Development of a scale to measure the psychological resources of grit in adults

Abstract Grit, a personality trait characterized by having passion and perseverance for long‐term goals, has been linked to increased performance, retention, and well‐being in various fields. In the 15 years since the original grit scale was published, many studies have investigated factors that promote grit. However, a scale has not yet been developed measuring characteristics that can contribute to higher levels of grit. This study aimed to develop a novel scale to measure the psychological resources of grit. The Delphi technique was used to obtain consensus from a national and international panel of academics and practitioners who are experts in grit and related constructs. A total of 30 participants rated 100 scale items over three rounds of online surveys. Experts agreed that items selected for the final scale (n = 20) were essential, with 85% rating them as important or very important. The scale, called the Grit Psychological Resources Scale, has promise as a helpful tool for practitioners involved in staff development and building capabilities that contribute to goal achievement.

experience better performance and health outcomes across disciplines, professions, and cultures (Datu, 2021). In nursing and medicine there has been considerable interest in grit and the positive role it likely plays in reducing burnout and turnover intention and improving performance and well-being (Cho & Kim, 2022;Dam et al., 2019;Jeong et al., 2019;Jumat et al., 2020;Terry & Peck, 2020). Studies have also investigated the factors that may precede, mediate, or moderate higher levels of grit (Schimschal et al., 2021). The extensive nomological network of characteristics that appear to play a role in achieving long-term goals can be explained by the conservation of resources (COR) theory (Hobfoll, 1989).
The COR theory was developed to provide a rigorous framework for conceptualizing and studying stress. Hobfoll's (1989) model is based on the view that people are inclined to conserve and build valuable resources, and any perceived or actual threat to such capital can result in stress. Resources include objects (e.g., housing, clothing), personal characteristics (e.g., attributes, skills), conditions (e.g., status, health), and energies (e.g., time, money) that people value and thus seek to retain or develop in the presence or absence of stressors (Hobfoll, 1989). As such, the motivation to conserve resources will vary between individuals according to the appraised value and contextual circumstances (Hobfoll, 2001). Building on this theory, Halbesleben et al. (2014) suggest that resources encompass "anything perceived by the individual to help attain his or her goals" (p. 1338).
Using this definition, it follows that many resources can facilitate motivational tendencies toward goal attainment, such as those associated with grit.
Previous research has linked various characteristics to higher levels of grit and, subsequently, improved outcomes. For example, grittier individuals exhibit higher levels of curiosity, self-awareness, and self-determination, which support commitment to one's interests (Dugan et al., 2019;Jin & Kim, 2017;Sheldon et al., 2015). Similarly, perseverance with goals has been linked to increased hardiness, resilience, self-efficacy, growth mindset, and self-regulation (Armstrong et al., 2018;Kannangara et al., 2018;Lovering et al., 2015). The growing body of literature identifying factors contributing to grit indicates an opportunity to develop programs that target underlying characteristics. However, to ensure the design and delivery of effective training, facilitators should first understand the current state of individual or team capability. This study helps to address this need by developing a novel scale that can provide information on characteristics that likely enable grit. Research on measuring the propensity for gritty behavior was anticipated to be of particular interest and benefit to a range of professionals, especially those involved in promoting staff development and achievement.

| BACKGROUND
In a companion paper , a model for the psychological resources of grit is proposed. The model was informed by previous work by Duckworth (2016), who identified interest, purpose, practice, and hope as key assets to increase grit. Following on from this work, a further 16 attributes were identified that could aid in accruing the four psychological resources of grit. Table 1 summarizes expected behaviors exhibited by individuals possessing high levels of these attributes in the context of grit. As indicated, exhibiting strength in the attributes associated with interest and purpose improves one's motivation to learn, adapt, and grow from experiences continuously.
These behaviors are thought to facilitate consistent interest in and thus the pursuit of valuable activities and goals. Equally, the attributes associated with practice and hope help individuals acquire the mindset to overcome challenges in developing expertise. This disposition enables one's perseverance with building capabilities to achieve goals.
However, a single scale has yet to be constructed measuring characteristics that can contribute to increased grit in adults. Existing scales that measure underlying qualities could be used; however, some aspects may not be relevant in the context of grit. That is, scales may measure attributes that are not directly applicable to improving one's level of grit. Furthermore, administering numerous questionnaires can overburden people, negatively impacting response rates and accuracy of answers (Olson, 2014). Without a single scale to measure behaviors thought to improve grit, it may be challenging to develop targeted programs and evaluate pre-and post-training data. Thus, this study aimed to develop a scale to measure the psychological resources of grit. Another study is underway to evaluate the resulting scale's psychometric properties, and these results will be reported in due course.

