Motivational-Cognitive Model of Maintaining Workplace Buoyancy through Career Goal-Progress Discrepancy

In organizations, individuals have careers that indicate their progression along a career path over years of employment, thereby fostering a sense of achievement along this path‌. On the career path, there are certain stages that mark an individual`s goals (Ott & Morgan, 2018), and their achievement brings about positive emotions‌, for instance, career satisfaction (Abele & Spurk, 2009)‌. In the quest to achieve career goals, an individual inevitably assesses their career goal-progress discrepancy, which is the perceived gap between an individual’s career goal (i.e., future self or situation) and the progress being made towards achieving that goal (i.e., current self or situation; Creed & Hood, 2015).

Drawn from the goal-setting literature, Akmal et al. (2024) identified four discrepancy domains: (i) achievement discrepancy (i.e., perceptions of achievements to date in relation to where one should be to achieve the career goal), (ii) effort discrepancy (discrepancy between effort expended so far and effort required to achieve the career goal), (iii) standard discrepancy (perception that the career goal is set too high), and (iv) ability discrepancy (perception that the individual might not have the capacity to achieve the desired goal. Bandura (2001) and Bandura and Locke (2003) have used the terms “toward”, “on track”, and “behind” to define discrepancy by which individuals may perceive goals closer or farther away. Once the current goal is within close range, individuals may establish farther goals in order to boost their motivational system by setting a more difficult goal. Career goal-progress discrepancy affects employees' emotions, such as buoyancy, through cognitive-motivational processes. Thus, the purpose of the present study is to examine a motivational-cognitive model of maintaining workplace buoyancy through career goal-progress discrepancy.

2 Theoretical Background

2.1 Career goal-progress discrepancy and emotion regulation

Large discrepancies between current progress and career goals evoke heightened emotional arousal (Holman et al., 2005). According to Watson and McKee’s (1996) model, when individuals consciously perceive this arousal, they label it as positive or negative affect. Building on Schachter’s (1964) two-factor theory of emotion, individuals then appraise and interpret this affect within a given context.

This appraisal process necessitates emotion regulation, defined as the internal and external processes involved in monitoring, evaluating, and adjusting emotional reactions—particularly in acute situations—to align with personal needs and goals (Gross, 2015). Emotion regulation often operates through cognitive strategies, which can be adaptive (e.g., reappraisal, problem-solving) or maladaptive (e.g., rumination, avoidance) (Bookhout et al., 2018; Garnefski & Kraaij, 2018). These strategies help individuals manage overwhelming emotions in stressful or threatening situations, thereby preventing emotional dysregulation and supporting goal pursuit (Berking et al., 2012).

Perceiving substantial goal-progress discrepancy intensifies feelings of failure (Williams et al., 2000). While traditional self-regulation theories focus primarily on cognitive aspects of performance and often overlook the emotional consequences of failure, the Motivational-Cognitive Life-Span Theory of Emotions (Oreyzi & Barati, 2020) explicitly addresses negative affect in career development and its regulatory impact on motivation.

Empirical evidence consistently links career goal discrepancies to negative emotional outcomes and the need for regulation. For instance, Creed et al. (2024) found that discrepancies between occupational aspirations and expectations are associated with career indecision, reduced confidence, and lower career maturity. Similarly, Ferguson et al. (2010) demonstrated in student samples that large ideal-actual self-discrepancies predict low self-esteem and depressive symptoms. Recent extensions to career contexts confirm that goal inconsistency elevates stress and activates emotion regulation to buffer emotional distress (Creed & Blume, 2013; Creed & Gagliardi, 2015; Hood et al., 2025; Widyowati et al., 2024).Thus, effective emotion regulation is essential to sustain motivation amid discrepancies. This leads to the first hypothesis:

Hypothesis 1: Career goal-progress discrepancy positively correlates with emotion regulation.

2.2 Career goal-progress discrepancy, emotion regulation and goal engagement/disengagement control strategies

A central challenge in goal management is determining when to intensify self-regulation for goal pursuit or when to disengage and redirect efforts toward more attainable alternatives (Haratsis et al., 2015). The life-span theory of control, originally proposed by Heckhausen & Schulz (1999) and subsequently updated by Spurk et al.(2022) provides a framework for this process by distinguishing between primary control (actions aimed at changing the external environment to match one’s needs) and secondary control (internal processes that protect motivation, self-esteem, and resources, such as positive reappraisal or goal adjustment). Shane & Hammack(2023) further emphasized the relevance of this framework in contemporary career development research.

Primary control is motivationally prioritized and involves active environmental engagement. When primary control fails (e.g., due to unattainable goals), secondary control compensates by mitigating losses in self-worth, conserving motivational resources, and facilitating disengagement if necessary (Hall et al., 2010; Heckhausen & Schulz, 1999). This compensatory dynamic is not oppositional; rather, secondary control supports and sustains primary control capacity over time (Oreyzi & Barati, 2014a, 2014b, 2015).

In career contexts, disengaging from unattainable goals reduces distress and preserves psychological health (Creed & Hood, 2014; Wrosch et al., 2002). Emotion regulation plays a pivotal role here, as it influences whether individuals persist with engagement strategies or shift toward disengagement (Haynes et al., 2009; Rozhkova & Murashkino, 2011). Accordingly, the following hypotheses are proposed:

Hypotheses 2 and 3: Emotion regulation correlates positively with goal engagement strategies (H₂) and negatively with disengagement control strategies (H₃).

2.3  Career goal-progress discrepancy,  emotion regulation, control strategies and buoyancy

Control strategies also influence workplace buoyancy, defined as the capacity to maintain positive mood, motivation, and engagement despite everyday career setbacks and challenges (Martin & Marsh, 2008a, 2008b; Oreyzi & Barati, 2020). Unlike general positive affect or wellbeing, buoyancy specifically involves sustained goal pursuit when feasible, or adaptive disengagement to preserve emotional energy when goals are blocked—aligning closely with the Conservation of Resources Theory (Hobfoll & Shirom, 2000).

