be mentioned. First, as mentioned previously, growth in scientific identity cannot be attributed to the effects of the program itself because a control/comparison group was not incorporated into the research design. Future research comparing SSP to science enrichment programs that are similar in content, scope, and duration would permit stronger inferences of causality in this regard. Second, measurement of identity across three time points limits the extent to which inferences can be drawn about the nature of the identity growth process. Measuring identity development across four or more time points would allow for estimation of nonlinear growth trends, thus permitting a more focused analysis of the points at which identification may increase or decrease more sharply. Finally, we were not able to fully establish temporal precedence of relatedness to the identity growth process because data for these variables were collected concurrently at T1. This limits the degree to which relatedness can be inferred as exerting a causal influence on research identity development. Researchers wishing to replicate the current study should consider measuring need for relatedness entirely as a temporal antecedent of identity within the framework of growth curve modeling, cross-lag panel modeling, or other related longitudinal designs.

Implications and Conclusion

The current findings have implications for career development interventionists and researchers that extend beyond the realm of scientific research. The authenticity of the research conducted by adolescents in SSP reflects a form of learning that would occur in an actual workplace, thus exposing them to the daily activities of a research scientist. This experience highlights the importance of engaging adolescents in work-based learning provided they are developmentally prepared to understand and execute tasks in simulated work situations that incorporate robust mentoring by experienced practitioners. Such experiences should help adolescents better clarify their vocational interests and aid them in making more informed career decisions. We recognize that our findings only represent short-term changes in domain-specific identity; more work is needed to link this bounded growth to more enduring changes in global vocational identity.

In sum, the current study illustrates the contribution of social bonds to the development of scientific research identity. Prior research had emphasized autonomy and competence-related constructs (e.g., self-efficacy) as the primary facilitators of this process but as we have shown, adolescent satisfaction of the need for relatedness appears to be associated with promoting a specific form of vocational identity. We encourage investigators to consider extending this line of research to vocational domains other than science. Because learning to develop and manage relationships is a critical developmental task in adolescence, understanding how such processes relate to other vocational identities would do much to advance career development research in this population.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Eric D. Deemer

Joseph P. Ogas

References

Albarello

Crocetti

Rubini

(2018). I and us: A longitudinal study on the interplay of personal and social identity in adolescence. Journal of Youth and Adolescence, 47(4), 689–702. https://doi.org/10.1007/s10964-017-0791-4

Alfeld

C. I.

Johnson

Watts

(2013). Work-based learning opportunities for high school students. National Research Center for Career and Technical Education. https://files.eric.ed.gov/fulltext/ED574519.pdf

Allen

Smith

J. L.

Thoman

D. B.

Walters

R. W.

(2018). Fluctuating team science: Perceiving science as collaborative improves science motivation. Motivation Science, 4(4), 347–361. https://doi.org/10.1037/mot0000099

Asparouhov

Muthén

(2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 329–341. https://doi.org/10.1080/10705511.2014.915181

Baumeister

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://doi.org/10.1037/0033-2909.117.3.497

Brown

E. R.

Smith

J. L.

Thoman

D. B.

Allen

J. M.

Muragishi

(2015). From bench to bedside: A communal utility value intervention to enhance students' biomedical science motivation. Journal of Educational Psychology, 107(4), 1116–1135. https://doi.org/10.1037/edu0000033

Bruun

Willoughby

Smith

J. L.

(2018). Identifying the stereotypical who, what, and why of physics and biology. Physical Review Physics Education Research, 14(2), 020125. https://doi.org/10.1103/PhysRevPhysEducRes.14.020125

Chávez

(2016). Psychosocial development factors associated with occupational and vocational identity between infancy and adolescence. Adolescent Research Review, 1(4), 307–327. https://doi.org/10.1007/s40894-016-0027-y

Chemers

M. M.

Zurbriggen

E. L.

Syed

Goza

B. K.

Bearman

(2011). The role of efficacy and identity in science career commitment among underrepresented minority students. Journal of Social Issues, 67(3), 469–491. https://doi.org/10.1111/j.1540-4560.2011.01710.x

10.

Cheryan

Plaut

Handron

Hudson

(2013). The stereotypical computer scientist: Gendered media representations as a barrier to inclusion for women. Sex Roles, 69(1–2), 58–71. https://doi.org/10.1007/s11199-013-0296-x

11.

Chiu

T. F. K.

(2023). Using self-determination theory (SDT) to explain student STEM interest and identity development. Instructional Science. Advance online publication. https://doi.org/10.1007/s11251-023-09642-8

12.

