Not Every Day is Monday for Employees Confined due to COVID-19: Anticipatory Happiness Matters

Introduction

As in other countries, the government in Spain declared a national state of alarm due to the situation created by COVID-19, and the population in this country suffered mobility restrictions and strict lockdown from March 15 to June 21, 2020. Citizens were forced to stay home and avoid family gatherings and being physical present at work, only essential activities outside the home being allowed. Investigating the experiences of people who faced this type of confinement could help in future crises, such as additional COVID-19 outbreaks and new epidemics (see Kuntz, 2021).

In this context, workers coped with important challenges during strict COVID19 confinement, such as remote work and risk of contagion at the workplace while performing essential tasks. Variability in their emotions throughout the work week could be affected as well. The anticipation of pleasure or happiness associated with leisure activities plays a relevant role in this type of dynamics (Seibel et al., 2020). It is reasonable to argue that typically excitement increases as the weekend (and expectations of pleasant activities) approaches. However, the routine of strict lockdown, where opportunities for leisure and pleasant weekend activities are limited, may have eroded the variability in anticipatory happiness throughout the week (current happiness but considering the rest of the week), also influencing in daily job satisfaction and affect in the workplace.

Our study concentrates on two types of employees who continued their work activity during the aforementioned confinement in Spain. The first group consisted of employees with home-based telework. The second group consisted of employees who performed essential activities and had to continue their work following strict health and safety measures, returning home after the workday. Our contribution is twofold. First, we examine the daily change pattern in anticipatory happiness in employees confined due to COVID-19. The consideration of temporal dynamics is necessary in order to build an organizational theory and accurately capture phenomena in the workplace (see George & Jones, 2000), including happiness at the workplace (Warr, 2013). Research has primarily focused on causal relationships, even in longitudinal studies, whereas time itself has been relatively neglected (see Roe, 2008). We investigate whether there is variability in anticipatory happiness throughout the week, despite confinement, through repetitive cycles or rhythms. Second, we test whether daily variations in both job satisfaction (daily fluctuations in the extent to which employees like their job) and job affect (daily fluctuations in the cumulative mood associated with the workday) in confined employees would depend in part on daily changes in general anticipatory happiness based on the proximity or distance of the weekend. Although some indirect evidence has been found (Nicholson & Griffin 2017; Sonnentag et al., 2008), this spill-over effect linking fluctuations in anticipatory happiness to daily job experiences has not yet been examined. Behind the possible appearance of monotony or emotional stability in confined employees, there may be significant anticipatory happiness variability associated with the distance or proximity of the weekend, contributing to daily fluctuations in work experiences. To achieve our goals, we conducted two quantitative diary studies. Using questionnaires, confined employees rated the study variables each day.

Workplace Happiness and Time

Workplace happiness refers to an affective experience that describes subjective well-being in the workplace domain (Warr, 2020). Therefore, it is a psychological experience rather than a financial, physical, or social view of well-being. Regarding time, it is generally assumed that workplace happiness is associated with events that are in the present and/or the past. Two influential theoretical approaches illustrate this temporal focus. First, the “adaptation process” has to do with fluctuations over time based on the way employees adapt to changes. Changes in the workplace can increase (or decrease) happiness, but after a temporary period, employees tend to return to their original level of happiness (called the “equilibrium state”) once they have adapted to the new situation (Boswell et al., 2005; Griffin, 1991). The longitudinal design in these studies made it possible to examine the evolution of happiness at the workplace from the past to the present. Second, Warr (2020) emphasized employees’ thoughts and interpretations underlying “emotion regulation” in understanding happiness at the workplace. Emotion regulation refers to “processes by which individuals influence which emotions they have, when they have them, and how they experience and express those emotions” (Gross, 1998, p. 275). Typically, emotion regulation is applied to events that have already happened, such as changes in a workplace situation to increase positive emotional experiences and reduce negative ones and recall of positive past events, among others (da Costa, 2014; Warr, 2020).

Although research on past and present events played an important role and significantly contributed to knowledge about happiness in the workplace, the anticipation of the future is also relevant. In addition, this orientation toward the future may have particular characteristics in employees confined due to COVID-19 because lockdowns changed the way most employees usually organized their job (e.g., Kumar et al., 2021; Shkoler et al., 2021; Wang et al., 2021). Unlike natural cycles in our lives (e.g., seasonal variations), the seven-day cycle is an artificial rhythm established in most countries (Tsai, 2019; Zerubavel, 1985). It helps to differentiate the weekday (Monday to Friday), when time is mainly dedicated to work, from the weekend (Saturday and Sunday), when people spend more time on leisure, family, and consumption. Typical weekend activities were quite limited during the COVID-19 lockdown, blurring the boundaries between weekdays and the weekend. This situation could have eliminated the excitement that is normally felt as the weekend approaches. In fact, some authorities have used the expression “every day is Monday” (20 minutos, 2020) to communicate the importance of confinement for citizens and the need to withstand the routine and avoid many pleasurable weekend activities.

