Research design is the master blueprint of any academic investigation. Before you upload a single document to ResearchLens, you need a clearly articulated research design that specifies what you are studying, why it matters, how you will gather evidence, and how you will interpret findings. The design choice is not stylistic — it is epistemological. It reflects your assumptions about reality, knowledge, and the legitimate ways of producing scholarly truth.
ResearchLens supports all four major research orientations used in social science, education, public health, military science, and the humanities. Each approach has rigorous standards, distinct workflows, and specific outputs that the platform is designed to produce.
Method 01
Qualitative
Explores meaning, experience, and lived reality through words, themes, and interpretation. Inductive. Constructivist roots.
Method 02
Content Analysis
Systematically codes and quantifies patterns within textual or visual communication. Can be qualitative or quantitative in nature.
Method 03
Quantitative
Tests hypotheses using numerical data, statistical inference, and objective measurement. Deductive. Post-positivist roots.
Method 04
Mixed Methods
Integrates qualitative and quantitative data within a single study to answer research questions that neither approach alone can fully address.
ResearchLens Platform Note
When you create a new analysis project, the platform asks you to select a research paradigm. This selection activates the correct coding workflows, interpretation engine, and report templates for your chosen method. Choose deliberately.
Choosing the Right Method
What is your primary research goal?
↓
Understand meaning / experience
Qualitative Research
Measure communication patterns
Content Analysis
Test hypotheses with numbers
Quantitative Research
Answer complex, multi-faceted questions
Mixed Methods
Common Mistake
Choosing your method after collecting data is called "post hoc rationalization." Your research question should determine your design — not the data you happen to have available. Define your RQ first, then select the method.
Knowledge Check: A doctoral student wants to understand how military veterans describe the psychological transition from active duty to civilian academic life. Which method is most appropriate? Select one
✅ Correct. When the research goal is to understand the lived experience — subjective meaning, personal narratives, psychological transitions — qualitative methodology (specifically phenomenology) is the appropriate design. The small, purposeful sample is characteristic of qualitative inquiry, not a weakness.
❌ Incorrect. This question asks about understanding lived meaning and personal experience — a qualitative research domain. Quantitative designs test hypotheses and measure variables numerically. The correct answer is B: qualitative phenomenological interviews.
Qualitative research is grounded in the assumption that reality is socially constructed and that meaning emerges through human interaction with the world. Rather than measuring phenomena, qualitative researchers seek to interpret them. The data are words, narratives, observations, and artifacts — not numbers.
The epistemological foundation is typically constructivism (Creswell & Poth, 2018) or interpretivism (Lincoln & Guba, 1985), which holds that the researcher and participant co-construct meaning. This means the researcher's positionality, reflexivity, and transparency are not flaws — they are methodological requirements that must be disclosed and addressed.
Core Qualitative Traditions
Phenomenology (lived experience) — Grounded Theory (theory generation from data) — Ethnography (culture and practice) — Narrative Inquiry (stories and identity) — Case Study (bounded context in depth). Each tradition has distinct data collection protocols, analytical procedures, and quality criteria.
Thematic Analysis: Step-by-Step
Braun and Clarke's (2006) thematic analysis is the most widely used qualitative analytical framework and is the primary engine behind ResearchLens's qualitative coding module.
-
1
Familiarize Yourself with the DataRead and re-read your corpus of data. Note initial ideas. In ResearchLens, this means uploading your documents and reviewing the extracted text in the context pane before coding begins. Mark initial observations as memos or annotations.
-
2
Generate Initial CodesSystematically code interesting features of the data. Codes are short labels that capture the essence of a data segment. In ResearchLens, add your codes in the "Theme Codes" input field. Each code should be grounded in the data, not pre-imposed from theory (unless you are conducting a theory-driven deductive analysis).
-
3
Search for ThemesCluster related codes into potential themes. A theme captures something important about the data in relation to your research question. Themes must be substantive — they are not simply topics but patterns of shared meaning. ResearchLens will show you code co-occurrence and frequency distributions to assist this process.
-
4
Review ThemesRefine themes against the full dataset. A theme is valid only if there is sufficient data to support it and it is clearly distinguishable from other themes. Some themes may collapse; new ones may emerge. This is iterative, not linear.
-
5
Define and Name ThemesWrite a clear definition of each theme and choose a compelling name that captures its essence. The name should convey what the theme is about to a reader who has not seen your data.
