Chronotype and organizational citizenship behavior during the COVID-19 restriction phase in Germany

ABSTRACT

Here, we researched the effects of the COVID-19 restriction measures on learning/work-related characteristics (working hours, creativity in problem-solving approaches/organizational citizenship behavior (OCB)) depending on chronotype of N = 681 German residents (mean age: 28.63 years, SD: 10.49 years). The data were collected with an online questionnaire from 18 May to 17 June 2020, during the most restrictive phase in Germany. We analyzed participants studying/working in home offices only. Morningness showed positive, while eveningness showed negative correlations to OCB. Morning types worked their usual working times, while evening types took more and longer breaks. In remote work, morning types felt more creative developing problem-solving approaches, while evening types reported the opposite. Our results suggest that remote working is not beneficial for evening types when performance components are concerned, even though they can choose their working time freely which benefits their biological rhythm. This study should be repeated with workers on a large scale to confirm these results.

KEYWORDS:

• Big Five
• new ways of working
• home office
• remote working
• morningness–eveningness

1. Introduction

Since the first known infection with the novel coronavirus induced disease COVID-19 in December 2019 (Wuhan, China), a global pandemic developed (Wu and McGoogan 2020). The consequences for the health system, economy and social life were immense. To stop the spreading, governments (starting with China) imposed restrictions and ultimately lockdowns on the populations. In Europe, the measures differed considerably. Italy called out a national health emergency and locked down the whole Italian territory beginning 10 March 2020. Until 3 May 2020, the population was placed in social isolation (Marelli et al. 2020). In Sweden, the measures mainly relied on voluntary compliance with the Public Health Agency’s recommendations (Hensvik and Skans 2020). In Germany restriction measures were set starting in the middle of March. From this point on, universities, schools, kindergartens, and non-essential businesses were closed. This resulted in a shift to home office and distant learning practices. The sudden changes in lifestyle had various physiological (e.g. absence of sport and leisure activities: Randler et al. 2020) and psychological consequences (e.g. altered sleep: e.g. Staller & Randler 2020).

Studies concerning sleep health during the global COVID-19 pandemic reported changes in contrary ways. Following increasing depression, anxiety and future uncertainty, many people faced sleep disturbances like insomnia, sleep loss, and poor sleep quality (Lai et al. 2020; Zhang et al. 2020; Cellini et al. 2020). For others, the change in sleep wake cycle due to working from home and the abolition of commuting times resulted in positive alterations of sleep parameters (Gao and Scullin 2020; Sinha et al. 2020; Leone et al. 2020; Cellini et al. 2020; Staller & Randler 2020). The latter findings support a modern approach to working called “New Ways of Working” (NWW) (Baane et al. 2011).

NWW as a concept tries to create work environments that are temporal and spatially flexible and focus on innovation and productivity while reducing costs (Nijp et al. 2016). It is proposed as a way to improve work time control by adjusting work to private life (Gajendran and Harrison 2007; Nijp et al. 2015) and biological needs (e.g., chronotype (Wittmann et al., 2006). Further advantages related to NWW are improved motivation due to autonomy (Pritchard and Payne 2003), cost reduction, increased efficiency and information sharing following digitalization and adapted use of office space (Peponis et al. 2007; Demerouti et al. 2014) as well as reduced commuting time/ costs and resulting decrease of environmental pollution (Manoochehri and Pinkerton 2003). Negative aspects of NWW might be the lack of collegial support and exchange (Halford 2005), increased stress due to the feeling of responsibility, loss of structure as well as missing boundaries between work and private life (Lundberg and Cooper 2010; Allvin et al. 2011; Mazmanian et al. 2013; Demerouti et al. 2014). Nevertheless, NWW is designed to influence the work output in a positive manner. The changes that the COVID-19 pandemic has brought about have enabled many people to integrate the NWW concept at least partially into their everyday lives.

This changeover was followed by decreasing social jetlag due to the temporal flexibility, underlining the positive adjusting effects of NWW (Staller & Randler 2020). In another study Staller et al. (2021) reported altered motivational regulation in biology teaching students, depending on chronotype and Big Five personality characteristics during the restriction phase in Germany. Herein it was concluded that the NWW concept might be more suitable for morning types and not/not very beneficial for evening types when only learning/working is concerned (not considering improved biological factors). It was stated that these relationships should be studied further.