People are happier on weekends than on weekdays (Stone et al., 2012), which is attributable to the fact that, compared to work, weekend activities provide higher autonomy and relatedness (Ryan et al., 2010). However, the affective experience associated with the weekend is not restricted to events that take place on Saturday and Sunday. One of the ubiquitous characteristics of humans is their capacity for mental time travel to the future, projecting themselves into pre-live events (Suddendorf & Corballis, 1997). This capacity has probably been a critical aspect of evolution and survival (Kabadayi & Osvath, 2017; Suddendorf & Corballis, 2007), and it also includes affective travels where humans are able to anticipate how future events will make them feel (e.g., Van Dijk et al., 2012). Although research on affective travel has focused mainly on forecasting – predictions of affective reactions to future events – (e.g., Michel et al., 2016), some scholars have proposed the existence of a complementary phenomenon called “savouring” (Loewenstein, 1987), anticipatory emotion (Loewenstein et al., 2001), or pleasant anticipation (Seibel et al., 2020), where hedonic anticipation of future events influences the current affective experience.

Anticipatory happiness throughout the week is especially relevant for understanding employees’ experiences during the week for two reasons. First, the repetitive seven-day cycle helps employees to anticipate (on workdays) the future positive weekend experience because people more accurately forecast events that they have experienced before and that will happen soon (Wilson & Gilbert, 2003). Second, and based on how proximity-distance from future events affects current experiences (Trope & Liberman, 2003), this anticipatory happiness could “colour” the way employees experience their workdays. For instance, the excitement caused by the proximity of the weekend could “infect” affective work experiences and allows employees to view their current job more positively.

The study of affective travel and anticipatory happiness in the workplace is still very limited. Dane and George (2014) made a conceptual effort to incorporate affective travel into the study of organizational behavior. Sonnentag et al. (2008) considered “weekend anticipation” as the anticipation of enjoying a nice weekend, but only as a control variable that was measured on Friday (their main goal was to analyze the relationships among work engagement, psychological detachment, and affect, measured twice a week). More recently, conducting an hourly online survey throughout one working day, Seibel et al. (2020) confirmed a positive relationship between pleasant anticipation of a planned leisure activity and thoughts about a planned leisure activity in the workplace.

The scarcity of research on anticipatory happiness or pleasure is surprising given the ubiquity of the future orientation as a human characteristic. Additionally, its consideration is especially relevant in lockdown situations. It allows us to observe to what extent confined employees lose the happiness variability associated with the anticipation of the weekend (“every day is Monday”) or, on the contrary, are capable of continuing to carry out weekend activities at home that allow them to maintain variability in anticipatory happiness and its possible impact throughout the work week (“not every day is Monday”).

Distinguishing Anticipatory Happiness from Life Happiness and Anticipated Happiness

Anticipatory happiness could be confounded with other related concepts, especially life happiness and anticipated (or forecasted) happiness. Life happiness refers to a global assessment of all aspects of a person’s life. Although it can focus on a specific life domain, this construct typically consists of an integrated judgment (Diener, 1984; Kahneman et al., 2006). Regarding time, life happiness covers past and present experiences (e.g., having lived wonderful family events) as well as future expectations (e.g., anticipating a satisfactory retirement) (Seibel et al., 2020). Nevertheless, happiness could refer to a more concrete and distinguishable experience associated with a specific moment in time, usually in the past (how a person felt yesterday or today) (Kahneman & Deaton 2010), making it possible to study the variability in happiness over time (Kahneman et al., 2004; Veenhoven, 2021). Anticipatory happiness also has a specific time referent, but it is in the future. Unlike the general nature of life happiness, anticipatory happiness is a current emotion that is not produced by past or present events, but specifically by events anticipated in the future (Loewenstein et al., 2001). In addition, anticipatory happiness requires an identifiable “target” event in the future (Seibel et al., 2020). In the present study, the target is the “rest of the week”. Therefore, participants had to declare their current happiness level, but taking the rest of the week into account.

Anticipatory happiness is also different from anticipated happiness, although both constructs share affective travel into the future. Anticipated happiness focuses on the human capacity to predict how the person will feel in a future event (Michel et al., 2016; Van Dijk et al., 2012), whereas anticipatory happiness emphasizes current experiences based on the anticipation of the future. The differentiation between these two approaches to human affective experiences was proposed by Loewenstein et al. (2001). In the case of anticipatory happiness, when people think about the future, they are not only able to predict how they will feel, but they also begin to “savour” that future event in the present moment. For example, a person who thinks of a great future vacation.

This conceptual particularity was especially relevant in understanding how employees experienced the work week during the lockdowns due to COVID-19. Employees who underwent confinement had more routine lives, and every day of the work week could be identical, reducing the affective variability associated with anticipation (greater distance or proximity) of the weekend. However, it is reasonable to imagine that, even in this situation, people engage in pleasant activities at home on weekends that break up the monotony and create variability in anticipatory happiness on workdays.