Originally conceptualized in academic settings (Martin & Marsh, 2008a, 2008b), buoyancy reflects the ability to “bounce back” from minor adversities (e.g., poor performance, pressure) without prolonged disruption (Datu & Yuen, 2018; Victoriano, 2016). It buffers negative emotions like stress (Putwain et al., 2012, 2016), predicts higher motivation and engagement (Martin et al., 2017), and contributes to positive outcomes when combined with skills such as communication (Tarbetsky et al., 2017). In workplace applications, including educational contexts, higher buoyancy is linked to greater engagement and lower burnout (Parker & Martin, 2009; Parker et al., 2012), and mediates relationships between emotion regulation and organizational commitment (Zeng et al., 2024).

Buoyancy is supported by five key motivational predictors: self-confidence (high self-efficacy), coordination (planning and self-regulation), commitment (persistence), composure (low anxiety), and control (low perceived uncertainty) (Martin & Marsh, 2009, 2008a; Toering et al., 2012). These factors enhance buoyancy by reinforcing motivation and engagement (Victoriano, 2016).

Given that goal engagement strategies promote persistence and the conservation of personal resources, whereas goal disengagement strategies protect individuals from prolonged distress and resource depletion, both types of control strategies are conceptually associated with career buoyancy in the face of career-related challenges. Thus, the following hypotheses are advanced:

Hypotheses 4 and 5: Goal engagement strategies are positively related to workplace buoyancy (H₄), whereas disengagement control strategies are negatively related to it (H₅).

2.4  Overview of the Present Study

Humans are motivated by goals and goal engagement. Goals that represent inner desires play a fundamental role in self-regulation and development over the life span (Ferrari et al., 2022). Based on this, perceived career goal-progress discrepancy produces changes in an individual’s emotions, especially when individuals perceive a large gap between their current state and their goals. In such cases, negative emotions may dominate, and the effect will be negative. This is where emotion regulation processes begin and the individual applies cognitive strategies to regulate their emotions. For instance, the individual may use adaptive emotion regulation strategies: refocusing on planning that is thinking about further steps to overcome or change a negative event. Alternatively, individuals may rely on emotion regulation strategies, such as acceptance of the situation and the regulation of negative emotions. Then, to maintain their motivation and engagement toward the goal, the individual may use the goal engagement control strategies‌, which can result in ‌a sense of control and commitment, leading to buoyancy. The aim of the current research was to investigate the motivational-cognitive model of maintaining workplace buoyancy through career goal-progress discrepancy (Figure 1). This study introduces several novel contributions to the literature. First, it is the first longitudinal examination of a motivational-cognitive model that integrates career goal-progress discrepancy, emotion regulation, goal engagement/disengagement strategies, and workplace buoyancy within a single framework. Second, by examining four distinct dimensions of career goal-progress discrepancy (achievement, ability, effort, and standard), this research provides a nuanced understanding of which aspects of perceived goal-progress inconsistency most strongly trigger emotion regulation processes. Notably, only the overall discrepancy score and the standard discrepancy sub-dimension demonstrated acceptable fit and were retained for hypothesis testing (see Results section).Third, the use of cross-lagged panel analysis over two time points allows for stronger inferences about the direction of causality among these variables, addressing a key limitation of prior cross-sectional studies. Fourth, this study uniquely examines the dual role of goal engagement and disengagement control strategies as parallel mediators linking emotion regulation to workplace buoyancy, thereby clarifying the motivational mechanisms through which employees maintain positive mood in the face of career setbacks. Although elements of this model have been theoretically suggested, no empirical study to date has tested the full longitudinal mediation model in an organizational context, to fill this gap. Although the model can be derived from the theoretical background, it has not yet been empirically examined; therefore, this study was conducted to address this gap.

Figure 1

Motivational-Cognitive Model of Maintaining Workplace Buoyancy through Career Goal-Progress Discrepancy

3 Method

3.1 Participants

The data were collected from a sample of 180 employees of a gas company, including 110 white-collar and 70 blue-collar workers from various departments, across two waves between 2020 and 2021. The second assessment was conducted 12 months after the first. Participants were selected on a voluntary basis. The sample size was determined based on a priori power analysis using G*Power software, with data from a pilot study of 50 employees from the same company. Assuming a significance level (α) of 0.05, a desired statistical power (1-β) of 0.80, and a medium effect size (f² = 0.15) for multiple regression analysis, the calculated minimum required sample size was 138 participants. The final sample of 180 participants at Time 1 exceeded this requirement, allowing for potential attrition and ensuring adequate statistical power for model testing. The attrition from T1 to T2 was 14 participants, leaving a total of 166 participants for Time 2. Following standard survey administration procedures (Gall et al., 1996), questionnaires were sent to all participants at their work addresses, and all participants returned their completed responses as instructed. The average work experience was 18.48 years (SD=6.59 years) and the participants’ average age was 44.17 years (SD=11.3 years). All participants provided written informed consent prior to participation. The study procedures were conducted in accordance with the ethical standards of the institutional review board, and participation was entirely voluntary.

3.2  Statistical procedure

Cross-lagged panel analysis was used to examine the prospective relationships between variables. Specifically, we assessed whether X at Time 1 (X₁) significantly predicted Y at Time 2 (Y₂), and whether Y at Time 1 (Y₁) significantly predicted X at Time 2 (X₂). It is important to note that while significant cross-lagged coefficients indicate predictive relationships over time, they do not by themselves establish causal direction, as causal inference from cross-lagged panel models requires stronger assumptions and additional evidence (Rogosa, 1980). The interpretation of cross-lagged effects is based on the following patterns: (a) If rₓ₁ᵧ₂ is significantly different from zero, this suggests that X predicts subsequent Y; (b) If rₓ₂ᵧ₁ is significantly different from zero, this suggests that Y predicts subsequent X; (c) If both coefficients are significant, this suggests bidirectional predictive relationships; (d) If both coefficients are non-significant despite correlated variables, this may indicate that the time lag between waves was either too short or too long to detect prospective effects (Dormann & Griffin, 2015), or that the association is attributable to common causes not included in the model. All statistical analyses were performed using IBM SPSS Statistics 26 and IBM SPSS AMOS 24.

