Unlocking STEM pathways: A person-centred approach exploring a teacher recruitment intervention

Abstract

Background and context of the study

This research employed a person-centred approach to evaluate the effectiveness of a recruitment intervention aimed at attracting STEM undergraduate students to the teaching profession. The study aimed to identify participant profiles based on changes of interest in teaching, examine the demographic factors associated with these profiles, and explore the outcomes associated with the identified profiles. A total of 267 participants from 18 universities in England were recruited for the study. The intervention involved presenting 12 vignettes that depicted different motivations for choosing teaching as a career. Participants rated their change of interest in teaching after reading each vignette.

Results

The latent profile analysis revealed four distinct profiles: dissuaded participants, unpersuaded participants, moderately persuaded participants, and highly persuaded participants. The highly persuaded profile reported the highest levels of self-efficacy, interest, perceived fit, and enjoyment in teaching. Participants from higher socioeconomic backgrounds were more likely to be persuaded by the recruitment intervention, but gender, ethnicity, or program levels did not significantly affect profile membership.

Conclusions

The findings demonstrate the potential of recruitment interventions to influence the interest of STEM undergraduate students in teaching and underscore the importance of considering individual characteristics and motivations when attracting prospective teachers to the profession.

Introduction

Teachers are in short supply in many jurisdictions around the world. According to UNESCO, nearly 69 million new teachers are needed to meet basic educational requirements by their 2030 goal (UNESCO, 2022). To achieve this target, there is a pressing need to attract and encourage individuals to apply for initial teacher training (ITT). In the UK, where this study is conducted, there has been a significant decline in ITT recruitment, with numbers falling from 94% of the government’s recruitment target in 2018–19 to only 44% of target in 2022–23. This decline in teacher recruitment is not limited to the UK but is also observed in other countries. For example, in Australia, interest in pursuing ITT has dropped by 20% in 2023 (Cassidy, 2023), while the number of students completing ITT in the US has been steadily declining since 2014 (Patrick, 2023). Shortages of new teachers have also been reported in the Middle East and Africa (Vargas-Tamez, 2022). Furthermore, the scarcity of teachers is particularly prominent in STEM subjects (Science, Technology, Engineering, and Mathematics) across various countries, including the UK (See et al., 2022; Sibieta, 2018), the US (Ingersoll & Perda, 2010; Kolbe & Strunk, 2012; McVey & Trinidad, 2019; Sutcher et al., 2016), Spain (De Vega, 2018), Germany, Luxembourg, Turkey (Carlo et al., 2013; Schleicher, 2012), Chile (Fajado, 2016), and Australia (Heffernan et al., 2019; McKenzie et al., 2014; Shah et al., 2020) .

In England’s secondary ITT programs, the STEM subjects most severely affected by the recruitment crisis are computer science, with only 30% of the target recruited in 2022–23, and physics, with only 17% of the target recruited (Department for Education, 2022). While biology recruitment numbers remain stronger than other subjects, there has been a decline in the percentage of the target recruited, dropping from 153% in 2018–19 to 85% in 2022–23. The only STEM subject that has seen an increase in teacher training recruitment in recent years is mathematics, with 70% of the target recruited in 2018–19 and 90% in 2022–23 (Department for Education, 2022). These statistics highlight the lack of interest among STEM undergraduates in the UK in pursuing a teaching career, with only 40% indicating an interest (Thompson-Lee et al., 2024).

Efforts to address the recruitment crisis in the UK have primarily focused on providing financial incentives to STEM graduates to pursue ITT. For example, starting from the 2023–2024 academic year, bursaries of £27,000 were made available to teacher trainees entering chemistry, computing, mathematics, and physics programs. In addition, candidates with a first-class or upper second-class (2:1) degree, master’s degree, or PhD were eligible for a £29,000 scholarship, which also offered additional benefits such as access to resources and specialized training. However, there is limited evidence to suggest that these bursaries effectively increase the number of ITT applicants (Allen et al., 2016; Noyes et al., 2019; See & Gorard, 2020). Despite these incentives, the UK continues to fall short of meeting the recruitment targets set by the government. These approaches have faced criticism for their primary focus on degree classification rather than the essential skills, attributes, and values that contribute to effective teaching (Noyes et al., 2019).

In addition, among newly qualified teachers who received bursaries for their initial teacher training, only 67% remain in teaching after 5 years (Fullard & Zuccollo, 2021). Research has shown that while financial incentives may increase the likelihood of individuals who are already interested in teaching applying for ITT, they do not have a significant impact on retention rates (See et al., 2022). Outside of the UK, efforts have been made to attract STEM undergraduates to consider ITT. STEM undergraduates typically have more career opportunities than students in other fields, making it important to make teaching an appealing option for them (Kunz et al., 2020). Like in the UK, globally, the most common approach to encouraging STEM undergraduates to pursue teaching is through financial incentives, as highlighted in a STEM recruitment review by Thompson-Lee et al. (2024). Moreover, most studies on recruitment interventions have been conducted using samples from the US, so there is limited information available on the effectiveness of these interventions in different countries and contexts.

In a longitudinal study conducted by Watt and Richardson (2013), which followed teachers from the start of their ITT through their first 8 years in the classroom, individuals who were motivated by a desire to make a social contribution (i.e., the social utility of the teaching career) were found to be more likely to stay in the profession compared to those primarily motivated by financial gains (i.e., the personal utility of the teaching career). Therefore, recruiting individuals who prioritize financial incentives over social motivations is unlikely to result in long-term retention within the teaching workforce (Watt et al., 2013). As a result, UK policymakers have acknowledged the limited effectiveness of financial incentives in recruiting teachers. Recent advertising campaigns, such as the UK Department for Education’s (DfE) Every Lesson Shapes a Life campaign in 2020, have shifted their emphasis away from financial rewards and instead, highlight the ‘social utility’ of teaching and the individual’s person–vocation ‘fit’ with a teaching career. However, there is still a lack of comprehensive research that evaluates the effectiveness of these relatively new recruitment initiatives in attracting individuals to the teaching profession. In addition, there are only a limited number of studies that specifically target STEM subject specialists’ interest in a teaching career.

Regarding intervention designs, research suggests that using realistic depictions of the teaching profession can effectively attract individuals who may find satisfaction in teaching (Baur et al., 2014). For instance, a study focused on increasing STEM undergraduates’ interest in teaching through realistic job previews (RJPs) found a positive correlation between students’ perception of how well teaching aligned with their skills and their interest in pursuing a teaching career (Klassen et al., 2023). This study demonstrated that providing undergraduates with accurate depictions of the teaching profession and feedback on their suitability for the role resulted in increased interest in teaching. Expanding on the findings of Klassen et al. (2023), this study aims to enhance recruitment efforts by using authentic classroom vignettes as a persuasive intervention to further increase the interest and perceived fit of STEM undergraduates in teaching.