Clark

S. L.

(2010). Mixture modeling with behavioral data. [Doctoral dissertation, University of California].

13.

Deci

E. L.

Ryan

R. M.

(1985). Intrinsic motivation and self-determination in human behavior. Plenum Press.

14.

Deci

E. L.

Ryan

R. M.

(2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4), 227–268. https://doi.org/10.1207/S15327965PLI1104_01

15.

Deci

E. L.

Ryan

R. M.

Gagné

Leone

D. R.

Usunov

Kornazheva

B. P.

(2001). Need satisfaction, motivation, and well-being in the work organizations of a former eastern bloc country: A cross-cultural study of self-determination. Personality and Social Psychology Bulletin, 27(8), 930–942. https://doi.org/10.1177/0146167201278002

16.

Deemer

E. D.

Marks

L. R.

Miller

K. A.

(2017). Peer science self-efficacy: A proximal contextual support for college students’ science career intentions. Journal of Career Assessment, 25(3), 537–551. https://doi.org/10.1177/1069072716651620

17.

Deemer

E. D.

Ogas

J. P.

Barr

A. C.

Bowdon

R. D.

Hall

M. C.

Paula

Capobianco

B. M.

Lim

(2022). Scientific research identity development need not wait until college: Examining the motivational impact of a pre-college authentic research experience. Research in Science Education, 52(5), 1481–1496. https://doi.org/10.1007/s11165-021-09994-6

18.

Diekman

A. B.

Clark

E. K.

Johnston

A. M.

Brown

E. R.

Steinberg

(2011). Malleability in communal goals and beliefs influences attraction to stem careers: Evidence for a goal congruity perspective. Journal of Personality and Social Psychology, 101(5), 902–918. https://doi.org/10.1037/a0025199

19.

Eccles

J. S.

(2007). Where are all the women? Gender differences in participation in physical science and engineering. In Ceci

S. J.

Williams

W. M.

(Eds.), Why aren’t more women in science? Top researchers debate the evidence (pp. 199–210). American Psychological Association. https://doi.org/10.1037/11546-016

20.

Erçelik

Dost-Gözkan

(2022). Basic psychological needs in relationships with parents and best friend and identity statuses among Turkish emerging adults. Current Psychology, 41(5), 2821–2834. https://doi.org/10.1007/s12144-020-00710-7

21.

Gottfredson

L. S.

(2002). Gottfredson’s theory of circumscription, compromise, and self-creation. In Brown

(Ed.), Career choice and development (4th ed., pp. 85–148). Jossey-Bass.

22.

Gushue

G. V.

Clarke

C. P.

Pantzer

K. M.

Scanlan

K. R. L.

(2006). Self-efficacy, perceptions of barriers, vocational identity, and the career exploration behavior of Latino/a high school students. The Career Development Quarterly, 54(4), 307–317. https://doi.org/10.1002/j.2161-0045.2006.tb00196.x

23.

Gushue

G. V.

Scanlan

K. R. L.

Pantzer

K. M.

Clarke

C. P.

(2006). The relationship of career decision-making self-efficacy, vocational identity, and career exploration behavior in African American high school students. Journal of Career Development, 33(1), 19–28. https://doi.org/10.1177/0894845305283004

24.

Hall

K. L.

Vogel

A. L.

Huang

G. C.

Serrano

K. J.

Rice

E. L.

Tsakraklides

S. P.

Fiore

S. M.

(2018). The science of team science: A review of the empirical evidence and research gaps on collaboration in science. American Psychologist, 73(4), 532–548. https://doi.org/10.1037/amp0000319

25.

Hirschi

(2012). Vocational identity trajectories: Differences in personality and development of well-being. European Journal of Personality, 26(1), 2–12. https://doi.org/10.1002/per.812

26.

Howard

J. L.

Bureau

Guay

Chong

J. X. Y.

Ryan

R. M.

(2021). Student motivation and associated outcomes: A meta-analysis from self-determination theory. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 16(6), 1300–1323. https://doi.org/10.1177/1745691620966789

27.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

28.

Jang

Kim

E. J.

Reeve

(2012). Longitudinal test of self-determination theory’s motivation mediation model in a naturally occurring classroom context. Journal of Educational Psychology, 104(4), 1175–1188. https://doi.org/10.1037/a0028089

29.

Kenny

M. E.

Walsh-Blair

L. Y.

Blustein

D. L.

Bempechat

Seltzer

(2010). Achievement motivation among urban adolescents: Work hope, autonomy support, and achievement-related beliefs. Journal of Vocational Behavior, 77(2), 205–212. https://doi.org/10.1016/j.jvb.2010.02.005

30.