-
6
Produce the Written ReportUse ResearchLens's APA Report Generator to produce a formatted report. Qualitative reports require vivid, analytic narrative that weaves together data extracts with interpretive commentary. Avoid description for its own sake — every data quote must be analytically justified.
🧪 Interactive Simulator: Qualitative Theme Coder
Try It
Paste a text excerpt (interview or document):
Enter theme codes to search for (comma-separated):
Rigor in Qualitative Research
Qualitative rigor is established through: credibility (member checking, prolonged engagement), transferability (thick description), dependability (audit trail), and confirmability (reflexive journaling). In ResearchLens, document your coding decisions in the Notes pane as an audit trail. Do not confuse frequency with significance — a theme appearing once can be as analytically powerful as one appearing 50 times.
What is the primary criterion for establishing a valid theme in Braun and Clarke's thematic analysis? Select one
✅ Correct. Thematic validity in Braun and Clarke's framework is determined by the theme's analytical merit relative to the research question — not by frequency thresholds, statistical tests, or theoretical pre-alignment. Frequency can inform but does not determine validity.
❌ Incorrect. Braun and Clarke explicitly caution against using frequency as the primary criterion for a valid theme. Validity is determined by the theme's meaningfulness and analytical relevance to the research question. The correct answer is B.
Content analysis is a systematic, replicable technique for compressing many words of text into fewer content categories based on explicit rules of coding (Krippendorff, 2018). It bridges qualitative and quantitative traditions — the process of identifying codes and themes is interpretive, but the resulting data (frequencies, proportions, co-occurrences) are quantitative.
Content analysis is appropriate when you want to analyze communication artifacts: policy documents, news articles, social media posts, transcripts, training manuals, advertisements, or published research. The unit of analysis can be a word, phrase, sentence, paragraph, document, or image.
Manifest vs. Latent Content
Manifest content is what is explicitly stated — surface-level, countable. Latent content is the underlying meaning, tone, or implication — requires interpretive judgment. ResearchLens's keyword frequency module captures manifest content. The interpretation engine applies rule-based heuristics to surface latent patterns. Both are required in a complete content analysis.
Content Analysis Workflow in ResearchLens
-
1
Define Your Unit of AnalysisDecide whether you are coding individual words, sentences, paragraphs, or whole documents. ResearchLens operates at the word/phrase level for frequency analysis and at the sentence/passage level for thematic coding. Document this decision in your Methods section — it must be replicable.
-
2
Develop Your Coding SchemeYour coding scheme maps content categories to text evidence. Categories must be exhaustive (every relevant unit can be coded), mutually exclusive (no unit belongs to two categories simultaneously), and reliably applicable by more than one coder. Enter your categories as theme codes in ResearchLens.
-
3
Run Frequency AnalysisResearchLens will compute keyword frequency, show the top N terms, and generate a frequency chart. Cross-reference with your coding scheme. High-frequency terms that fall outside your scheme may indicate gaps in your categories.
-
4
Calculate Inter-Rater ReliabilityIf two coders are applying the same scheme, measure agreement. Cohen's Kappa (κ) above .70 indicates acceptable reliability; above .80 is strong. ResearchLens provides export functions — use the CSV export to calculate κ in a spreadsheet or statistical tool. Document your IRR in the Methodology section.
-
5
Interpret and Report FindingsInterpret what the frequency distribution and coding results mean in relation to your research question. Content analysis findings require both description (what the data show) and interpretation (what the patterns mean). Use the ResearchLens APA report generator to structure these sections correctly.
🧪 Interactive Simulator: Keyword Frequency Analyzer
Try It
Paste document text for frequency analysis:
Filter: minimum word length (characters)
Critical Error: Conflating Frequency with Significance
A word appearing 50 times does not automatically mean it is a significant finding. Context, co-occurrence, and sentiment matter equally. Content analysis requires triangulating frequency data with close reading of actual passages. ResearchLens's "Context Review" panel shows each keyword in context — always review it before drawing conclusions from the frequency chart alone.
A researcher is coding Army recruitment documents for themes of patriotism, career development, and financial incentive. Two coders independently code the same 100 sentences and agree on 78 of them. Applying Cohen's Kappa, this indicates: Select one
✅ Correct. 78% raw agreement with three categories approximates Cohen's Kappa of around .67–.72 depending on marginal distributions. This falls in the "acceptable" range. The researcher should review and resolve the 22 disagreements through discussion and refine the codebook definitions before proceeding.
