Reading across sources to find agreement, disagreement, evidence and method differences, gaps, contradictions, and the patterns that prepare you for synthesis
Imagine two students working on the same research question: "How does social media use affect adolescent mental health?" The first student reads five sources, summarizes each one in a paragraph, and moves directly to writing. Her paper reads like a tour of five separate opinions. The second student reads the same five sources, then asks where they agree, where they disagree, what kinds of evidence each one offers, and whether any patterns emerge. His paper reads like an analysis.
The difference between these two students is not the number of sources. It is what they did with the sources once they had them. This chapter teaches source comparison and pattern recognition — the bridge between collecting evidence and understanding it.
Comparing sources across multiple readings to identify agreement, disagreement, emphasis, omission, evidence differences, method differences, gaps, contradictions, and the recurring patterns that give research its analytical depth.
Comparing sources transforms research from collection into understanding. It is the cognitive work that turns a pile of notes into a coherent intellectual position. It is also what separates research from opinion: when you compare sources carefully, you stop asking "What do I think?" and start asking "What does the evidence show, where does it agree, where does it disagree, and what do those patterns mean?" That shift from personal impression to evidential reasoning is one of the most important intellectual moves a student researcher can make — and one of the most transferable skills you will carry into medicine, law, policy, journalism, business, and nearly any other field that requires weighing competing studies, reports, or expert opinions.
Research is not primarily a collection activity. It is, at its core, a comparison activity. You are not simply gathering everything written on a topic. You are gathering sources so that you can compare what they say, understand where the evidence is strong or weak, and form a responsible conclusion about what is known, what is debated, and what remains unclear.
When you collect sources without comparing them, you have information. When you compare sources carefully, you begin to have understanding. The difference matters enormously.
Comparison requires reading sources with specific questions in mind: What claim does this source make? What evidence does it offer for that claim? How does that claim relate to what other sources say? Where do sources agree? Where do they disagree? What does each source emphasize, and what does each omit? What patterns emerge across multiple sources? What is still unclear, contradicted, or unresolved?
These questions transform reading from passive absorption into active inquiry. You are not simply taking in information; you are interrogating it against other information.
| Activity | What It Involves |
|---|---|
| Summary | Recording what one source says |
| Comparison | Placing two or more sources in relationship to one another |
| Pattern recognition | Identifying recurring claims, gaps, or contradictions across sources |
| Synthesis | Explaining what patterns mean and what the evidence as a whole shows |
Chapter 36 covered organizing sources by theme. This chapter builds directly on that skill. You now have organized notes. The task is to read across those notes and begin comparing what sources say.
Source comparison is genuinely challenging. It requires holding multiple positions in mind simultaneously, recognizing similarities and differences that are not always obvious, and tolerating uncertainty — because sources often disagree, and sometimes neither is clearly right. It also requires resisting the temptation to simply pick the source that confirms what you already believe.
These difficulties are reasons to practice comparison deliberately and systematically, not reasons to avoid it. The frameworks in this chapter give you specific questions to ask and structures to use so that comparison becomes a learnable skill.
When sources that were produced independently — by different researchers, using different methods, at different times — reach similar conclusions, that convergence is meaningful. Agreement across independent sources is one of the most powerful forms of evidence in research.
Agreement does not make a claim automatically true. But it does strengthen the case that a claim is well-supported. The more independent the agreeing sources are — different institutions, different methodologies, different time periods, different disciplinary perspectives — the more significant the agreement becomes.
Agreement is not always obvious. Sources may reach the same conclusion while using different language, data, examples, or analytical frameworks. Part of your task is to recognize agreement even when sources do not use identical words. Look for shared conclusions, overlapping evidence, consistent framing, and compatible recommendations.
Finding agreement is valuable, but it does not complete your analysis. Ask: What is the quality of the agreeing sources? Do they share a common methodological flaw? Are they all drawing from the same original data, which would make their agreement less independent than it appears? Are there dissenting voices worth examining? Agreement sets a baseline; the rest of your comparison work refines and complicates it.
