見出し画像

論文指導 3 Literature Reviewに向けて(1)

前回まででリサーチプロポーザルを書く練習をした。3000語くらいの論文形式で提出して、博士論文執筆の承認を獲得する。基本的な論文執筆の概念とその意味、そしてChatGPTと対話して論文を添削してもらうことである程度のものにはなっていると思う。ここまで出来たら、次の段階に入る。それはLiterature Reviewである。現在の学問を実践して研究をするためにまず要求されるのがこのLiterture Reviewを書く能力である。英米式の博士論文を書いて博士号を獲得して帰国した研究者であれば例外なくこの訓練を受けている。日本の博士号の仕組みは学術博士が導入されてから本来はこの訓練をきちんとするべきなのだが、その前の慣行があり、徹底されていない。ドイツもフランスもPhDの博士号を出していない限り、博士と言っても修士と大差ないことが多い。

Literature Reviewという名前も誤解を読む。もちろん文学を読むことではない。本や論文を読む。だがそのサマリーを羅列してもだめである。自分の研究テーマに関していままでどのような研究がおこなわれて、どのような書籍や論文が発表されているのか。そして、ここからが大事なのだが、自分の研究は過去の様々な論文をレビューして分析した結果、誰も手をつけていないことがわかった、と論述を展開して、従ってここを研究トピックあるいはテーマとして研究することで学問の発達に貢献したい、と述べる。これがLiterature Reviewである。以降訳語として文献レビューを使っていく。

博士論文を書くにはこの文献レビューが書けなくてはいけない。このやり方はなかなか文章では伝わらない。実際指導教授との面談での論文指導のなかで学んでいくものである。論文の書き方の大枠もかなり独学で身につけるのは難しいが、文献レビューの執筆はほぼ独学では不可能なくらいハードルが高い。この文献レビューを書くプロセスをChatGPTへのプロンプトエンジニアリングで再現して、文献レビューを完成させたい。

Okude: explain literature review in detail

まずは文献レビューとは何かを聞いてみた。ChatGPTに質問することで、これからの議論は文献レビューについての議論だよ、とコンテキストを与えるのだ。

以下が答えであるが、非常に上手に説明しているので、英文のあとで、日本語訳も載せておく。
 
ChatGPT
A literature review is a comprehensive analysis and synthesis of existing research and knowledge on a specific topic. It involves the critical evaluation of relevant published work, including books, articles, theses, and other scholarly materials, to provide a coherent understanding of the subject. The primary goal of a literature review is to identify, assess, and summarize the most significant findings, theories, and arguments related to a research question or problem.

筆者コメント:上記の説明はコンパクトで非常にうまい。まさにこの通りであり、問題はそれを以下に実行するか、である。ChatGPT は次のように続ける。
 
The literature review process can be divided into several steps:
 
Define the research question or problem:
Clearly identify the research question or problem you intend to address. This will guide the scope of the literature review and help determine the relevant materials to include.
 
Conduct a literature search:
Search for relevant published works, using databases, search engines, reference lists, and other resources. The search should be systematic, comprehensive, and guided by keywords and phrases related to your research question.
 
Evaluate and select sources:
Assess the quality, relevance, and credibility of the sources you find. Prioritize recent publications, peer-reviewed articles, and works from respected authors or institutions. Make sure the sources cover different perspectives and aspects of the research question.
 
Organize and categorize the literature:
Group the sources by themes, methodologies, theories, or other relevant criteria. This will help you identify patterns, gaps, and contradictions in the literature.
 
Analyze and synthesize the literature:
Critically evaluate the selected sources, and summarize their main arguments, findings, and contributions. Compare and contrast the different perspectives, identify areas of consensus and disagreement, and note any gaps or limitations in the existing research.
 
Write the literature review:
Structure the review with a clear introduction, body, and conclusion. The introduction should present the research question or problem, and explain the purpose and scope of the review. The body should be organized thematically, discussing the main findings and debates in the literature. The conclusion should summarize the key insights, highlight any gaps or limitations, and suggest directions for future research.
 
Cite and reference sources:
Properly cite and reference all the sources used in the review, following a consistent citation style (e.g., APA, MLA, or Chicago).
 
