Original Research

Exploring the prevalence of workplace flourishing amongst teachers over time

Kleinjan Redelinghuys, Sebastiaan Rothmann

Received: 20 Nov. 2019; Accepted: 02 Mar. 2020; Published: 08 June 2020

Copyright: © 2020. The Author(s). Licensee: AOSIS.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Orientation: Research indicates that teachers are more vulnerable to work-related stress, psychological distress and burnout than many other occupational groups. Despite these hindrances, and against all odds, some teachers are able to feel and function well at work. As positive teacher functioning is an achievable objective, it is important that more studies focus on the positive aspects associated with teacher functioning.

Research purpose: The aim of this was to determine whether workplace flourishing is non-static in nature and that employees’ functioning levels may fluctuate positively or negatively over time.

Motivation for the study: Studies predominantly looked at workplace flourishing from a cross-sectional viewpoint. This is problematic, as it provides little information on how employee well-being develops over time.

Research approach/design and method: A longitudinal survey design was used with 202 secondary school teachers. The Flourishing-at-Work Scale and Turnover Intention Scale were administered. A series of analyses (confirmatory factor analysis, longitudinal measurement invariance, cross-tabulations) were performed to achieve the study objectives.

Main findings: The results showed that teachers experienced notable changes in their classification categories (non-flourishing vs. flourishing) over time and that these changes were related to their intention to leave.

Practical/managerial implications: Because of the impact workplace flourishing may have on individual performance and retention, it is important for organisations to continuously monitor their employees’ levels of functioning.

Contribution/value-add: This study results should offer new insights into how employee well-being develops over time, the complexity of individual uniqueness and evidence for individualising well-being interventions.

Keywords: flourishing; longitudinal; prevalence; teachers; well-being; workplace flourishing.

Introduction

Incidences of burnout, depression and stress are frequently reported amongst teacher samples (Capone, Joshanloo, & Park, 2019; Van Droogenbroeck, Spruyt, & Vanroelen, 2014; Yu, Wang, Zhai, Dai, & Yang, 2015). Kinman, Wray and Strange (2011) stated that ‘A body of research indicates that teachers are more vulnerable to work-related stress, psychological distress and burnout than many other occupational groups’ (p. 843). The findings of Kinman et al. (2011) painted a bleak picture for the teaching profession. This is, however, not surprising when one considers the ever-increasing challenges teachers face daily. These challenges include unfavourable teacher–learner ratios, student–teacher bullying, fellow teacher absenteeism, constantly increasing administrative burdens, diminished autonomy and under-resourced classroom facilities (Ballet, Kelchtermans, & Loughran, 2006; Pyhältö, Pietarinen, & Salmela-Aro, 2011; Skaalvik & Skaalvik, 2010). Despite these challenges, and against all odds, teachers do show resilience and are able to feel and function well at work. For example, studies suggest that some teachers flourish (Redelinghuys, 2016), are engaged (Hakanen, Bakker, & Schaufeli, 2006) and experience a sense of enthusiasm and satisfaction with their jobs (Roth, Assor, Kanat-Maymon, & Kaplan, 2007).

As positive teacher functioning is an achievable objective, it is important that more studies focus on the constructive and/or progressive aspects associated with teacher functioning. Therefore, a stronger emphasis on constructs, such as workplace flourishing, should be invested in this vocation. Workplace flourishing is a multidimensional construct that focuses on employees’ emotional, psychological and social well-being (SWB) (Rothmann, 2013), providing cues on how employees feel and function in the workplace (Rautenbach, 2015). The emotional well-being (EWB) component includes positive affect and job satisfaction, reflecting the degree to which employees experience frequent positive emotions at work and whether they derive a sense of enjoyment from their job (Rautenbach, 2015). The psychological well-being (PWB) component includes autonomy, competence, relatedness, engagement, meaningful work and learning (Rothmann, Van Zyl, & Rautenbach, 2019). It reflects the extent to which employees feel confident to express themselves in their work, are competent in managing their work responsibilities, find themselves connected to colleagues, experience their work meaningful and purposeful, are engaged at work and find themselves learning often (Rautenbach, 2015).

Lastly, the SWB component includes acceptance, actualisation, coherence, contribution and integration (Redelinghuys, 2016). It reflects the degree to which employees feel part of their organisation, whether their organisation is developing, if the way their organisation is functioning makes sense to them, feel they have something important to contribute and whether they experience a sense of belonging at work (Rautenbach, 2015). Studying and offering workplace flourishing tools for teachers is vital, as it strengthens the prospective for healthy in-role performance, organisational citizenship behaviour and retention, (Janse van Rensburg, Rothmann, & Diedericks, 2017; Redelinghuys, Rothmann, & Botha, 2019b). Accordingly, when employees feel and function well, they are more inclined to perform better and remain with their organisation. Although most studies indicate that only a few individuals flourish at work, we support hopefulness.