❌ Incorrect. 78% raw agreement is generally acceptable for content analysis (κ ≈ .70 threshold). The correct action is to treat this as acceptable but use the disagreements as an opportunity to refine the coding scheme. The correct answer is C.
Quantitative research operates within a post-positivist paradigm that assumes a single, objective reality that can be measured with precision. The goal is to test hypotheses derived from theory, measure relationships between variables, establish causation or correlation, and generalize findings to a broader population (Creswell & Creswell, 2023).
In quantitative research, rigor is established through validity (does your instrument measure what it claims?), reliability (does it produce consistent results?), and statistical power (is your sample large enough to detect real effects?). These are non-negotiable — ResearchLens's quantitative module is designed to help you surface these metrics from your survey CSV data.
Descriptive Statistics
- Mean, median, mode (central tendency)
- Standard deviation, variance (spread)
- Range, min, max, percentiles
- Frequency distributions and histograms
- Skewness and kurtosis for normality
Inferential Statistics
- t-tests (compare two group means)
- ANOVA (compare three or more groups)
- Chi-square (categorical relationships)
- Pearson/Spearman correlation
- Regression analysis (prediction)
Using ResearchLens with CSV Survey Data
Upload your cleaned survey CSV file. The platform will compute descriptive statistics for each numeric column, generate distribution charts for Likert-scale items, and identify variables with the highest frequency of extreme responses. Use these outputs as the foundation for your quantitative findings section.
Quantitative Analysis Workflow
-
1
Define Variables and HypothesesIdentify your independent variable (IV), dependent variable (DV), and any covariates. Write null and alternative hypotheses in formal notation: H₀: μ₁ = μ₂ vs. H₁: μ₁ ≠ μ₂. Enter these as Research Questions in ResearchLens — the platform's RQ Alignment Score will later tell you how strongly your data responded to each question.
-
2
Clean and Prepare Your DataRemove duplicates, handle missing values, check for outliers, and verify that all numeric fields are coded consistently. ResearchLens reads CSV files directly. Ensure column headers are clean — no special characters, spaces replaced with underscores, numeric Likert scales encoded as integers not text (e.g., 1–5, not "Strongly Agree").
-
3
Run Descriptive StatisticsRun frequency analysis to see distributions. Review the automated charts. Look for skewed distributions that may violate assumptions of parametric tests. A heavily skewed Likert distribution may require non-parametric alternatives (e.g., Mann-Whitney U instead of t-test).
-
4
Interpret the RQ Alignment ScoreResearchLens computes a Research Question Alignment Score (0–100) that measures how strongly your document or data content aligns with each research question. In quantitative work, treat this as a preliminary alignment check. Scores below 30 may indicate the data set does not adequately address the RQ and warrants methodological reconsideration.
-
5
Export and Complete Statistical AnalysisExport your cleaned dataset from ResearchLens as CSV, then run inferential statistics in SPSS, R, or Python. ResearchLens handles descriptive, frequency, and content-level quantitative analysis. Full inferential testing requires a dedicated statistical environment. Document the entire pipeline in your methodology section.
p-Value Misconception Warning
A statistically significant result (p < .05) does not mean your finding is practically important or large in magnitude. Always report effect size (Cohen's d, eta-squared, Cramér's V) alongside significance values. Report confidence intervals. A tiny effect can be statistically significant with a large enough sample.
A researcher surveys 312 military distance learners and finds that students who used video content scored an average of 12 points higher on final exams than those who used text-only content (p = .03, d = 0.18). What is the most accurate interpretation? Select one
✅ Correct. Cohen's d = 0.18 is classified as a small effect (small ≈ 0.2, medium ≈ 0.5, large ≈ 0.8). The finding is statistically significant — meaning it is unlikely due to chance — but the practical magnitude is modest. Researchers must report and interpret both statistics to avoid overstating results.
❌ Incorrect. The correct answer is C. p = .03 is statistically significant. However, d = 0.18 is a small effect size by Cohen's conventions. Statistical significance and practical significance are separate concepts. Always report and interpret effect sizes alongside p-values.