Daily Job Satisfaction and Daily Job Affect: Past and Future Events

Job satisfaction is a key attitude defined as the “evaluative judgment one makes about one’s job or job situation” (Weiss, 2002, p. 175). Although this attitude is relatively stable over time, research has concluded that it can also vary considerably over time (Bowling et al., 2005; Loi et al., 2009). Regarding job affect, research is usually based on the distinction between trait (long-term predispositions to feeling good or bad) and state (temporary fluctuations in moods) affect (Watson & Tellegen, 1985). Focusing on variability over time (states), scholars have confirmed, as in the case of job satisfaction, the existence of significant fluctuations in job affect over time (e.g., Ilies & Judge, 2002; Kwon et al., 2019). The time referent in the present study is the “day”. Therefore, daily job satisfaction and daily job affect focus on day-to-day changes.

Research has usually attributed this daily variability to the circumstances employees face each day (Ilies & Judge, 2002), which is congruent with the affective events theory. This is a general theory that describes fluctuations in affective states (Weiss & Cropanzano, 1996) based on workplace stimuli (events) (see Daus et al., 2020). Traditionally, scholars have concentrated on how employees react to events already experienced in the workplace. However, as mentioned above, humans are oriented toward the future, and their current experiences are also based on the anticipation of future events (Seibel et al., 2020).

Variability in Anticipatory Happiness

We expect to find variability in employees’ anticipatory happiness despite the COVID-19 lockdown. Although many traditional weekend activities were impossible, and the monotony throughout the days of the week could be greater than before the confinement, the proximity or distance from the weekend should allow them to anticipate an affective response that would vary during the work week. It is reasonable to argue that employees anticipate that the weekend involves breaking with the work discipline, engaging in pleasant individual and/or family activities at home, and having remote contact (electronic) with other family members and friends. Regarding the change pattern, Trope and Liberman (2003) proposed that proximity-distance from future events influences the way people mentally represent these events. According to the affect-dependent time-discounting hypothesis, temporal proximity increases the importance of affective experiences associated with the expected future event (Trope & Liberman, 2003). Similarly, Loewenstein (1987) proposed that the intensity of anticipatory present affective experiences increases as the future event approaches, accelerating right before the event. Transferring this argument to a normal work week (from Monday to Friday), current happiness – but considering the rest of the week – would follow a curvilinear (quadratic) pattern. Anticipatory happiness during the week would increase as workdays go by, and it would accelerate on Friday, right before the weekend.

Hypothesis 1: Time has a curvilinear relationship with anticipatory happiness during the work week, beyond the linear relationship. Specifically, anticipatory happiness accelerates as the weekend approaches.

Linking Anticipatory Happiness to Daily Job Satisfaction and Daily Job Affect

As mentioned above, employees’ experiences in the workplace are usually associated with events experienced in the past. Constructs such as daily job satisfaction and daily job affect are usually connected to the events an individual has experienced throughout the day or that have happened previously. However, employees’ current experiences in the workplace also depend on the future, and one relevant factor is their current happiness but considering hedonic anticipation (anticipatory happiness). Time is inextricably linked to current human experiences, not only in the past, but also in a future that, although it does not yet exist, is characterized in the present by expectations, possibilities, hopes, and dreams (Cernas-Ortiz et al., 2018; George & Jones, 2000; James, 1952; Mead, 1934). There are future events that are especially relevant to employees. The anticipation of the weekend is important for employees who have a normal weekly work schedule from Monday to Friday. Additionally, in a unique crisis situation, such as the one experienced by employees confined due to COVID-19, happiness stemming from the anticipation of the weekend (despite the restrictions) could be a positive stimulus. The daily variability in anticipatory happiness, as a consequence of the distance or proximity of the weekend, should affect employees’ daily experiences at work (job satisfaction and job positive affect) as a cross-domain spill-over mechanism from nonwork to work domains. This mechanism is implicit in some arguments from previous studies (Nicholson & Griffin 2017; Sonnentag et al., 2008), but there is a lack of empirical evidence linking variability in anticipatory happiness during the week to daily experiences in the workplace. To do so, we considered it important to control for daily emotional exhaustion as a central indicator of stress experienced by the employee throughout the day (Robbins et al., 2012). In fact, previous efforts have related stress at work to both job satisfaction (Eatough et al., 2016) and affect (Hillhouse et al., 2000). Based on these arguments, we propose the following hypothesis:

Hypothesis 2: Daily changes (from Monday to Friday) in anticipatory happiness during the week (current happiness but considering the rest of the week) are positively related to daily changes in job satisfaction and job positive affect, after controlling for daily emotional exhaustion.

Data Analysis

In both studies, all the analyses were computed using SPSS 24. To model the change in anticipatory happiness during the week, and to find out whether daily changes in anticipatory happiness over time were related to changes in daily job satisfaction and job positive affect, we computed hierarchical linear models (Heck et al., 2013) using growth modelling (Duncan et al., 2013). Specifically, we tested the nested models, considering the data structure: Level 1 (occasions) and Level 2 (individuals). As a complementary analysis to capture specific differences in anticipatory happiness on workdays, we computed one-way ANOVAs with repeated measures. Furthermore, we tested whether the average values between the study variables were similar for women vs. men by computing t-tests for independent samples. In addition, we estimated Pearson correlations separately by gender, and we computed a hypothesis contrast test using the Fisher r to z transformation to analyze whether the correlations between the study variables were similar in men and women.

