Geriatric health promotion scale (GHPS): The authors own
design of a health promotion scale for seniors was applied. An
18-item version of the GHPS was established, and explained
63.8% of the total variance. A 4-point Likert scale was used to
score the frequency of any behavior performed, which covered
“never”, “sometimes”, “usually” and “always”. The higher scores
indicated more health promoting behavior. The questions
concerning elderly health promoting behavior were based on
those found in a literature review. A content validity index
(CVI) was computed using the proportion of experts agreeing
on item relevance. The CVI was 0.89 after evaluation by 5
nursing faculties in the field of elderly care. The internal
consistency of reliability was measured by Cronbach's α (=
0.86), and the construct validity was established using factor
analysis. The simplified version of the GHPS was made up of 4
dimensions of behavior: Self-protection (7 items, e.g. wearing
non-slip shoes, getting enough sleep, taking three meals
regularly), community participation (5 items, e.g. participating
in township- or village-held activities), health responsibility (3
items, e.g. measuring blood sugar, blood pressure), and healthy
diet (3 items, e.g. minimum intake of one and a half bowls of
vegetable, 5 groups or balanced diet).
error, before the project was conducted, the research assistants
were separated into 5 pairs to interview an elder in the community
activity center, and a 90% correct response rate of inter-rater
reliability among the 5 pairs was confirmed.