The chronotype is a personality trait, which cannot be influenced externally (Tsaousis 2010) but changes during the lifespan (Randler et al. 2017). Morning types get up earlier and go to bed earlier than evening types and reach their peak performance earlier during the day (Adan et al. 2012). This trait has a biological basis, which can be measured by questionnaires, hormones, or daily fluctuations of the body temperature (see Adan et al. 2012, for a review). In general, women are more morning oriented than men (Randler and Engelke 2019), children are more morning oriented, while a shift toward eveningness occurs during puberty, and back to morningness at the end of puberty (Randler et al. 2017). The relationship of morningness to different personality dimensions, for example, agreeableness and conscientiousness, has been replicated repeatedly (see, e.g. Tsaousis 2010; Adan et al. 2012). Among others, these personality characteristics mark an important influencing factor of working success, since, for example, they correlate to constructs, which regard an employee’s contribution to the employing company’s success (performance components, e.g. Organizational Citizenship Behavior (OCB)). Therefore, the circadian preference might be an understudied influencing factor of learning/working success and work models like NWW. Hence, to achieve the best possible results in learning/working, the chronotype’s influences should be explored more in relation to work success variables.

To measure the subjective assessment of participants’ own involvement to their learning/working duties we used the conduct of Organizational Citizenship Behavior in its German version from Staufenbiel and Hartz (2000). OCB has increasingly become a subject of organizational psychological research (see Staufenbiel 2000 for an overview). The construct was originally created to resolve the contradiction between the belief that there is a positive correlation of job satisfaction and performance, on the one hand, and empirical data, on the other hand (Organ 1988). Organ (1988) assumed the way of measuring the performance may be inadequate for the expected results because of the small influence employees have to this value. He named other factors like technology or workflows as potentially limiting. If, instead, performance components such as OCB are surveyed, higher correlations with job satisfaction result (see meta-analysis from Organ and Ryan 1995). Important definition criteria of the OCB construct are voluntary helpfulness with absence of any reward, and that the employee’s behavior must be conducive to the effectiveness of the organization (Organ 1988). Based on this definition, related constructs can be delimited. For example, the conduct of prosocial organizational behavior (Brief and Motowidlo 1986) also includes behavior that is formally required, such as helping customers. Personality dimensions that are favorable for OCB were also linked to morningness (e.g. conscientiousness, agreeableness, see, e.g. Tsaousis 2010; Adan et al. 2012). Further aspects of individual differences that are both relevant for circadian preference and OCB are, for example, proactivity. This characteristic was identified as an important factor for workplace success. Randler (2009) reported that proactivity was related to morningness, while procrastination was related to eveningness (Díaz-Morales et al. 2008).

NWW should benefit evening types from a biological viewpoint because they can choose their work start times freer in concordance with their intrinsic biological rhythms, which results in lower social jetlag. However, in terms of performance components, we hypothesize NWW is less beneficial for evening types, because they need collegial support and default structure more than morning types. Morning types in comparison have more favorable personality characteristics, such as higher conscientiousness, to succeed in learning/working settings with less supervision. Here we examine how work-related characteristics (e.g. creativity in problem-solving approaches) are affected by the home office orders and how these changes relate to circadian preference.  

2. Methods

2.1. Data collection

This study was performed in accordance with the Declaration of Helsinki for experiments involving humans, approved by the ethics committee of the faculty for economy and social sciences of the Eberhard Karls university Tuebingen (Nr. A.Z.: A2.5.4–124_kr). The Data was collected during the most restrictive phase in Germany, from 18 May 2020 until 17 June 2020. Participants were recruited through an online link to the survey they received via Mail (electronic mailing list) or postings on different social media platforms (Facebook/Instagram). Data collection was limited to a single point in time (when the respondents visited the questionnaire via the link). The survey was hosted on an online platform (SoSciSurvey), fulfilling the European Union’s data privacy rules. On average, it took 12 min ± 5 min (SD) to complete the questionnaire. Formal consent was inquired in advance. Participants were informed about the theoretical background and study goals (not the hypothesis), voluntariness of the participation, the option to stop the data collection at any point without consequences, and that participation will not be remunerated. The total number of evaluable cases amounts to 681.

