近年、コミュニケーション能力の育成を目指して四技能を総合的に学ぶことの重要性が日本の英語教育において指摘されている。大学における言語テストや授業実践でも技能統合型タスクの実施が進む一方、タスクの高度な要求に学習者が対応できるよう習熟度に応じた足場がけを検討することが急務となっている。そこで、日本人大学生を対象に講義視聴と筆記要約からなる技能統合型タスクを実施し、事前タスクとして三種類の異なる足場がけを学習者に提供した上で、文法能力に基づき習熟度別に教育効果を比較した。その結果、講義の「構成」に関する足場がけは習熟度にかかわらず適切な要約作成を促すものの、「重要語彙」や「低頻度語彙」を用いた足場がけの働きは習熟度によって異なることが示された。本稿では学習者の要約例を議論した上で、低習熟度の学習者にはトップ・ダウン型、高習熟度の学習者にはボトム・アップ型の足場がけを活用することを提案する。Aiming at fostering students' communicative skills in the target language, English education in Japan now emphasizes integrated teaching of the four skills. Skill-integrated tasks, which require students to utilize more than one of the four skills, have been implemented at a university level, in classroom tasks and standardized language tests. Students tend to struggle, however, with the high demands of such tasks, so further examinations of devising appropriate scaffoldings for the students' different proficiency levels would be anticipated. This paper focuses on one type of skill-integrated tasks: an integrated writing task that consists of listening to a lecture on global issues and writing its summary. The aim of the study is to explore whether or not the students' performance in the task is affected by different pre-task scaffoldings and/or by their proficiency levels. A total of 66 Japanese first-year students were assigned to the integrated writing task. Before completing the task, students were divided into three groups and provided with one of the following three types of pre-task scaffolding: lecture organization, keywords, or difficult words. Completed summaries were compared in terms of their length and the coverage of the content of the original lecture. Each group was also subdivided into two proficiency levels, depending on their grammatical knowledge, in order to achieve the aim of the study. The results thus indicated that providing lecture organization was most effective in enhancing the quality of the summary's content, regardless of the students' proficiency. The functions of keywords and difficult words scaffolding seemed to differ according to the students' proficiency levels. From the discussion using example summaries created by each group, we propose employing top-down scaffolding for low-proficiency students and bottom-up scaffolding for high-proficiency students.