兒童與青少年長期追蹤研究之未回應趨勢及其社會人口學因素

目標：兒童與青少年行為之長期發展研究（簡稱CABLE計畫）因面臨研究樣本隨教育階段改變（國小、國中、高中、大學）而調整資料收集方法，本研究目的為描述兒童與青少年長期追蹤未回應之趨勢及原因，並分析其相關社會人口學因素。方法：CABLE計畫自2001年起針對台北及新竹國小一年級（世代一）和四年級學童（世代二）進行追蹤，以獲有父母同意書之學童為研究樣本（分別為2,852、2,663人）。未回應係指每年追蹤調查的應施測名單中未能回收問卷者。以世代分層Group-based Trajectory Model探討未回應趨勢及相關社會人口學因素。結果：經軌跡分析，世代一、二均呈現四個未回應軌跡：持續回應（世代一、二之百分比分別為73.4%、76.0%）、後期未回應（11.2%、5.4%）、漸增持續未回應（9.3%、8.2%）、及早期未回應後期回應（6.1%、9.7%）。相對於持續回應者，居住於台北、父母教育為高中及以下、父母婚姻為非結婚者，分別有較高之機會屬於不同狀況之未回應軌跡。結論：以CABLE持續追蹤資料進行推論時宜處理未回應之社會人口學差異。

關鍵字

長期追蹤；未回應趨勢；群體軌跡分析

並列摘要

Objectives: The Child and Adolescent Behaviors in Long-term Evolution (abbreviated as CABLE) project had to change data collection methods for these subjects as they advanced to higher level schools including elementary school, junior high school, senior high school, and college or university. The purpose of this study was to describe the trajectories of non-response during the 9 year follow-up and to analyze the socio-demographic factors related to those trajectories. Methods: CABLE commenced in 2001 and subjects were followed every year. They were 1st and 4th grade students (sample sizes were 2853 and 2663 respectively) with parental consent in Taipei City and Hsin-Chu County. Non-response was defined as not responding to a questionnaire every year. We used the Group-based Trajectory Model to find non-response trajectories and related factors as stratified by cohorts. Results: Both cohorts showed four trajectories: continuing response (percentages in cohort 1 and cohort 2 were 73.4% and 76.0%, respectively), late non-response (11.2% and 5.4%), increasing non-response (9.3% and 8.2%), and early non-response but late response (6.1% and 9.7%). With continuing response as the reference group, those who lived in Taipei City, those whose parental education was lower than senior high school, and those whose parents were not married were more likely to be non-responsive. Conclusions: Using CABLE long term data to make implication should consider these socio-demographic differences with non-response trajectories.

並列關鍵字

Long term follow-up study ； Non-response trajectory ； Group-based trajectory model

參考文獻

Thygesen, L.C.,Johansen, C.,Keiding, N.,Giovannucci, E.,Gronbaek, M.(2008).Effects of sample attrition in a longitudinal study of the association between alcohol intake and all-cause mortality.Addiction.103,1149-59.

Google Scholar

Goldberg, M.,Chastang, J.F.,Zins, M.,Niedhammer, I.,Leclerc, A.(2006).Health problems were the strongest predictors of attrition during follow-up of the GAZEL cohort.J Clin Epidemiol.59,1213-21.

Google Scholar

Schneider, K.L.,Clark, M.A.,Rakowski, W.,Lapane, K.L.(2012).Evaluating the impact of non-response bias in the Behavioral Risk Factor Surveillance System (BRFSS).J Epidemiol Community Health.66,290-5.

Google Scholar

Frojd, S.A.,Kaltiala-Heino, R.,Marttunen, M.J.(2011).Does problem behaviour affect attrition from a cohort study on adolescent mental health?.Eur J Public Health.21,306-10.