Mixed methods research is not simply using both a survey and an interview in the same project. It is a methodological paradigm in its own right, grounded in pragmatism, that intentionally integrates qualitative and quantitative data in ways that produce insights neither strand could generate alone (Creswell & Plano Clark, 2018). Integration — the philosophical core of mixed methods — must occur at the design level, data collection level, analysis level, or all three.
The choice of design is determined by the sequence of data collection (concurrent or sequential), the priority of each strand (equal weight or one strand dominant), and the point of integration (data collection, analysis, or interpretation). These decisions must be explicit in your dissertation or research proposal.
Core Mixed Methods Designs
🔢 → 💬 Explanatory Sequential Design (QUAN → qual)
Sequential
▼
Quantitative data is collected and analyzed first. Results that are surprising, significant, or unclear are then explained through qualitative follow-up. Priority sits with the quantitative strand. This is the most common design in dissertation research.
Example: Survey 500 veterans on retention rates (QUAN). Follow up with 12 interviews to explain why high-scoring regions unexpectedly had lower retention (qual).
ResearchLens Workflow
Upload survey CSV → run frequency + RQ alignment → export results → upload interview transcripts → code themes → generate integrated APA report comparing both strands
Integration Point
Quantitative results inform which qualitative questions to ask. The point of integration is in the interpretation — qualitative findings explain quantitative patterns.
💬 → 🔢 Exploratory Sequential Design (qual → QUAN)
Sequential
▼
Qualitative data is collected first to explore an understudied phenomenon. Findings then inform the development of a survey instrument or intervention that is tested quantitatively with a larger sample. Used when validated instruments do not yet exist for the population or construct.
Example: Interview 15 distance learning coordinators to identify technology adoption barriers (qual). Develop and administer a validated survey to 400 coordinators based on the emergent themes (QUAN).
ResearchLens Workflow
Upload interview transcripts → code themes → use themes to build survey → upload survey CSV → run quantitative frequency analysis → integrate findings in report
Integration Point
Qualitative themes directly inform the design of the quantitative instrument. Integration occurs at the instrument development stage.
⚡ Convergent Parallel Design (QUAN + qual simultaneously)
Concurrent
▼
Both strands are collected independently at the same time with equal priority. Results are compared, merged, or contrasted in the interpretation phase to create a fuller picture. Used when the researcher wants multiple perspectives on the same phenomenon simultaneously.
Example: Simultaneously survey 300 Army recruits on enlistment motivations (QUAN) and interview 20 recruits about their decision-making experience (qual). Merge to assess convergence and divergence.
ResearchLens Workflow
Upload both document types → run parallel analyses → export both sets of results → note convergence and divergence in the integrated interpretation section of the APA report
Integration Point
Analysis is kept separate. Integration occurs at the interpretation stage through side-by-side comparison, joint displays, or meta-inferences.
Using ResearchLens for Mixed Methods
ResearchLens supports mixed methods by allowing multiple documents and multiple research questions in a single project. Upload qualitative documents alongside your CSV survey data. Set separate research questions for each strand. The platform's RQ Alignment Score will run across all documents, and the APA report generator produces integrated findings that discuss both strands in relation to each overarching research question.
A researcher uses qualitative interviews to develop survey items, then administers the survey to 600 participants. This is an example of which mixed methods design? Select one
✅ Correct. The defining feature of exploratory sequential design is that qualitative findings are used to build or refine the quantitative instrument. The sequence is qual → QUAN, with qualitative providing the foundation for quantitative expansion.
❌ Incorrect. When qualitative data is collected first specifically to inform the development of a quantitative instrument, this is an exploratory sequential design (qual → QUAN). The correct answer is C.
Platform Architecture
ResearchLens runs 100% in your browser. No data is sent to a server. All document processing, OCR, theme coding, frequency analysis, and report generation happen locally on your device. This ensures FERPA compliance and protects the privacy of any participant data in your research documents.
Step-by-Step: From Upload to APA Report
1
Upload Your Documents
Drag and drop or click to upload. ResearchLens accepts PDF (text-layer and scanned/image-based), DOCX, and CSV files. For scanned PDFs, Tesseract.js OCR runs automatically. For DOCX files, Mammoth.js extracts content including headers, body, and table text. CSV files are parsed for quantitative and survey analysis.