Sources disagree. This is not a problem with your research — it is a feature of the research landscape that you must learn to navigate rather than avoid. Disagreement exists because researchers use different methods that reveal different aspects of a phenomenon, analyze different populations or time periods, operate from different theoretical frameworks, or because evidence genuinely points in multiple directions. Some questions are inherently contested — not because evidence is absent, but because values, interpretations, and priorities legitimately differ.
Understanding why sources disagree is often as important as noticing that they disagree. Not all disagreement is the same kind.
| Type of Disagreement | What It Means | What to Do |
|---|---|---|
| Factual | Sources report different data or findings | Evaluate methodological quality; look for replication |
| Interpretive | Sources read the same facts differently | Identify the assumptions behind each interpretation |
| Value-based | Sources prioritize different outcomes | Acknowledge the values at stake; do not present as a factual dispute |
| Scope-based | Findings apply differently in different contexts | Specify the conditions under which each finding holds |
Sources may disagree about what actually happened, what the data shows, or what a study found. This can often be partially resolved by examining the quality of each source's evidence, the size and representativeness of samples, the rigor of methods, and whether independent replication has occurred.
Sources may agree on the facts but disagree about what those facts mean. Two historians might agree an event occurred but disagree about whether economic pressures or political decisions caused it primarily.
Some disagreements are fundamentally about priorities rather than evidence — should a city prioritize affordable housing or green space? Evidence can inform these trade-offs but not resolve them.
Sources may disagree about how broadly a finding applies. A study showing an intervention works in urban schools is not necessarily contradicted by one showing weaker effects in rural settings — both may reflect genuine variation in context.
Suppose that while most sources agree early reading interventions produce positive effects, you also find a study reporting more modest effects in certain low-income urban districts, and a critique arguing that some widely cited studies have methodological limitations in how literacy outcomes were measured.
Two sources can reach different conclusions not because one is right and one is wrong, but because each is looking at different evidence. Different types of evidence illuminate different aspects of a question. When you compare sources, pay close attention not only to what they conclude but to what kind of evidence they used to reach those conclusions.
A large-scale quantitative study measuring outcomes across ten thousand students tells you something different from an ethnographic study closely observing fifteen children in one classroom. Both offer legitimate evidence, but they are answering somewhat different versions of the research question.
| Evidence Type | What It Tends to Show | What It May Miss |
|---|---|---|
| Large quantitative studies | Broad patterns across many cases | Individual variation, context, mechanism |
| Small qualitative studies | Depth, context, lived experience | Generalizability across larger populations |
| Historical documents | What happened and how it was understood at the time | Long-term outcomes, alternative perspectives |
| Expert opinion | Experienced professional judgment | Empirical data, systematic evidence |
| Experimental data | Causal relationships under controlled conditions | Real-world applicability outside the experimental setting |
| Survey data | Self-reported attitudes, beliefs, and behaviors | What people actually do versus what they say they do |
When two sources reach different conclusions, trace the difference back to evidence type before assuming one source is simply better. Sometimes the right conclusion is: "Source A, using large-scale outcome data, finds X. Source B, using in-depth interviews, finds Y. These findings are not contradictory — they describe different dimensions of the same phenomenon."
Large survey studies often find that students who participate in peer tutoring report higher satisfaction and more confidence. Controlled experimental studies show more mixed results when looking specifically at grade outcomes. Interview-based studies reveal that benefits depend on how programs are structured, the subject matter, and the tutor-student relationship.
Research is a family of activities using different methods — and different methods reveal different things. Method includes how a researcher gathered data, what kinds of sources they analyzed, how large their sample was, how they defined their variables, what controls they used, and how they analyzed what they found. Two studies asking roughly the same question may reach different conclusions because they used different methods — and both findings can be valid within the scope of what each method can reveal.
| Method | Strength | Limit |
|---|---|---|
| Experimental | Powerful for establishing causation through random assignment and controlled variables | Controlled conditions may not reflect real-world complexity |
| Observational / Survey | Captures real-world behavior and attitudes | Cannot establish causation as definitively as experiments |
| Qualitative | Illuminates context, meaning, and process | Small numbers of cases; does not generalize statistically |
| Historical / Archival | Essential for understanding historical questions | Limited by survival and completeness of records |
| Computational / Data Analysis | Identifies patterns across enormous datasets | Depends heavily on data quality and representativeness |
| Meta-analysis / Systematic Review | Provides the most comprehensive picture of a field's knowledge | Depends on the quality of the studies it reviews |
One of the most important things you can learn from comparing sources is not what they say, but what they leave out. Gaps in the research literature are meaningful signals about the limits of current knowledge, the populations and questions that have not been adequately studied, and the perspectives that have historically been excluded from academic conversation.