A well-conducted literature review provides a solid foundation for your research by highlighting the current state of knowledge on a topic, identifying gaps and limitations, and suggesting areas where further research is needed. It demonstrates your familiarity with the relevant literature and your ability to critically evaluate and synthesize complex information.

以上であるが、翻訳を載せておく。ChatGPTにお願いしたが、専門的な言葉は今後の議論にも関係するので筆者の方で英語を並記、あるいは英語のままにした。これらの言葉は次回に説明する。
(翻訳)
文献レビュー(literature review)は、特定のトピックに関する既存の研究と知識の包括的なanalysis と synthesisです。それは、主題に関する綿密な理解を提供するために、本、記事、論文、およびその他の学術資料を含む関連する公開資料のcritical evaluationを含みます。文献レビューの主な目的は、研究の問題または問題に関連する最も重要な調査結果、理論、および論争を特定し、評価し、要約することです。文献レビューのプロセスは、いくつかのステップに分けることができます。

文献検索を行う:
データベース、検索エンジン、参考文献リスト、およびその他のリソースを使用して、関連する公開作品を検索します。検索は、研究問題に関連するキーワードやフレーズによって案内されるように、systematicでcomprehensiveである必要があります。

資料を評価し、選択する:
見つけた資料の品質、関連性、および信頼性を評価します。最近の出版物、査読済みの記事、および尊敬される著者または機関からの作品を優先してください。資料が研究の問題の異なる視点と側面をカバーしていることを確認してください。

文献を整理し、カテゴリーに分ける
テーマ、方法論、理論、またはその他の関連する基準によって資料をグループ化します。これにより、文献のパターン、ギャップ、および矛盾を特定できます。

文献を分析し、統合する(analysis と synthesis)
選択した資料を批判的に評価し、その主要な論点、調査結果、および貢献を要約します。異なる視点を比較・対照し、合意と不一致の分野を特定し、既存の研究のギャップや制限を注意してください。

文献レビューを書く:
明確な導入部、本文、および結論でレビューを構成します。導入部では、研究の問題または問題を提示し、レビューの目的と範囲を説明します。本文は、テーマ別に整理され、文献の主要な調査結果と論争を議論する必要があります。結論では、主要な洞察を要約し、ギャップや制限を強調し、今後の研究の方向性を提案します。

引用および参考文献のソース:
レビューで使用されるすべてのソースを、一貫した引用スタイル(APA、MLA、またはシカゴなど)に従って適切に引用および参照します。

適切に実施された文献レビューは、トピックに関する現在の知識の状態を強調し、ギャップと制限を特定し、さらなる研究が必要な分野を示唆することで、研究のための堅固な基盤を提供します。それは、関連する文献に対するあなたの熟知度と、複雑な情報を批判的に評価し、統合する能力を示します。

大体文献レビューの概要はわかったと思う。次に参考文献を聞いてみた。

Okude: tell me the best or most recognized book on the literature review
 
ChatGPT
One of the most recognized and highly recommended books on conducting a literature review is:
 
Title: "The Literature Review: A Step-by-Step Guide for Students"
Authors: Diana Ridley
Publisher: SAGE Publications Ltd.
Publication Date: 2008 (1st edition), with updated editions available


 
This book is aimed at guiding students through the literature review process by providing a clear, step-by-step approach. It covers all aspects of the literature review, from defining the research question and conducting a literature search to organizing, analyzing, and synthesizing the literature. The book also includes practical tips and examples from a wide range of disciplines, making it a valuable resource for students across various academic fields.
 
Keep in mind that there are other excellent books on literature reviews as well, but Ridley's book is widely recognized for its accessibility and practical guidance.

  CjataGPT4で翻訳してみよう。

この本は、明確で段階的なアプローチを提供することにより、学生を文献レビュープロセスを通して導くことを目的としています。研究問題の定義や文献検索の実施から、文献の整理、分析、統合まで、文献レビューのすべての側面をカバーしています。また、さまざまな学術分野からの実用的なヒントや例を含んでいるため、多岐にわたる学術分野の学生にとって貴重なリソースとなっています。リドリーの本はそのアクセシビリティと実践的な指導で広く認知されていますが、文献レビューに関する他の優れた本もあることに注意してください。

とある。アマゾンのサイトで見てみたがべたぼめだ。

This second edition of Diana Ridley′s bestselling book provides a step-by-step guide to conducting a literature search and literature review, using cases and examples throughout to demonstrate best practice. Ridley outlines practical strategies for conducting a systematic search of the available literature, reading and note taking and writing up your literature review as part of an undergraduate research project, Masters dissertation or PhD thesis.

