Discussion

This research has looked at the effect of sleep quality on the relation between momentary job demands and momentary work engagement. A positive effect of self-perceived sleep quality and recovery on the relation between momentary job demands and momentary dedication was only found for poor self-perceived sleep quality and recovery. This is counterintuitive, as poor self-self-perceived sleep quality and recovery from the previous day were not expected to lead to more momentary dedication. This might be explained by the data, which show a relatively high mean for self-perceived sleep quality (3.67) and recovery (3.47), which may be an indication that such poor sleep quality was still sufficiently high to experience dedication. This already relatively high value for poor self-perceived sleep quality and recovery may also explain why no effect was found for higher self-perceived sleep quality and recovery.

To speculate, these variables subjectively measured sleep quality and recovery, which may explain why participants no longer reported an increase in dedication under higher levels of self-perceived sleep quality and recovery. They may have already achieved their self-perceived saturation point for sleep quality and recovery, which is why their dedication under high levels of self-perceived sleep quality and recovery may no longer increase their dedication as a result. Another explanation for this finding may also be possible. It may be speculated that even for poor self-perceived sleep quality and recovery, an employee could become so strongly dedicated to its task under the influence of challenging momentary job demands that the expected hampering effect of poor self-perceived sleep quality and recovery on dedication no longer exists.

Scientific literature described that high levels of recovery and sleep quality were expected to affect the relation between job demands on work engagement. It was expected that if an employee experiences high recovery and sleep quality, the employee should be better able to cope with the job demands and experience more work engagement as a result (Kühnel et al., 2017; Salanova et al., 2005; Schaufeli et al., 2009; Schaufeli & Taris, 2013; Sonnentag, 2003). However, the findings of the current research suggest that momentary job demands positively affect dedication and that this effect is increased even under the conditions for poor self-perceived recovery and self-perceived sleep quality. Thus, this finding suggests that even when sleep is perceived as being of lower quality, this leads to an increase in momentary dedication as challenging momentary job demands also increase. Indirectly, this is thus expected to lead to more work engagement. Literature for the positive relation between daily job demands and work engagement moderated by self-perceived recovery, suggested that self-perceived recovery could turn daily job demands into challenges when employees have sufficiently recovered from the previous day (Bakker et al., 2010). Moreover, the results of this thesis are consistent with the effort and recovery model (Meijman & Mulder, 1998), by showing that sufficient self-perceived recovery could benefit an employees’ daily work engagement (Bakker et al., 2010). Furthermore, this thesis adds to this literature by showing that daily self-perceived recovery not only benefits momentary work engagement, but also enhances the relation between momentary job demands and momentary dedication, and therefore indirectly momentary work engagement.

Momentary job autonomy and momentary work engagement

This research has found that more momentary job autonomy leads to more dedication and vigor. This effect means that higher momentary job autonomy could positively affect dedication and vigor. This finding is in line with the literature, which describes that higher job autonomy, a job resource, should lead to more work engagement (Bakker & Demerouti, 2007; Demerouti et al., 2001; Hakanen et al., 2006; Schaufeli & Bakker, 2004; Vera, Martínez, Lorente, & Chambel, 2016). As this literature researched general levels of job autonomy and work engagement, this research showed the same results for

momentary levels of job autonomy and work engagement. Thus, ensuring that an employee experiences more momentary job autonomy should be beneficial for its experience of work engagement. This may be achieved by providing the employee with the freedom to plan its own activities and time schedule (Hackman & Oldham, 1975).