Prevalence of workplace flourishing

In South Africa, the Mental Health Continuum–Short Form (MHC-SF; Keyes, 2002) is often utilised to assess the prevalence of workplace flourishing. For example, Diedericks and Rothmann (2014) found that, amongst 205 information technology specialists, 37.6% flourished, 58.5% were moderately mentally healthy and 3.9% languished. In a study conducted amongst the adult population in the North West Province, Khumalo, Temane and Wissing (2012) found that 20% of the participants flourished, 67.8% were moderately mentally healthy and 12.2% languished. In a managerial context, Swart and Rothmann (2012) established that 48.5% of the participants flourished, 48.5% were moderately mentally healthy and 3% languished. Boshoff, Potgieter, Van Rensburg and Ellis (2014) established that, amongst 200 black secondary/high school teachers in the North West Province, 28% were flourishing, 70.5% were moderately mentally healthy and 1.5% were languishing.

Apart from the MHC-SF, another instrument that assesses workplace flourishing is the SF of the flourishing at work scale (FAWS-SF). Applying the criteria of flourishing at work, Rautenbach (2015) established that 35.9% of employees in the fast-moving consumer goods industry flourished, 65.15% were moderately mentally healthy and 8% languished. In addition, within the tertiary education sector, Janse van Rensburg (2016) discovered that 43.1% of employees flourished, 44.5% were moderately mentally healthy and 12.4% languished.

Although there have been numerous studies that looked at the prevalence of workplace flourishing amongst South African teacher (e.g. Boshoff et al., 2014) and non-teacher (Khumalo et al., 2012) samples, these studies predominantly looked at workplace flourishing from a cross-sectional viewpoint. This is problematic, as it provides little information on how employee well-being develops over time. Although some studies (Cummins, Gullone, & Lau, 2002; Headey & Wearing, 1989) suggested stability of employee well-being over time, others (Fredrickson, 1998; Hobfoll, 1989; Mäkikangas, Kinnunen, Feldt, & Schaufeli, 2016) indicated that employee well-being may increase or decrease. Therefore, one’s level of functioning (flourishing, moderately mentally healthy and languishing) could be susceptible to change, and thus numerous cross-sectional studies did not take the dynamic and progressive nature of workplace flourishing into consideration. This study contested the latter limitation by demonstrating that workplace flourishing is non-static in nature and that employees’ functioning levels may fluctuate positively or negatively over time.

Research design

Research sites and participants

A longitudinal survey design through convenience sampling enabled us to collect data from high school teachers in both Sedibeng districts (East and West) in the Gauteng Province. A total of 800 questionnaires were issued, of which 258 were completed in a satisfactory manner in the first period of administration (time 1). This accounted for a response rate of 32%. However, a further 56 participants dropped out of the study and did not complete the follow-up questionnaire in the second period of administration (time 2). Thus, a total of 202 participants completed the questionnaire at both time points. As per Table 1, most of the participants were female (74.3%), had more than 10 years’ educational experience (52%) and fell within the 35+ age category (51%). Over 40% of the participants were in possession of a tertiary degree.

Measuring instruments

A short version (17 items) of the FAWS (FAWS-SF; Rautenbach, 2015) was used to measure workplace flourishing. The scale measured how frequently participants experienced certain phenomena at work by recording their responses on a six-point scale ranging from 1 (never) to 6 (every day). Emotional well-being was measured by three items (e.g. ‘During the past month at work, how often did you feel happy?’), PWB by nine items (e.g. ‘During the past month at work, how often did you feel confident to think or express your own ideas and opinions?’) and SWB by five items (e.g. ‘During the past month at work, how often did you feel you really belong to this school?’). More sample items from the questionnaire are available (see Redelinghuys, Rothmann, & Botha, 2019a). The FAWS-SF has been shown to be valid and reliable (>0.70) in the South African context (Rautenbach & Rothmann, 2017).

The Turnover Intentions Scale (TIS; Sjöberg & Sverke, 2000) was used to measure intention to leave. The TIS comprises three items rated on a five-point response scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Covering a single dimension, a sample item is: ‘If I was completely free to choose, I would leave this job’. The Raykov’s rho coefficients for the current study were 0.91 for time 1 and 0.92 for time 2.