3.3   Instruments

This study used the following instruments for data collection:

1-Workplace Buoyancy Scale: The Workplace Buoyancy Scale was developed by the authors for the purpose of the present study. The scale was initially constructed based on the conceptual definitions of buoyancy (Martin & Marsh, 2008b) and the motivational-cognitive framework outlined in the present research. Since the original scale was developed in Persian, no translation was required. However, to ensure cultural appropriateness for the specific context of gas company employees, the scale was reviewed by a panel of three organizational psychology experts who assessed the clarity and relevance of all items. Minor wording modifications were made based on their feedback to enhance comprehension for both blue-collar and white-collar workers. The scale consists of nine items measuring the extent to which employees maintain their buoyancy when encountering negative workplace events. Participants responded on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree). A sample item is: "If the manager evaluates my performance negatively, this will not ruin my day."In a preliminary study conducted by the authors with a sample of 320 employees from various organizations in Isfahan, the factorial validity of the scale was supported through exploratory and confirmatory factor analysis. All factor loadings were above the acceptable threshold of 0.51, and model fit indices indicated good fit (χ²/df = 2.84, CFI = 0.91, RMSEA = 0.06, GFI = 0.92). The scale demonstrated acceptable internal consistency with a Cronbach's alpha coefficient of 0.79. In the current study, the psychometric properties of the scale were re-examined. Confirmatory factor analysis supported the unidimensional structure of the scale with acceptable fit indices (χ²/df = 2.91, CFI = 0.90, RMSEA = 0.06, GFI = 0.89). All factor loadings were statistically significant and ranged from 0.53 to 0.78. The internal consistency reliability was assessed using Cronbach's alpha, yielding coefficients of 0.69 at Time 1 and 0.72 at Time 2. Although the Time 1 alpha is slightly below the conventional threshold of 0.70, it remains within the acceptable range for research purposes, particularly given the relatively small sample size and the exploratory nature of applying this newly developed scale in a specific occupational context (Hair et al., 2010). The test-retest reliability over the 12-month interval was 0.68, indicating moderate stability.

2-Goal Engagement Control Strategies Questionnaire: The Goal Engagement Control Strategies Questionnaire, developed by Tomasik et al. (2010), was used to measure employees' strategies for maintaining goal pursuit. The questionnaire consists of nine items measuring three dimensions of goal engagement control strategies based on the life-span theory of control (Heckhausen & Schulz, 1995): primary control (e.g., investing time and effort), selective secondary control (e.g., enhancing goal value), and compensatory primary control (e.g., seeking external help). Participants responded on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree). A sample item is: "I am also prepared to make a big effort in order to find a good solution." In the original development study by Tomasik et al. (2010) with a sample of 1,761 German adults, the questionnaire demonstrated good internal consistency (Cronbach's α = 0.87) and factorial validity, with all factor loadings exceeding 0.60. In the Iranian context, Oreyzi and Barati (2014a) examined the psychometric properties of the Persian version using a sample of 280 employees from various organizations in Isfahan. Confirmatory factor analysis supported the three-dimensional structure with acceptable fit indices (χ²/df = 2.76, CFI = 0.92, RMSEA = 0.06, GFI = 0.90). They reported a Cronbach's alpha coefficient of 0.84, test-retest reliability of 0.83 over a four-week interval (n = 50), and concurrent validity coefficient of 0.52 with a measure of positive affect.In the current study, the internal consistency reliability was assessed using Cronbach's alpha, yielding coefficients of 0.82 at Time 1 and 0.84 at Time 2. Test-retest reliability was examined over a 12-month interval using data from 166 participants who completed the questionnaire at both waves, resulting in a coefficient of 0.84. Concurrent validity was assessed by examining the correlation with workplace buoyancy at Time 1, yielding a coefficient of 0.50 (p < .01), indicating acceptable convergent validity. Although confirmatory factor analysis was not re-examined in the present sample due to sample size limitations relative to the number of parameters to be estimated (three factors with nine items), the three-dimensional structure has been well-established in previous research (Tomasik et al., 2010; Oreyzi & Barati, 2014a). Furthermore, the pattern of intercorrelations among the three subscales in the present study (ranging from 0.48 to 0.62) was consistent with theoretical expectations and previous findings, providing indirect support for the dimensional structure.

3-Goal Disengagement Control Strategies Questionnaire: Goal disengagement control strategies were measured using the purpose-made questionnaire developed by Tomasik et al. (2010).The questionnaire consists of six items measuring goal disengagement control strategies, which in the life-span theory of control (Heckhausen & Schulz, 1995) are related to compensatory secondary control . These strategies involve self-protective mechanisms that help individuals disengage from unattainable goals to preserve motivational and emotional resources . Participants responded on a 5-point Likert scale ranging from 1 (totally disagree) to 5 (totally agree). A sample item is: "If I can't find a solution then I put the problem to the back of my mind." In the original development study by Tomasik et al. (2010) with a sample of 1,761 German adults, the questionnaire demonstrated good internal consistency (Cronbach's α = 0.83). In the Iranian context, Oreyzi and Barati (2014a) examined the psychometric properties of the Persian version using a sample of 280 employees from various organizations in Isfahan. They reported a Cronbach's alpha coefficient of 0.79, test–retest reliability of 0.81 over a four-week interval (n = 50), and a concurrent validity coefficient of 0.49 with the Positive and Negative Affect Schedule (PANAS; Watson et al., 1988), indicating that higher use of disengagement strategies was associated with lower positive affect and higher negative affect. In the current study, the internal consistency reliability was assessed using Cronbach's alpha, yielding coefficients of 0.80 at Time 1 and 0.82 at Time 2. Test–retest reliability was examined over the 12-month interval using data from 166 participants who completed the questionnaire at both waves, resulting in a coefficient of 0.82. Concurrent validity was assessed by examining the correlation with workplace buoyancy at Time 1, yielding a coefficient of -0.50 (p < .01), indicating acceptable convergent validity, as higher disengagement was associated with lower workplace buoyancy as theoretically expected.