2.2. Questionnaire

2.2.1. Demographic data

We asked for age (mean: 28.63 SD: 10.49), sex (n = 197 males; n = 484 females), household size, number of children living in participants’ household during the lockdown measures, profession and the option of flex time working. We later dichotomized the answers to “profession” into student (N = 400) and non-student (N = 281). N = 545 participants reported no children in their household, while N = 136 stated one or more. We explicitly asked for the number of children in the household (not the number of own children) because students may live with their families and consequently younger siblings during the lockdown measures. Thus, “children in the household” is a better measure than “own children”, because children in general may impair quality of learning/ working/ sleeping regardless of relationship (own children/siblings/ other cases).

2.2.2. Organizational citizenship behavior

OCB is usually measured using external assessment questionnaires. The first and a frequently used questionnaire originated from Smith et al. (1983), using 16 items assigned to the personality characteristics “altruism” and “conscientiousness” (Staufenbiel and Hartz 2000). It was later revised by Organ (1988), who added three more dimensions of personality characteristics (sportsmanship, civic virtue, courtesy; Staufenbiel and Hartz 2000). Although the instruments of the group around Podsakoff & MacKenzie (see, e.g. Podsakoff et al. 1990; MacKenzie et al. 1991), using these five personality characteristics (altruism, conscientiousness, sportsmanship, civic virtue, courtesy), found the largest distribution, no questionnaire has gained general acceptance in American research (Staufenbiel and Hartz 2000). Hence, as no questionnaire was suitable for translation, Staufenbiel and Hartz (2000) developed a new questionnaire for German research. Items from the existing OCB-instruments were selected which met the following criteria: (1) broad coverage of the five OCB-subscales, (2) not too specific to individual areas of activity, (3) clear compliance with Organs (1988) definition of OCB (Staufenbiel and Hartz 2000). Of the five facets (altruism, conscientiousness, sportsmanship, civic virtue, courtesy) the facet “courtesy” was not empirically supported and therefore deleted in the new German questionnaire (Staufenbiel and Hartz 2000). For the other four facets, the five items that loaded the highest on the factor and showed lowest cross loading were selected (Staufenbiel and Hartz 2000). Additionally, the scale of “in-role behavior” of Williams and Anderson (1991) was used. Therefore, the questionnaire has a five-factor structure (altruism, conscientiousness, sportsmanship, civic virtue, in-role behavior) with every factor being represented with five items. The five-factor model showed a ${\chi }^2$ of 640,5 (p < 0.001), df of 265 and CFI of 0.91 (Staufenbiel and Hartz 2000). The questionnaire is available in external- or self- assessment form. Here, we used the self-assessment version. Staufenbiel and Hartz (2000) proposed a confirmatory factor analysis to assess the fit of the data structure. In our current sample, we found an RMSEA of 0.053, with the CI ranging from 0.049 to 0.057, and a PClose of 0.111, suggesting a good fit of our data with the theoretical construct.

2.2.3. Additional freeform questions concerning the working behavior during the home office phase

We asked for changes in creativity in the development of problem-solving approaches and working hours additionally to the OCB questionnaire. These questions aim to address the main changes that occurred due to the transfer from workplace/university to home office: (1) the lack of collegial support → other perspectives and approaches help to find creative problem-solving strategies; (2) the possibility of a freer allocation of time, which can improve the ability to concentrate, but can also provide distractions. A freeform design questionnaire was used to uncover these changes: (1) Do you find it easier/equal/ more difficult to develop creative solutions and solve problems in your home office? Describe. (2) Do you split your working hours more than usual (e.g. more/longer breaks) or do you work as usual? Describe. The answers were later grouped in (1) 1 = easier/2 = equal/3 = more difficult; (2) 1 = same working hours/2 = more and/or longer breaks.

2.2.4. Chronotype

In this study, we used circadian preference as a measure of chronotype. This differs from the clock-based measurement of chronotype, such as the MCTQ (Roenneberg et al. 2003), but in fact, both measures are closely related (and are correlated with each other with an r = 0.73; Zavada et al. 2005). In the present study, we used the Morningness-Eveningness Stability Scale improved (MESSi; Randler et al. 2016; Díaz-Morales et al. 2017) to determine circadian preference. The MESSi consists of three subscales represented by five items each, the Morning Affect (MA), Eveningness (EV) and Distinctness (DI). Each of those queries a delimited facet of the M/E trait. MA deals with the energy level in the morning (e.g. “How alert do you feel during the first half hour after having awakened in the morning?”) while EV is concerned with the situation in the evening (e.g. “In general, how is your energy level in the evening?”). The DI subscale asks for changes in active phases during the day (e.g. “There are moments during the day where I feel unable to do anything.”). High scores in the MA or EV subscale indicate high orientation toward this facet, while high DI scores represent higher daytime fluctuations. MESSi’s validity has already been confirmed repeatedly in different languages, using a broad spectrum of methods (Díaz-Morales and Randler 2017; Rahafar et al. 2017; Rodrigues et al. 2018; Faßl et al. 2019; Tomažič and Randler 2020) and is used on a global scale (see, e.g. Arrona-Palacios (2020) for a Mexican sample). Cronbach’s α in the current sample was 0.899 for MA, 0.889 for EV, and 0.775 for DI.