Momentary job demands and momentary work engagement moderated by momentary job autonomy

The relation between momentary job autonomy and momentary job demands was further supported by the findings of hypothesis 4. This hypothesis found a positive effect for low and high levels of momentary job autonomy on the relation between momentary job demands and dedication. This means that more momentary job autonomy would lead to an increased effect of momentary job demands on one

dimension of momentary work engagement, namely dedication. Thus, it is expected that as momentary job demands increase, under the assumption of either low or high levels of momentary job autonomy, momentary levels of dedication also increase for an employee. Moreover, results of this thesis showed that high levels of momentary job autonomy are more beneficial for experiencing momentary

dedication. This enhanced effect of the presence of high levels of momentary job autonomy is in line with the literature, which suggests that daily work engagement may be enhanced by the presence of job resources (Breevaart et al., 2014). Furthermore, the presence of sufficient momentary job autonomy could make it easier to perform the work duties, which in turn could make the employee more successful and encourage more positive feelings towards the job (Reis et al., 2015). This increase in resources could spark a repeating effect when the resources are accumulated, turning the resource gain into a gain cycle (Schaufeli & Bakker, 2013).

One might argue that the presence of the resource momentary job autonomy, in combination with momentary job demands that are perceived as a challenge (which then act as an intrinsic motivator), could lead to more work engagement (Sawang, 2012). Thus, when an employee experiences sufficient levels of momentary job autonomy and interprets his momentary job demands as challenges, together they might have a positive effect on work engagement (Breevaart et al., 2014; LePine et al., 2005). As the availability of sufficient job resources could spark a repeating effect, this might also be the case for the effect of momentary job autonomy on the relation between momentary job demands and momentary work engagement. The positive effect that momentary job demands exert on momentary dedication in combination with sufficient levels of momentary job autonomy, might enable a similar repeating effect.

Thus, when an employee is dedicated to his/her work, this dedication might grow through the presence of momentary job demands and momentary job autonomy and this might enable an employee to experience more work engagement (Meijman & Mulder, 1998; Schaufeli & Bakker, 2013; Sonnentag et al., 2012). A reinforcing effect of job autonomy and job demands has been described in scientific

literature. Such research has found that in daily job crafting, employees are more likely to seek resources and less likely to reduce demands on days that are characterized by high work pressure and high

autonomy (Petrou, Demerouti, Peeters, Schaufeli, & Hetland, 2012). This is further explained by the finding that on days where employees have the autonomy to craft their work environment, they are better able to accommodate their job demands, and thus able to experience more daily work engagement (Bakker, 2011). Therefore, high daily job demands and high daily job autonomy could together lead to high daily work engagement.

When an employee experiences low momentary job autonomy, it was also found that dedication increases under the influence of high momentary job demands. This finding could possibly be explained by the notion that employees need to experience a bit of a challenge to stimulate dedication. When an employee has little to no momentary job autonomy, the employee is thus not given the ability to craft the work environment. As a result, the employee has to obtain dedication solely through its given job demands. When these momentary job demands are low, they may not be perceived as a challenge by the employee, resulting in lower dedication. When momentary job demands are higher and they are perceived as a challenge, this leads to an increase in dedication in the employee.

Momentary job autonomy and momentary work engagement moderated by sleep quality

It was expected that high daily sleep quality would enhance the relation between momentary job autonomy and momentary work engagement. This means that when an employee experiences higher sleep quality, he or she is also expected to be able to make more use of the resource momentary job autonomy, which in turn is expected lead to more momentary work engagement. Since momentary job autonomy is a job resource, the employee would be expected to need recovery to optimally make use of this resource. Therefore, it was expected that high levels of recovery and sleep quality would reveal the positive effect of momentary job autonomy on work engagement (Ebert et al., 2015; Meijman & Mulder, 1998; Schaufeli et al., 2009). However, for all variables of sleep quality, no effect on the relation between momentary job autonomy and momentary work engagement was found.

To speculate, an explanation for not finding an effect of daily sleep quality on the relation between momentary job autonomy and momentary work engagement may reside in the notion that momentary job autonomy for employees might not fluctuate as much as other job resources. Thus, momentary job autonomy might not be strongly influenced by the sleep quality of the day before. A reason for this may reside in the fact that an employee is also dependent on other factors that determine their job

autonomy, which might play a more important role when measuring job autonomy at a momentary level. For instance, when an employee experiences less momentary job autonomy because he/she is in a mandatory meeting, this reduction in momentary job autonomy is less likely to be caused by the sleep quality of the previous night. Therefore, measuring job autonomy at a momentary level might not reveal the same results as a measurement of job autonomy during a longer period.