Supports: PDF, DOCX, CSV
2
Enter Research Questions
Click "Add Research Question" and type each RQ in the text field. You may add up to the platform's maximum. The RQ Alignment Score system evaluates how well your uploaded documents address each question. Write your RQs using standard academic framing: "To what extent does X influence Y in the context of Z?"
RQ Alignment Score: 0–100
3
Select Research Paradigm
Select Qualitative, Quantitative, Content Analysis, or Mixed Methods. This selection changes which analysis tabs are foregrounded and which interpretation rules are applied in the report engine. You can run multiple paradigms on the same document set — useful for mixed methods.
Critical: match your research design
4
Add Theme Codes
Enter your a priori or emergent codes in the Theme Codes panel. Separate with commas. Codes are case-insensitive. The platform will scan all uploaded text for occurrences and co-occurrences of your codes. Use short, precise code labels. Avoid synonyms that refer to the same concept — group them as variants of a single code.
Qualitative + Content Analysis
5
Run Analysis
Click "Analyze" to trigger the full pipeline: keyword frequency, theme matching, context extraction, RQ alignment scoring, chart generation, and interpretation. Review each tab: Overview, Frequency, Themes, Context Review, Charts. Check the sentiment tagging in Context Review to identify whether coded passages carry positive, negative, or neutral valence.
Full pipeline: ~5–15 seconds
6
Review Context Passages
The Context Review panel shows each coded segment with surrounding text. Use this for close reading. In qualitative research, these are your candidate quotes for the findings section. In content analysis, these passages verify that frequency counts reflect genuine conceptual occurrences, not spurious keyword matches.
Essential for validity
7
Generate APA Report
Click "Generate Report" to produce a structured APA-style research report including: Abstract, Introduction (with RQs), Methodology, Findings (descriptive + interpretive), Discussion, and Conclusion. Download as HTML or RTF. Use the HTML export for clean web viewing; use the RTF export to open in Word for final formatting before submission.
Export: HTML, RTF, CSV, JSON
8
Export Raw Data
Export your frequency table as CSV for external statistical processing. Export your coded theme data as JSON for use in specialized qualitative software (ATLAS.ti, NVivo). The full analysis state can also be exported as a JSON snapshot for reproducibility — critical for dissertation and IRB documentation.
Reproducibility + IRB compliance
NEW: Codebook Builder v2 — Full 8-Field Academic Export
What makes this dissertation-ready?
The Codebook Builder stores and exports all 8 fields academic committees require: Theme Name, Definition, Related Keywords, Example Excerpts, Number of Matches, Linked Research Question, Interpretation, and Researcher Notes. These transform a keyword list into a rigorous, auditable codebook.
A
Fill All 8 Fields — Especially Interpretation and Researcher Notes
After clicking + New Code, the full editor is visible. The Interpretation field captures what the code means in relation to your RQ. Researcher Notes are your reflexivity memo — coding decisions, audit trail entries, revisions. Write these after running analysis when data context is clear.
Qualitative + Content Analysis
B
Link Each Code to Its Research Question(s)
Each code can be linked to one or more RQs. This creates a clear methodological trail showing which codes address which research questions — a requirement in most dissertation codebook appendices. Linked RQs appear in all four export formats.
Required for dissertations
C
Export in 4 Academic Formats
Use the Academic Export panel at the bottom of the Codebook tab: Full CSV (all 8 fields plus per-doc counts), JSON Package (full metadata for reproducibility), HTML Codebook (formatted printable document), and Word-Ready RTF (paste directly into your dissertation appendix in Microsoft Word).
Export: CSV, JSON, HTML, RTF
NEW: Manual Coding Mode v2 — Interactive Two-Pane Workspace
Why manual coding matters
Keyword frequency counts are a starting point. Genuine qualitative research requires human interpretive judgment. Manual Coding Mode v2 puts that judgment at the center with a two-pane interface keeping document and coding form side by side at all times.
A
Select Document — Coded Passages Highlight Automatically
The left pane loads your document in reading-optimized Georgia serif with generous line spacing. Any passages you have already coded are highlighted in their code color, giving you an instant visual map of coverage and gaps in your coding.
Visual coding map on load
B
Highlight Text — Floating Toolbar Appears Instantly
Select any passage. A floating toolbar appears above the selection with a Code This button. It follows your selection precisely — no scrolling to buttons at the top or bottom of the page. Click it to open the form with the selected text pre-populated.
