PM Proofreading
Level 41, Vista Tower, Intermark,
Jln Tun Razak, 50400 KL
Work Hours
Daily: 10AM - 7PM
From Excel to Narrative: Best LLM Prompts for Presenting Thesis Results & Graphs
ByDr. Sara
Table of Contents show
You have spent months distributing surveys, running experiments, and compiling data. Your Excel spreadsheets are full, and your SPSS output viewer is packed with tables. However, translating those raw numbers into a cohesive and academically sound Chapter 4 (Results) is often where postgraduate researchers freeze.
Staring at a blank page with a scatter plot in hand can be overwhelming. Many students attempt to copy-paste their entire data set into an LLM like ChatGPT, Claude, or Gemini, and hoping that the AI will instantly write the chapter. This usually results in academic disaster. The AI might hallucinate p-values, inject subjective opinions, or write in a robotic tone that guarantees a high Turnitin AI score.
To successfully write your Results chapter using AI, you must use precise Academic Prompt Engineering. You must act as the lead researcher directing a junior data analyst. This guide provides the exact prompts you need to translate your charts, graphs, and statistical outputs into a flawless and IPS-compliant narrative.
1. The Golden Rule of Chapter 4: Objectivity
Before prompting the AI, you must understand the primary function of a Results chapter: Objective Reporting.
Your goal here is to state what you found, not why you found it. Speculation, interpretation, and comparison to previous literature belong strictly in Chapter 5 (Discussion). The AI does not know this automatically. It will naturally try to “help” by explaining the data. You must forcefully constrain it.
Additionally, results should almost always be written in the past tense, as you are describing completed actions and finalized data.
2. Prompts for Quantitative Data (SPSS/Excel)
When dealing with inferential statistics (like t-tests, ANOVAs, or regressions), precision is non-negotiable. Missing a degree of freedom or misstating a significance level can lead to immediate rejection by your examiners.
Use these prompts to force the LLM to format your raw software output into flawless APA 7th style prose.
The Descriptive Statistics Synthesizer
“Act as an academic statistician. I will paste the raw descriptive statistics from my SPSS output below. Write a concise, 2-paragraph summary of this data for my Results chapter. Focus only on the central tendencies (Mean, Median) and dispersion (Standard Deviation). Write in the past tense and maintain an objective, formal tone. Do not attempt to interpret what these numbers mean for the broader research topic. Raw data: [Paste Data]”
The Inferential Statistics Reporter
“I have conducted an independent samples t-test. The raw output is pasted below. Write a single, highly technical paragraph reporting these results in strict APA 7th Edition format. You must explicitly state the t-value, degrees of freedom, and exact p-value in parentheses. State whether the null hypothesis was rejected or retained. Crucial Constraint: Do not hallucinate any numbers or variables. Only use the exact figures provided in the text below. Raw data: [Paste Data]”
3. Prompts for Qualitative Data (NVivo/Transcripts)
Qualitative results chapters are that are entirely different beasts. You are not reporting p-values; you are reporting human experiences, recurring themes, as well as narrative patterns.
The challenge here is using direct participant quotes verbatim into your academic prose without the chapter looking like a messy interview transcript.
The Thematic Quote Weaver
“I am writing the Results section for my qualitative thesis. The current theme I am addressing is: [Insert Theme, e.g., ‘Resistance to Remote Work Technologies’]. Here are three direct quotes from my participants: [Paste Quotes]. Write a cohesive, 250-word paragraph that introduces the theme, summarizes the participants’ consensus, and smoothly integrates short snippets of these quotes as supporting evidence. Do not alter the participants’ words. Use the past tense to describe what the participants expressed.”
The Pattern Identifier
“Review the following summarized responses from my focus group regarding [Topic]. Write an objective paragraph detailing the recurring points of agreement and the primary points of disagreement among the participants. Do not analyze why they disagreed; merely report the divergence in opinion clearly and formally.”
4. Prompts for Chart & Graph Descriptions
Tables present the exact values, while graphs and charts illustrate trends. When you insert a figure into your thesis, you must provide a narrative description in the text above it.
You should never expect the reader to interpret the chart themselves. The AI can help you write “at-a-glance” summaries that guide the examiner’s eye to the most important data point.
The Visual Trend Describer
“I have created a line graph showing [Variable A] over a period of [Timeframe]. The data shows a steady increase from [Starting Point] to [Peak Point], followed by a sharp decline. Write a clear, 3-sentence descriptive paragraph introducing this figure to the reader. Focus only on the overarching trend and the most significant peak/valley. Do not list every single data point. End the paragraph with: ‘These trends are illustrated in Figure X below.'”
5. Humanizing the Output for Turnitin
Turnitin’s 2026 AI detection models are exceptionally aggressive when scanning technical writing. Because statistical reporting is already highly formulaic, AI-generated results chapters frequently trigger false positives or high AI similarity scores.
If the LLM uses robotic transitions like “It is interesting to note that…” or “The data underscores the pivotal nature of…”, your examiner will be suspicious.
To protect your manuscript:
- Vary Sentence Length: Break up the perfectly symmetrical blocks of text the AI generates.
- Remove Fluff: Erase all adverbs and empty transition phrases. Technical writing must be dense and direct.
- Cross-Check Every Number: Never trust the AI to carry over a decimal point correctly. Verify every single stat against your original Excel or SPSS file.
If your data is sound but your narrative feels stiff or robotic, send your Chapter 4 to our editors at PM Proofreading. Our Academic Editing and AI Humanizing Services will refine the text, lower the Turnitin AI score, and ensure your tables and text comply perfectly with your university’s IPS formatting guidelines.
Further Reading: The Researcher’s Prompt Library: Tested LLM Prompts for Thesis Writing & Journal Publishing
