Framing Analysis

Framing analysis is a method for studying how issues are presented — in news media, political communication, policy documents, or other texts. It focuses not on whether coverage is true or false, but on which aspects of an issue are made salient and how that shapes interpretation.

This page explains the method, walks through how to apply it, and provides guidance on structuring a framing analysis thesis. It draws on the core literature in the field and addresses practical questions that come up repeatedly in supervision.

Where does framing analysis fit? It is one of the most widely used methods in media studies, political communication, and international relations research. If your research question asks how an issue is presented, represented, or constructed in texts, framing analysis is likely a good fit. For related approaches, see Discourse Analysis.

What Is Framing Analysis?

At its core, framing analysis examines how communication sources define and construct issues. The foundational definition comes from Entman (1993): to frame is to “select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation.”

This definition identifies four framing functions:

Function	What it does	Example
Problem definition	Identifies what the issue is	“Immigration is a security threat” vs. “Immigration is an economic opportunity”
Causal interpretation	Attributes cause or responsibility	“Government policy caused the crisis” vs. “Global forces caused the crisis”
Moral evaluation	Makes a normative judgment	“This policy is unjust” vs. “This policy is necessary”
Treatment recommendation	Suggests what should be done	“Borders should be tightened” vs. “Integration programs should be expanded”

Not every frame performs all four functions in every text. But Entman’s framework gives you a systematic way to identify what a frame is doing.

Framing analysis differs from content analysis in that it goes beyond counting topics or themes. It asks how an issue is constructed — what is emphasized, what is left out, and what interpretive lens is offered to the audience.

Key Concepts

Frame. A central organizing idea that provides meaning to an issue by selecting, emphasizing, and connecting certain aspects of reality while downplaying others (Gamson & Modigliani, 1989).

Framing devices. The textual elements that signal a frame — metaphors, catchphrases, exemplars, visual images, and appeals to principle (Gamson & Modigliani, 1989). These are what you look for when coding.

Generic vs. issue-specific frames. Generic frames (e.g., conflict, human interest, economic consequences, responsibility, morality) appear across many issues and can be compared across contexts (Semetko & Valkenburg, 2000). Issue-specific frames are tailored to a particular topic and emerge from your data or prior literature.

Episodic vs. thematic framing. Iyengar (1991) distinguished between episodic frames (focusing on specific events, individuals, or cases) and thematic frames (placing issues in a broader context with systemic causes). This distinction is especially useful for analyzing news coverage, where episodic framing tends to individualize responsibility while thematic framing points to structural explanations.

Frame building vs. frame setting. Frame building is the process by which frames emerge in media or political discourse (how do certain frames come to dominate?). Frame setting is the process by which media frames influence audience perceptions (Scheufele, 1999; de Vreese, 2012). Most thesis-level framing analyses focus on identifying frames in texts (frame building), not on measuring audience effects (frame setting), since the latter requires experimental or survey data.

Salience. A key mechanism in framing: making certain pieces of information more noticeable, meaningful, or memorable (Entman, 1993). A frame does not have to fabricate information — it works by emphasizing some elements and downplaying others.

Frames are not the same as topics. A topic is what a text is about (e.g., climate change). A frame is how the text presents that topic (e.g., climate change as an economic burden, a moral imperative, a scientific controversy, or a security threat). A single topic can be framed in multiple, competing ways. This is one of the most common points of confusion in student work.

When to Use It

Framing analysis is appropriate when your research question asks how an issue is presented, constructed, or represented in communication. It works well for:

Media coverage analysis — How does the press frame a political crisis, policy debate, or social issue? Do frames differ across outlets, countries, or time periods?
Political communication — How do political actors frame issues to build support? How do competing frames interact in public debate?
Policy discourse — How are policy problems defined and solutions justified in official documents?
Comparative studies — How does framing of the same issue differ across countries, media systems, languages, or time periods?

Framing analysis is not the best choice when:

You want to measure the effects of framing on audiences (that requires experimental or survey methods)
Your interest is in language structure, power relations, or identity construction at a deep level (consider discourse analysis instead)
You want to count word frequencies or topic occurrence without analyzing how issues are constructed (that is closer to basic content analysis)

Ask yourself

Does your research question ask how something is presented, not just what is covered?
Can you identify a specific set of texts (news articles, speeches, policy documents) to analyze?
Are you interested in comparing how different sources, countries, or time periods present the same issue?
Is the framing of this issue contested — do different actors present it in meaningfully different ways?