| Study design
The Delphi technique was used to gain consensus on items to be included in the Grit Psychological Resources Scale (GPRS) from a panel of experts through three rounds of online surveys (Keeney et al., 2011). Following each round, responses were analyzed and used to construct subsequent surveys. Delphi studies have been used to gain consensus among experts on a wide range of matters such as defining concepts (e.g., Raine, 2006), developing guidelines (e.g., Cox et al., 2016), and scale development (e.g., Bauer et al., 2019). The technique has been used in various fields and disciplines, including nursing and health sciences (Sanna et al., 2022;Shimazaki et al., 2021). This study was approved by the University of Tasmania Social Sciences Human Research Ethics Committee, reference number S0020355 (H-70103). The three-round modified e-Delphi was undertaken from July to October 2020.

| Participants
The inclusion criteria for recruitment included being English-speaking researchers, leaders and professionals over 18 years of age who are experts in grit and related constructs. A participant was considered an "expert" if they met at least one of the following criteria: (1) has published or presented on grit/related constructs; (2) has recent experience in a leadership role that involves building capabilities to improve performance; and (3) regularly uses grit/related constructs in a professional setting to help others set and achieve long-term goals.
Purposeful and snowball sampling was used for recruitment. Panel members were identified from reference lists in papers relating to grit and contacts in the first author's professional network on LinkedIn.
This study aimed to recruit 30 participants and have at least 23 complete all three rounds, which is considered adequate to T A B L E 1 Expected behaviors from individuals with high levels of attributes associated with the psychological resources of grit Psychological resource Attribute Individual behaviors contextualized for grit Interest Curiosity (Berlyne, 1954) • Pursues new experiences • Embraces challenges and uncertainty • Slow to lose interest Self-awareness (Duval & Wicklund, 1972) • Often uses self-reflection • Analyzes choices and behavior • Uses insights to learn and adapt behavior Courage (Rate et al., 2007) • Takes considered risks • Not afraid to try new things • Open to change and failure Patience (Schnitker, 2012) • Is patient when exploring new things • Remains calm when learning is hard • Can focus on complex problems Purpose Self-determination (Deci & Ryan, 1985) • Has control over actions • Motivates self to work hard • Pursues meaningful goals Self-concordance (Sheldon & Elliot, 1999) • Pursues activities because they are valuable • Follows through with plans to gain experience • Persists with goals to stay motivated Self-compassion (Neff, 2003) • Can learn from negative experiences • Tolerant of choices and direction • Accepting of setbacks and failure Emotional intelligence (Salovey & Mayer, 1990) • Notices how emotions influence behavior • Is aware of things that bring happiness • Can control emotions and persist with goals Practice Hardiness (Kobasa, 1979) • Remains composed under learning pressure • Feels in control of expertise development • Keeps trying when things are hard Resilience (Carver, 1998) • Good at dealing with challenges • Can overcome setbacks and failure • Improves from adversity Self-efficacy (Bandura, 1977) • Confident with achieving goals • Likes to set hard goals • Attributes failure to effort Flow (Csikszentmihalyi, 1990) • Goals slightly exceed capabilities • Has a clear direction • Can perform seamlessly Hope Goal orientation (Dweck, 1986) • Failure is a motivation • Leans in to challenges and learning • Uses feedback to improve Growth mindset (Dweck, 2006) • Works to change traits • Ties success to hard work • Failure is seen as opportunity Optimism (Abramson et al., 1978) • Does not take setbacks personally • Can overcome problems • Learns from the past Self-regulation (Baumeister & Heatherton, 1996) • Can change behavior • Usually avoids distractions • Adjusts pace to be effective produce reliable results (Akins et al., 2005). Recruitment was conducted during the COVID-19 pandemic, and as such, people were experiencing changes to working conditions and employment.
Accordingly, a large number of invitations (n = 201) were sent out to offset an anticipated low response rate. Potential participants were sent a personal email inviting them to participate and were encouraged to suggest other people from their networks. Invitations included an Internet link that directed participants to the first online survey in REDCap, a secure web-based application that enables data collection for research studies. All participants selfrated themselves against the inclusion criteria and had an opportunity to discuss any questions with the researchers before providing informed consent to participate.