New to this edition are:

  • Examples drawn from a wide range of disciplines

  • A new chapter on conducting systematic reviews

  • Increased guidance on evaluating the quality of online sources and online literature

  • Enhanced guidance in dealing with copyright and permissions issues.

This book also comes with a companion website containing a wide range of examples of successful literature reviews from various academic disciplines.

翻訳:
この第2版は、ダイアナ・リドリーのベストセラー本で、文献検索と文献レビューを行うための段階的なガイドを提供し、実践的な事例と例を使って最善の方法を示しています。リドリーは、利用可能な文献の体系的な検索、読書とノートの取り方、学部の研究プロジェクトや修士論文、博士論文の一部として文献レビューを書くための実用的な戦略を概説しています。

この新版には以下の内容が追加されています:

さまざまな学問分野からの例

システマティックレビューを行うための新しい章

オンラインソースとオンライン文献の品質を評価するための強化された指針

著作権と許可に関する問題に対処するための強化されたガイダンス。

これは良さそうだ。もう少し知りたいので質問した。
 
Okude: what are other books ?
 
ChatGPT
Here are a few other well-regarded books on conducting literature reviews:
 
Title: "Doing a Literature Review: Releasing the Research Imagination"
Author: Chris Hart
Publisher: SAGE Publications Ltd.
Publication Date: 1998, with updated editions available


 
Title: "Systematic Approaches to a Successful Literature Review"
Authors: Andrew Booth, Anthea Sutton, and Diana Papaioannou
Publisher: SAGE Publications Ltd.
Publication Date: 2011, with updated editions available

 Title: "The Literature Review: Six Steps to Success"Authors: Lawrence A. Machi and Brenda T. McEvoy Publisher: CorwinPublication Date: 2008, with updated editions available Title: "Synthesizing Research: A Guide for Literature Reviews"Author: Harris M. CooperPublisher: SAGE Publications Inc.Publication Date: 1989, with updated editions available 



These books offer different perspectives and approaches to conducting a literature review, catering to different disciplines and research contexts. It is important to choose a book that aligns with your specific research needs and academic field. Additionally, consulting multiple sources can provide a broader understanding and help you develop a more comprehensive and well-informed literature review. 

  翻訳:
これらの本は、異なる分野や研究の文脈に対応する文献レビューの実施方法に関する異なる視点とアプローチを提供しています。特定の研究ニーズと学術分野に合わせた本を選ぶことが重要です。さらに、複数の情報源を参照することで、より広い理解が得られ、より包括的で十分に情報に基づいた文献レビューを作成するのに役立ちます。

とある。ちなみに筆者は"The Literature Review: Six Steps to Success"Authors: Lawrence A. Machi and Brenda T. McEvoy
を出版されたときから指導に使っている。トゥールミン型の論述をつかって文献レビューを行っていく本だ。

いろいろな文献レビューの書き方の本に目をとおして、考え、自分なりの文献レビューを書いていくことが大切なのだ。

Okude:I like doing an trial of writing a literature review for my PhD where should I start? 