To summarize, this research has found that momentary job demands lead to more momentary work engagement. An explanation for this finding may be that momentary job demands can be perceived as job challenges. Furthermore, it was found that self-perceived sleep quality and recovery play a role in the achievement of momentary work engagement when faced with momentary job demands. This finding was expected, as sleep quality was expected to enable the experience of momentary work engagement.

The observed variables of sleep quality were found to impact the relation between momentary job demands and momentary work engagement less strongly than self-perceived aspects of sleep quality.

Thus, it may be so that experiencing good sleep quality is more important than the actual observed sleep quality itself. Moreover, momentary job autonomy was also found to play an important role in achieving momentary work engagement. When momentary job demands are high, momentary job autonomy further stimulates the experience of momentary work engagement among employees. Lastly, sleep quality of the employee was not found to strongly affect the relation between momentary job autonomy and momentary work engagement. An explanation for this finding is that momentary job autonomy may not fluctuate as much as other job resources and may thus not be strongly influenced by the sleep quality of the day before.

Limitations and Directions for Future Research

This study has certain limitations that should be considered when interpreting the findings. Due to the use of self-report measures, common method variance might be a concern, as the measurement method might influence the results. Common method variance is “variance that is attributable to the

measurement method rather than to the constructs the measures represent” (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003, pp. 879). For example, when measures from the dependent and independent variable are obtained from the same participant at the same moment in time, common method variance might become a problem. The common method variance could then create a false internal consistency, which could create a seeming correlation among variables that are generated by their common source (Chang, Van Witteloostuijn, & Eden, 2010). A remedy for this could be by asking the questions that measure the independent and dependent variables at different moments in time.

Another limitation of the study may reside in the interpretation of the construct variable work engagement. Since that variable consisted out of separate items (i.e., vigor and dedication) and the effects of the items were also interpreted separately, it was difficult to assess the overall effect of the predictor variables on work engagement. Therefore, it is uncertain whether the effects of the

explanatory variables on work engagement were really measured in this study. Future studies might therefore consider using different measures of work engagement, for example by using items derived from the UWES instead of the OLBI. Moreover, the variables job demands and job autonomy were measured using only one item. This may make it more difficult to assess the interpretation of the question by the participant and to check whether questions were answered consistently. Due to a one item measure, its internal consistency is unknown. For future studies, it would thus be recommended to use multiple items to measure one variable, because it is then more certain that they are measured correctly. However, this would result in more questions that would need to be answered several times a day, making the study less accessible to participate in. Furthermore, the use of the OLBI may be a weakness of this study. The OLBI is most widely used to measure burnout, but can be altered to measure work engagement as well, as work engagement is seen as the opposite of burnout (Demerouti et al., 2003). Using the UWES may have also been suitable in this study, because it is the more commonly used instrument for measuring work engagement (Schaufeli et al., 2002). Therefore, a future study could consider using items from the UWES.

A strength of this study resides in the lack of clear consensus on a measure of sleep quality in the

literature. Therefore, this research focused on five measures of sleep quality, which were both subjective and objective in nature. This choice was made based on the available data and on literature that

described the importance of both objective and subjective experiences of sleep quality (Buysse,

Reynolds, Monk, Berman & Kupfer, 1989; Crivello et al., 2019; Krystal & Edinger, 2008; Pilcher, Ginter, &

Sadowsky, 1997). Thus, this research has attempted to investigate which role different measures of sleep quality may play in the experience of work engagement, and thus has attempted to make a contribution to the literature. However, more extensive research on how to measure sleep quality on a day-to-day basis could shed more light on the best measurement of sleep quality.

Furthermore, the timing of the research may also have affected its results, as the measurements were performed during autumn, which may have caused different experiences of participants than during other seasons. Moreover, this research used the data measured on workdays and thus not only measurements made during work hours. This is a strength of the research, because it has been found that work engagement does not lose its influence when the working day is over (Bakker, Petrou, &

Tsaousis, 2012). However, the measurements made during the weekend were not taken into account in the current research, even though the weekend may also have had an impact on the experience of work engagement. Furthermore, participants may have also worked during the weekend. Therefore, it could be an interesting direction for future research to look into this and assess the impact of leaving out the data of the weekend. Also, it could be interesting to contrast data from the weekend and the workweek.