Klimstra

T. A.

Hale

W. W.

Raaijmakers

Q. A. W.

Branje

S. J. T.

Meeus

W. H. J.

(2010). Identity formation in adolescence: Change or stability? Journal of Youth and Adolescence, 39(2), 150–162. https://doi.org/10.1007/s10964-009-9401-4

31.

Lapan

R. T.

Tucker

Kim

S.-K.

Kosciulek

J. F.

(2003). Preparing rural adolescents for post-high school transitions. Journal of Counseling and Development, 81(3), 329–342. https://doi.org/10.1002/j.1556-6678.2003.tb00260.x

32.

Lent

R. W.

Brown

S. D.

Hackett

(1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal of Vocational Behavior, 45(1), 79–122. https://doi.org/10.1006/jvbe.1994.1027

33.

León

Núñez

J. L.

Liew

(2015). Self-determination and STEM education: Effects of autonomy, motivation, and self-regulated learning on high school math achievement. Learning and Individual Differences, 43, 156–163. https://doi.org/10.1016/j.lindif.2015.08.017

34.

Fan

Cheung

F. M.

Wang

(2019). Reciprocal associations between career self-efficacy and vocational identity: A three-wave longitudinal study. Journal of Career Assessment, 27(4), 645–660. https://doi.org/10.1177/1069072718796035

35.

Mendell

N. R.

Rubin

D. B.

(2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. https://doi.org/10.1093/biomet/88.3.767

36.

Luhtanen

Crocker

(1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18(3), 302–318. https://doi.org/10.1177/0146167292183006

37.

Luyckx

Duriez

Klimstra

T. A.

De Witte

(2010). Identity statuses in young adult employees: Prospective relations with work engagement and burnout. Journal of Vocational Behavior, 77(3), 339–349. https://doi.org/10.1016/j.jvb.2010.06.002

38.

Luyckx

Vansteenkiste

Goossens

Duriez

(2009). Basic need satisfaction and identity formation: Bridging self-determination theory and process-oriented identity research. Journal of Counseling Psychology, 56(2), 276–288. https://doi.org/10.1037/a0015349

39.

Macht Jantzer

Stalides

D. J.

Rottinghaus

P. J.

(2009). An exploration of social cognitive mechanisms, gender, and vocational identity among eighth graders. Journal of Career Development, 36(2), 114–138. https://doi.org/10.1177/0894845309345841

40.

Marcia

J. E.

(1993). The status of the statuses: Research review. In Marcia

J. E.

Waterman

A. S.

Matteson

D. R.

Archer

S. L.

Orlofsky

J. L.

(Eds.), Identity: A handbook for psychosocial research (pp. 22–41). Springer-Verlag.

41.

Marcia

J. E.

(1966). Development and validation of ego-identity status. Journal of personality and social psychology, 3(5), 551–558. https://doi.org/10.1037/h0023281

42.

Markus

Nurius

(1986). Possible selves. American Psychologist, 41(9), 954–969. https://doi.org/10.1037/0003-066X.41.9.954

43.

Mead

Metraux

(1957). Image of the scientist among high-school students: A pilot study. Science, 126(3270), 384–390. https://doi.org/10.1126/science.126.3270.384

44.

Muthén

(2004). Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. In Kaplan

(Ed.), Handbook of quantitative methodology for the social sciences (pp. 345–368). Sage.

45.

Muthén

L. K.

Muthén

B. O.

(2016). Mplus user’s guide (7th ed.). Muthén & Muthén.

46.

Negru-Subtirica

Pop

E. I.

(2018). Reciprocal associations between educational identity and vocational identity in adolescence: A three-wave longitudinal investigation. Journal of Youth and Adolescence, 47(4), 703–716. https://doi.org/10.1007/s10964-017-0789-y

47.

Negru-Subtirica

Pop

E. I.

Crocetti

(2015). Developmental trajectories and reciprocal associations between career adaptability and vocational identity: A three-wave longitudinal study with adolescents. Journal of Vocational Behavior, 88, 131–142. https://doi.org/10.1016/j.jvb.2015.03.004

48.

Nylund-Gibson

Grimm

Quirk

Furlong

(2014). A latent transition mixture model using the three-step specification. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 439–454. https://doi.org/10.1080/10705511.2014.915375

49.

Peel

McLachlan

G. J.

(2000). Robust mixture modelling using the t distribution. Statistics and Computing, 10(4), 339–348. https://doi.org/10.1023/A:1008981510081

50.