では書き始めてみたい。どうすればいいかを質問してみた。

ChatGPT: Starting a literature review for your PhD can be an exciting and challenging endeavor. Here are some steps to help guide you through the process: Define your research question or problem: Begin by clearly identifying the research question or problem that you want to address in your PhD. This will shape the focus of your literature review and determine the type of sources you'll need to include. Consult your supervisor or advisor: Discuss your research question or problem, and your plans for the literature review with your supervisor or advisor. They can provide valuable guidance, suggestions, and feedback to ensure your review is on the right track. Conduct a literature search: Search for relevant published works using academic databases, search engines, reference lists, and other resources. Be systematic and comprehensive in your search, using keywords and phrases related to your research question. Make a list of potential sources to include in your review. Evaluate and select sources: Assess the quality, relevance, and credibility of the sources you find. Prioritize recent publications, peer-reviewed articles, and works from respected authors or institutions. Make sure the sources cover different perspectives and aspects of the research question. Organize and categorize the literature: Group the sources by themes, methodologies, theories, or other relevant criteria. This will help you identify patterns, gaps, and contradictions in the literature. Analyze and synthesize the literature: Critically evaluate the selected sources, and summarize their main arguments, findings, and contributions. Compare and contrast the different perspectives, identify areas of consensus and disagreement, and note any gaps or limitations in the existing research. Write the literature review: Structure the review with a clear introduction, body, and conclusion. The introduction should present the research question or problem, and explain the purpose and scope of the review. The body should be organized thematically, discussing the main findings and debates in the literature. The conclusion should summarize the key insights, highlight any gaps or limitations, and suggest directions for future research. Revise and edit: Review your literature review for clarity, coherence, and consistency. Check for grammatical errors, citation mistakes, and any other issues that may detract from the quality of your review. Share your draft with your supervisor or advisor: Seek feedback from your supervisor or advisor on your literature review. They can provide suggestions for improvement, identify potential issues, and help ensure that your review aligns with your overall PhD research. Remember that a literature review is an iterative process, and you may need to revise and update it as your research progresses. Be prepared to invest time and effort in conducting a comprehensive, well-informed, and critical review that provides a solid foundation for your PhD research. 

翻訳を載せておく。

ChatGPT
PhDの文献レビューを始めることは、興味深く、やりがいのある試みです。以下は、そのプロセスを案内するためのいくつかのステップです:
研究問題を定義する:
PhDで取り組む研究問題を明確に特定し、それを文献レビューの焦点として、含めるべき情報源のタイプを決定します。
指導教員やアドバイザーに相談する:
研究問題や文献レビューの計画について指導教員やアドバイザーと話し合います。彼らは貴重なアドバイスや提案、フィードバックを提供して、レビューが正しい方向に進んでいることを確認できます。
文献検索を行う:
学術データベース、検索エンジン、参考文献リスト、その他のリソースを使って、関連する出版物を探します。研究問題に関連するキーワードやフレーズを使って、検索が体系的で包括的になるようにします。レビューに含める可能性のある情報源のリストを作成します。
情報源を評価し、選択する:
見つけた情報源の質、関連性、信頼性を評価します。最近の出版物、査読済みの論文、尊敬される著者や機関からの作品を優先します。情報源が研究問題の異なる視点や側面をカバーしていることを確認します。文献を整理し、分類する:情報源をテーマ、方法論、理論などの関連する基準に基づいてグループ化します。これにより、文献の中のパターン、ギャップ、矛盾を特定できます。文献を分析し、統合する:選択した情報源を批判的に評価し、その主要な議論、調査結果、貢献を要約します。異なる視点を比較対照し、合意と不一致の領域を特定し、既存の研究のギャップや制限を指摘します。
文献レビューを書く:
明確な序論、本文、結論でレビューを構成します。序論では、研究問題を提示し、レビューの目的と範囲を説明します。本文は、テーマ別に整理され、文献における主要な調査結果や議論を取り上げます。結論では、重要な洞察をまとめ、ギャップや制限を強調し、今後の研究の方向性を示唆します。
校正・編集を行う:
文献レビューの明瞭さ、一貫性、統一性を確認します。文法の誤り、引用の誤り、レビューの質を損なうその他の問題をチェックします。
指導教員やアドバイザーに下書きを共有する:
文献レビューについて指導教員やアドバイザーからフィードバックを求めます。彼らは改善の提案や潜在的な問題の特定を行い、レビューがPhD研究全体と整合していることを確認するのに役立ちます。文献レビューは繰り返しのプロセスであり、研究が進むにつれて改訂や更新が必要になることを覚えておいてください。総合的で、十分に情報に基づいた、批判的なレビューを行うために、時間と労力を投資する準備が整っていることが重要です。これにより、PhD研究のための確固たる基盤が提供されます。

さて始めよう!!


この記事が気に入ったらサポートをしてみませんか?