比較日誌與直接詢問法對勞力時間的估計:以已婚花農婦女為例
A Comparison of Labor Time Estimates between Diaries and Direct Questions: An Example of Married Women in the Flower Production Industry
作者:杜素豪(Su-Hao Tu) | 首次發表於 2020-07-01 | 第 33 期 December 2004
DOI:https://dx.doi.org/10.6786/TJS.200412_(33).0005
研究紀要(Research Notes)
論文資訊 | Article information
摘要 Abstract
在西方社會學及經濟學分析中,比較日誌與直接詢問法對勞力時間的估計已有二十年歷史,這類研究卻未見於台灣。本文以已婚花農婦女為例,分別比較時間日誌與兩種直接詢問法,在估計無酬勞力時間的差異。此比較著重在確定時間日誌法之有效性,直接詢問法的高估情形,與高估現象的可能解釋。面訪調查資料輔以深入訪談質化資料的分析結果顯示,時間日誌法適合於估計已婚花農婦女的農務與家務工作時間。工作時間的高估與婦女所回答的工作時間長短息息相關。儘管在時間高估的社會心理解釋上仍分歧,且研究對象侷限在特殊產業的農家婦女,本文的研究架構與分析方法卻適切的提供台灣社會科學領域中時間測量方法研究的重要參考,尤其是時間運用、生活型態、兩性無酬勞力分工等相關研究。對未來研究的建議包括修改社會讚許性的測量方法,重建對高估時間的模式分析,更多時問日誌的信度與效度的研究,多元方法綜合運用於時間測量方法的比較分析,鼓勵對其他自營職業婦女與其他學科領域從事類似的時間運用研究,以及從事具代表性樣本與追蹤樣本的反覆檢驗。

關鍵字:時間日誌、直接詢問、社會讚許效應、時間壓力感效應、迴歸平均效應 
In the West, time estimates between direct questions and time diaries have long been debated in sociological and economic analyses of household labor and market labor; no similar studies have been conducted in Taiwan. The author interviewed self-employed married women in the flower production industry to examine differences in estimates of unpaid labor time between time diaries and direct questions. Survey and ethnographic data were used to validate the time diaries by investigating differences in diary estimates between weekdays and weekends, overestimates from direct questions, and possible explanations for overestimates. The results generally support the idea that time-diaries-keeping is an appropriate means of measuring labor time. The participants who reported long work hours tended to exaggerate their time spent on farm and household tasks. Given the divergent findings on social-psychological effects in a small sample of farm women, the results demonstrate a good frame of reference for subsequent methodological studies, particularly those related to lifestyle, time-use, and gender divisions in markets and household labor. The study limitations point to a need for revising measures of social desirability, refining analytical models of overestimates, conducting more research on time diary reliability and validity, making comparative analyses of multiple methods for time data collection, and replicating past studies using a representative national sample and longitudinal design.

Keywords: time diaries, stylized direct questions, social desirability, time-pressure perceptions, regression to the mean