4- Cognitive Emotion Regulation Questionnaire (CERQ): The CERQ, developed by Garnefski and others (2002), is a 36-item questionnaire consisting of nine conceptually distinct subscales, each with four items and referring to what someone thinks after experiencing a threatening or stressful life event. Cognitive emotion regulation strategies were measured on a 5-point Likert scale ranging from 1 (almost never) to 5 (almost always). Individual subscale scores were obtained by summing the scores belonging to the particular subscale (ranging from 4 to 20). The psychometric properties of the CERQ have been extensively examined across diverse cultural contexts. A recent systematic review of cross-cultural adaptations and validations of the CERQ (Fekar Gharamaleki & Fathipour-Azar, 2024), covering studies published between 2000 and 2024, confirmed that the instrument demonstrates strong test–retest reliability, excellent internal consistency, and reasonable construct validity across multiple languages, affirming the usability of translated versions in different linguistic contexts . The review identified 13 validated versions of the CERQ, with the nine-factor structure consistently supported through confirmatory factor analyses .The Persian translation of the CERQ was validated by Abdi et al. (2012) using a sample of Iranian university students . Confirmatory factor analysis supported the nine-factor structure with acceptable fit indices, and Cronbach's alpha coefficients for the subscales ranged from 0.64 to 0.82 . More recently, Hasani et al. (2024) examined the psychometric properties of the Persian version of the CERQ-short and confirmed its reliability, validity, factor structure, and measurement invariance across gender and age groups . The Persian version has been subsequently used in various organizational and clinical studies in Iran, demonstrating adequate psychometric properties.In the current study, given that only four subscales (self-blame, other-blame, rumination, and catastrophizing) were used, we re-examined the psychometric properties of these subscales in our sample. Confirmatory factor analysis was conducted to test the four-factor structure of the negative cognitive emotion regulation strategies. The results supported the four-factor model with acceptable fit indices: χ²/df = 2.68, CFI = 0.91, RMSEA = 0.05, GFI = 0.90. All factor loadings were statistically significant and ranged from 0.52 to 0.81. Internal consistency reliability was assessed using Cronbach's alpha, yielding the following coefficients at Time 1 and Time 2 respectively: self-blame (α = 0.76, 0.78), other-blame (α = 0.74, 0.75), rumination (α = 0.79, 0.81), and catastrophizing (α = 0.77, 0.79). Test–retest reliability over the 12-month interval was examined using data from 166 participants who completed the questionnaire at both waves, yielding the following coefficients: self-blame (0.70), other-blame (0.68), rumination (0.73), and catastrophizing (0.71), indicating acceptable stability. These findings demonstrate that the four negative cognitive emotion regulation strategies subscales possess adequate psychometric properties in the current sample, supporting their use in subsequent analyses.

5-Career Related Discrepancies Scale: The Career Related Discrepancies Scale, developed by Creed and Hood (2015), was used to measure employees' perceived gap between their career goals and their current progress toward those goals. The English version was translated into Persian using forward-backward translation (Brislin, 1980). Content validity was established through review by three organizational psychology experts, and face validity was confirmed via cognitive interviews with five employees. The 12-item unidimensional scale covers four underlying domains: achievement, ability, effort, and standard discrepancy (three items each). Responses are on a 6-point Likert scale (1 = strongly disagree to 6 = strongly agree). Sample items: "My plans are not working out to get the career I really want" and "I doubt I can meet the standards of entry to my ideal career." Creed and Hood (2015) confirmed the unidimensional structure with four facets (N = 615), yielding good fit: χ²/df = 2.84, CFI = 0.94, RMSEA = 0.06, GFI = 0.92. Internal consistency was excellent (α = 0.95–0.96), and construct validity was established through expected relationships with career goal commitment and distress. CFA supported the unidimensional structure with four domains: χ²/df = 2.74, CFI = 0.91, RMSEA = 0.04, GFI = 0.91. Factor loadings ranged from 0.53 to 0.79. Internal consistency for the total scale was α = 0.92 (Time 1) and 0.91 (Time 2). Subscale reliabilities (Time 1/Time 2): achievement (0.81/0.77), ability (0.79/0.81), effort (0.73/0.75), standard (0.77/0.74). The effort discrepancy alpha (0.73) is acceptable for research purposes given the small number of items (Hair et al., 2010). Test–retest reliability over 12 months (n = 166) was 0.84. Intercorrelations among subscales (r = 0.37–0.68) aligned with theoretical expectations.

4  RESULTS

In Table 1, the measures of central tendency and dispersion for research instruments are presented.

Table 1.
 Central Tendency and Dispersion for Research Instruments

Career goal-progress discrepancy has the maximum mean and standard deviation; effort discrepancy shows the minimum mean and achievement discrepancy indicates the minimum standard deviation.

Table 2.

Research variables internal correlations (T₁)

*=p <0.05 ,**=p <0.01

The internal correlations between research variables are presented in Table 2. Correlation coefficients between career goal-progress discrepancy and its four areas, namely, achievement discrepancy, ability discrepancy, effort discrepancy, and standard discrepancy, with workplace buoyancy were negative. That is, as career goal-progress discrepancy increases, workplace buoyancy decreases, and vice versa. Additionally, the relationship between workplace buoyancy and cognitive emotion regulation strategies, as well as goal engagement control strategy was positive, whereas its relationship with the goal disengagement control strategy was negative. In other words, participants who applied cognitive emotion regulation strategies and goal engagement control strategies experienced increased buoyancy, whereas those who applied goal disengagement strategies experienced decreased buoyancy. To define the causal direction, cross-lagged correlation (CLC) analysis (Figure 2) was used. In Figure 2, if r_y2x1 < r_x1y2, then the directions of the relationships depicted in Figure 2 are correct, otherwise, the direction of relationships needs to be reversed.  X and Y are the variables under investigation, and their subscripts are related to time T₁ and T₂.

Figure 2.

 Cross-lagged Correlation (CLC) Pattern

A comparison of the correlation coefficients is made by one of these four tests: Hendrickson et al. (1970), Hotelling (1940), Meng et al. (1992), Williams (1959). Statistical significance indicates the difference in correlation coefficients. If the test result is not significant, then there is a two-way relationship direction.

Table 3. 

Cross-Lagged Correlations Among Study Variables

Note. This table presents cross-lagged correlation coefficients between cognitive emotion regulation strategies, workplace buoyancy, goal engagement, and goal disengagement control strategies measured at Time 1 and Time 2. The tests of significance (Hotelling, Williams, Hendrickson–Stanley–Hills, and Meng–Rosenthal–Rubin) compare the relative strength of cross-lagged paths to infer temporal precedence. All coefficients are Pearson correlations; ∗ p < 0.05, ∗∗ p < 0.01.