Zero-friction text capture
C
Assign Codes via Chip Picker, Sentiment Buttons, RQ Link, and Memo
Click code chips to assign multiple codes simultaneously. Click sentiment buttons (Positive, Negative, Neutral, Mixed). Set evidence status. Link to a RQ. Write your research memo. Save. The excerpt appears in the Evidence Library immediately and syncs to the Codebook's example excerpt field automatically.
7-field excerpt record + real-time sync
NEW: Codebook Builder v2 — Full 8-Field Academic Export
What makes this dissertation-ready?
The upgraded Codebook Builder stores and exports all 8 fields that academic reviewers and dissertation committees require: Theme Name, Definition, Related Keywords, Example Excerpts, Number of Matches, Linked Research Question, Interpretation, and Researcher Notes. These fields transform a keyword list into a rigorous, auditable codebook.
1
Create or Edit a Code — Fill All 8 Fields
After clicking + New Code, you will see the full code editor. The new fields are: Interpretation (what this code means in relation to your RQ) and Researcher Notes (reflexivity memos, coding decisions, audit trail). Fill these after running the analysis when the data context is clear.
Qualitative + Content Analysis
2
Link Codes to Research Questions
Each code can be linked to one or more of your project's Research Questions. This produces a clear methodological trail showing which codes address which RQs — a requirement in most dissertation codebook appendices. Linked RQs appear in both the card view and all four export formats.
Required for dissertations
3
Run Analysis — Example Excerpts Auto-Populate
After running the analysis AND saving coded excerpts in Manual Coding Mode, the Codebook automatically pulls the first linked excerpt as an Example Excerpt for each code. This eliminates manual copy-paste and ensures your codebook is always synchronized with your evidence library.
Auto-synced from Evidence Library
4
Export in 4 Formats
The Academic Export panel at the bottom of the Codebook tab provides four export options: Full CSV (all 8 fields plus per-document counts and keyword breakdown), JSON Package (machine-readable with full metadata for reproducibility), HTML Codebook (formatted academic document you can print or submit), and Word-Ready RTF (open in Microsoft Word to paste into your dissertation appendix).
Export: CSV, JSON, HTML, RTF
NEW: Manual Coding Mode v2 — Full Interactive Workspace
Why manual coding matters
Keyword frequency counts are a starting point, not the destination. Genuine qualitative research requires a human researcher making interpretive judgments about meaning, context, and evidence. Manual Coding Mode v2 puts human judgment at the center with a two-pane interface that keeps your document and coding form side-by-side at all times.
1
Select a Document — It Loads in the Reading Pane
The left pane displays your document text in reading-optimized typography (Georgia serif, generous line spacing). Existing coded excerpts are automatically highlighted with their code color when you load the document, giving you an instant visual map of what has already been coded.
Visual coding map on load
2
Highlight Text — Floating Toolbar Appears
Select any passage of text in the reading pane. A floating toolbar immediately appears above the selection with a Code This button. Click it to open the coding form with the selected text pre-populated. The floating toolbar requires no mouse movement to the top or bottom of the screen — it follows your selection.
Zero friction text capture
3
Assign Codes — Click Chips, Not Dropdowns
The coding form shows all your codebook codes as clickable chips. Click one or more to select them — no holding Ctrl/Cmd, no scrolling a multi-select list. Each chip shows the code's color dot for visual identification. This chip-based picker is significantly faster than the traditional multi-select dropdown for heavy coding sessions.
Multi-code chip picker
4
Mark Sentiment, Evidence Status, RQ Link, and Memo
Click one of four sentiment buttons (Positive, Negative, Neutral, Mixed), set the evidence status (Supports, Contradicts, Neutral, Outlier, Needs Review), link to a research question, and add a research memo. The memo field is where your analytical thinking lives — note why the excerpt matters, what it suggests, or how it challenges your assumptions.
Full 7-field excerpt record
5
Save to Evidence Library — Searchable, Filterable
Click Save to Evidence Library. The excerpt appears in the right-pane library card. Filter by code, evidence status, or research question. All excerpts feed the Codebook Builder automatically — your example excerpts, frequency of use per code, and RQ linkages stay current without manual entry.