Motivation theories pertinent to teacher attraction, recruitment, and retention

Recruitment strategies aimed at targeting individuals with specific motivations for pursuing a career in teaching can attract suitable candidates who are more likely to thrive in, and remain committed to, the profession. From a theoretical point of view, there are two motivation-related theories that hold significant importance in the context of teacher recruitment interventions: FIT-choice motivation framework and the person–vocation fit framework.

The FIT-choice motivation framework

Watt & Richardson (2007) FIT-choice motivation framework suggests that teachers choose their career based on perceived benefits or utility. In the teaching profession, there are two primary utility values: personal and social. Personal utility refers to factors like job security, job transferability, and stable income. Social utility relates to making a positive impact on society, supporting students, and promoting social equality (Klassen et al., 2021a; Watt & Richardson, 2007). Personal utility focuses on how teaching enhances a teacher’s quality of life, while social utility emphasizes the broader contributions to society and the community.

The influence of personal and social utility values on teacher outcomes can vary significantly. Teachers who prioritize personal benefits often show less commitment to collective development or group benefits (Lönnqvist et al., 2009; Spain et al., 2014). For example, if a teacher highly values pedagogical autonomy but cannot fulfil that value, they may become unmotivated and dissatisfied, potentially leading to job or career dissatisfaction (Gillet et al., 2013). Similarly, teachers who prioritize job security or a stable salary may be initially more inclined to stay in their roles (Wang & Klassen, 2023). However, this inclination does not always translate to organizational or professional commitment (Mohsin et al., 2013). In fact, individuals who prioritize personal utility values are more likely to demonstrate poorer career intentions and leave if their personal needs are not met by the organization or profession (Haivas et al., 2013).

In contrast, social utility values prioritize societal contributions over personal gains. Individuals with higher social utility values show greater career commitment (Grant, 2007), improved job performance (Van Loon et al., 2018), and are less likely to leave their jobs (Thompson & Gregory, 2012). This trend is also observed in the teaching profession. For instance, a study of Taiwanese teachers found that those with stronger altruistic values were more likely to foster trust among colleagues and share knowledge within their groups (Chen et al., 2014). Another study showed that Canadian teachers who put more emphases on making social contributions and building strong student relationships than others tended to report higher job satisfaction, experience lower burnout, and were less inclined to leave the profession (Wang & Hall, 2019; Wang & Klassen, 2023).

While the influence of personal and social utility values on teacher outcomes can vary significantly, it is important to recognize that these values often coexist within teachers. It is unlikely for a teacher to solely endorse either social or personal utility values in their teaching career. Instead, they typically endorse a combination of both, choosing to enter the profession because it allows them to make significant social contributions while also promoting their own quality of life. Therefore, recruitment campaigns that target both the social and personal utility of teaching are likely to be the most effective (Wang & Klassen, 2023).

The person–vocation fit framework

While individual differences such as teachers’ personal or social utility values can influence career intentions, the interaction between these factors, particularly how teachers perceive their ‘fit’ in the profession, can have great impact on their career outcomes (Van Vianen, 2018). According to Schneider et al.’s (1998) attraction–selection–attrition (ASA) framework, individuals are drawn to, chosen by, and retained in organizations or professions that align with their personal attributes. A better fit between teachers’ perception of their abilities, values, and norms and the expectations of the teaching profession is crucial in boosting their work motivation, commitment, and career intentions while reducing attrition (Hayes & Stazyk, 2019; Li et al., 2015; Qiao & Hu, 2024).

The concept of person–vocation fit falls within the broader concept of person–environment fit, originated from industrial and organizational psychology (Cable & Judge, 1996; Kristof-Brown & Billsberry, 2013). This overarching term emphasizes the importance of an individual’s compatibility with their environment in general. Person–environment fit encompasses various sub-components, or “sub-fits,” such as person–organization fit (alignment between an individual and an organization), person–job fit (matching between a person’s ability and the specific job requirement), and person–vocation fit (matching of interests/abilities with the characteristics of the vocation). The impact of each sub-fit on individual outcomes can vary (Kristof et al., 2005; Vogel & Feldman, 2009). For instance, teachers who perceive a poor person–organization fit are more likely to leave their schools, potentially seeking employment elsewhere with different leadership styles or working with different colleagues (Ellis et al., 2017; Miller & Youngs, 2021). On the other hand, teachers experiencing a poor person–vocation fit are more inclined to exit the teaching profession entirely (Klassen et al., 2021a). In the context of attracting teachers, especially those without prior teaching experience (e.g., STEM undergraduate students), the construct of person–vocation fit becomes particularly relevant. Only when potential candidates perceive an alignment between their personal characteristics and the teaching profession’s requirements can they seriously consider pursuing a career in education or obtaining a teaching degree.

The strong associations between teachers’ person–vocation fit and their career intentions suggest that recruitment strategies focusing on this “power of fit” can effectively influence attitudes and behaviours. By leveraging this approach, it becomes possible to attract more suitable candidates to the teaching profession and retain high-quality teachers for longer periods (Ehrhart & Markransky, 2007; Klassen et al., 2023). Essentially, while the utility value factor explains who is more likely to have strong career intentions, enjoy teaching, and remain in the profession long-term, the person–vocation fit factor offers a situated explanation of who is likely to stay in teaching specifically, taking into account the interplay between individual and occupational characteristics.

Teacher recruitment intervention design

There is a lack of research exploring teacher recruitment interventions that go beyond financial incentives or traditional advertisements. One study, conducted by Klassen et al. (2023), investigated the impact of utilizing realistic job previews (RJPs) with STEM undergraduates. RJPs are a recruitment approach where potential applicants are provided with an accurate portrayal of what to expect in a specific job role in a specific organization (Wanous, 1973). This approach aims to present a balanced view of the job, encompassing both the rewarding and challenging aspects of the role (Baur et al., 2014).