Diary Study 1

Method

Procedure and Participants

Our Diary Study 1 focused on employees with home-based telework during the aforementioned national confinement in Spain. They reported on the assessed variables on five consecutive workdays (from Monday to Friday). Data collection was carried out during the strict confinement period in Spain (March 14-June 21, 2020). It began on March 30, 2020 to avoid the possible influence of the impact of the initial confinement on employees. To be eligible, employees had to telework at home in Spain during the aforementioned COVID-19 lockdown. They also had to have a weekly work schedule from Monday to Friday and do not work on weekends. Researchers used social networks to recruit employees. Due to confinement, communication with participants and data collection were carried out remotely. Participants received an email every day at 2.30 (pm) with a reminder that they had to answer the questionnaire at the end of the workday, from Monday to Friday. To ensure anonymity and group each individual’s five questionnaires together, participants created an individual code that was written on the questionnaire each day. The research team identified and invited a total of 86 employees to participate, all of whom fulfilled the eligibility criteria. Of them, 71 employees (82.55 %, 43.14 years old on average, SD = 9.23, 59.2% women) agreed to participate and answer the questionnaire daily on five consecutive workdays. By contrast, 15 employees (17.45 %) were excluded from the analyses because they declined to participate or because they did not answer the questionnaires over time according to the procedure despite having received reminders. This sampling plan resulted in 355 observations in all (71 employees x 5 days). They were working in a variety of sectors: public administration, computer programming, administrative support in health care services, banking services, insurance, teaching, and professional services as freelancers. The Ethical Committee on Research in Humans at the University of the corresponding author evaluated and approved the design and procedures. Researchers informed participants about the study objectives and participants gave their informed consent regarding the procedure. Anonymity and confidentiality were guaranteed. Participants were free to leave the study at any time.

Measures

We followed the recommendations for diary studies, where maintaining participation every day for a period of time is a great challenge. This was especially difficult during the COVID-19 lockdown, given the special circumstances. Bearing this in mind, the instrument was a combination of valid short forms and single items (Ohly et al., 2010), allowing participants to answer the questions in less than five minutes each day (Reis & Gable, 2000). We used single-item measures for daily job satisfaction, daily emotional exhaustion, and anticipatory happiness. One of the reasons for this decision is the need to have an agile assessment that would allow us to collect data on a daily basis in a difficult situation for employees given the strict COVID-19 confinement. As Diamantopoulos et al. (2012) pointed out, these difficulties in data collection justify the use of single-item measures. In addition, statistical information on using single-item scales is provided below.

Regarding “positive job affect” associated with each workday, we used the validated character-based pictorial scale for reporting moods, which is available for research purposes (Desmet et al., 2016). This measure has very positive aspects for diary studies. It requires little effort from participants to report their moods, focuses on universal human moods, allowing broad applicability, can be adapted to the research question, and reflects cumulative and combined affect, such as the affect corresponding to a workday. In addition, measures based on characters have been used to assess variability in affective experiences (Hilbolling et al., 2012). We used the subset of 4 positive moods to create our positive affect measure based on cumulative workday moods (“How did you feel at work TODAY?”): excited, cheerful, relaxed, and calm. Participants responded on a 7-point Likert-type scale ranging from 1 = not at all to 7 = very. Reliability was good for every workday, with alpha coefficients ranging from .75 to .82.

To measure “daily job satisfaction”, we used a single-item with the typical format to assess this variable (e.g., Zablah et al., 2016): “TODAY I have felt satisfied at work” (1 = strongly disagree to 7 = strongly agree). Research shows that single-item measures of overall job satisfaction have good validity (Nagy, 2002; Wanous et al., 1997). These authors found significant correlations (.60-.72) between single-item measures and multiple-item measures. In addition, single-item measures may be better to measure changes in job satisfaction (Wanous et al., 1997), which is quite useful given the dynamic nature of our study.

To control for “daily emotional exhaustion”, we used a central single item from the revised Maslach Burnout Inventory (Schaufeli et al., 2002), focusing on emotional exhaustion symptoms: “TODAY I have felt emotionally drained” (1 = strongly disagree to 7 = strongly agree). Evidence indicates that these types of single-item measures of emotional exhaustion provide meaningful information at work (West et al., 2009). These authors found statistical correlations (.76-.83) between single-item measures and multiple-item measures. In addition, the single-item measure was useful to stratify participants at risk of experiencing high burnout.

We also controlled for “sex” and “age”. Men and women in the workplace could experience different roles that affect their daily mood states (Williams et al., 1991), and this could be accentuated during the COVID-19 confinement. We also controlled for age because differences in variables such as job satisfaction (Clark et al., 1996) and job stress (Birdi et al., 1995) have also been observed depending on the age of the person.