Table 3 shows the results of the cross-lagged correlation (CLC) test for a comparison of the relationships between cognitive emotion regulation strategies, workplace buoyancy, goal engagement control strategy and goal disengagement control strategy in Time 1 and Time 2. The cross-lagged correlation (CLC) coefficient between cognitive emotion regulation strategies in Time 1 and Time 2 and goal engagement control strategy at Time 1 and Time 2 is significant and negative. In addition, the cross-lagged correlation (CLC) coefficient between cognitive emotion regulation strategies at Time 1 and Time 2 and goal disengagement control strategy at Time 1 and Time 2 is significant and positive. Accordingly, the cross-lagged patterns provide evidence for the hypothesized temporal precedence and predictive direction in these relationships. This supports the model's proposed sequence, suggesting that cognitive emotion regulation strategies predict subsequent goal engagement and disengagement control strategies, rather than the reverse. Moreover, the cross-lagged correlation (CLC) coefficient between workplace buoyancy at Time 1 and Time 2 and goal engagement control strategy at Time 1 and Time 2 is significant and positive. In addition, the cross-lagged correlation (CLC) coefficient between workplace buoyancy at Time 1 and Time 2 and goal disengagement control strategy at Time 1 and Time 2 is significant and negative. Similarly, the cross-lagged coefficients support the predictive ordering proposed in Figure 1, indicating that goal engagement control strategies at Time 1 predict subsequent goal disengagement control strategies at Time 2. However, as noted in the Method section, these findings reflect predictive associations rather than definitive causal inference.

Table 4. 
Fit Indices for Structural Equation Models

Note. Fit indices are reported for five separate SEM models, each with a different independent variable (total career goal‑progress discrepancy, achievement discrepancy, ability discrepancy, effort discrepancy, or standard discrepancy) predicting negative cognitive emotion regulation strategies. Abbreviations: GFI = Goodness‑of‑Fit Index, CFI = Comparative Fit Index, PNFI = Parsimonious Normed Fit Index, RMSEA = Root Mean Square Error of Approximation, RMR = Root Mean Square Residual. Fit evaluation: Excellent (RMSEA ≤ 0.05, CFI ≥ 0.90, GFI ≥ 0.90), Acceptable (RMSEA ≤ 0.06, CFI ≥ 0.85, GFI ≥ 0.85), Poor (rejected) (RMSEA > 0.06 or CFI/GFI < 0.85).

Table 4 illustrates the results of structural equation modeling (SEM; Figure 1) for five independent variables—career goal-progress discrepancy (total), achievement discrepancy, ability discrepancy, effort discrepancy, and standard discrepancy—each influencing cognitive negative emotion regulation strategies. The fit indices for the five models are presented in Table 4. Among these, the model with standard discrepancy as the independent variable demonstrated the best fit to the data (χ²/df = 2.56, GFI = 0.92, CFI = 0.91, RMSEA = 0.04), indicating that employees' perception of their career goals being set too high is the strongest predictor of subsequent emotion regulation processes. The models for achievement discrepancy (χ²/df = 2.93, GFI = 0.87, CFI = 0.85, RMSEA = 0.06) and ability discrepancy (χ²/df = 2.84, GFI = 0.89, CFI = 0.87, RMSEA = 0.05) showed acceptable fit based on RMSEA values (≤ 0.06), although their GFI and CFI values fell slightly below the conventional threshold of 0.90. This suggests that these subdimensions may not independently capture the full complexity of the emotion regulation process. Specifically, achievement discrepancy—reflecting perceived past failures—may require additional contextual variables (e.g., organizational support or feedback) to more accurately predict emotional and motivational outcomes. In contrast, the effort discrepancy model demonstrated poor and unacceptable fit indices (χ²/df = 3.28, GFI = 0.85, CFI = 0.83, RMSEA = 0.07), falling below the conventional cut-off criteria (CFI ≥ 0.90, RMSEA ≤ 0.06; Hu & Bentler, 1999). Based on these results, the effort discrepancy sub-dimension was excluded from the final structural model (Table 5), and only the total career goal-progress discrepancy score and the standard discrepancy sub-dimension were retained for hypothesis testing. A possible explanation for this poor fit is that perceptions of insufficient effort may be highly dynamic and context-dependent, influenced by fluctuating factors such as daily fatigue, workload variations, or momentary self-efficacy beliefs, making them less suitable for stable longitudinal prediction over a one-year interval. Overall, these results suggest that different dimensions of career goal-progress discrepancy vary in their capacity to predict cognitive-emotional pathways to workplace buoyancy. Future research should consider testing these models with additional moderators or shorter time intervals (e.g., daily diary designs) to better capture within-person variability. Consequently, only the total career goal-progress discrepancy score and the standard discrepancy sub-dimension were used in the final structural model (Table 5) to test the study hypotheses.

Based on the superior fit indices observed for the standard discrepancy model (see Table 4), the final SEM analysis was conducted with this subdimension as the independent variable (Table 5). To evaluate the predictive power of the final model, the squared multiple correlation (R²) for the endogenous variable (workplace buoyancy at T2) was calculated based on the structural equation modeling analysis. This value is reported in Table 5.

Table 5.
 Path Coefficients for the Final SEM Model

Note:  The final model uses standard discrepancy as the independent variable. T1 = Time 1 (baseline), T2 = Time 2 (12 month follow up). Unstd. B = unstandardized coefficient, SE = standard error, Std. β = standardized coefficient. All paths are significant at p < 0.01. All coefficients represent predictive relationships over the 12‑month interval. Since the mediating variables (CER, GE, GD) were measured at T2, concurrently with the outcome (WB), the full mediation model was not strictly tested longitudinally. Consequently, the results are interpreted as supporting the hypothesized directional sequence based on the temporal precedence of each adjacent pair of paths, rather than as confirming a causal mediation chain. R² value indicates the proportion of variance explained in Time 2 workplace buoyancy by the full set of predictors in the model.

Table 5 presents the unstandardized and standardized path coefficients for the final structural model, using standard discrepancy as the independent variable. All hypothesized paths were statistically significant. The strongest direct effect was observed from goal engagement control strategies to workplace buoyancy (β = 0.46, p < 0.001), followed by the effect of standard discrepancy on negative emotion regulation strategies (β = 0.41, p < 0.001). These findings confirm that perceiving one's career goals as excessively high triggers negative cognitive emotion regulation, which in turn influences goal management strategies and ultimately affects employees' workplace buoyancy.