Research procedure and ethical consideration

Before the commencement of the study, permission was obtained from the Gauteng Department of Education’s (GDE) research division and both Sedibeng district offices (East and West). Also, scientific and ethical clearance was secured from a reputable institution for higher education. Ethical approval was granted by the Humanities and Health Research Ethics Committee (HHREC) at the North-West University (NWU-HS-2015-0193). Approval letters from the GDE to conduct the study were issued from 08 February 2016 (the first school quarter) to 30 September 2016 (the third school quarter) to ensure minimal disruption of school activities. Meetings with key stakeholders from mainstream secondary schools in the stipulated regions (principals, gatekeepers and probable participants) were arranged to discuss the logistics and purpose of the study. Private schools and learners with special educational needs (LSEN) schools were excluded. Participants were allowed 2 weeks to complete the paper-based English questionnaires, taking up approximately 30 min of their time. English as an instructional language was deemed appropriate for the target audience because of their educational level and occupation. The second round of administration took place exactly 6 months after the questionnaires were first delivered to the respective schools. In employee well-being studies, 6-month time intervals are common (Mäkikangas et al., 2016).

Accordingly, in line with the GDE’s approval letter, the first data collection took place during February 2016, whilst the second data collection occurred in August 2016. The same participants completed both waves of assessments. Arrangements were made with participants for the secure return of their questionnaires. Every participant had the option to voluntarily withdraw from the study without any consequences.

Statistical analysis

Mplus 8.2 (Muthén & Muthén, 1998–2019) and the 25th version of the Statistical Package for the Social Sciences (SPSS 25) (IBM Corp., 2017) were used for analysing the dataset.

A measurement model for each time interval was tested with the maximum likelihood estimation with robust standard errors (MLR) indicator in Mplus. To assess model fit, the following indices were used: the chi-square statistic, root mean square error of approximation (RMSEA), Tucker–Lewis index (TLI), comparative fit index (CFI), standardised root mean square residual (SRMR), Akaike information criterion (AIC) and the Bayesian information criterion (BIC). Values of 0.90 and higher on the CFI and TLI, as well as values lower than 0.08 on the RMSEA and SRMR, were considered acceptable (Cudeck & Browne, 1993; Van de Schoot, Lugtig, & Hox, 2012). Lower AIC and BIC values indicated better fit (Wang & Wang, 2012). The statistical significance confidence interval (CI) was set at 95%. Thus, p-values lower than, or equal to, 0.05 showed evidence of statistical significance. To assess the reliability of the scales, the Raykov’s (2009) rho coefficient was used.

Values of ≥0.70 were considered sufficient (Nunnally & Bernstein, 1994).

Analysis of longitudinal measurement invariance (inclusive of configural, metric and scalar invariance) was conducted to assess whether workplace flourishing and intention to leave was measured in a consistent manner over the time intervals (Horn & McArdle, 1992; Vandenberg & Lance, 2000). To assess participants’ levels of feeling and functioning (flourishing, moderately mentally healthy and languishing), the authors used an adapted SPSS syntax in line with the approach followed by Keyes (2009). To work with a full dataset, the authors used listwise deletion to handle missing data.

Cross-tabulations were used to assess the relationships between nominal variables. The chi-squared test statistic (c²) was used to test for the statistical significance of relationships. Cramer’s V was used to test the practical significance of effects (Cohen, 1992). Steyn (2012) pointed out that the size of the table has an influence on the value of V. However, the maximum number of columns in this study was three. Cut-off values of 0.10 (small effect), 0.30 (medium effect) and 0.50 (large effect) were used for the practical significance of V (Steyn, 2012).

Results

Means, standard deviations, reliability coefficients and correlations

From Table 2, it is evident that participants experienced lower mean scores during time 2 on all the flourishing dimensions. All correlations were practically and statistically significant. All the reliability coefficients also comfortably exceeded the acceptable reliability threshold (>0.70) (Nunnally & Bernstein, 1994).

Testing measurement models

To confirm the three-factor structure of workplace flourishing (as established by previous studies), confirmatory factor analysis via Mplus was performed for each time interval. Workplace flourishing was specified as a second-order factor, containing emotional, psychological and SWB factors. Each latent factor (emotional, psychological and SWB) was specified, with its corresponding questions as observed indicators. Thus, EWB included three observed variables, PWB nine observed variables and SWB five observed variables. The original measurement models did not yield acceptable fit. For time 1, the following fit statistics were produced: χ² = 238.198; df = 116; χ²/df = 2.05; p < 0.001; TLI = 0.88; CFI = 0.90; RMSEA = 0.07 (90% CI, 0.06–0.09); SRMR = 0.06; AIC = 9769.74; BIC = 9948.39. For Time 2, the fit statistics were as follows: χ² = 279.441; df = 116; χ²/df = 2.41; p < 0.001; TLI = 0.87; CFI = 0.89; RMSEA = 0.08 (90% CI, 0.07–0.10); SRMR = 0.06; AIC = 9555.51; BIC = 9734.15.