Real-time codebook sync
Supported File Types and Their Use Cases
| File Type | Best For | Method Alignment | Notes |
|---|---|---|---|
| PDF (text-layer) | Academic articles, policy documents, reports, transcripts | Qualitative Content | Direct text extraction. Formatting stripped. Best quality extraction. |
| PDF (scanned) | Historical documents, handwritten notes, printed forms | Qualitative Content | OCR via Tesseract.js. Quality depends on scan resolution. 300 DPI minimum recommended. |
| DOCX | Interview transcripts, research memos, literature reviews | Qualitative Mixed | Full content including tables. Best format for research documents created in Word. |
| CSV | Survey data, Likert scales, behavioral logs, enrollment records | Quantitative Mixed | Numeric columns auto-detected. First row must be column headers. UTF-8 encoding required. |
The Research Integrity Checklist
- Declare your positionality — In qualitative and mixed methods research, disclose your relationship to the topic, population, and data. Your perspective shapes interpretation. This is not a flaw; undisclosed positionality is the flaw.
- Obtain IRB approval — Any research involving human participants (including analysis of their documents or survey responses) requires Institutional Review Board review. Complete this before uploading any participant data to ResearchLens.
- De-identify data before uploading — Remove names, student IDs, social security numbers, and any direct identifiers from documents before uploading. ResearchLens runs locally, but data hygiene is your responsibility as the researcher.
- Match your method to your epistemology — If you believe knowledge is constructed socially, using a purely quantitative design creates an internal contradiction. Your worldview, methodology, and methods must be philosophically consistent (Creswell & Poth, 2018).
- Triangulate your findings — Never rely on a single data source or method. Cross-reference your ResearchLens outputs with external literature, participant feedback, or peer review of your coding scheme.
- Audit your coding trail — Document every decision you make about code creation, modification, and deletion. ResearchLens's export functions allow you to capture snapshots of your analysis at multiple stages. Maintain these as appendices.
- Do not cherry-pick findings — Report disconfirming data, outliers, and alternative interpretations. Academic integrity requires presenting findings that challenge your hypothesis with the same rigor as findings that support it.
- Do not generalize beyond your design — Qualitative findings from 12 interview participants cannot be generalized to a population. Quantitative findings from a convenience sample cannot be generalized beyond the sample's characteristics. State limitations explicitly.
Methodology Comparison at a Glance
| Criterion | Qualitative | Content Analysis | Quantitative | Mixed Methods |
|---|---|---|---|---|
| Paradigm | Constructivism / Interpretivism | Both / Pragmatic | Post-positivism | Pragmatism |
| Data Type | Words, narratives | Text, communication artifacts | Numbers, variables | Both |
| Sample Size | Small, purposeful | Variable (corpus-based) | Large, representative | Depends on design |
| Rigor Standard | Trustworthiness (Lincoln & Guba) | IRR, systematic coding | Validity, reliability, power | Integration quality |
| ResearchLens Tool | Theme coder, context review | Frequency analysis, coding scheme | CSV analysis, RQ scoring | All modules combined |
| Typical Output | Themes, narratives, quotes | Frequency tables, coding maps | Statistics, charts, p-values | Integrated findings |
Citing ResearchLens in Your Research
When using ResearchLens as your analytical tool, cite it in your Methods section: Rodas, R. A. (2025). ResearchLensPRO™: Browser-based academic content analysis platform (Version 10.0) [Software]. Rontechmedia. https://rontechmedia.com/researchlens
Core Methodological References (APA 7th Edition)
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
Creswell, J. W., & Creswell, J. D. (2023). Research design: Qualitative, quantitative, and mixed methods approaches (6th ed.). SAGE Publications.
Creswell, J. W., & Plano Clark, V. L. (2018). Designing and conducting mixed methods research (3rd ed.). SAGE Publications.
Creswell, J. W., & Poth, C. N. (2018). Qualitative inquiry and research design: Choosing among five approaches (4th ed.). SAGE Publications.
Krippendorff, K. (2018). Content analysis: An introduction to its methodology (4th ed.). SAGE Publications.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. SAGE Publications.
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research: A guide to design and implementation (4th ed.). Jossey-Bass.
Neuendorf, K. A. (2017). The content analysis guidebook (2nd ed.). SAGE Publications.
Patton, M. Q. (2015). Qualitative research and evaluation methods (4th ed.). SAGE Publications.
Yin, R. K. (2018). Case study research and applications: Design and methods (6th ed.). SAGE Publications.