RJPs are not limited to mere job descriptions, as is typical in traditional job advertisements; instead, they strive to capture the essence of daily job routines, work environment, organizational culture, and potential difficulties one might encounter. This is often accomplished through various mediums, such as written materials, video presentations, or even simulation activities. By offering this transparency, RJPs help prospective employees to self-assess their suitability for the job, thus facilitating an informed decision-making process. Most importantly, a thorough understanding of a job role through RJPs can help the candidates set realistic expectations, which in turn can enhance job satisfaction, improve job performance, heighten person–vocation fit, thereby increasing retention rates and job success on a long run (Baur et al., 2014; Ellis et al., 2017; Roberson et al., 2005). In the context of teacher recruitment, Klassen et al. (2023) found that person–vocation fit measured by the RJPs and interest in a teaching career were positively associated with each other and that the scores on the RJPs predicted interest in teaching. This suggests that the interventions providing realistic depictions of teaching could increase potential applicants interest in a teaching career.

One established method of providing potential applicants with realistic depictions of a job is experimental vignette methodology (EVM). EVM involves the use of carefully constructed and realistic scenarios to evaluate intentions, attitudes, and behaviours (Aguinis & Bradley, 2014; Skilling & Stylianides, 2020). Vignettes, in this context, are concise and detailed descriptions of a person, object, or situation that represent a systematic combination of characteristics (Atzmüller & Steiner, 2010). By using vignettes, researchers can measure attitudes and behaviours in response to hypothetical scenarios, which can be more reliable than survey methodologies that focus solely on current personal circumstances (Schoenberg & Ravdal, 2000). In education research, vignettes can be useful to explore value-laden constructs such as teaching-related beliefs and motivations (Skilling & Stylianides, 2020). Well-written and appropriate vignettes should adequately reflect realistic teaching scenarios, thereby illustrating the key motivations, such as perceived utility and person–vocation fit in teaching.

Using the person-centred approach to understanding intervention effectiveness

Researchers often use variable-centred approaches to evaluate the effectiveness of interventions that promote teacher motivation (e.g., Daniels et al., 2021; Horowitz et al., 2018). This research method assumes that the population is homogeneous in terms of how certain variables, such as an intervention and its components, influence individuals. The aim is to examine the individual or partial effects of predictor variables on outcome variables, such as career intentions. The advantage of using a variable-centred approach is that it clearly demonstrates the impact of an intervention, facilitates comparison between different intervention components, and identifies the most effective component. However, this method has limitations in terms of providing insights into the specific percentage or groups of participants that can be positively or negatively influenced by the intervention. It also overlooks the interaction among multiple components of an intervention within groups of individuals, such as the combined effects of personal and social utility of teaching (Tetzlaff et al., 2023).

In contrast, the person-centred approach assumes that the population is heterogeneous in terms of how variables operate within groups of individuals (Morin et al., 2018). This approach focuses on different combinations of variables and classifies individuals into subgroups that share similar characteristics. The objective of this research approach is to gain empirical insights into the size and characteristics of each group, as well as their differences in outcome variables. With this information, researchers can gain a better understanding of the effectiveness of their interventions and how various intervention components function collectively within each group of individuals (Tetzlaff et al., 2023).

Contrasting utility values and fit in the person-centred research

While utility values and person–vocation fit are closely linked to teachers’ career intentions, they are separate constructs that both play essential roles in interventions aimed at fostering career intentions in teaching. It is important to note that individuals who solely recognize the social utility of teaching may not see themselves as fitting well within the profession. As a result, they may still be reluctant to pursue a career in teaching. Conversely, individuals who feel a strong alignment between themselves and the teaching profession may not necessarily embrace the appropriate social utility values. They might be drawn to teaching primarily for personal gain, such as financial incentives or job security.

Previous research has demonstrated that these individuals who are attracted to the teaching profession by personal gains are less likely to remain in the profession (e.g., Wang & Hall, 2019). However, since personal utility and social utility values are likely to coexist and be positively correlated with each other (Wang & Klassen, 2023), the optimal intervention should aim to persuade potential candidates to adopt a balanced level of these utility values. This includes predominantly emphasizing the social utility of teaching while also acknowledging some degree of personal utility. Simultaneously, it is important for these candidates to perceive a strong person–vocation fit in the teaching profession.

To further assess the effectiveness of the recruitment intervention based on the frameworks of utility values and person–vocation fit, the use of the person-centred approach helps to provide valuable insights into the specific percentage and profiles of individuals who are differently impacted by the intervention and thus, fall into possibly different categories. Specifically in the present study, based on a 2 × 2 dichotomy that considers both individuals’ utility values (high vs. low) and perceived fit in the teaching profession (high vs. low), there might be four distinct groups of intervention participants: (a) those who are persuaded by both the value and fit aspects (value and fit; highly persuaded), (b) those who are convinced of the utility values of the teaching profession but do not see themselves as a good fit (value but not fit; moderately persuaded), (c) those who are not convinced of the utilities of teaching but perceive a strong personal fit in teaching (fit but not value; moderately persuaded), and (d) those who are not convinced by either the values or fit aspects of the recruitment intervention (no value, nor fit; unpersuaded). By obtaining this information, researchers can obtain a comprehensive picture of the true effectiveness of the intervention and subsequently revise and update the intervention contents based on the characteristics of each profile.

The present study

Informed by both the FIT-choice motivation framework and person–vocation fit theory, this research aims to assess the effectiveness of a recruitment intervention that is innovatively designed to use Experimental Vignette Methodology to attract potential teachers. In addition, the study employs a person-centred approach to explore the collective contributions of the utility of teaching and person–vocation fit in a recruitment intervention. It investigates the specific profiles of intervention participants, their characteristics, and how these profiles differ in terms of their self-efficacy, interest, enjoyment, and perceived fit in the teaching profession after the completion of the recruitment intervention. Given the exploratory nature of this study, the focus is on addressing the following research questions rather than investigating specific hypotheses:

Research Question 1: How many participant profiles can be identified based on changes in interest in the teaching profession throughout the recruitment intervention? This research question is generated based on the FIT-choice model (Watt & Richardson, 2007) and person–vocation fit theory (Kristof-Brown & Billsberry, 2013) which indicate that individuals are motivated by different combinations of values and a perception of person–vocation fit when considering a teaching career (e.g.Klassen et al., 2023; Uggerslev et al., 2012). Perception of fit (in terms of utility values and personal attributes) is positively associated with career choices, including the choice to enter teaching. Individuals who perceive a fit between their personal attributes and teaching, or between the utility values associated with teaching, are likely to have a strong interest in pursuing a career in teaching. They are also more likely to be influenced by an intervention that emphasizes this fit. In contrast, individuals who do not see themselves as fitting into the teaching profession, or who perceive a mismatch in values, are expected to have lower interest in teaching and be less affected by interventions promoting fit. Specifically, based on the 2 (high/low utility values) × 2 (high/low perceived fit) model, we propose there are four distinct profiles of participant responses to the scenarios: (a) highly persuaded by value and fit, (b) moderately persuaded by value, but not fit, (c) moderately persuaded by fit, but not value, and (d) unpersuaded by value and fit.