A gap may be a population rarely studied, a time period or geographic region lacking coverage, a perspective systematically excluded, a methodological approach not yet applied, or a dimension of a topic researchers have tended not to examine.
Gaps are visible only through comparison. If you have read only one source, you cannot see what is missing. Ask: Who is studied, and who is not? What time periods or places are represented, and which are absent? Are there stakeholders, disciplines, or communities that do not appear to have been consulted? Do the sources themselves acknowledge limitations or call for further research?
Do not assume absence means a question has been answered negatively. Absence means the question has not been adequately studied — those are different things. In your writing you might say: "The research on this question has focused primarily on college-aged populations. Less is known about how these effects play out among adults in the workforce."
In fields such as history, social science, public health, and education, the absence of certain perspectives from the research literature is not only an intellectual limitation but an ethical one. Research conducted without affected communities' meaningful participation may be systematically distorted in ways that affect both its accuracy and its usefulness. When you identify missing perspectives, note them as invitations to ask whose knowledge counts and what a more complete research picture would require.
Disagreement is normal in research. Contradiction is a stronger and more specific condition. A contradiction exists when two sources make claims that cannot both be true — when Source A asserts X and Source B asserts not-X, using comparable populations, methods, and definitions.
True contradictions are rarer than they appear, because apparent contradictions often dissolve when you examine exactly what each source studied, how it defined its terms, and what kinds of evidence it used. But genuine contradictions do exist, and they signal that the evidence is genuinely ambiguous, that methodological limitations are affecting results, or that the question is more complex than it first appears.
1. Check definitions — are key terms used the same way? 2. Check populations — are the same groups being studied? 3. Check methods — could methodological differences explain the contradiction? 4. Check timeframes — are conditions comparable across periods? 5. Check replication — has either finding been independently confirmed? 6. Consult secondary literature — has a review article already analyzed this contradiction?
Many students feel that acknowledging uncertainty makes their paper look weak. The opposite is true. Responsible researchers do not overclaim — they report what the evidence shows with appropriate confidence and honestly acknowledge what remains unclear, disputed, or unresolved. "Research on this question remains divided, with some studies finding X and others finding Y; the methodological differences between these studies suggest the answer may depend on [contextual factor]" is a stronger intellectual position than a confident but inaccurate overclaim.
The central principle of this textbook applies with particular force here: Do not merely find information. Earn trust in information. Earning that trust requires honesty about what the evidence does and does not establish.
A pattern is a recurring feature that appears across multiple independent sources. Once you have compared sources carefully — noting agreement, disagreement, evidence type, method, and omission — you are in a position to identify the patterns running through your source collection. Patterns are not manufactured; you read carefully and repeatedly, asking what keeps appearing, what is consistently absent, and what relationships multiple sources observe.
| Pattern Type | What It Tells You |
|---|---|
| Consensus pattern | Multiple independent sources reach similar conclusions using different methods — suggests robustness |
| Consistent qualification pattern | Sources agree but consistently note conditions or exceptions — the finding is real but not universal |
| Persistent disagreement pattern | Sources consistently disagree even after examining method — indicates genuine uncertainty |
| Coverage pattern | Sources consistently address some dimensions and ignore others — reveals what the field has prioritized |
| Change-over-time pattern | Earlier and more recent sources differ — understanding may be evolving or conditions changing |
| Magnitude variation pattern | Sources agree on direction but disagree on size of effect — often reflects methodological variation |
Sources include a meta-analysis of thirty studies, a city-level quantitative study, a qualitative study of low-income neighborhoods, a public health equity report, a park-redesign study, a survey of urban planners, a historical study of park design, and a recent study addressing income variation.