2.2.5. Big Five personality

To measure Big Five personality dimensions, we used the German short version of the Big Five inventory (BFI-10; Rammstedt and John 2007; Rammstedt et al. 2013). This scale is based on the BFI-44 (Benet-Martínez and John 1998) and shortened to a 10-item questionnaire. Two items represent each personality dimension (extraversion, agreeableness, openness, neuroticism, and conscientiousness) with one item per dimension reverse coded. The items are answered in a five-point-Likert format. The BFI-10 showed a clear five factor structure and good external validity (Rammstedt and John 2007). Extraversion, Neuroticism, and Conscientiousness had average correlations with the BFI-44 scales of .89, .86, and .82, while Agreeableness and Openness had .74 and .79. In our sample, the correlation between the two items was 0.666 for Extraversion, 0.482 for Neuroticism, 0.298 for Conscientiousness, 0.157 for Agreeableness, and 0.452 for Openness.

3. Results

The independent predictor variables circadian preference and Big Five personality dimensions were related with each other (see Table 1). Morningness was positively associated with agreeableness and conscientiousness, and negatively with neuroticism and openness. Eveningness, in turn, was negatively related to conscientiousness, but positively to openness. Distinctness was positively related to neuroticism and openness but negatively to extraversion, conscientiousness and agreeableness.

Table 1. Correlation matrix chronotype and Big Five personality dimensions.

		Extraversion	Agreeableness	Conscientiousness	Neuroticism	Openness
MA	r	0.075	.123**	.272**	−.183**	−.079*
	P	0.052	0.001	<0.001	<0.001	0.039
EV	r	0.026	−0.038	−.182**	−0.052	.102**
	P	0.502	0.327	<0.001	0.176	0.008
DI	r	−.104**	−.107**	−.234**	.341**	.086*
	P	0.007	0.005	<0.001	<0.001	0.025

MA = Morningness; EV = Eveningness; DI = Distinctness¸** denotes p < 0.01; * denotes p < 0.05

The correlation matrix between OCB and work-related characteristics on one side and personality/circadian preference on the other showed that morningness was positively related to conscientiousness, civic virtue, and in-role behavior (see Table 2). Furthermore, morningness related positively to creativity in problem-solving approaches and negatively to splitting working hours (see Table 3). Eveningness correlated negatively to conscientiousness, in-role behavior (see Table 2) as well as creativity and positively to splitting working hours (see Table 3).

Table 2. Correlation matrix of chronotype and Big Five Personality dimensions with OCB-variables.

		OCB altruism	OCB conscientiousness	OCB sportsmanship	OCB civic virtue	in-role behavior
Morning Affect	r	0.052	.291**	0.066	.092*	.184**
	P	0.173	0.001	0.084	0.016	0.001
Eveningness	r	0.015	−.189**	0.003	0.001	−.092*
	P	0.693	0.001	0.942	0.97	0.017
Distinctness	r	−.098*	−.123**	−.184**	−.187**	−.168**
	P	0.01	0.001	0.001	0.001	0.001
Extraversion	r	.264**	−0.024	0.075	.284**	.086*
	P	0.001	0.535	0.051	0.001	0.025
Agreeableness	r	.244**	0.061	.301**	0.006	0.01
	P	0.001	0.112	0.001	0.876	0.786
Conscientiousness	r	.200**	.334**	.163**	.258**	.412**
	P	0.001	0.001	0.001	0.001	0.001
Neuroticism	r	−.168**	0.035	−.270**	−.242**	−.155**
	P	0.001	0.368	0.001	0.001	0.001
Openness	r	.148**	−0.026	0.056	.111**	0.01
	P	0.001	0.492	0.145	0.004	0.797