Research Question 2: What roles do demographic factors which have been found to influence interest in teaching and interest in STEM, including gender (see Fuchs et al., 2022; Gore et al., 2016; Lai et al., 2005; Van Rooij et al., 2020), ethnicity (see Demie & See, 2023), and socioeconomic background (see Gorard & See, 2009), play in determining these intervention profiles? Specifically, we considered gender as a potential influencing factor. This is especially important when aiming to attract STEM undergraduates to the teaching profession. STEM fields tend to have a higher proportion of male students (73% according to 2021/22 data from HESA), yet only a small percentage of them choose to pursue teaching as a career (28% according to the ITE Census for 2021/22). In addition, we also examined ethnic groups as a factor, given the underrepresentation of ethnic minorities in the teaching profession in the UK (Demie & See, 2023). Understanding their utility values and perception of fit in teaching is essential for research investigations. Finally, we included socio-economic background as a potential antecedent, as previous studies have shown that individuals from lower socio-economic backgrounds are less likely to pursue science-related fields in higher education (Gorard & See, 2009).

Research Question 3: What are the outcomes associated with these identified profiles in terms of the following key indicators of interest in a future teaching career—self-efficacy, interest, enjoyment, and perceived fit in teaching—after the completion of the intervention? We hypothesize that individuals who are persuaded by the utility of teaching or perceive a better fit in the teaching profession are more likely to report high levels of self-efficacy, interest, enjoyment, and overall fit in teaching. Specifically, self-efficacy is associated with positive experiences and the perception of fit (Bandura, 1997). Interest in any career, including teaching, has been found to be associated with the perception of fit (Uggerslev et al., 2012). Given that the intervention aims to highlight values associated with teaching, we would expect participants who are persuaded by these values to experience an increase in fit.

Method

Participants and procedures

Two hundred and sixty-seven participants were recruited from 18 universities across England through direct contact with universities and the UK Institute of Mathematics. The distribution of participants across regions was as follows: the North East (25%), Yorkshire (22%), Wales (14%), the Midlands (10%), the South East (9%), the North West (8%), London (5%), the South West (5%), Scotland (2%), and Northern Ireland (0.2%). The average age of the participants was 24.50 (SD = 10.19). We included individuals of all ages in our sample to align with the current student population, where 37% of undergraduates in 2019/20 were classified as “mature” students (defined in the UK as over 21 years; House of Commons Report, 2021). In our sample, participants over 21 years accounted for 27% of the total. The inclusion of individuals over 40 years constituted 8% of the sample, consistent with data published by The Office for Students in 2017 (12% of undergraduates over 40 years).

The mean age of our sample was higher than the majority of students’ ages in our sample (19–21) due to the representation of mature students to reflect the UK university population. The inclusion of mature students is essential as 47% of postgraduate teacher trainees in England are aged over 25 (Department for Education, 2022). Therefore, mature students play a significant role in researching undergraduates’ motivations to pursue teaching. Fifty-five percent of our sample identified as female, aligning closely with 56% of females in the undergraduate population in 2023. In addition, 78% of participants identified as White, reflecting 75% of the undergraduate population that is White in the academic year 2021/22.

The participants were studying for the following degrees: mathematics (49%), physics (7%), computing/computer science (7%), chemistry (1%), biology (1%). Fifteen percent were studying combinations of these STEM subjects. Twenty-one percent were studying “other” STEM subjects, mostly psychology (48%). We included participants studying psychology in our definition of STEM because graduates of psychology degrees in the UK can teach science and mathematics in secondary schools when they complete a short subject knowledge enhancement (SKE) course (Department for Education, 2023a, 2023b). Psychology is also included in the definition of STEM by the UK Government and the Universities and Colleges Admission System (UCAS; Circuit, 2023; House of Lords, 2012).

The questionnaire was administered using the online survey platform Qualtrics through an anonymous link. This allowed participants to complete the study in their natural environment this allows the vignette methodology to have more realism (Aguinis & Bradley, 2014). The participants completed the survey in one sitting which enhances the validity of the results (Aguinis & Bradley, 2014). The first section of the questionnaire collected information on sociodemographic characteristics, including gender, ethnicity, program level, and socioeconomic background. After completing this section, participants were presented with the intervention, which included 12 vignettes focusing on the social utility, personal utility, and person–vocation fit of teaching. Following the vignettes, participants were asked to complete a 17-item questionnaire that assessed their interest in teaching, imagined teaching self-efficacy, perceived fit for a future teaching career, and imagined enjoyment of teaching (see Fig. 2 for more details regarding the study design). Each participant who completed the questionnaire received a £5 shopping voucher.

Measures

Vignettes and motivational messages

When designing the experimental vignette intervention, we followed Aguinis and Bradley (2014) comprehensive guide for EVM construction and we also referred to Skilling & Stylianides framework for vignettes in education research (2020). In line with Skilling and Stylianides, the content of vignettes was informed by theoretical frameworks (FIT-choice, personal–vocation fit), existing literature about teacher motivation and PV fit, and the practical experiences of past teachers. The current intervention comprised 12 brief vignettes that depicted 12 different aspects of teaching, representing three motivations for choosing teaching: personal utility (salary and pension, job security, job transferability, time for holidays and family, mastery, and autonomy), social utility (shaping the future of children/adolescents, enhancing social equity, making a social contribution), and person–vocation fit in teaching (conscientiousness/organization/planning, resilience/adaptability, empathy/communication/emotion management).

The vignettes were co-written by the researchers, including an educational consultant with 12 years of experience as a secondary school teacher. When designing the intervention, we considered the length of the vignettes. Research suggests that short vignettes can increase attention and response rates (Hughes & Huby, 2004), while longer vignettes tend to be more engaging and potentially persuasive (Skilling & Stylianides, 2020). To compare the effectiveness of short, direct vignettes versus longer, more narrative vignettes, we conducted a small-scale pilot study. The results indicated that participants rated the long narrative vignettes as more persuasive than the shorter alternatives. The length of the vignettes ranged from 81 to 122 words, falling within the recommended parameters for vignette length in recent educational research frameworks (50–200 words, Skilling & Stylianides, 2020). To ensure consistency, we controlled the length of each vignette so that the variation in length was less than two standard deviations from the mean length (M = 97.58, SD = 15.38).

