Abstract
This study aimed to establish a method for quantitative description concerning preschool children's eating behavior.
(1) Two time-sampling techniques were compared with each other : one was carried out by observing the behavior at an interval of 15 seconds (Method A) and the other, at an interval of 1 minute (Method B). From the observation recorded by Method A on 12 children of 3 years old, 210 actions recorded were categorized into 36 subcategories and subsequently into 5 categories such as“0”, rejection of eating; “1”, play not relevant to eating; “2”, play relevant to eating; “3”, actions associated with eating and“4”, eating itself. When the result was expressed as the rate of number of actions belonging to each category to total number of actions observed for individual child as well as the sum of whole subjects, the result by Method B was comparable with that by Method A.
(2) Next, two observation periods (10 days and 20 days) were compared with each other by using Method B, 10-day observations did not necessarily yield the identical result to that by a 20-day observation in terms of the quantitative rate of categorized actions, but a good accordance between the results by two observation periods was obtained for the rank of individual child in the rate of categorized actions to the total.
From these, Method B is judged to be applicable as similar as Method A, but a 20-day observation is rather preferable than a 10-day observation.