Porfeli

E. J.

Lee

Vondracek

F. W.

Weigold

I. K.

(2011). A multi-dimensional measure of vocational identity status. Journal of Adolescence, 34(5), 853–871. https://doi.org/10.1016/j.adolescence.2011.02.001

51.

Reis

H. T.

Sheldon

K. M.

Gable

S. L.

Roscoe

Ryan

R. M.

(2000). Daily well-being: The role of autonomy, competence, and relatedness. Personality and Social Psychology Bulletin, 26(4), 419–435. https://doi.org/10.1177/0146167200266002

52.

Roe

(1953). The making of a scientist. Dodd, Mead.

53.

Roe

(1961). The psychology of the scientist: A definite personality pattern, encompassing a wide range of traits, characterizes the creative scientist. Science, 134(3477), 456–459. https://doi.org/10.1126/science.134.3477.456

54.

Ryan

R. M.

Deci

E. L.

(2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25(1), 54–67. https://doi.org/10.1006/ceps.1999.1020

55.

Ryan

R. M.

Deci

E. L.

(2003). On assimilating identities to the self: A self-determination theory perspective on internalization and integrity within cultures. In Leary

M. R.

Tangney

J. P.

(Eds.), Handbook of self and identity (pp. 253–272). Guilford Press.

56.

Sharkawy

(2009). Moving beyond the lone scientist: Helping 1^st-grade students appreciate the social context of scientific work using stories about scientists. Journal of Elementary Science Education, 21(1), 67–78. https://doi.org/10.1007/BF03174716

57.

Sheldon

K. M.

Elliot

A. J.

(1999). Goal striving, need satisfaction, and longitudinal well-being: The self-concordance model. Journal of Personality and Social Psychology, 76(3), 482–497. https://doi.org/10.1037/0022-3514.76.3.482

58.

Skinner

Saxton

Currie

Shusterman

(2017). A motivational account of the undergraduate experience in science: Brief measures of students’ self-system appraisals, engagement in coursework, and identity as a scientist. International Journal of Science Education, 39(17), 2433–2459. https://doi.org/10.1080/09500693.2017.1387946

59.

Skorikov

Vondracek

F. W.

(1998). Vocational identity development: Its relationship to other identity domains and to overall identity development. Journal of Career Assessment, 6(1), 13–35. https://doi.org/10.1177/106907279800600102

60.

Skorikov

V. B.

Vondracek

F. W.

(2007). Vocational identity. In Skorikov

Patton

(Eds.), Career development in childhood and adolescence (pp. 143–168). Sense Publishers.

61.

Smith

J. L.

Deemer

E. D.

Thoman

D. B.

Zazworsky

(2014). Motivation under the microscope: Understanding undergraduate science students’ multiple motivations for research. Motivation and Emotion, 38(4), 496–512. https://doi.org/10.1007/s11031-013-9388-8

62.

Spell

R. M.

Guinan

J. A.

Miller

K. R.

Beck

C. W.

(2014). Redefining authentic research experiences in introductory biology laboratories and barriers to their implementation. CBE-Life Sciences Education, 13(1), 102–110. https://doi.org/10.1187/cbe.13-08-0169

63.

Steiger

J. H.

(2007). Understanding the limitations of global fit assessment in structural equation modeling. Personality and Individual Differences, 42(5), 893–898. https://doi.org/10.1016/j.paid.2006.09.017

64.

Steinberg

Diekman

A. B.

(2017). Elevating positivity toward STEM pathways through communal experience: The key role of beliefs that STEM affords other-oriented goals. Analyses of Social Issues and Public Policy, 17(1), 235–261. https://doi.org/10.1111/asap.12135

65.

Stets

J. E.

Burke

P. J.

(2000). Identity theory and social identity theory. Social Psychology Quarterly, 63(3), 224–237. https://doi.org/10.2307/2695870

66.

Stets

J. E.

Burke

P. J.

(2014). Self-esteem and identities. Sociological Perspectives, 57(4), 409–433. https://doi.org/10.1177/0731121414536141

67.

Thoman

D. B.

Brown

E. R.

Mason

A. Z.

Harmsen

A. G.

Smith

J. L.

(2015). The role of altruistic values in motivating underrepresented minority students for biomedicine. BioScience, 65(2), 183–188. https://doi.org/10.1093/biosci/biu199

68.

Vondracek

F. W.

Schulenberg

Skorikov

Gillespie

L. K.

Wahlheim

(1995). The relationship of identity status to career indecision during adolescence. Journal of Adolescence, 18(1), 17–29. https://doi.org/10.1006/jado.1995.1003

69.

Wenger

(1998). Communities of practice: Learning, meaning, and identity. Cambridge University Press.