Each vignette was crafted from the perspective of a newly qualified STEM teacher starting their first teaching position. The level of immersion participants would experience was considered when we designed the vignettes (following Aguinis & Bradley, 2014). Specifically, we opted for a second-person narrative voice in which participants could view themselves as the central character, as previous research has suggested that this approach enhances reflection and immersion (Iacovides et al., 2022). To provide clarity and generate interest, participants were given a brief introductory text about the character and the school before encountering the vignettes (refer to Fig. 1). In addition, each vignette was accompanied by a classroom image that complemented its content. Using images helps to enhance the realism and level of immersion created by the vignettes (as suggested by Aguinis & Bradley, 2014). In line with Aguinis and Bradley’s (2014) recommendations on transparency we provide all the materials used, including pictures at the following link: https://osf.io/yu8mq/.

Fig. 1

Example trial: Social utility—Shape the future of children. Read the following scenario and then answer the following question: You notice that your tutee, Muhammad is struggling: he is the last to leave for class and arrives to registration on his own. His teachers say he is focused, bright, but has no clear friendship group. You meet with Muhammad to ask how he is settling. He shares that outside of school, he really likes robotics: some Sixth Form pupils are starting a robotics club, you ask if he would be happy to help them plan the club with some other members of his year group: he is really keen! It is a first step, but you can see Muhammad looking brighter already; it feels great knowing you can have such a positive impact. Compared to your interest in teaching before reading the scenario, are you more or less interested in teaching now?

To ensure the validity of the vignette content, we conducted a validation exercise involving 12 participants, consisting of eight teachers and four non-teachers. They were asked to match each vignette with one of the three motivation types. The validation exercise resulted in a high agreement rate, with 11 out of 12 vignettes correctly aligned with their intended motivation (91% agreement). The vignette that showed relatively lower levels of matching, specifically related to a subdimension of social utility value called shaping the future of children and adolescents (55% initial correct matching), was subsequently revised to ensure accurate classification. Table 1 provides an overview of the three motivations, their corresponding motivational messages, and an example vignette.

Table 1 Motivations and example vignettes

After reading each vignette, participants were asked to rate their level of interest in teaching compared to their interest before reading the scenario. This was done using a follow-up question: “Are you more or less interested in teaching now?” (see Fig. 1). Participants could rate their interest on a scale ranging from − 2 (much less interested) to 2 (much more interested), with − 1 indicating a little less interest, 0 indicating no change in interest, and 1 indicating a little more interest. In this context, a positive score in participants’ change in interest indicated that they were persuaded by the vignette and became more interested in teaching after reading it. Conversely, a negative score suggested that participants were dissuaded by the vignette and became even less interested in teaching; and a score of zero meant that the participants were neither persuaded nor dissuaded by the vignette. To mitigate any potential order effects, each trial, consisting of a vignette and its corresponding follow-up question, was presented in a randomized order to each participant.

Participants were asked to complete a brief questionnaire immediately after the intervention. The phrase “After reading the scenarios…” was presented before each item’s wording, to prompt participants to reflect on their interest, self-efficacy, perceived fit, and enjoyment in teaching that may have specifically resulted from participating in the intervention and reading all the vignettes. For each question, participants were asked to select a response on a 7-point Likert scale (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly disagree). The questionnaire measured the following outcomes.

First, to measure participants’ interest in teaching, we used two items that were adapted from Klassen et al. (2021b) and based on Hackett et al.’s occupational commitment scale (Hackett et al., 2001). The two items were as follows: “I am interested in exploring teaching as a career,” and “I am considering training as a teacher.” Second, to assess participants’ self-efficacy for teaching after the intervention, we employed three items adapted from the Teachers’ Sense of Efficacy Scale (TSES, Tschannen-Moran & Woolfolk Hoy, 2001). This short form, consisting of three items, has demonstrated reliability and validity as a measure of teacher self-efficacy in previous research (e.g., Klassen & Durksen, 2014; Klassen et al., 2021b). The specific items used in this study were: “If I were a teacher, I would be confident that I could connect well with my students,” “If I were a teacher, I would be confident that I could create a positive classroom atmosphere,” and “If I were a teacher, I would be confident that I could become an effective teacher.”

Third, to assess perceived fit, we used three items adapted from Klassen et al. (2021b), which were originally derived from Chuang et al. (2016). The items included in this study were: “There is a close match between my skills, knowledge, and abilities and those required for a teaching career,” “There is a close match between my personal characteristics (e.g., personality) and those required for a teaching career,” and “There is a close match between my interests and those required for a teaching career.” Finally, participants’ perceived enjoyment in teaching was measured using two items adapted from Frenzel et al. (2016). The items included “I think I would enjoy teaching” and “I can imagine being happy as a teacher.” (Fig. 2).

Fig. 2

Overview of the study design

Data analyses

We conducted data analyses using a three-step procedure. First, we conducted preliminary analyses to calculate descriptive statistics and examine correlations among the variables under investigation. In addition, confirmatory factor analysis (CFA) was performed to establish the construct validity of the variables used in this study. In the next step, a latent profile analysis was conducted to identify different profiles based on participants’ change of interest in teaching after each vignette. The characteristics of each profile were then examined. Finally, we compared the profiles in terms of participants’ demographics, including gender, ethnicity, program level, and socioeconomic background. Finally, we examined their interest, self-efficacy, perceived fit, and enjoyment in teaching.

We used MPlus 8 software (Muthén & Muthén, 1998–2017) to conduct the latent profile analysis (LPA). The aim of the LPA was to determine the appropriate number of profiles based on participants’ changes of interest in teaching. The profile indicators consisted of the three motivational constructs: personal utility, social utility of teaching, and person–vocation fit in teaching. To identify the ideal number and pattern of profiles, we considered several factors, including fit indices and theoretical interpretability (Lubke & Muthén, 2005; Nylund et al., 2007). Multiple indices were monitored, including the Akaike information criterion (AIC; Akaike, 1973), Bayesian information criterion (BIC; Schwarz, 1978), sample-adjusted Bayesian information criteria (SABIC; Sclove, 1987), Vuong–Lo-Mendell–Rubin Likelihood Ratio Test (Asparouhov & Muthén, 2012), and Lo-Mendell–Rubin test (LRT; Lo et al., 2001). Lower scores on AIC/BIC/SABIC indicate more optimal profile solutions (Nylund et al., 2007). In addition, a significant p value on the Vuong–Lo-Mendell–Rubin Likelihood Ratio Test or LRT indicates that the k-class model fits significantly better than the k-1 model and should be selected. In cases where different fit criteria suggested multiple potential profile solutions, we chose the optimal solution based on theoretical interpretability.

