To make your results section consistent, you need to build a direct, traceable path from your research questions to your data tables and then to your written analysis. Consistency isn't about making the data look perfect; it's about ensuring that a reader can find the same answer regardless of whether they are looking at a chart, a quote, or a summary paragraph. If your committee says you lack consistency, they usually mean there is a gap between what you promised to investigate in your introduction and what you actually reported in your findings.
Aligning Your Data with Your Research Questions
The most common reason for a "lack of consistency" is drift. You start with three specific research questions, but by the time you finish your analysis, you've found ten interesting things. You end up reporting the interesting things while neglecting the original questions. This creates a structural mismatch that confuses your reviewers.
Start by creating a mapping document. This isn't for your committee; it's for your sanity. List your research questions in one column and the specific table or page number that answers that question in the next. If you have a research question that doesn't have a corresponding table or a dedicated section of text, you've found your first inconsistency. You either need to find that data or admit the data didn't yield a clear answer.
Consider the case of a third-year Sociology student who spent months interviewing urban residents about gentrification. Her first research question focused on economic displacement, but her results section spent ten pages discussing cultural identity and only two pages on rent hikes. When her committee flagged "inconsistency," she realized she was writing about what she liked most about her data rather than what she promised to investigate. She fixed this by restructuring her headings to mirror her research questions exactly.
Synchronizing Tables and Narratives
A frequent mistake is treating tables as appendices that sit next to the text rather than integrated evidence. Consistency fails when your table says one thing (e.g., a p-value of .04) but your narrative says another (e.g., "the results were not statistically significant"). These contradictions are what make a committee nervous because they suggest a lack of attention to detail.
When you revise, read your results section in a "ping-pong" fashion. Read a paragraph of analysis, then immediately look at the table it references. Ask yourself if the numbers in the table actually support the claim in the sentence. If you describe a trend as "sharp" or "significant," the table should show a clear, visible jump in data. If the table shows a marginal increase, change your wording to "slight" or "moderate."
APA style requires a specific relationship between text and visuals. You should never simply say "see Table 1." Instead, you should summarize the key finding from Table 1 in the text and then refer to the table for the full data set. The text provides the meaning, and the table provides the proof. If the meaning and the proof don't match, the consistency is broken.
"I spent three days obsessing over my wording, but the problem was that my Table 3 used different category names than my Table 4. My committee didn't care about my prose; they cared that I called a group 'Low Income' in one place and 'Economically Disadvantaged' in another."
Standardizing Terminology and Formatting
Consistency also lives in the small details. If you refer to a group as "participants" on page 10, don't switch to "subjects" on page 15 or "respondents" on page 20. In sociology, these terms can imply different theoretical frameworks. Switching them mid-chapter makes it look like you're pulling from different drafts or sources without integrating them.
Check your decimal points and rounding. If you report a mean as 4.52 in one table, don't report it as 4.5 in the text. APA style is rigid about this for a reason: it prevents the reader from wondering if you've changed the data or if you're being imprecise. This is a mechanical task, but it's the kind of "polish" that signals to a committee that you are ready for your defense.
Use this framework to decide how to handle conflicting data points during your final polish:
| Scenario | The Inconsistency | The Fix |
|---|---|---|
| Table shows a trend, but text ignores it | Omission of data | Add a sentence explaining why this trend is or isn't relevant to the RQ. |
| Text claims a result, but table is missing | Unsupported claim | Create a summary table or move the data from your raw notes into the draft. |
| Different terms for the same variable | Terminology drift | Pick one term and use "Find and Replace" to standardize it throughout. |
| Contradictory findings across chapters | Narrative conflict | Acknowledge the tension in the text; explain it as a nuance of the data. |
Overcoming the "Forest for the Trees" Block
When you've been staring at 40 pages of sociology data for months, you stop seeing the errors. Your brain fills in the gaps because you know what you meant to say. This is why you feel confused even after highlighting themes. You are reading your intentions, not the actual words on the page.
To break this, try reading your results section backward. Start with the last paragraph of the chapter and work your way to the beginning. This disconnects the narrative flow and forces you to look at each paragraph as a standalone claim. When you do this, you'll notice more easily when a paragraph lacks a supporting table or when the terminology suddenly shifts.
Another strategy is to use a "dummy" reader. Find a peer who isn't in your specific niche of sociology. Give them your research questions and one of your results tables. Ask them, "Based on this table, does the answer to the research question seem clear?" If they struggle to find the answer, your presentation is inconsistent. You don't need them to edit your grammar; you need them to test the logic of your data delivery.
Common Pitfalls to Avoid
- Over-explaining in the results: Don't start interpreting why something happened in the results section. Save the "why" for the discussion. Consistency is ruined when you mix reporting (what happened) with interpretation (what it means).
- Redundant tables: If a table shows the exact same information as a previous one, delete it. Too much data creates noise, and noise leads to perceived inconsistency.
- Ignoring "Null" results: If your data didn't support your hypothesis, don't try to hide it by being vague. State clearly that the result was not significant. Being honest about a lack of a relationship is more consistent than trying to imply one that isn't there.
Your Immediate First Step
Stop reading the whole chapter. Open a blank document and create a three-column table: Research Question, Supporting Table Number, and Page Number of the Narrative Summary. Fill this out for every single research question in your dissertation. If any cell remains empty, that is exactly where your inconsistency lies. Focus your remaining time only on those gaps.