If you answered yes to most of these, framing analysis is likely a good fit. If your interest is more in the deep structure of language and power, look at discourse analysis.

Framing analysis can also be applied to visual media (photographs, political cartoons, infographics) and social media content (tweets, posts, short-form video), though the analytical techniques differ from those used with traditional textual sources. This guide focuses on textual framing of print and online news, policy documents, and similar written materials. If you are working with visual or social media data, discuss the necessary adaptations to your coding approach and analytical framework with your supervisor.

How to Apply It

Step 1: Define Your Approach

The first decision is whether to use a deductive or inductive approach — or a combination.

Deductive (theory-driven). You start with a set of frames drawn from existing literature and look for them in your data. This is common when using generic frames. Semetko and Valkenburg (2000) identified five generic news frames — conflict, human interest, economic consequences, morality, and responsibility — that have been widely applied across contexts. A deductive approach is more structured, more replicable, and easier to manage for BA students.

Inductive (data-driven). You develop frames from the data itself through iterative reading and coding. This is necessary when studying an issue where no established frame typology exists, or when you want to capture issue-specific frames. It is more demanding and requires careful documentation of how frames were identified. Matthes and Kohring (2008) offer a rigorous cluster-analytic approach to inductively identifying frames.

Combined approach. Many strong theses use a hybrid strategy: start with frames from the literature, then allow new frames to emerge from the data. This gives you theoretical grounding while remaining open to what the material actually contains.

BA vs. MA expectations. BA students are generally well served by a deductive approach using established generic frames, or a small set of issue-specific frames drawn from prior research. MA students are expected to engage more critically with the choice of approach and may be expected to develop their own frame typology, justify it methodologically, and reflect on its limitations.

Step 2: Build Your Corpus

Your corpus is the collection of texts you will analyze. Corpus construction is itself a methodological decision that must be justified. Key questions:

Source selection. Which outlets, speakers, or document types? Why these and not others? Justify your selection in terms of your research question (e.g., choosing outlets that represent different editorial positions, or official documents from specific institutions).
Time period. What period does your analysis cover, and why? Are you analyzing a specific event, a policy cycle, or long-term trends?
Sampling. Will you analyze all relevant texts in your time period (census approach), or a sample? If sampling, what is your sampling strategy?
Size. There is no universal minimum, but your corpus must be large enough to support the claims you make. A BA thesis might analyze 50-100 articles; an MA thesis might analyze 100-200 or more, depending on the method. See the Building a Corpus page for detailed guidance on corpus size, including how to balance analytical depth against volume for different methods.

For detailed guidance on source selection, search strategies, and data management, see Building a Corpus.

Step 3: Develop a Codebook

The codebook is the backbone of your analysis. It defines exactly what you are looking for and how you will identify it. A well-developed codebook makes your analysis systematic, transparent, and replicable.

Your codebook should include:

Frame definitions. A clear, concise definition of each frame you are coding for. Each definition should be specific enough that someone else could apply it consistently.
Indicators. The concrete textual markers (keywords, phrases, arguments, metaphors, narrative structures) that signal each frame. List multiple indicators per frame.
Coding rules. Explicit instructions for how to handle ambiguous cases, including:
- What counts as the unit of analysis (full article, paragraph, sentence)?
- Can multiple frames be assigned to a single text? If so, do you code a primary/dominant frame, or all frames present?
- What is the threshold for assigning a frame — must it be the central organizing idea, or can it be secondary?
Examples. At least one clear example of each frame from your data or from prior literature.

Pilot your codebook. Before coding your full corpus, test the codebook on a small sample (10-15 texts). This will reveal definitions that are too vague, categories that overlap, or frames that never appear. Revise the codebook based on this pilot, and document what you changed and why. This is a sign of methodological rigor, not weakness.

Step 4: Code Your Material

With a tested codebook in hand, code your full corpus systematically.

Coding procedure:

Read each text carefully — in full, not just the headline (see the section on full article vs. headline framing below).
Identify which frame(s) are present using the indicators in your codebook.
Record your coding decisions in a structured format (spreadsheet, coding software, or a systematic table).
Keep notes on difficult or ambiguous cases. These notes are valuable for your methodology section and for discussing limitations.