OCB = Organizational Citizenship Behavior; ** denotes p < 0.01; * denotes p < 0.05

Table 3. Correlations of chronotype and Big Five Personality dimensions with the additional freeform questions concerning the working behavior during the home office phase.

		creative problem solving approaches	splitting working hours
Morning Affect	r	.082*	−.123**
	P	0.036	0.002
Eveningness	r	−.086*	.114**
	P	0.028	0.004
Distinctness	r	0.063	.142**
	P	0.109	0.001
Extraversion	r	0.039	−0.023
	P	0.324	0.559
Agreeableness	r	0.004	−0.05
	P	0.927	0.211
Conscientiousness	r	0.012	−0.073
	P	0.759	0.064
Neuroticism	r	.078*	−0.029
	P	0.048	0.459
Openness	r	−0.063	−0.021
	P	0.108	0.59

** denotes p < 0.01; * denotes p < 0.05

4. Discussion

4.1. Chronotype and Big Five

In our sample, we replicated the relationship of morningness to conscientiousness and agreeableness as well as of eveningness to openness (compare Tonetti et al. 2009; Adan et al. 2012). We failed to replicate a connection of eveningness to extraversion and neuroticism but showed a negative correlation of eveningness to conscientiousness and morningness to neuroticism and openness. This might be due to the method determining the chronotype. Here, we used the MESSi to examine the chronotype. The MESSi is a rather new questionnaire, which was created to challenge the disadvantages of clock time-based questionnaires and to provide a consistent Likert-scale rating from 1 to 5, which is not available in most previous chronotype questionnaires (Di Milia et al. 2013). Further, the MESSi portrays chronotype in three dimensions rather than only one (for discussion, see Randler et al. 2016). Even so, to confirm these results further, it would be useful to replicate the study design with a standard instrument to assess circadian preference.

4.2. Chronotype and OCB

The results of both morning and evening types in the Big Five personality dimension “conscientiousness” were in accordance with the results of the OCB-variable “conscientiousness”. Morning types showed a positive relationship to the OCB “conscientiousness”, while evening types showed a negative one. This was strengthened further by the correlation of Big Five conscientiousness to OCB conscientiousness. Moreover, morningness related positively to the variables OCB “civic virtue” and “in-role behavior”, without relating to an OCB variable negatively. Eveningness correlated negatively to in-role behavior, not relating positively to one of the OCB variables. Morningness showed a positive while eveningness showed a negative connection to OCB and therefore conduct at the workplace. These results show that morning types invest more energy in job/studies than evening types do. This adds another point of view to the considerations on why evening types show worse academic performance than morning types throughout their whole education (compare Randler and Frech 2006; Arbabi et al. 2015; Tonetti et al. 2015; Kolomeichuk et al. 2016). Because of the socially demanded working hours and school starting times, evening types live with a sleep deprivation during the week, which is called social jetlag (Wittmann et al. 2006). Their internal biological clock differs from the social clock. So far, this was a main reason believed to cause worse academic performance in schools and universities not least because the differences in success of university students were mitigated by the level of free time allocation and therefore social jetlag (Tonetti et al. 2015). Here we show that evening types invest less in their job and educational duties on a self-aware level, which might be another important reason. However, since we have not received an OCB evaluation from participants’ supervisors to reconcile the self-assessment and external assessment data, the results could also be interpreted differently. Evening types might feel not having reached their full potential yet, feeling capable of investing more and achieving better results. They therefore may have assessed themselves as not meeting the expected standards.

4.3. Chronotype and changes in working behavior during the restriction phase

The OCB questionnaire in our survey queried the self-assessment of one’s general work behavior, not limited to the home office phase. We therefore additionally asked for changes in creativity in the development of problem-solving approaches and working hours during the restriction phase to specifically address possible alterations.

Morningness was positively correlated, while eveningness was negatively correlated with creative problem-solving approaches. Both relationships might be explained by the lack of colleagues around to engage with. For morning types who are less open (Big Five personality dimension) this change might be beneficial as far as work-related creativity is concerned, because they can concentrate on their own thinking and do not have to consider other arguments into their problem-solving approach. Therefore, they might feel less held back by others, which results in a higher self-assessed rating of creativity of problem-solving approaches. Evening types, in contrast, are more open (Big Five personality dimension) and may relate more on conversations and brainstorming with colleagues to build strategies for problem solving. Therefore, they might feel less creative during home office because of the lack of input from others. In contrast, Giampietro and Cavallera (2007) found evening types to be more likely to use divergent thinking strategies and show creative thinking than morning types.