In time 1, the TLI, and in time 2, the CFI and TLI values were below the recommended value of 0.90. The researchers continued in an exploratory manner to identify avenues through which fit could be obtained.

During time 2, high modification indices between two pairs of questions were established: 36.52 between ‘Feel you are a key member of this school?’ and ‘Feel you really belong to this school?’ on the SWB subscale, and 22.97 between ‘Feel that you understand how your work contributes to your life’s meaning?’ and ‘Feel that the work you do serves a greater purpose?’ on the PWB subscale. To improve fit, the item errors of these pairs of items were allowed to correlate. The same changes were made to the measurement model for time 1 for consistency purposes. The fit statistics for the revised models are reported in Table 3.

As per Table 3, acceptable fit according to all the relevant criteria was achieved. Competing measurement models or structural models were not tested, as the emphasis was not on testing associations between variables but rather on classifying participants according to their level of functioning (flourishing, moderately mentally healthy and languishing).

Longitudinal measurement invariance

Through longitudinal measurement invariance, the authors assessed whether workplace flourishing and intention to leave were measured consistently across time intervals (Horn & McArdle, 1992). Results are indicated in Table 4.

As shown in Table 4, no significant differences were found in the comparison of the configural, metric and scalar models of the two samples across the two time intervals. Configural invariance suggested a consistent factor structure over time; metric invariance indicated consistent factor loadings over time and scalar invariance implied consistent item intercepts over time. These findings provided evidence for strong longitudinal measurement invariance.

Prevalence and classification of workplace flourishing

Workplace flourishing is organised into three categories: flourishing at work, moderately mentally healthy and languishing at work. To be grouped into the flourishing at work category, participants needed to display a minimum of one of the three signs of EWB, a minimum of eight of the 14 signs of PWB and SWB ‘every day’ or ‘almost every day’. In contrast, to be grouped into the languishing at work category, participants had to display a minimum of one of the three signs of EWB, a minimum of eight of the 14 signs of PWB and SWB ‘never’ or ‘once or twice’. Participants fitting into neither of the preceding categories were deemed moderately mentally healthy (adapted from Rautenbach, 2015).

During time 1, 48.5% of the participants flourished, 47% were moderately mentally healthy and 4.5% languished. During time 2, however, the number of flourishing participants decreased to 43.1%, whilst the participants who were moderately mentally healthy and languishing increased to 51% and 5.9%, respectively (see Table 5).

Regarding the participants who dropped out of the study, the moderately mentally healthy (57.1%) constituted the largest percentage, followed by flourishing (28.6%) and then the languishing individuals (14.3%) (see Table 6).

Interestingly, most participants who dropped out were not in the lower spectrum of functioning as one might expect. Participants who dropped out were predominantly female (71.4%), older than 35 years (46.4%), had more than 10 years’ educational experience (51.8%) and possessed a degree (44.6%).

As per Table 7, 64.85% of the participants demonstrated no change in the classification category in which they found themselves. A total of 20.30% of the participants, however, showed a negative change in the classification category. Accordingly, these participants went from flourishing to moderately mentally healthy, from moderately mentally healthy to languishing and from flourishing to languishing. In contrast, 14.85% of the participants demonstrated a positive change in the classification category. Accordingly, these participants went from languishing to moderately mentally healthy, from moderately mentally healthy to flourishing and from languishing to flourishing.

To assess the significance of the preceding results, a series of cross-tabulations were performed in SPSS. In the first computation, flourishing versus non-flourishing groups were compared over time. Although the categories were initially labelled as languishing, moderately mentally healthy and flourishing, these were changed to flourishing and non-flourishing categories to ensure that each cell had more than five entries for statistical purposes. The results showed that 77 of the participants who did not flourish in time 1 (74%) did not flourish in time 2 as well. However, 27 of the participants who did not flourish in time 1 (26%) flourished in time 2. A total of 60 of the participants who flourished in time 1 (61.2%) flourished in time 2 as well. In contrast, 38 of the participants who flourished in time 1 (38.8%) did not flourish in time 2. A significant Pearson chi-square (25.59; df = 1; p = 0.000) and Cramer’s V value (0.36, medium effect) indicated that there were indeed notable changes in classification categories of teachers over time. This illustrated the dynamic nature of workplace flourishing and how individuals’ functioning levels might change over a 6-month period.