In this study, we investigated various LPA models by specifying class-specific and class-invariant means, variances, and covariances. This approach enabled us to identify the most optimal profile solutions (Lubke & Muthén, 2007). We selected the profile solution with class-specific means and class-invariant variances based on its comparatively low AIC/BIC/SABIC scores, high stability, and infrequent occurrence of local maxima for the best log likelihoods. To ensure accuracy and avoid convergence on a local solution, we conducted 600 starts with 120 iterations in the initial stage and 120 final stage optimizations for each model.

In the next step of our analysis, we compared the profiles in terms of participants’ demographics and the outcomes of the intervention on persuading the STEM undergraduates to enter the teaching profession (i.e., their interest, self-efficacy, perceived fit, and enjoyment in teaching). Specifically, we tested whether gender (male vs. female), ethnicity (white vs. person of color), program level (STEM undergraduate years 1–4), and family socioeconomic background was systematically associated with profile membership using the Mplus automated three-step procedures (R3STEP command; Asparouhov & Muthén, 2014a). To assess differences in the intervention outcomes (i.e., self-efficacy, perceived fit, interest, and enjoyment in teaching) among the profiles, we employed the automatic Bolck–Croon–Hagenaars (BCH) approach in Mplus (Asparouhov & Muthén, 2014b).

Results

Preliminary analyses

Table 2 displays the descriptive statistics and latent correlations among the study variables. The correlation analyses revealed that participants who reported higher interest after reading the vignettes on the personal utility of teaching were also more likely to report similar responses after reading vignettes about the social utility of teaching and person–vocation fit. Likewise, participants who were influenced by the vignettes on the social utility of teaching were also more likely to be influenced by the vignettes on person–vocation fit. Furthermore, participants who reported increased interest after reading all three types of motivation vignettes were also more likely to report higher levels of self-efficacy, interest, perceived fit, and enjoyment in teaching. Lastly, all four types of outcomes (self-efficacy, interest, perceived fit, and enjoyment in teaching) exhibited positive and significant correlations with each other.

Table 2 Psychometric properties of study variables and latent correlations between variables (N = 267)

To establish the construct validity of the study measures as latent variables, a comprehensive confirmatory factor analysis (CFA) was conducted. This analysis encompassed all the study variables, including three motivation types assessed in the recruitment intervention (personal utility values, social utility values, and person–vocation fit), and four intervention outcomes (self-efficacy, interest, perceived fit, and enjoyment in teaching). The findings indicated an excellent model fit: CFI = 0.977, TLI = 0.972, RMSEA = 0.038 [0.026, 0.049], SRMR = 0.036. The standardized factor loadings were all significant, ranging from 0.419 to 0.975.

Determination of profile solutions

We performed latent profile analysis (LPA) using models ranging from one to eight profile solutions (see Table 3 for details). The results indicated that the four-profile solution exhibited the lowest BIC compared to all other models, although it did not have the lowest AIC or SABIC. Figure 3 visually represents the fit criteria, clearly showing that the BIC is lowest at the four-profile solution, after which the line went up. Although the AIC and SABIC continued to decrease after the four-profile solution, the decrease was minimal. Furthermore, both the Vuong–Lo-Mendell–Rubin Likelihood Ratio Test and Lo–Mendell–Rubin Adjusted LRT Test supported the four-profile solution over the three-profile solution, and subsequent models tested after the four-profile solution did not exhibit significant improvement. Thus, all these criteria strongly indicated that the four-profile solution should be adopted. Further investigation into the four-profile solution confirmed it as the optimal model, as it displayed high classification probabilities for the most likely latent class membership (ranging from 87% to 99%). These probabilities exceeded the suggested threshold of 70% for satisfactory classification precision, as proposed by Nylund et al. (2007).

Table 3 Model fit indices for the latent profile analysis

Fig. 3

Results of latent profile analysis

The first profile, consisting of only 2% of participants, was labeled “dissuaded participants.” These individuals reported decreased levels of interest in the teaching profession after reading all three types of motivation vignettes (personal utility, social utility, and person–vocation fit). They were particularly dissuaded by the social utility aspects of the teaching profession, experiencing the sharpest decline in teaching interest after reading vignettes focused on the social contributions of the career, followed by person–vocation fit in teaching and personal utility of teaching.

The second profile, representing 45% of participants, was labeled “unpersuaded participants.” These individuals reported very slight increases in interest in teaching after reading the personal utility and social utility vignettes, but a slight decrease in interest after reading the person–vocation fit vignettes. Given the minimal changes in teaching interest, we considered them unpersuaded by the intervention.

The third profile, comprising 36% of participants, was labeled “moderately persuaded.” Participants in this profile were persuaded by the vignettes addressing the personal utility and social utility values of the teaching profession. However, they were not influenced by the person–vocation fit vignettes, resulting in minimal changes of their interest in teaching. Therefore, we considered them moderately persuaded.

The final profile, labeled “highly persuaded,” represented 17% of the participants. These participants reported the highest changes in teaching interest and were persuaded by all three motivation dimensions of the vignettes. They were particularly influenced by the social utility values of the teaching profession, showing the strongest increases in their interest in teaching after reading vignettes focused on the social contributions that teachers can make, followed by the personal utility vignettes and person–vocation fit vignettes (refer to Table 4 and Fig. 4).

Table 4 Latent profiles of the intervention and the outcomes of the intervention

Fig. 4

Intervention profiles and their post-vignette change in teaching interest

In the subsequent step, we examined whether there was an association between the demographic variables (i.e., gender, ethnicity, program level, and socioeconomic background) and the four identified profiles. This was done using the three-step approach in Mplus (Asparouhov & Muthén, 2014a). No discernible differences were observed in the classification across profiles regarding participants’ gender, ethnicity, and program level. This suggests that the likelihood of being dissuaded, unpersuaded, or persuaded by the intervention was similar for male and female students, white students or students of color, and students throughout all 4 years of the undergraduate program. However, it was observed that students from families with a higher socioeconomic background were more likely to fall into the highly persuaded profile than the unpersuaded profile (see Table 5 for more details).

