Why the Same Job Feels Different from Day to Day

The broad workplace model describes relationships across studies, but it can make employees look like stable bundles of motivation and well-being. Diary research asks a different question: how much can autonomy, competence, and relatedness change within the same person as the workday changes? This chapter follows that time-sensitive question so the reader can distinguish an employee-level difference from a day-level condition, and can see why autonomy-supportive work has to be renewed in the actual flow of work.

From average employee to changing workday

A between-person comparison might tell us that one employee reports more autonomy satisfaction than another. That is useful, but it does not tell us whether either employee feels the same way every day. A within-person comparison asks whether the first employee experiences more or less ownership, mastery, or connection on different occasions. The difference is not a statistical technicality. It changes where we look for an explanation.

Suppose Maya and Leon have similar roles and report similar average autonomy over a month. Maya’s high average could conceal a pattern of very good days and very bad days. Leon’s average could reflect a steadier, moderate experience. A manager who sees only the average may conclude that the two employees need the same intervention. A manager who looks at days can ask whether Maya’s difficult periods coincide with an unpredictable handoff, an unavailable decision-maker, or a week of conflicting priorities. The same job title does not guarantee the same temporal pattern.

The diary review examined this within-person question through quantitative diary studies of work-related basic psychological needs. The search began with 2,251 records. After duplicates and screening, 21 articles were included, representing an aggregated 2,114 employees. Fourteen studies measured needs daily, for periods ranging from three to ten days, while seven measured them weekly, over periods ranging from three to 48 weeks. The review was systematic in its search and screening, but it used a narrative synthesis because the studies were too diverse for a meta-analysis.

To estimate how much experience varied within people, the review used intraclass correlations. An intraclass correlation estimates the proportion of variance that lies between people; one minus that value gives the proportion attributed to within-person variation. In the reviewed studies, daily within-person variation ranged from 41.00% to 72.70% across autonomy, competence, relatedness, and combined need satisfaction. The median values were roughly 54.00% to 56.40%. In weekly studies, the range was 21.00% to 61.00%, with medians of about 39.00% to 47.00%.

The cautious conclusion was that need satisfaction varied both daily and weekly, and appeared more dynamic from day to day. Daily variation accounted for somewhat more of the fluctuation than between-person differences, while the reverse appeared at the weekly level. This does not mean that employees are unstable or that a single survey is worthless.