However, this was beyond the scope of the current study.

A final limitation of this research might be the study’s time period. Measurements were only made for a period of seven days and it could be interesting to investigate whether longer measurements per participant might have led to different results. By adopting a momentary approach, this may offer interesting insights on how momentary job autonomy, momentary job demands and daily sleep quality relate to momentary work engagement. For instance, a momentary approach may reveal how

momentary job demands may lead to episodes of momentary work engagement. However, it is unknown if and how these findings might be generalized to other timeframes. Therefore, longer time frames, such as studies with longitudinal designs, might for instance be important to shed light on the role of

momentary job autonomy, as no momentary effect was found in this study.

Practical implications and conclusions

Notwithstanding the limitations that were mentioned previously, there are some potentially important practical implications of this thesis. This research has found that momentary job demands positively related to dedication. Short-term job demands may thus be perceived as a challenge, instead of a strain.

To achieve this experience, it might be beneficial for employees to experience their job demands as short-term. This experience of job demands as a challenge might be stimulated by assigning tasks to employees that take a shorter timeframe to complete. This might prevent them from experiencing long-term job demands and may thus lead to more work engagement (Bakker, 2014). Furthermore, reducing the number of hindrance demands, such as administrative squabbles, role conflicts and politics, and increasing the number of challenged demands should aid the experience of work engagement (Crawford et al., 2010).

The research has found several other interesting effects, which may have practical implications. Firstly, a positive direct effect between perceived sleep quality and perceived recovery on dedication and vigor was found. Furthermore, it was found that as the total sleep time of a participant decreased, this also led to a decrease in dedication. These results suggest that higher levels of sleep quality can positively affect momentary work engagement during the (next) workday. To achieve higher self-perceived recovery, it might be important to experience a mental detachment from work and to ensure good sleep (Sonnentag et al., 2012). Secondly, the current study has found support for the enhancing effect of self-perceived sleep quality and self-perceived recovery on the relation between momentary job demands and dedication. This finding indicates that self-perceived sleep quality and self-perceived recovery might be important for stimulating dedication, which is an aspect that is expected to be important for achieving a state of work engagement. Since the results also show an effect for the self-perceived variables of sleep quality, intervening with the self-perceived variables is expected to be valuable. Therefore, it might be crucial for an employee to monitor his/her self-perceived sleep quality and self-perceived recovery and to intervene where necessary. An example of an intervention could be for the employee to go to bed earlier or work less hours, so that the employee has more time to recover. An employer could intervene by stimulating the self-perceived recovery by not sending any emails or other disturbances outside work

The research has found several other interesting effects, which may have practical implications. Firstly, a positive direct effect between perceived sleep quality and perceived recovery on dedication and vigor was found. Furthermore, it was found that as the total sleep time of a participant decreased, this also led to a decrease in dedication. These results suggest that higher levels of sleep quality can positively affect momentary work engagement during the (next) workday. To achieve higher self-perceived recovery, it might be important to experience a mental detachment from work and to ensure good sleep (Sonnentag et al., 2012). Secondly, the current study has found support for the enhancing effect of self-perceived sleep quality and self-perceived recovery on the relation between momentary job demands and dedication. This finding indicates that self-perceived sleep quality and self-perceived recovery might be important for stimulating dedication, which is an aspect that is expected to be important for achieving a state of work engagement. Since the results also show an effect for the self-perceived variables of sleep quality, intervening with the self-perceived variables is expected to be valuable. Therefore, it might be crucial for an employee to monitor his/her self-perceived sleep quality and self-perceived recovery and to intervene where necessary. An example of an intervention could be for the employee to go to bed earlier or work less hours, so that the employee has more time to recover. An employer could intervene by stimulating the self-perceived recovery by not sending any emails or other disturbances outside work