Table 5 Demographic predictors of the intervention profiles

Associations with occupational well-being

We conducted an automatic BCH analysis (Asparouhov & Muthén, 2014b) to examine the relationships between the four identified profiles and participants’ reported interest, self-efficacy, perceived fit, and enjoyment in teaching after the intervention and exposure to all vignettes. The results revealed significant differences among all four profiles in all assessed outcomes. Specifically, participants who were “dissuaded” by the intervention vignettes reported the lowest levels of the assessed outcomes, followed by those who were “unpersuaded” and then, “moderately persuaded.” Those who were “highly persuaded” reported the highest levels in all four assessed outcomes (refer to Table 4 and Fig. 5 for more detailed information on means and standard errors).

Fig. 5

Outcomes of the intervention profiles concerning students’ self-efficacy, interest, perceived fit, and enjoyment in teaching

Discussion

A brief summary of results

In this study, we conducted a recruitment intervention aimed at persuading STEM undergraduate students in the UK to pursue a teaching career, trying to address the long-term STEM teacher shortage crisis. The recruitment intervention was developed based on the FIT-choice motivation framework (Watt & Richardson, 2007), which suggests that individuals are primarily attracted to the teaching profession due to the personal or social utility factors associated with teaching. In addition, we incorporated the concept of person–vocation fit (Uggerslev et al., 2012; Vogel & Feldman, 2009), situated in the psychological theory of person–environment fit (Kristof et al., 2005), which suggests that individuals are more likely to be drawn to the teaching profession if they perceive a strong alignment between their own characteristics and those of the vocation. To develop the intervention content, we used the Experimental Vignette Methodology. Furthermore, to examine the specific profiles of participants who were differentially impacted by the recruitment intervention, we employed a person-centred approach in data analysis. Finally, we investigated how the identified profiles differed at the end of the intervention in terms of participants’ interest, self-efficacy, perceived fit, and enjoyment in the teaching profession.

The distribution of participants across the four identified profiles offers valuable insights into the effectiveness of our recruitment intervention in influencing undergraduate STEM students’ interest in teaching. The “dissuaded” profile, comprising 2% of participants, experienced a decline in teaching interest after the intervention. These individuals were particularly dissuaded by the social utility values of the teaching profession. Conversely, the “unpersuaded” profile, representing 45% of participants, exhibited minimal changes in teaching interest, indicating that the intervention had a limited impact on this group. The “moderately persuaded” profile, which accounted for 36% of participants, demonstrated some level of persuasion by the personal and social utility vignettes but not by the person–vocation fit vignettes. Finally, the “highly persuaded” profile, consisting of 17% of participants, showed the greatest increases in teaching interest after exposure to all three types of vignettes. These participants were particularly influenced by the social utility values of the teaching profession. Across all four profiles, the “highly persuaded” group displayed the highest levels of enjoyment, interest, perceived fit, and self-efficacy in teaching after the intervention, followed by the “moderately persuaded” group. The “unpersuaded” group had lower levels in these outcomes, and the “dissuaded” group demonstrated the least favourable outcomes.

Initially, we hypothesized that there would be four profile types based on the 2 × 2 dichotomy: (a) those who are persuaded by both the values and fit aspects of the recruitment intervention (value and fit), (b) those who are convinced of the utility values of the teaching profession but do not perceive a good fit (value, but no fit), (c) those who are unconvinced of the utility values of teaching but perceive a strong personal fit in teaching (no value, but fit), and (d) those who are unpersuaded by either the value or fit aspects (no value, nor fit). Our results suggest that profiles (a) value and fit (the “highly persuaded” profile), (b) value but not fit (the “moderately” persuaded), and (d) no value nor fit (the “unpersuaded” profile), do exist in our sample, but profile (c) differed from our expectations. There does not seem to be a profile of participants who were solely convinced by the “fit” scenarios, but not the “value” scenarios (no value, but fit). Instead, our results suggest that there was an alternative profile of participants—the “dissuaded” profile. These participants reported a decrease in interest in teaching after participating in the intervention.

Furthermore, it is important to note that we did not identify any gender differences among the profiles. This means, both males and females were equally represented in each group. Considering that gender imbalances may contribute to the shortage of STEM teacher trainees, as STEM fields tend to have more male students while teaching attracts more females, the lack of gender disparity in our current findings is a promising result. Similarly, the lack of impacts of ethnicity and grade level on profile membership are noteworthy. However, socioeconomic status (SES) emerged as a significant factor for membership in the highly persuaded group. Research indicates that individuals from higher SES backgrounds tend to have greater access to extracurricular activities and attend schools with better quality and resources (Tompsett & Knoester, 2023). While this finding does not directly explain why STEM undergraduates from high SES backgrounds in our sample were more likely to be influenced by our intervention, it is possible that their positive learning experiences, stemming from better school quality, have shaped a more favourable or open-minded attitude toward teaching.

Interpretations and implications of the results

The distribution of participants across the identified profiles highlights the heterogeneity in participants’ responses to the intervention. Therefore, it is important to consider these profiles and their characteristics when designing future interventions to attract STEM undergraduate students into teaching. Specifically, more than 50% of the students who participated in our recruitment intervention were moderately or highly persuaded, reporting a change of their interest in teaching after reading the vignettes. These findings suggest that our intervention was generally effective in persuading STEM undergraduates to develop a greater interest in the teaching profession. Furthermore, those who became more interested in teaching were most strongly influenced by the social utility aspect of the profession, such as the opportunity to make social contributions, build relationships with students, and promote social equity.

Based on these findings, and previous research, it appears to be both appropriate and promising to use social utility as a means to attract STEM undergraduates to pursue teaching careers. Previous research, employing a variable-centred approach, consistently demonstrates that people who strongly endorse social utility values exhibit higher career intentions, job satisfaction, and lower emotional exhaustion once they enter a job (Grant, 2007; Levontin & Bardi, 2018; Richardson & Watt, 2014; Thompson & Gregory, 2012; Van Loon et al., 2018; Watt & Richardson, 2007; Willis-Shattuck et al., 2008), this has also been found to be true for the teaching profession (e.g., Chen et al., 2014; Wang & Hall, 2019; Wang & Klassen, 2023). Therefore, our intervention successfully achieved its objective by highlighting the social utility of teaching to potential teachers. Specifically, those participants who endorse values relating to the social impact of a teaching career are more likely to be positively influenced by the intervention.

Furthermore, participants who were “moderately persuaded” and “highly persuaded” by the intervention were also strongly influenced by the personal utility of teaching. This finding is consistent with previous research suggesting that individuals are drawn to the teaching profession not only for social reasons but also because they recognize the personal benefits the teaching job can offer (Watt & Richardson, 2007). These benefits may include job autonomy to make pedagogical decisions and longer holidays to spend quality time with family and children. However, when comparing the personal and social utility factors, this study, along with numerous others (e.g., Klassen et al., 2021a; Wang & Klassen, 2023; Watt & Richardson, 2007), has consistently shown that social utility values have a stronger association with (potential) teachers’ (imagined) career satisfaction.