| Survey development
The construction of survey items involved an extensive literature review and input from colleagues with expertise in constructs of interest and scale development. First, using the model for the psychological resources of grit , definitions were developed to define how each psychological resource likely enabled individuals to attain their goals (see Table 1 for resources).
Namely, interest was defined as a psychological resource for grit that enables an individual to explore and deepen their interests via the attributes of curiosity, self-awareness, courage, and patience.
Likewise, purpose enables an individual to make their interests more meaningful via the attributes of self-determination, self-concordance, self-compassion, and emotional intelligence. Together, interest and purpose are resources that likely help people remain consistent with their focus and direction (i.e., demonstrate passion for long-term goals). Practice was defined as a resource that enables an individual to persist with skill development via the attributes of hardiness, resilience, self-efficacy, and flow. Similarly, hope enables an individual to develop and stick with goals via the attributes of goal orientation, growth mindset, optimism, and self-regulation. Collectively, practice and hope are resources that may aid with overcoming setbacks and challenges (i.e., demonstrate perseverance for long-term goals).
Second, definitions were developed for each attribute associated with the psychological resources of grit (see Table 1 for attributes). These definitions then informed the identification of behaviors that would likely be associated with scoring a low and a high score for each attribute. For example, an individual with low self-regulation may struggle to change behavior, be distracted easily, and find it hard to adjust their pace. Conversely, an individual high in self-regulation can change behavior, usually avoid distractions, and adjust their pace to be effective. Third, the behavioral descriptions were used to develop positively and negatively worded items. Care was taken to ensure the items were focused on behavior and were worded in an unbiased way using plain language. Existing scales were also reviewed to elicit insights on item wording, instructions, and response format. Initially, 200 positively and negatively worded items  If I fail it is rarely due to my abilities

| Data collection and analysis
Surveys were set up in REDCap to ensure participants would be required to answer each question, thereby reducing the risk of missing data. Participants were asked to rate items according to the importance of inclusion in the scale using a 5-point Likert scale (not important, slightly important, moderately important, important, very important). After each round, the number of items was reduced according to consensus and a predetermined level of reduction. This approach was chosen to create a short form survey with a low response burden that could be readily applied in a practice setting (Beatty et al., 2020). Consensus was based on the rank ordering of items according to mean rating scores. The number of items retained after each round was predetermined: 60 after Round 1, 40 after Round 2, and 20 after Round 3. A cut-off score for retention was not used. Accordingly, after Round 1, items were rank ordered, and 60 items with the highest scores were selected (ranging from 3.7 to 4.7 for retained items). Following the same process, 40 items were retained after Round 2 (ranging from 3.5 to 4.5), and 20 after Round 3 (ranging from 3.9 to 4.5). When two items had the same mean rating score, the item with the smaller standard deviation was retained.
In Round 1, participants were asked to answer demographic questions for group analysis and rate 100 items. Responses were imported into Microsoft Excel for Mac (version 16.46), and descriptive statistics (mean and standard deviation, median and interquartile range, and frequencies) were generated to determine panel characteristics and group responses to items. Using the approach described above, 60 items, including at least two for each attribute, were retained for the second survey. In Round 2, participants were sent unique links to access the second survey and asked to re-rate the items retained. Responses were analyzed and 40 items, including at least two for each attribute, were retained for the final survey. In Round 3, participants were asked to re-rate the remaining items one final time. Following analysis, 20 items, including at least one for each attribute, were retained for scale construction. After each round, sensitivity tests were conducted to assess the potential for selection bias due to item reduction (Raine, 2006). Retaining a minimum number of items for each attribute did not appear to bias the results.

| RESULTS
Of the 201 experts invited to participate, 30 completed the Round 1 survey (15% response rate), 24 completed Round 2 (80%), and 27 completed Round 3 (90%). In Delphi studies, participants are normally required to respond to all rounds. However, given the design of this study (reduction of items), participants were not required to complete every round, hence the higher participation rate for Round 3. Table 2 shows the panel characteristics in each round. In Round 1, the panel was multidisciplinary and comprised of both national and international experts working in academia and professional practice. The participants' mean age was 49 years, and most had completed higher education, with 20% (n = 6) holding a Bachelor's degree, 30% (n = 9) a Master's degree, and 33.3% (n = 10) a Doctoral degree. Participants worked in a variety of industry sectors, including 43.3% (n = 13) from education and training and 30% (n = 9) from manufacturing. In subsequent rounds, the panel characteristics did not change markedly. Table 3 presents mean rating scores for each item across all rounds. After each round, the number of items was progressively reduced from 100 to 20 according to group responses. After Round 1, 60 items were retained for Round 2, 15 for each resource of grit.
Mean rating scores for retained items ranged from 3.7 to 4.7 out of 5, with 41 (68%) scoring ≥4. After Round 2, 40 items were retained for Round 3, 10 for each resource of grit. Mean rating scores for retained items ranged from 3.5 to 4.5, with 22 (55%) scoring ≥4. After Round 3, 20 items were retained to construct the final 20-item scale with mean rating scores ranging from 3.9 to 4.5, with 17 (85%) scoring  item was retained after Round 2. The mean level of agreement for attributes ranged from 3.9 to 4.5 in the final scale, with self-awareness, self-efficacy, and flow receiving slightly lower ratings. Items linked to overarching purpose and self-determination received the highest ratings of 4.3 and 4.5, respectively. Overall, the mean rating scores for resources and attributes increased with item reduction after each round.

