A new statistical model for the Day Reconstruction Method

Hong Lee, Andy C.Y. Tse, Ka Yiu Lee

Research output: Journal article publicationJournal articleAcademic researchpeer-review

5 Citations (Scopus)


The Day Reconstruction Method (DRM) is a method to measure one's subjective affective status by soliciting information in a questionnaire about the previous day's activities. We developed a new model to examine the association of daily activities, the friendliness of interacting partners, and time-of-day on net affect scores among 10,377 adults participating in the World Health Organization's Study on global ageing and adult health (SAGE). A multilevel regression was fitted and the time-of-day effect was modeled by restricted cubic spline. The net affect score was a serpentine curve; stable from 4 a.m. to 6 a.m., increased from 7 a.m. to 12 noon, and became stable onwards. Participants had the highest net affect scores during religious activities (0.48, 95% confidence interval [CI]: 0.44, 0.53), and they enjoyed leisure activities, exercising, and household responsibilities more than work. Compared with events that lacked interacting partners, activities with very friendly interacting partners were associated with higher net affect scores (0.21, 95% CI: 0.19, 0.22), but events with slightly friendly interacting partners, slightly irritating or very irritating partners had lower net affect scores. To conclude, researchers using DRM for assessing well-being status across time should include the type of activities and the friendliness of the interacting partners.
Original languageEnglish
Article numbere1547
JournalInternational Journal of Methods in Psychiatric Research
Issue number4
Publication statusPublished - 1 Dec 2017


  • data analysis
  • non-linear
  • questionnaire
  • well-being
  • World Health Organization

ASJC Scopus subject areas

  • Psychiatry and Mental health


Dive into the research topics of 'A new statistical model for the Day Reconstruction Method'. Together they form a unique fingerprint.

Cite this