5 Discussion

The aim of the current research was to investigate the motivational-cognitive model of maintaining workplace buoyancy through career goal-progress discrepancy. For this purpose, a model was developed based on emotion regulation and goal engagement/disengagement control strategies that promote workplace buoyancy. In this model, career goal-progress discrepancy first affects emotion regulation strategies; these strategies then lead to goal engagement/disengagement control strategies, which ultimately result in buoyancy. By selecting a sample of employees, the variables were measured at two time points over one year. The results of structural equation modeling confirmed the hypothesized model, demonstrating good fit indices: χ²/df = 2.74, GFI = 0.91, CFI = 0.90, RMSEA = 0.04, and RMR = 0.05. Among the four discrepancy domains, standard discrepancy demonstrated the best fit (χ²/df = 2.56, GFI = 0.92, CFI = 0.91, RMSEA = 0.04). However, the effort discrepancy model exhibited unacceptable fit indices (χ²/df = 3.28, GFI = 0.85, CFI = 0.83, RMSEA = 0.07), falling below conventional cut-off criteria (CFI ≥ 0.90, RMSEA ≤ 0.06; Hu & Bentler, 1999). Consequently, effort discrepancy was excluded from the final structural model (Table 5) and only the total career goal-progress discrepancy and the standard discrepancy sub-dimension were retained for hypothesis testing. Given that the effort discrepancy model did not meet acceptable fit criteria, the study hypotheses were tested using only the total career goal-progress discrepancy score and the standard discrepancy sub-dimension, both of which demonstrated good to excellent fit (Table 4).

The finding that career goal-progress discrepancy (CGPD) positively predicts negative cognitive emotion regulation strategies (β = 0.29, p < .05) aligns with the Motivational-Cognitive Life-Span Theory of Emotions (Oreyzi & Barati, 2020), which posits that perceived failure in career progress automatically triggers regulatory processes. According to Bandura's Social Cognitive Theory (2001), when individuals perceive a significant gap between their current state and valued goals, this discrepancy generates emotional arousal that necessitates cognitive management. In line with Gross's (2015) extended process model, situations appraised as goal-relevant threats—such as perceiving oneself as "off-track" in career advancement—activate cognitive emotion regulation strategies. However, when the discrepancy is perceived as large or unmanageable, individuals tend to adopt maladaptive strategies (e.g., rumination, self-blame, catastrophizing) rather than adaptive ones. This finding is consistent with previous research (Creed & Hood, 2015; Widyowati et al., 2024), demonstrating that career goal inconsistency elevates distress and activates emotion regulation to buffer emotional strain. From a practical perspective, these findings underscore the importance of collaborative goal-setting and regular progress feedback within organizations to prevent the accumulation of perceived discrepancies that may trigger maladaptive cognitive patterns.

The path analysis revealed significant direct and indirect effects. Career goal-progress discrepancy significantly predicted cognitive emotion regulation strategies (β = 0.29, p < 0.05), supporting Hypothesis 1. Cognitive emotion regulation strategies, in turn, significantly predicted goal engagement control strategies (β = -0.31, p < 0.01) and goal disengagement control strategies (β = 0.32, p < 0.01), confirming Hypotheses 2 and 3. Furthermore, goal engagement control strategies positively predicted workplace buoyancy (β = 0.42, p < 0.01), while goal disengagement control strategies negatively predicted workplace buoyancy (β = -0.35, p < 0.01), supporting Hypotheses 4 and 5. The cross-lagged panel analysis provided further support for the hypothesized temporal ordering of these relationships, indicating that emotion regulation at Time 1 predicted the subsequent utilization of control strategies at Time 2, which in turn predicted later workplace buoyancy. While these findings are consistent with a causal interpretation, they should be interpreted with caution given the correlational nature of the data. The final structural model (Table 5), which included standard discrepancy at T1, negative cognitive emotion regulation strategies at T2, and goal engagement/disengagement control strategies at T2, explained 38% of the variance in Time 2 workplace buoyancy (R² = 0.38), indicating a moderate-to-large effect size (Cohen, 1988).

Results indicated a positive relationship between workplace buoyancy and emotion regulation cognitive strategies; that is, when individuals apply emotion regulation cognitive strategies, their buoyancy increases. Conversely, increased reliance on maladaptive emotion regulation strategies (e.g., rumination and catastrophizing) is associated with a greater tendency toward goal withdrawal. This finding aligns with the life-span theory of control (Heckhausen & Schulz, 1995); when individuals fail to cope with negative emotions through adaptive strategies, they resort to secondary control mechanisms to protect themselves against failure. In other words, ineffective emotion regulation can lead to goal abandonment (Wrosch et al., 2002). This is also consistent with Martin et al. (2010), who stated that a core component of buoyancy is coordination (planning), and that self-regulation supports effective planning. Therefore, as self-regulation increases, buoyancy is enhanced. Applying cognitive emotion regulation strategies represents an effort toward self-regulation, which in turn fosters workplace buoyancy. Specifically, when an employee applies these strategies, they enhance their coordination (high planning), perceive themselves as prepared for specific tasks, and consequently experience increased mood and buoyancy.

Furthermore, the relationship between negative cognitive emotion regulation strategies and goal disengagement (r = 0.32, p < 0.01) suggests that the more individuals engage in negative cognitive strategies, the more likely they are to withdraw from pursuing their goals. The theoretical explanation for this finding pertains to the compensatory role of secondary control strategies; when a goal is perceived as unattainable, individuals unconsciously distance themselves from the goal to preserve self-esteem and motivational resources (Hall et al., 2010). In these cases, negative cognitive strategies not only fail to regulate emotions effectively but also amplify tendencies toward goal withdrawal.