Suppose you are researching whether requiring students to use laptops in college classrooms improves learning outcomes, with three sources: a quantitative study comparing exam scores in laptop-permitted versus laptop-prohibited sections (200 students); a review article summarizing twenty studies finding highly variable effects depending on task structure; and a qualitative interview study of thirty students describing both benefits and distraction.
| MGU Source Evaluation Model | Assessment |
|---|---|
| Relevance | All three sources are directly relevant to the question |
| Authority | Peer-reviewed journal study; review article; interview-based qualitative research |
| Evidence | Quantitative, synthesized quantitative, and qualitative evidence types |
| Accuracy | No implausible findings; the review and interviews add contextual accuracy missing from the single study alone |
| Purpose | All three appear to be scholarly research without obvious advocacy motivation |
| Currency | Laptop research evolves rapidly; older studies may not reflect current technology |
| Context | The single-course sample of 200 is limited; the review and interviews broaden the context |
The weight of evidence suggests that unrestricted laptop use in classrooms is associated with reduced academic performance, primarily because many students use devices for off-task activities. However, structured use of laptops appears to have neutral or positive effects. The equity dimension — students who depend on laptops for accessibility — is largely absent from this literature, marking a clear gap.
Use the MGU Research Note Model as the foundation for individual source notes. When comparing sources, add a comparison layer on top of your individual notes.
Keeping comparison notes separate from individual source notes makes your thinking visible and builds the analytical foundation you will need for synthesis in Chapter 38.
One of the most common misinformation strategies is the false claim of scientific or expert consensus. A source may claim that "researchers agree" or "studies show" a particular position when the research picture is actually far more mixed or contested.
Check who is described as agreeing. "Scientists agree" is too vague to evaluate — which scientists, in which field, with what credentials, publishing where?
Check whether dissenting evidence exists. Genuine consensus does not mean unanimous agreement, but it does mean the preponderance of credible, peer-reviewed evidence points consistently in one direction.
Check whether agreement is independent. If multiple sources all draw on the same original dataset, their apparent agreement is far less significant than true convergence from independent sources.
Check who is funding research. Industry- or interest-funded studies are not automatically wrong, but they deserve additional scrutiny.
Be wary of cherry-picking. A source presenting a confident conclusion about a complex topic without acknowledging any contrary evidence is likely leaving something important out.
Who made it? What is the claim? What evidence is offered? What is missing? Can another reliable source confirm it? Is the emotional reaction stronger than the evidence? Has the source been altered, misrepresented, or taken out of context?
When multiple sources support the same claim, cite all of them — do not cite only one as though it were the only evidence. Citing multiple sources for a well-supported claim demonstrates that you are drawing on a broad evidence base, not a single study.
1. Begin with the Encyclopedia to build context and identify major debates. 2. Use subject guides to identify key databases and journals. 3. Use academic database search to fill gaps identified through comparison. 4. Use Reference pages for quick methodological or evaluation guidance. 5. Return to your comparison notes and update them as new sources add information.
Do not merely find information. Earn trust in information. Earning that trust requires honest comparison, accurate pattern-finding, and intellectual humility about what the evidence does and does not establish.
Think of a time when reading multiple sources on the same topic changed your understanding — perhaps you started believing one thing and ended up with a more complicated view. What sources were involved? What made you reconsider? How does that experience relate to the comparison process described in this chapter?
Identify a topic where you have seen claims of "scientific consensus" in the media or online. Using the questions from the Misinformation Alert, evaluate whether that claimed consensus appears genuine, independent, and well-supported, or whether it may be overstated or cherry-picked.
Choose a topic connected to your intended major or an area of interest. What populations, time periods, or perspectives do you suspect are underrepresented in the research on that topic? How might you go about confirming whether that gap genuinely exists?
Chapter 37 taught you how to compare sources and identify the patterns running through them — agreement, disagreement, evidence and method differences, gaps, and contradictions. Chapter 38 builds directly on this foundation. It will teach synthesis: the process of explaining what the patterns you have identified actually mean, combining multiple sources into a coherent account of what the evidence shows, where genuine debate exists, and what remains uncertain. Synthesis is more than summary — it is the intellectual work of turning a comparison into understanding, and it is one of the most important skills this textbook will teach.