Survey data underlying the publication: Crafting through financially demanding times

Method Participants and Procedures Data were collected among Dutch entrepreneurs who own private companies and who employ less than 250 employees (cf. Jayasekara et al., 2020). After completing the informed consent, respondents filled out a general questionnaire, followed by a weekly questionnaire sent out on Thursday afternoon with a reminder e-mail on Monday morning for four consecutive weeks. Due to its enormous societal impact, a so-called ‘COVID-19 lockdown period’ (October 2020 till February 2021) was chosen to collect data because more entrepreneurs experienced financial demands like income loss (cf. Torrès et al., 2022). During this lockdown period, the Dutch government implemented restrictions concerning traveling (i.e., not allowed to travel unless absolutely necessary) and meeting other people (i.e., not allowed to meet in groups and mandatory self-isolation if a person tested positive for the COVID-19 virus). Confidentiality and anonymity of responses were emphasized and assured, and participation was voluntary. The Ethics Review Board of the university approved the study. A total of N = 91 entrepreneurs signed up for the study and received a weekly invitation to participate in the dairy study. Of these N = 91, N = 69 entrepreneurs finished one or more of the weekly surveys (75.8%), resulting in N = 189 weekly inputs (M = 3.0 per participant). The sample included 32 men (46.4 %) and 37 women (53.6%). Their mean age was 54 years (SD = 11.38). The average age of their business was 13.9 years (SD = 11.42), and most participants were active in business services (31.7%), the culture and leisure sector (16.4%), and health and social work (11.6%). Weekly Survey For all constructs, we used shortened scales to minimize the time needed to fill out the weekly surveys (cf. Ohly et al., 2010). All responses were given on 5-point Likert scales ranging from 1 ((almost) never) to 5 ((almost) always) referring to the current week. All variables were adapted to the week-level, in Dutch, and rewritten to apply to entrepreneurs. Weekly income loss was measured by asking the entrepreneurs to compare their current weekly income to a workweek before the Covid-19 pandemic. We asked the participants to rate their income loss on a scale of -100 (complete loss of income) to +100 (complete gain of income). Weeks with income gain (rating > 0) were excluded in this study. This led to the exclusion of N = 6 entrepreneurs and N = 40 weeks. Weekly business crafting was measured with adapted job crafting scales by Demerouti and Peeters (2018) and Petrou et al., (2012). In the process of scale adaptations, we interviewed (N = 5) and consulted (N = 4) entrepreneurs from different industries to make sure the items were written in the “language” of entrepreneurs. We selected three items for each business crafting strategy (Table 1). For increasing resources (e.g., “I have tried to learn new things for my business”), Cronbach’s α ranged from .76 to .86 (M = .81). For increasing challenging demands (e.g., “I have tried new approaches”), Cronbach’s α ranged from .80 to .87 (M = .82). For optimizing demands (e.g., “I look for ways to do my work more efficiently”), Cronbach’s α ranged from .89 to .96 (M = .93). Weekly leisure crafting was measured with the four-item scale of Petrou and Bakker (2016) (e.g., “I have tried to build relationships through leisure activities.”). Cronbach’s α ranged from .83 to .90 (M = .86).Weekly well-being was distinguished in motivation and fatigue severity and measured using shortened subscales of the Checklist Individual Strength (CIS; Bültmann et al., 2000). We used four items to measure motivation (e.g., “I was full of plans”). Cronbach’s α ranged from .71 to .87 (M = .81). Another four items were used to measure fatigue severity (e.g., “I felt tired”), Cronbach’s α ranged from .91 to .94 (M = .93). Weekly goal attainment was measured with three items from Grebner et al. (2010) (e.g., “I completed my tasks”). Cronbach’s α ranged from .84 to .89 (M = .87). Strategy of Analysis We expected differences within entrepreneurs because some weeks may be more demanding than others. Therefore, we used multilevel analyses to test all hypotheses. Within-person analyses compare the differences within one individual (e.g., differences in weekly behavior). Between-person analyses compare the differences between different individuals (e.g., differences between the behavior of different entrepreneurs). We analyzed the data using lmer in RStudio. In multilevel analysis, the intraclass correlation (ICC) decomposes the variance in two components (variance at the within-person level and at the between-person level). There were medium amounts of variance to be explained by between-person variations (between .50-.75; Hox & Maas, 2006), justifying the multilevel approach. The week-level predictor variables were centered around the person-mean for the within-person analyses. For the between-level analyses, the variables were centered around the grand mean.