Additionally, the results showed a positive relationship between workplace buoyancy and goal engagement control strategies. When positive goal engagement control strategies are applied, the employee's buoyancy increases. The path coefficient of 0.42 (p < 0.01) indicates a strong positive relationship. Goal engagement enhances workplace buoyancy by strengthening perceived control and motivational persistence. Employees who remain committed to their goals, even when facing obstacles, employ primary control strategies (such as increased effort) and selective secondary control strategies (such as enhancing goal value), both of which contribute to maintaining positive mood (Martin et al., 2010). This finding confirms that goal commitment not only facilitates progress but also functions as an affective resource against daily work-related stressors.  This finding is consistent with Creed and Hood's (2014) finding that where goal engagement exists, control mechanisms are triggered and the individual seeks resources for assistance (Compensatory Primary Control), makes efforts (Selective Primary Control), and is highly motivated to achieve the goal. Although achieving a goal is difficult, the individual remains prepared placing greater subjective value on their goals (Heckhausen & Schulz, 1995). Considering the above, the individual possesses a feeling of control which leads to buoyancy (Martin et al., 2010).

The results also showed a negative relationship between workplace buoyancy and goal disengagement control strategies. The path coefficient of -0.35 (p < 0.01) confirms this inverse relationship. When individuals apply goal disengagement control strategies, their workplace buoyancy decreases. Although goal disengagement may temporarily shield individuals from the distress of failure, it ultimately undermines workplace buoyancy in the long term. When individuals withdraw from a goal, they not only lose a sense of progress but also experience diminished intrinsic motivation to engage with subsequent challenges (Creed & Hood, 2014). This finding is consistent with Conservation of Resources Theory (Hobfoll, 2000), as goal disengagement represents the loss of a critical motivational resource, resulting in reduced psychological energy for coping with adversities. When there is no goal engagement (goal disengagement), the secondary compensatory control strategy is activated, which includes self-directed support strategies. For instance, the individual reduces the perception of control over a situation and attributes failure and deprivation to external forces to reduce the potential negative effects of failure on major motivational resources, such as emotional balance and self-esteem. Goal disengagement can also lead to downward social comparisons, which facilitate self-enhancement and provide short-term protection of self-esteem under situational stress (de Souza et al., 2010). Nonetheless, the long-term application of goal disengagement has negative consequences. This strategy affects commitment—a key resource of buoyancy—resulting in decreased engagement. Commitment is defined as resilience and the tendency to maintain engagement in a specific task (Martin & Marsh, 2006). When an individual applies goal disengagement control strategies, they reduce their engagement toward the goal; therefore, commitment is reduced, which in turn reduces buoyancy.

Furthermore, the present results demonstrated a negative relationship between workplace buoyancy and career goal-progress discrepancy. The correlation coefficient of -0.27 (p < 0.05) indicates that as career goal-progress discrepancy increases, workplace buoyancy significantly decreases. In other words, when employees perceive a large gap between their current status and their desired career goals, their capacity to maintain a positive mood in the face of workplace adversities diminishes. According to Bandura's social-cognitive theory (2001), goals perceived as excessively distant undermine self-efficacy and deplete motivational resources. Furthermore, consistent with emotional control theory, this perception activates negative cognitive appraisals (such as self-blame or catastrophizing), which directly diminish workplace buoyancy (Creed & Hood, 2015).  An individual continuously evaluates goals along their career path, and when their goals are perceived as far-fetched and unattainable, they begin managing their engagement to achieve their goals. According to goal-oriented theories, such as social cognitive theory (Bandura, 2001), control theory (Carver & Scheier, 1998), and life-span control theory (Heckhausen & Schulz, 1995), this management is provided by acting on the individual's internal needs and inclination. As noted earlier, based on the life-span theory of control, employees may vary their needs and inclinations by decreasing engagement in a specific goal. Such a reduction in goal engagement leads to lower commitment—a key source of workplace buoyancy (Martin & Marsh, 2006).

Furthermore, Cross-Lagged Correlations (CLC) provided evidence for the hypothesized predictive relationships among the variables. The CLC analysis showed that r (cognitive emotion regulation T1 with goal engagement T2) = -0.31 (p < 0.01) was significantly different from r (cognitive emotion regulation T2 with goal engagement T1) = -0.15 (p < 0.05), supporting the temporal precedence of emotion regulation in predicting subsequent goal engagement strategies. Similarly, the predictive path from goal engagement to workplace buoyancy was supported, with a stronger cross-lagged correlation from T1 goal engagement to T2 buoyancy (r = 0.52, p < 0.01) compared to the reverse path. The cross-lagged results suggest a directional flow primarily from emotion regulation to goal control strategies, as the predictive coefficient from Time 1 emotion regulation to Time 2 control strategies was significantly stronger than the reverse path. This pattern is consistent with the theoretical model, but we acknowledge that alternative explanations (e.g., unmeasured third variables) cannot be ruled out. This finding supports the notion that individuals’ capacity to effectively manage negative emotions is a prerequisite for selecting adaptive goal management strategie (Oreyzi & Barati, 2020). In other words, employees who can regulate their emotions effectively are better equipped to make informed decisions about whether to persist in or withdraw from a goal, which ultimately ensures their workplace buoyancy.  While these findings do not establish definitive causality, they are consistent with the proposed directional sequence in the motivational-cognitive model: Emotion regulation predicts the subsequent utilization of goal engagement/disengagement control strategies, which in turn predict later workplace buoyancy. These results are also consistent with recent findings in academic contexts, showing that emotion regulation strategies facilitate academic buoyancy by reducing anxiety and enhancing coping in challenging environments (Putwain et al., 2022), supporting the generalizability of our model beyond workplace settings.

One of the most intriguing findings of this study was the significant positive correlation between negative cognitive emotion regulation strategies and workplace buoyancy at Time 1 (r = 0.37, p < 0.01; Table 2). At first glance, this positive bivariate association appears inconsistent with the theoretical model, which predicts that maladaptive strategies such as rumination, self‑blame, and catastrophizing should undermine employees' capacity to maintain positive mood and motivation. However, this paradox may be explained by distinguishing between the total (bivariate) effect and the indirect (mediated) effects. Drawing on activation theories of emotion (Carver & Scheier, 1998), negative cognitive strategies can paradoxically increase cognitive arousal and alertness, which in the short term may mobilize energy and prompt corrective actions to reduce goal discrepancies. Thus, the positive correlation may reflect a temporary 'functional tension' or heightened emotional awareness, particularly among employees who are deeply invested in their careers and simultaneously experience both high negative affect and high engagement.