“Anticipatory happiness” was at the end of the questionnaire to keep participants from thinking about the rest of the week because it might influence their answers to the other questions. It was measured with a single item, allowing participants to report how happy they are now, but considering the rest of the week. Single items for happiness are commonly used in research, showing good psychometric properties (Diener, 1984; Fordyce, 1988; Wilson & Gilbert, 2003). In general, single-item measures have strong correlations with multiple-item scales, with scores above .80 (Gilbert et al., 1998), thus supporting the use of single items to assess happiness. Our measure was designed by adapting the procedure by Seibel et al. (2020) to assess pleasant anticipation related to future leisure plans. Accordingly, there were two steps. First, participants were asked to think about the future “target”, but instead of thinking about a future leisure plan our study asked participants to think about the rest of the week. Second, they had to report their current happiness, but thinking about the future. More specifically, every day we asked participants to think about the rest of the week in order to report their current happiness: “The week has seven days (from Monday to Sunday). Think about the rest of this week. With this in mind, indicate how happy you feel now” (1 = not happy to 7 = very happy).

Results

Preliminary Results

Table 1 shows the aggregated descriptive results and bivariate correlations among the variables for the total sample and differentiating between women and men. In general, in the total sample, there were positive correlations between anticipatory happiness, job positive affect, and job satisfaction. In addition, these variables were negatively correlated with daily emotional exhaustion. In the case of gender, results showed that there were significant differences between women and men in the mean average for anticipatory happiness (t = 1.97, p < .05) and emotional exhaustion (t = 3.09, p < .01). Both anticipatory happiness and emotional exhaustion were higher in men than in women. By contrast, there were no significant differences in job positive affect (t = -0.43, p > .05) and job satisfaction (t = -1.04, p > .05). In general, there were no significant differences between women and men in the correlations among the variables. The only exception was the correlation between job satisfaction and emotional exhaustion (z = 2.32, p < .05), which was higher for women than for men.

In the relationship between daily anticipatory happiness, on the one hand, and daily job satisfaction and daily job positive affect, on the other, we assumed the existence of fluctuations in the individual scores over time. Therefore, it was necessary to confirm within-person variance in our variables. Table 2 shows that within-person variance over time was statistically significant for all variables, ranging from 58% (job positive affect) to 70% (emotional exhaustion), justifying the study of the links from daily changes in anticipatory happiness to daily changes in job satisfaction and job positive affect.

Testing the Change Pattern and Mean Differences in Anticipatory Happiness

Our findings supported the curvilinear (quadratic) pattern of change. There was a significant curvilinear relationship between time and anticipatory happiness (γ₂₀ = .07, p < .01), after controlling for the linear relationship. As expected (H1), this result confirmed the acceleration in the increase in anticipatory happiness right before the weekend (see Figure 1).

One-way ANOVAs with repeated measures confirmed statistically significant differences among mean values in anticipatory happiness, F(4, 280) = 14.78, p < .01, η² = .17, across the workdays (from Monday to Friday). Bonferroni pairwise comparisons indicated that anticipatory happiness average scores remained stable from Monday to Wednesday (p > .05), but they began to increase on Thursday (p = .02) and reached the highest value on Friday (p = .03).

Testing the Relationship between Changes in Daily Anticipatory Happiness and Changes in Daily Experiences at Work

Growth modelling findings supported H2. Daily changes in anticipatory happiness were positively and significantly associated with daily changes in both job positive affect (γ50 = .28, p < .01) and job satisfaction (γ50 = .33, p < .01), after controlling for daily emotional exhaustion, sex, and age (Table 3).

Diary Study 2: Replication

Diary Study 2 was designed to replicate the findings observed in Diary Study 1 but implementing three relevant modifications. First, we measured our variables over the course of two weeks. This allowed us to extend the time frame and more accurately examine the existence of a repetitive cycle or rhythm, which was only implicit in Diary Study 1. Therefore, a cubic change pattern is expected where anticipatory happiness decreases on the second Monday and then accelerates its increase again with the arrival of the weekend. Second, we considered employees who performed an essential activity in Spain during the confinement due to COVID-19. Therefore, they had to go to the workplace, follow health and safety measures, and return home after the workday. Third, we added a measure of daily job negative affect. This made it possible to check whether daily changes in anticipatory happiness are also related to changes in daily negative affect. This is especially relevant for employees who were forced to go to the workplace in a crisis situation created by COVID-19 (e.g., Kotera et al., 2021) and with the possibility of contagion. Therefore, our hypotheses for Diary Study 2 are as follows:

Hypothesis 3: Time has a cubic relationship with anticipatory happiness over the course of two weeks, beyond the linear and curvilinear relationships. Specifically, a repetitive weekly cycle or rhythm is expected where anticipatory happiness decreases on Monday and then accelerates its increase with the approach of the weekend.

Hypothesis 4: Daily changes in anticipatory happiness over two work weeks are positively related to daily changes in job satisfaction and job positive affect, and negatively related to daily changes in job negative affect, after controlling for daily emotional exhaustion.