| DISCUSSION
This Delphi study used ratings from a panel of 24 experts to identify items for inclusion in the GPRS, a scale to measure the psychological resources of grit. After three rounds, strong agreement was achieved for 20 of the 100 items proposed for the scale. The mean rating scores for psychological resources of interest, purpose, practice, and hope were similar across all rounds, indicating participants viewed these assets as equally important for grit (Duckworth, 2016). Furthermore, participants rated items related to each psychological resource as important for inclusion in the final scale. Only the overarching item for purpose ("I am committed to my goals") was retained. Overarching items for the psychological resources of interest ("I can usually find ways to remain committed"), practice ("I usually practice to achieve a specific goal"), and hope ("I actively use various tactics to persist with my goals") had good agreement in Round 1 but were dropped in Round 2. The overarching item for interest and items linked to curiosity saw the largest change with a 0.6-point difference. This suggests participants may have had some difficulty rating one item higher than another. Additionally, participants may not have been able to differentiate between overarching items related to a psychological resource and those relating to associated attributes.
After the first round, a preference emerged for one to two attributes linked to each resource. These results largely persisted in the final round. Specifically, no items for self-awareness were removed after the first round. Similar results were seen for four other attributes. Only one item each was removed after Round 1 for self-determination, hardiness, resilience, and goal orientation. The preference for items linked to resilience and goal orientation persisted into the third round, suggesting participants viewed these attributes as more important for grit. Alternatively, the items may have resonated more strongly with participants, especially since the survey was conducted during the COVID-19 pandemic. That is, job losses and changes to working conditions required many people to retrain in new industries and work flexibly, which in turn likely required increased resilience and goal orientation (ILO and OECD, 2020).
Items for some attributes were viewed less positively after the first round. For example, four items each for curiosity, self-compassion, self-efficacy, flow, and optimism were removed after the first round. As with those viewed more positively, some items linked to these five attributes may have resonated less with participants. Alternatively, participants had less understanding of or experience working with these attributes. Nevertheless, items retained after the final round had median scores of ≥4.0, indicating that there was a good level of agreement for items at both the resource and attribute level, which lends support for the structure of the final scale.

| Strengths and limitations
A panel of experts reached consensus on important items for inclusion in the GPRS. The inclusion criteria for participation in this study enabled the research team to gather expert feedback from an academic and practitioner standpoint. Of the 30 participants who began the study, only six did not complete all rounds, enabling reliable results (Akins et al., 2005). The participation and response rates were positive, given that this study was conducted during a global pandemic, which may have limited communication due to forced leave and severances (Hales, 2021). Conversely, working conditions may have provided people with more time and flexibility to participate in the study (de Lucas Ancillo et al., 2021). One limitation of the study is the similar mean rating scores seen across items making it hard to select items for reduction. These results could suggest that all of the items being considered were thought to be important. Alternatively, participants may have all had a very positive disposition, making it more difficult to rate some items as more important than others.
Nonetheless, the good level of agreement gave support to retaining a minimum number of items for each attribute through the process and, therefore, to the final structure of the GPRS.

| Practical implications
The 20-item GPRS resulting from this Delphi study has several practical implications. The scale could be used as a tool to identify opportunities to strengthen psychological resources that enable consistent interest in and perseverance with ideas, projects, and goals. As the context will likely influence the extent to which an individual accrues and applies the psychological resources of grit, several variations of scale instructions have been developed so users can get more accurate results. For example, an individual may be interested in their current psychological resources in general. Alternatively, information may be wanted in an employment or sports context when working toward specific individual or team projects and goals. Thus, the separate scores for interest, purpose, practice, and hope, as well as the total score, may provide insights into the reasons for satisfaction and engagement in different areas of life. Scores could also provide leaders and those responsible for developing individual and team capabilities with feedback on potential training needs. Depending on the nature of work and psychological resources of grit potentially contributing more or less to goal achievement, various interventions could be considered, such as self-directed learning and professional coaching. Finally, the GPRS could be used together with other reliable existing scales, such as scales for grit (Duckworth et al., 2007), passion (Vallerand et al., 2003), and persistence (Howard & Crayne, 2019), to enhance needs analysis further.

| CONCLUSION
Utilizing a national and international panel of experts in grit and related constructs, this study obtained a consensus on items to include in the GPRS. It is expected that this scale will provide a wide variety of professionals across various industries an effective tool to assess the psychological resources of grit at an individual or team level and in different contexts. Additionally, professionals could use the scale at different time points to identify the changing needs of individuals and businesses. Future research could test the reliability and validity of the scale.