To further emphasise the importance of monitoring employees’ functioning, a second series of cross-tabulations were performed. Flourishing and non-flourishing groups in time 1 were compared regarding their intention to leave (low, moderate and high) in time 2. Significant Pearson chi-square (21.72; df = 2; p = 0.000) and Cramer’s V values (0.33, medium effect) were established. Regardless of the time under consideration, non-flourishing participants predominantly found themselves in the high-to-moderate intention to leave categories, whilst flourishing participants predominantly found themselves in the low-to-moderate intention to leave categories. Thus, non-flourishing employees were significantly more likely to report intentions to leave the organisation, compared to their flourishing counterparts. Because of the effect that employees’ functioning has on their intentions to leave, it is important to take into account fluctuating functioning levels over time.

Discussion

The current study aimed to demonstrate that workplace flourishing is non-static in nature and that employees’ functioning levels may fluctuate positively or negatively over time. To ensure validated findings for longitudinal research, the constructs under scrutiny should be reliable and invariant. The results supported the configural, metric and scalar invariance of workplace flourishing over time. Furthermore, it showed good test–retest reliability for the short version of the FAWS. This confirmed the value of a construct, such as workplace flourishing, not merely from a cross-sectional viewpoint but also longitudinally.

Similar to previous studies (e.g. Boshoff et al., 2014; Diedericks & Rothmann, 2014), more than 50% of the sample did not flourish. These participants mainly fell in the moderately mentally healthy and languishing groups. These findings were mostly in line with studies that assessed flourishing in the South African context.

Surprising results were found regarding the participants who dropped out of the study. Most participants who dropped out of the study were not in the lower spectrum of functioning, as one might expect.

One can only speculate on the reason behind this. Perhaps, these participants felt that their functioning levels were at a satisfactory level and that there was, thus, no need for them to assess their functioning levels again.

The results showed that 64.85% of the participants demonstrated no change in the classification category in which they found themselves, whilst 20.30% indicated a negative change and 14.85% a positive change.

Thus, in the context of change, participants’ functioning levels were more likely to change negatively than positively. As direct comparisons to previous studies are impossible, one can only speculate about the reason behind this.

In addition, the results showed that teachers experienced notable changes in their classification categories (non-flourishing vs. flourishing) over time and that these changes were related to their intention to leave. This supports previous research findings, suggesting that employee well-being may change over time (Fredrickson, 1998; Hobfoll, 1989; Mäkikangas et al., 2016). This also illustrates the importance of considering fluctuating employee functioning levels.

The results also showed that there were always exceptions to the rule, specifically referring to those participants who went from languishing to flourishing (one participant) and those who changed from flourishing to languishing (four participants). This illustrated the complexity of individual uniqueness and the reason why companies should never have a one-size-fits-all approach.

Limitations

The study did not come without limitations. The current study only looked at the study phenomena from a quantitative perspective. Because of the quantitative design, this study could not identify the reasons as to why certain employees switched between well-being categories. Furthermore, only two time intervals over a 6-month period were assessed. Having more than two time periods with different time intervals may or may not yield different findings. The current study did not use state-of-the-art missing data handling techniques.

Conclusion and recommendations

Because of the impact workplace flourishing may have on individual performance and retention, it is important for organisations to continuously monitor their employees’ levels of functioning. Categorising employees according to their levels of functioning once at the beginning of the year is not enough, as these levels may be susceptible to change. Accordingly, organisations should not treat this as a once-off process but rather as a continuous investment to ensure the productivity and retention of key staff members. Because of the complexity of individual uniqueness, employees might also move from languishing to flourishing in certain instances, and vice versa. Therefore, organisations should never have a one-size-fits-all employee management approach, as what works for one employee might be highly irrelevant for another. This calls for individual-tailored strategies, and although these might be tedious and timely to set up, they should pay off in the long run. After all, employees are the most valuable organisational and societal assets.

In terms of recommendations for future research, mixed-method or qualitative studies may provide a deeper understanding regarding as to why certain employees switched between functioning categories, whilst others remained in the same category. Assessing the functioning levels over different time periods (as opposed to the 6-month period of the current study), across more than two time intervals, as well as studies with an intervention component may provide potentially interesting results. As the current study only provides a glimpse of how employee well-being develops over time, more research is needed in this regard.

Acknowledgements

The authors would like to acknowledge the schools who participated in the study.

The authors have declared that no competing interest exist.

K.R. coordinated the data collection process. S.R. assisted with data analysis and interpretation. K.R. and S.R. made contributions regarding the conceptualisation of the article, as well as writing and editing.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The authors confirm that the data supporting the findings of this study are available within the article, with only the primary researchers having access to the raw data.

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.

References