Method

Procedure and Participants

Data collection was carried out during the strict confinement period in Spain (March 14-June 21, 2020). The process began in April 2020 to avoid the possible influence of the impact of the initial confinement on employees. The research team selected a company that was carrying out an essential activity during a very strict COVID-19 confinement in Spain: packaging and selling fruits and vegetables. Because this was considered as essential during the COVID-19 lockdown, employees continued to go to work at their company’s facilities but following strict health and safety measures. In addition, these employees also had a weekly work schedule from Monday to Friday during the data collection. We contacted the Human Resources Manager of the company to explain the project and ask permission to collect diary data for two consecutive weeks. We invited the 90 employees who worked at this company every day during the confinement to participate. Of them, 83 (92.22 %, 46.95 years old on average, SD = 9.91, 71.1% women) agreed and answered the questionnaire each day. They had to answer at the end of each workday. One of the researchers had permission (following the required health and safety measures) to go to the company every day to control the process and collect the questionnaires. Again, each participant created an individual code and write it on the questionnaire each day. This process allowed us to ensure anonymity and group each participant’s questionnaires. On ten consecutive workdays (two cycles of 5 days, from Monday to Friday) employees answered all the questions. This resulted in 830 observations in all (83 employees x 10 days). They were also invited to report on their anticipatory happiness on the two weekends, on both Saturday and Sunday. This resulted in 332 observations (83 employees x 4 days). They received a reminder at 7 (pm) on Saturday and Sunday using an anonymized WhatsApp group. These questionnaires were collected on Monday at the company’s facilities. The member of the research team who was responsible for the data collection asked the participants on Monday for confirmation that they had answered the questionnaires on Saturday and Sunday towards the end of the day. The participants confirmed that they had answered after 7(pm). All the participants answered a paper questionnaire every day of the study. The Ethical Committee on Research in Humans at the University of the corresponding author evaluated and approved the design and procedures. Again, researchers informed participants about the study objectives, and participants gave their informed consent regarding the procedure. Anonymity and confidentiality were guaranteed. Participants were free to leave the study at any time.

Measures

We used the same measures of “daily job negative affect” as in Diary Study 1, but adding the subset of 4 negative moods (tens, irritated, sad, and bored) from the validated character-based pictorial scale (Desmet et al., 2016). This allowed us to create the daily job negative affect measure (“How did you feel at work TODAY?”) with a 7-point Likert-type scale ranging from 1 = not at all to 7 = very. Reliability was good for both positive and negative job affect measures in Diary Study 2, with alpha coefficients – on the 10 consecutive workdays – ranging from .70 to .85. As in Study 1, anticipatory happiness was at the end of the questionnaire to keep participants from thinking about the rest of the week because it might influence their answers to the other questions.

Results

Preliminary Results

Table 4 shows the aggregated descriptive results and bivariate correlations among the variables for the total sample and differentiating between men and women. In the total sample, there were positive correlations between anticipatory happiness, job positive affect, and job satisfaction. There were also positive correlations between job negative affect and emotional exhaustion. Finally, two groups of variables were negatively related to each other: anticipatory happiness, job positive affect, and job satisfaction versus job negative affect and emotional exhaustion. In the case of gender, results showed that there were significant differences between women and men in the mean average on all the variables. Job negative affect (t = -10.01, p < .01) and emotional exhaustion (t = -9.25, p < .01) were higher in women than in men. By contrast, anticipatory happiness (t = 4.60, p < .01), job positive affect (t = 3.97, p < .01), and job satisfaction (t = 3.77, p < .01) were higher in men than in women. Furthermore, several correlations among the variables were statistically different in women vs. men. Four correlations were higher in women than in men: anticipatory happiness-job positive affect (z = 4.96, p < .01); job positive affect-job satisfaction (z = 5.69, p < .01); job positive affect-emotional exhaustion (z = 4.06, p < .01); and job negative affect-emotional exhaustion (z = 4.59, p < .01). By contrast, the correlation between anticipatory happiness and job negative affect (z = -2.14, p < .05) was higher in men than in women.

We also compared the mean averages in Study 1 vs. Study 2 on the variables measured in both studies. We observed statistical differences in all the variables: anticipatory happiness, M_{(Sample Study 1)} = 5.09, SD = 0.79, M_{(Sample Study 2)} = 4.84, SD = 1.18, t = 3.35, p < .01); job positive affect, M_{(Sample Study 1)} = 4.54, SD = 0.80, M_{(Sample Study 2)} = 4.12, SD = 1.02, t = 6.28, p < .01; job satisfaction, M_{(Sample Study 1)} = 5.30, SD = 0.89, M_{(Sample Study 2)} = 5.76, SD = 0.93, t = -6.91, p < .01; and emotional exhaustion, M_{(Sample Study 1)} = 2.63, SD = 1.07, M_{(Sample Study 2)} = 3.41, SD = 1.52, t = -8.31, p < .01. In general, and except for job satisfaction, employees in the second sample (Study 2), who continued to go to work at their company’s facilities, had more negative experiences during the lockdown than those who teleworked (Study 1).

Again, it was necessary to confirm within-person variance in our variables. Table 5 shows that within-person variance over time was statistically significant for all variables (ranging from 36% – job positive affect – to 72% – job satisfaction), justifying the study of the links from daily changes in anticipatory happiness to daily changes in job satisfaction and job affect.