Crucially, despite this positive bivariate correlation, the longitudinal structural model (Table 5) provides clearer evidence that negative cognitive emotion regulation strategies at Time 1 predict reduced goal engagement (β = -0.31) and increased goal disengagement (β = 0.32) at Time 2, ultimately contributing to a decline in workplace buoyancy over time (via disengagement: β = -0.35). Thus, while the cross‑sectional correlation appears positive, the longitudinal pathways suggest that the cumulative effect of relying predominantly on maladaptive emotion regulation is detrimental to motivational persistence and affective well‑being in the long run. This discrepancy between bivariate and mediated effects highlights the critical importance of employing longitudinal designs and mediation analyses to uncover the nuanced processes underlying employee adaptation to career challenges. Future research using within‑person designs (e.g., daily diary or ecological momentary assessment) is needed to disentangle the reciprocal dynamics between emotion regulation and buoyancy over shorter time intervals.

There are some limitations in this study that need to be noted. First, the current study focused on a specific cultural context in Iran. Previous research (e.g., Azimi et al., 2018) has shown that in some cultural contexts such as religious cultural contexts compared to non-religious contexts, people are more likely to use negative emotion regulation strategies such as self-blame. Therefore, the findings may not be generalizable to other cultural contexts, and future research should examine the model among employees in diverse cultural settings. Second, to validate the scale in other contexts, this study suggests researchers conduct Confirmatory Factor Analysis on the workplace buoyancy scale, which was addressed in the present research for the first time. The confirmatory factor analysis in this study supported the factorial validity of the workplace buoyancy scale with all factor loadings above 0.51 and acceptable reliability (α = 0.69 at T1, 0.72 at T2). Third, the current study is limited in assuming stability when examining causal directions from control strategies to workplace buoyancy and from control strategies to negative cognitive emotion regulation. Individuals may experience different growth trajectories in control strategies over time, which poses a challenge to internal validity in field research. Fourth, the relatively weaker fit indices for the effort discrepancy model (χ²/df = 3.28, GFI = 0.85, CFI = 0.83, RMSEA = 0.07) suggest that the perception of insufficient effort may be more dynamic and influenced by fluctuating factors such as fatigue, workload, or self-efficacy beliefs. Future research should test these models using additional moderators or longitudinal designs that capture within-person variability. Fifth, the study did not account for emerging factors such as the impact of artificial intelligence on career goals and discrepancies, which recent reports indicate may exacerbate goal-progress inconsistencies for younger workers (Deloitte, 2025). Future research should incorporate these technological influences. Given the relatively small sample size in this study, the findings should be interpreted with caution, and future research is encouraged to replicate these results with larger and more diverse samples. As the sample consisted of volunteers, the findings may be influenced by self-selection effects, and caution should be exercised when generalizing the results to the broader employee population. Importantly, demographic variables such as age, work experience, education level, and marital status were not controlled for in this study; thus, the observed relationships may be partially influenced by these unmeasured factors. Future studies are recommended to control for such demographic variables in their analyses to better isolate the effects of career goal–progress discrepancy, emotion regulation, and control strategies on workplace buoyancy. Sixth, a significant limitation of this study is that the proposed mediation model was not strictly longitudinal. While the cross-lagged panel analysis established temporal precedence for adjacent paths (e.g., T1 emotion regulation → T2 control strategies; T1 control strategies → T2 buoyancy), the mediators and outcomes were measured simultaneously at Time 2. This precludes strong inferences about the indirect (mediating) effects of control strategies in the pathway from emotion regulation to buoyancy. Future studies should employ three-wave longitudinal designs with appropriate time lags between measurement occasions to rigorously test the full mediation model.

Based on the results of the current research, it is recommended that organizations and managers take proactive steps to reduce employees' career goal-progress discrepancy. This can be achieved by designing realistic, attainable career goals in collaboration with employees, ensuring that goals are challenging yet achievable given the individual's abilities, resources, and effort. Managers should regularly review and adjust goals based on employees' progress and changing circumstances, thereby preventing the accumulation of perceived failure and the subsequent decline in workplace buoyancy.

Furthermore, employees should be trained in adaptive goal management strategies, such as goal revision and goal reprioritization, as alternatives to premature goal disengagement. Rather than withdrawing effort and becoming disengaged when facing obstacles, employees can learn to reframe or modify their goals to align more closely with their current progress and capacities. This approach helps maintain motivation and preserve emotional energy, thereby sustaining workplace buoyancy. In addition, training in cognitive emotion regulation strategies—such as positive refocusing, refocusing on planning, and positive reappraisal (Diefendorff et al., 2008)—can equip employees with the tools to manage negative emotions arising from goal discrepancies, further supporting their ability to remain buoyant.

Given that the model explained 38% of the variance in workplace buoyancy, organizations should prioritize interventions that target both goal-setting practices and employees’ goal regulation skills. Specifically, the strong relationship observed between standard discrepancy (the perception that career goals are set too high) and the model variables suggests that managers should ensure performance standards. Besides, career expectations are perceived as realistic and attainable. Regular feedback and open communication about goal feasibility can mitigate the negative effects of unrealistic standards on employees’ emotional well-being and motivation.

6 Conclusions

In summary, the current study examined a motivational–cognitive model of maintaining workplace buoyancy through career goal–progress discrepancy (CGPD). The model demonstrated that a large career goal–progress discrepancy triggers cognitive emotion regulation strategies, which in turn influence goal engagement and disengagement strategies to maintain buoyancy. The findings provide the first longitudinal evidence that standard discrepancy (the perception that goals are set too high) is the strongest predictor of this motivational–cognitive pathway, and that goal engagement and disengagement strategies appear to function as sequential predictors in the pathway from emotion regulation to workplace buoyancy. While the cross-lagged patterns support the hypothesized temporal ordering of these variables, the mediating role of control strategies could not be fully established longitudinally due to the simultaneous measurement of mediators and outcomes at Time 2. Future research employing three-wave longitudinal designs (e.g., T1: discrepancy, T2: emotion regulation, T3: control strategies and buoyancy) is needed to rigorously test the proposed mediation model. Together with previous research, these results highlight the critical role of career goals in shaping employees’ work-related emotions, motivation, and adaptive functioning.