How to Fix 20% Similarity on CNKI? | Targeted Rewriting for High-Similarity Paragraphs

What this page helps you do first

• Characteristics and common distribution of 20% similarity
• Rewriting strategies for high-similarity paragraphs
• Core principle of reducing similarity without losing quality

What level is 20% similarity

Most universities require undergraduate similarity below 20%-30%, masters below 15%-20%, and doctoral below 10%-15%. 20% is generally safe for undergraduate but may be borderline for masters.

First confirm your degree level and specific school requirements, then set similarity reduction goals.

Typical distribution of 20% similarity

• Literature review typically contributes 30%-40% of similarity
• Research methods description is the second major contributor
• Conclusion and discussion sections are relatively controllable
• Self-written parts (introduction, data analysis) usually have lowest similarity

Core approach to literature review similarity reduction

Literature review similarity mainly comes from two sources: extensive paraphrasing of existing research and overly close general expressions.

Rewriting is not translation or word replacement, but reorganizing literature through your own argument logic and expressing understanding in your own words.

Practical rewriting techniques

• Change "this paper argues" type paraphrases to "existing research shows" summary-style expressions
• Merge consecutive short sentences to increase information density
• Reorganize literature order from your research perspective
• Add your own understanding and evaluation after paraphrasing

Parts that must be protected during reduction

• Research data and experimental results sections should not be modified
• Original viewpoints and discoveries must be retained
• Necessary citations should not be deleted; maintain proper citation format
• Technical terms and definitions should not be arbitrarily replaced

Frequently asked questions

Will the school directly reject a paper with 20% similarity?

Most schools allow one retest. However, if it exceeds the school threshold (such as 30% for undergraduate, 20% for masters), failure in retest may affect thesis defense.

Can translation method be used for similarity reduction?

Translation has some effect on similarity reduction, but carries higher risk. Translated Chinese may be ungrammatical and may create new similarity problems. Recommend as supplementary method, not primary.

Will similarity reduction affect academic quality?

Possibly, if rewriting is improper it may damage argument logic. Safe approach: first understand each paragraph argument task, then rewrite sentence structure and expression while maintaining logic.