Testing the Change Pattern and Mean Differences in Anticipatory Happiness

In Diary Study 2, we measured anticipatory happiness every day for two weeks. Thus, a cubic pattern of change is expected (H3). Our results confirmed this cubic pattern of change (see Figure 2), after controlling for the linear and curvilinear forms. This cubic pattern was confirmed for the work weeks (10 consecutive workdays from Monday to Friday) (g₃₀ = 0.01, p < .01) and for the two full weeks, including weekends (14 days, γ₃₀ = .01, p < .01). Focusing the attention on workdays (10 days), there was a repetitive cycle where anticipatory happiness remains relatively stable from Monday to Wednesday, begins to increase on Thursday, and reaches its highest average score on Friday. Anticipatory happiness drops sharply on Monday, and then the cycle begins again, describing a rhythm. When also considering the two weekends (see Figure 2, dotted lines), a repetitive cycle or rhythm also exists. The results confirmed that anticipatory happiness continues to grow on Saturday, but, not surprisingly, it begins to decrease on Sunday when the end of the weekend approaches and Monday draws near.

Two ANOVAs were also carried out, each with the data for each week (seven days, from Monday to Sunday). Our results confirmed statistically significant differences among mean values in anticipatory happiness across the different days for both the first week, F(6, 492 = 26.91, p < .01, η² = .25, and the second week, F(6, 492) = 32.80, p < .01, η² = .29. Regarding the first week, Bonferroni pairwise comparisons indicated that anticipatory happiness remained stable from Monday to Thursday (p > .05), and it showed a statistically significant increment on Friday (p < .001) but remained stable on the weekend (p > .05), decreasing again right before Monday arrived. Bonferroni pairwise comparisons for the second week indicated that anticipatory happiness remained stable from Monday to Wednesday (p > .05), then began to increase on Thursday (p = .01) and again on Friday (p = .03), achieving its highest average score on Saturday (p = .001). It dropped on Sunday when the end of the weekend was approaching (p = .01).

Testing the Relationship between Changes in Daily Anticipatory Happiness and Changes in Daily Experiences at Work

Our findings supported H4. Testing the nested models (Level 1 – occasions – and Level 2 – individuals), our results also corroborated that daily changes in anticipatory happiness were positively and significantly associated with daily changes in job positive affect (γ₅₀ = .32, p < .01) and job satisfaction (γ₅₀ = .20, p < .01), and negatively associated with daily changes in job negative affect (γ₅₀ = -.28, p < .01), after controlling for daily emotional exhaustion, sex, and age (Table 6).

Discussion

Our results were not congruent with the idea that “every day is Monday” for employees suffering from lockdowns due to COVID-19. Daily variability in anticipatory happiness is a persistent phenomenon related to the human capacity for affective travel, which exists even in very special circumstances. The speculation that confined employees experience a boring emotional life throughout the week was not supported in reality. Our data suggest that affective stability is probably not a concern because people are able to combine mobility restrictions with pleasant weekend activities at home that allow them to avoid routine, differentiate workdays from the weekend, and maintain a certain degree of excitement associated with the upcoming weekend that contributes to daily experiences in the workplace.

Theoretical Implications

One of the main contributions of the current study is the confirmation of a change pattern in the variability of anticipatory happiness during the week. Research has mainly focused on developing constructs and examining their relationships, for instance, in terms of their nomological network. Even when using longitudinal designs, researchers are usually interested in causal relationships rather than in time itself (see Roe, 2008). However, time is a relevant aspect of everyday life, including organizational behaviour (Bartunek & Woodman 2015). Regarding workplace affect, scholars urge researchers to study facets such as the duration of emotional experiences in order to have a better picture of employees’ behaviour (Dudenhöffer & Dormann, 2013; Venz et al., 2020). Our investigation answers this call, building a time-based theory of anticipatory happiness that is meaningful in crisis situations such as the one created by COVID-19. Based on the affect-dependent time-discounting hypothesis (Trope & Liberman, 2003), we corroborated the existence of a recurrent weekly rhythm where anticipatory happiness increases with the arrival of the weekend to decrease drastically on Monday, starting the cycle again.

It is also quite relevant that daily fluctuations in anticipatory happiness were able to contribute (“colour”) to daily changes in job satisfaction and job affect, beyond daily emotional exhaustion. Despite the ubiquity of affective travel and its impact on current experiences through hedonic anticipation (e.g., Loewenstein et al., 2001), its study in organizations is practically non-existent. As far as we know, only conceptual analyses (Dane & George, 2014) and implicit or indirect arguments (Nicholson & Griffin 2017; Sonnentag et al., 2008) have been considered. Therefore, empirical efforts generally assume that employees’ current experiences are related to previously experienced events, such as emotional exhaustion stemming from daily stressful demands at work. However, our findings confirmed that humans are also oriented toward the future, and this travel is related with our current job experience – through a spill-over effect. Thus, to capture the essential nature of human behaviour as a phenomenon, anticipation of the future should also be considered at any single moment in time (George & Jones, 2000). This could be especially relevant in crisis situations. When, due to the COVID-19 confinement, employees have to telework at home, or have to go to their workplace because they carry out an essential activity (even increasing the risk of contagion), the anticipation of the weekend turns into a positive stimulus to break the monotony or affective stability of the workdays during the lockdown.