When splitting working hours are concerned, morningness showed a negative while eveningness showed a positive correlation. Morning types by themselves are on an advantage by socially demanded working hours and school starting times. Their usual working hours therefore might be beneficial followed by no need to change them. Furthermore, morning types are more conscientious, proactive, and behaving according to the OCB criteria, which might raise the desire to fulfill the expected availability. Eveningness, on the other hand, relates negatively to conscientiousness and positively to procrastination. Evening types, therefore, may take more breaks because they procrastinate and have no remorse because of this (Díaz-Morales et al. 2008).

4.4. Big Five and OCB

Agreeableness and conscientiousness as two of the Big Five personality dimensions are well-known predictors of OCB (Organ and Ryan 1995; Ilies et al. 2009). Here, we could show the Big Five personality dimension “conscientiousness” to correlate with all OCB measures and in-role behavior on a highly significant level. Agreeableness was shown to relate to OCB altruism and sportsmanship also highly significant. Together with Emotional Stability, which is understood as the opposite of neuroticism in the Big Five personality dimension, these three characteristics are clustered as socially desirable behavior (Chiaburu et al. 2011). This might be explained by the responsible behavior (conscientiousness), sensitivity in dealing with others (agreeableness) and the absence of negatively directed emotions (emotional stability as opposite of neuroticism) (Chiaburu et al. 2011). Here, we could show every OCB scale and the in-role behavior except for OCB conscientiousness to negatively correlate to neuroticism (which can be understood as correlating positively to emotional stability). In our sample, neuroticism was the best predictor of OCB (all scales combined) second only to conscientiousness, while agreeableness showed to only relate to two scales. Extraversion and openness, on the other hand, are assigned as qualities of dynamic tendencies and proactivity (Chiaburu et al. 2011). Openness is associated with learning orientation and curiosity and extraversion with dominance, while both relate to proactivity (Chiaburu et al. 2011). In our sample, extraversion correlates with the OCB scales of altruism and civic virtue as well as in-role behavior, which underlines the based characteristics, which are accounted to it (dominance/proactivity). The proactivity as a dynamic tendency might guide extraverted employees to help new colleagues during the training period (altruism), while the dominance facet might account for civic virtue. They might want to be leading and first, therefore showing personal initiative. Openness related to OCB altruism and civic virtue as well, which is also in line with this argument. Employees scoring high on the openness scale might be curious and want to be involved with new colleagues or problems of others. The learning orientation facet of openness might reflect in civic virtue since this scale, for example, appoints to constantly upgrading their qualifications or voluntarily taking on additional tasks

5. Conclusion

NWW benefits evening types in terms of biological necessities (internal biological rhythm; decrease of social jetlag). When performance components are concerned, the relationships to the Big Five personality dimensions as well as the OCB scales suggest evening types needing the personal contact to colleagues more than morning types. The analysis of the changes during the restriction phase underlines this connection further. Morning types, on the other hand, might have benefitted from the transition to home office.

6. Limitations and strengths

Our results are very important for employers and companies because they have the potential to increase an employee’s work productivity by improving the work performance components. By implementing the results presented here, suitable working conditions can be tailored to the personality of the employee without incurring additional costs for the employer. To further confirm these results, this experiment should be repeated with workers on a large scale, since we here examined only N = 281 workers. The low number of workers but high number of students may be a limitation of the study execution or more specific for transferability to the working population. Although universities implemented distance learning, the transferability to working in home office may not be given. Furthermore, the sample was non-representative of the German population since the sample size is relatively small and most participants had an academic background. In addition, the sample was relatively unbalanced with about twice as many respondents being female. The recruitment might be limiting as well, since consciously perceived changes in sleep timing or productivity at work could have been accompanied by an increased interest in participation. Another limitation may be the self-assessed study design. Also, participants’ attribution of changes in their productivity to external (changes due to the pandemic) rather than internal factors (changes due to altered supervision) may be problematic (data might be biased). A better alternative to challenge the latter two limitations would be having the participants’ productivity assessed by a supervisor, which still does not guarantee objective data.

Disclosure statement

No potential conflict of interest was reported by the author(s).