In this sense, our research findings are consistent with previous research using the Watt and Richardson’s (2007) FIT-choice motivation framework, indicating that teachers who prioritize social utility values are more likely to remain in or express a stronger intention to stay in the teaching profession (e.g., Chen et al., 2014; Klassen et al., 2021a). However, contrary to prior studies that suggest a negative association between personal utility values and teachers’ career intentions (e.g., Wang & Klassen, 2023), our present research employing a person-centred approach proposes that individuals considering a career in teaching may be drawn to the profession by both personal and social utilities of the teaching career. In addition, the present recruitment intervention integrates both the personal utility values, social utility values, and the person–vocation fit in teaching, thus extending beyond previous research which only focuses on one aspect of these motivation constructs. Specifically, when comparing the three motivation factors assessed and addressed in this study, it was discovered that perceived person–vocation fit was the most challenging factor to improve, as compared with personal utility or social utility values in teaching. Although we were successful in persuading over half of the participants to develop a greater interest in teaching through the use of social and personal utility vignettes (the “moderately persuaded” and “highly persuaded” profiles), the perceived fit among individuals in the “unpersuaded” and “moderately persuaded” profiles remained unchanged. This finding suggests that while these participants recognized the value of the teaching profession, they still had reservations about their own suitability for or fit in this profession. On the other hand, the most persuaded group consists of participants who not only endorse the values of teaching but also perceive themselves fitting well with the teaching profession. This alignment results in the greatest increase in their interest in the profession, compared to participants in the other three profiles.

In consequence, the current intervention is most likely to attract these highly persuaded participants to the teaching profession by highlighting the social utility of the profession along with confirming fit with the profession and potentially increasing teaching self-efficacy. To a certain extent, these findings align with prior research that has adopted the person–vocation fit framework (Vogel & Feldman, 2009) in suggesting that STEM undergraduates’ perception of their fit in teaching is strongly linked to their attraction to and interest in the profession.

However, it is crucial to acknowledge that a small percentage (2%) of participants were dissuaded from pursuing a career in teaching after undergoing the current intervention. These individuals reported even lower interest in teaching after reading the vignettes, particularly the ones related to social utility and person–vocation fit. This pattern could be the result of the scenarios highlighting a lack of perceived fit with teaching, resulting in reduced self-efficacy and a lack of change in interest in pursuing teaching. Although the intervention failed to change the views of this 2% of participants, it might be the case that the intervention has ultimately assisted these individuals in making an informed decision about their career path. Specifically, the intervention may have inadvertently helped these participants by confirming their lack of alignment with the expected teaching values, thereby suggesting that a teaching career may not be the best fit for them. Consequently, this would prevent them from entering an unsuitable profession and potentially experiencing a sense of misfit and dissatisfaction in the long run.

Therefore, our results suggest that an intervention highlighting the personal and social utility values of teaching, alongside person–vocation fit, could effectively promote STEM undergraduates’ interest in the teaching profession. By identifying different response profiles to an intervention using realistic depictions of teaching, we can tailor recruitment strategies to target specific groups. We can also be realistic about the potential outcomes of such interventions; that is, a group of people who do not perceive teaching to fit their values are very unlikely to be persuaded to enter teaching. Moreover, our results also imply that emphasizing the social contribution of a teaching career, as well as the person–vocation fit within the profession, may hold the key to addressing teacher shortages and the retention crisis in the UK.

Limitations of the present study

It is important to acknowledge the limitations of the present study. First, the sample size may have limited the generalizability of the findings. Future research with larger and more diverse samples is needed to validate these findings. Moreover, as the study focuses on attracting STEM undergraduates into the teaching profession in the UK context, it may not be equally effective in attracting teachers in other subjects (e.g., language, arts, etc.), or in other countries. In addition, the study relied on self-reported measures, which may be subject to response biases. Future studies could incorporate objective measures or observational data to enhance the validity of the findings.

Moreover, the current study does not provide information on the long-term effects of the intervention. For example, we are unable to determine whether participants in the “highly persuaded” group will ultimately choose to enter the teaching profession. In the present study, we focused solely on assessing the participants’ self-reported interest in teaching, rather than examining their actual behavioural tendencies. In addition, we did not investigate the long-term effects of the recruitment intervention. Participants completed the post-survey immediately after the intervention, so we lacked information on how long the intervention’s impact may persist. It remains unknown whether the intervention’s effectiveness is short-lived, lasting only a few minutes, or if it has a more enduring impact that could significantly alter their perceptions of the teaching profession. Therefore, future recruitment interventions should consider following participants for a longer period, such as until they graduate from their degrees and make a career decision, to gain insights into whether any participants will ultimately find a fit in the teaching profession.

Furthermore, it is important to note that no causal references can be drawn from these analyses. Specifically, we cannot conclude that the positive outcomes observed in the “highly persuaded” group were solely due to the intervention itself or if these participants already present high levels of the outcomes and were, therefore, more likely to be interested in teaching and persuaded by the intervention. Although we attempted to address this issue by specifically asking participants about the impact of reading the vignettes, the study still cannot make causal inferences from the analyses.

Finally, there are also limitations in experimental vignette methodology. Specifically, vignettes involve hypothetical scenarios, which may not fully capture participants’ real-world experiences or reactions. There is a potential for participants to respond differently when faced with actual situations compared to the hypothetical scenarios presented in vignettes. Vignettes may also oversimplify complex classroom phenomena; hence, participants’ responses may not accurately reflect their actual behaviours in real-life contexts.

Conclusion

The results of this study highlight the possibilities of a low-cost and scalable intervention designed to influence STEM undergraduates’ perceptions of teaching as shown by interest in teaching, self-efficacy for teaching and perceived person–vocation fit for teaching. We used a novel experimental vignette intervention to expose STEM undergraduates to authentic teaching situations, and were able to show increases, for some participants, in interest in pursuing a teaching career. By adopting a person-centred approach, the present study identified four profiles of participants, analysed their different characteristics in response to the intervention, and compared their demographic predictors and post-intervention outcomes. By emphasizing the utilities of a teaching career and the person–vocation fit in teaching, the current intervention provides valuable information for educational institutions and policymakers seeking strategies to attract more STEM students into teaching careers. Ultimately, scalable and low-cost interventions such as this one can contribute to addressing the shortage of qualified STEM teachers alongside current advertising and recruitment interventions.