Influential theoretical approaches to understanding the dynamics of happiness, satisfaction, and affect in the workplace (affective events theory, adaptation process, emotion regulation) illustrate this predominant focus on past and present events (Warr, 2020; Weiss & Cropanzano, 1996). Accordingly, current experiences depend on the daily circumstances (events) people had to face, how they adapted to changes, and how they regulated their emotions. Our findings suggest, however, that future events also matter. In addition to events already lived, employees are able to think about the future. The proximity of hedonic experiences for employees (typically the weekend) is also an identifiable “target”. In the same way that employees are able to interpret and appraise the events that are happening in their lives, they also have dreams and expectations (George & Jones, 2000; James, 1952; Mead, 1934) that affect their happiness and the way they experience their daily work.

Practical Implications

Our findings have practical implications. One of the challenges for today’s organizations and societies during strict confinement periods (and also in those without crisis) is to achieve adequate emotional states in employees who continue to work. Enjoying pleasurable weekend activities is desirable in itself, but also because it contributes to variability in anticipatory happiness during the week and to the subsequent level of excitement at work (and a lower level of negative affect) when the weekend approaches. Instead of promoting routine (“every day is Monday”), managers and policy makers can enhance the acceptance and reinforcement of weekend activities that, respecting the confinement measures, maintain the normal affective rhythm throughout the week. The employees who participated in our studies did this spontaneously. However, different types of organizational and societal authorities could promote and authorize weekend activities that facilitate appropriate and sustainable affective variability over time during the lockdown, while respecting restrictions.

One of the challenges is what to do on days when employees’ anticipatory happiness suffers, especially at the beginning of the week. To approach this issue, effective positive interventions are a promising area that can also address situations of crisis (see Salanova et al., 2013). Donaldson et al. (2019) evaluated different types of positive interventions in the workplace. At least two types of interventions were effective in improving well-being. The first type, employee strengths interventions (e.g., Meyers & Van Woerkom, 2017), typically have two steps: a) the identification of employee strengths and b) training in meaningfully incorporating these strengths in the workplace. This type of intervention can be implemented on days when anticipatory happiness typically suffers, in order to mitigate this experience by achieving a stimulating work environment that considers the employee’s strengths. The second type, employee well-being interventions, are oriented toward improving aspects such as positive emotions and engagement, for instance, by involving employees in the practice of aspects such as gratitude and visualizing their best-self (Neumeier et al., 2017). Again, this intervention can be considered on work days when anticipatory happiness can deteriorate.

Limitations and Future Research

As is true for all studies, the present one has some limitations that are insights for future research. First, our studies concentrated on confined employees who had a weekly work schedule from Monday to Friday, but other employees did not necessarily have this schedule (e.g., police, firefighters, medical staff). It is reasonable to expect that these employees also feel variability in anticipatory happiness based on the proximity-distance of their days off, which would affect their work daily experiences. However, their situation is somewhat particular (e.g., they usually cover weekend emergencies) because it is likely that a number of them cannot share weekend activities with relatives who live with them in the same house during confinement. Future studies could examine this type of situation to obtain a better picture of anticipatory happiness and its contribution to workplace experiences. Second, we used quantitative diary studies. This allowed us to examine changes in employees’ scores and their association with daily working experiences. However, conducting qualitative diary studies with self-recorded information in combination with quantitative studies could also be an adequate strategy for future studies. This type of mixed-method approach (e.g., Colombo & Landoni, 2014) could provide deeper, complementary knowledge about employees’ experiences in the workplace. Third, we measured some of the constructs (job satisfaction, emotional exhaustion, and anticipatory happiness) with single-item scales. This did not reduce the validity of our measures. In addition to special circumstances (diary studies during strict confinement due to COVID-19) that recommend the use of single-items (Diamantopoulos et al., 2012), statistical analyses support this type of scale to measure the aforementioned constructs. Nevertheless, the consideration of multiple-item measures, based on specific facets, could add information about how different aspects of the constructs behave in understanding the dynamics of anticipatory happiness and experiences at work. Fourth, exploring the impact of the aforementioned positive interventions in the workplace can provide a picture of how the shape of fluctuations in anticipatory happiness can vary depending on the actions implemented. For example, it would be possible to examine the extent to which positive interventions at the beginning of the work week mitigate the abrupt change in anticipatory happiness. Finally, we observed significant differences between employees who teleworked (Study 1) and those who continued to go to work at their company’s facilities (Study 2). Except for job satisfaction, employees in the first sample had more positive experiences during the lockdown than those who continued to go to work. This difference could be based on different factors such as the control enjoyed by employees who telework or the lower cost of working at home (e.g., commuting time). Future studies could confirm this difference and the possible mechanisms involved.

Conclusion

Despite these limitations, the current study represents an initial step in understanding daily variability in anticipatory happiness during the week (confirming the existence of repetitive cycles or rhythms) and its role in daily fluctuations in job satisfaction and job affect (as a cross-domain spill-over mechanism). Experiencing lockdowns due to pandemics is compatible with performing weekend activities that avoid monotony during the week and on workdays because employees are able to anticipate and enjoy future positive events today. Anticipatory happiness fluctuations, as a facet of human affective travels, persist despite COVID-19 confinement and contribute to daily work experiences.