This research studies whether the moment of occurrence of a task or contextual behaviour with a low performance produces a primacy or recency effect and whether it causes changes in performance appraisal. We also analyzed whether the nature of assessment questionnaire items affects raters’ assessments and how the sequence of questionnaire presentation and completion may do so. Participants were 146 undergraduate students. We used a design with two inter-subject variables (questionnaire presentation and performance sequence) and one within-subject variable (global versus specific questionnaires). Findings show that if a low performance is presented at the beginning of the assessment period, the performance assessment will be more negative. Also, results indicate that task performance appraisals and contextual behaviour assessments are higher and less accurate when performed with a questionnaire that includes global items.