Handling multi-frame texts. Many texts contain more than one frame. You need a clear decision rule, established in advance:

Dominant frame approach. Assign each text the single frame that is most prominent overall. This simplifies analysis but loses information about secondary frames.
All-frames approach. Code every frame present in each text. This captures more complexity but requires you to define how you determine frame presence (e.g., a frame must appear in at least two paragraphs, or be supported by at least two distinct indicators).
Hierarchical approach. Code a primary frame and one or more secondary frames. This balances simplicity with nuance.

Whichever approach you choose, state it explicitly in your methodology chapter and apply it consistently.

Intercoder reliability. If you are the sole coder (as most thesis students are), you cannot calculate formal intercoder reliability. Acknowledge this as a limitation. You can partially mitigate it by: (a) having your supervisor or a peer code a small subsample to check your interpretations, (b) documenting your coding decisions thoroughly, and (c) discussing borderline cases transparently in your analysis.

Step 5: Analyze and Interpret

Once coding is complete, move from description to analysis:

Frame distributions. What are the most and least common frames in your corpus? Present this quantitatively (tables, charts showing frequencies and percentages).
Patterns across sources. Do different outlets, speakers, or document types favor different frames? This is where comparative analysis becomes powerful.
Patterns over time. Do frames shift during the period you are studying? Are there critical moments where framing changes?
Qualitative analysis of frames. For each frame, present representative examples from your data. Show how the frame works in practice — what language, arguments, and narrative structures are used? This qualitative dimension is essential; numbers alone do not tell the full story.
Absent frames. What perspectives are missing from the discourse? What is not said can be just as significant as what is. If a frame that you expected to find based on the literature is absent or marginal, that is a finding worth discussing.

Exercise: Draft your codebook

Develop a preliminary codebook for your framing analysis. For each frame, write:

A name and one-sentence definition
At least three textual indicators (keywords, phrases, argument types, or narrative patterns)
One example — real or hypothetical — showing how the frame would appear in a text from your corpus

Then write a short paragraph (150-200 words) explaining your coding rules: What is your unit of analysis? How will you handle texts that contain multiple frames? What is the threshold for assigning a frame?

Bring this codebook to your next supervision meeting.

Full Article vs. Headline Framing

Students frequently ask whether they should analyze full articles or only headlines. This is not a minor technical question — it is a methodological decision with real consequences for your findings.

Headlines and full articles often frame issues differently. Headlines are crafted to attract attention and are often written by editors, not journalists. They may emphasize conflict, drama, or controversy in ways that the article body does not sustain. A headline might frame an event as a crisis while the article provides a more nuanced treatment. Conversely, a neutral headline might sit atop an article with strong framing throughout.

When to analyze headlines only:

Your research question is specifically about how issues are presented at first glance — what audiences encounter when scanning a newspaper or news feed
You have a very large corpus and need to manage scope (headline analysis allows a bigger sample)
You are studying the “attention-grabbing” dimension of framing — how editors choose to package stories

When to analyze full articles:

You want to understand the complete framing of an issue, including causal attributions, supporting arguments, and recommended solutions — most of which appear in the body text
Your corpus is manageable in size (under ~150 articles for a BA thesis, under ~200-250 for an MA thesis)
You are interested in the depth and complexity of framing, not just the initial signal

When to analyze both:

You want to compare headline framing with article-body framing as part of your research design (this can be a research question in itself)
You suspect that editorial packaging diverges from journalistic treatment of the issue

Practical recommendation. For most thesis-level projects, analyze full articles. Headlines alone rarely capture the full framing structure — particularly causal interpretations and treatment recommendations, which tend to appear in the body text. If your corpus is too large to analyze in full, it is usually better to narrow the corpus (fewer articles, tighter time period) than to analyze only headlines from a larger sample.

If you do analyze headlines, be transparent about what this choice captures and what it misses.

Structuring Your Thesis

A framing analysis thesis follows a standard structure, but each chapter has specific content requirements. Below is a guide for how your framing analysis maps to thesis chapters.

Introduction

Present the issue and explain why its framing matters
State your research question clearly
Preview your approach and the structure of the thesis

Literature Review

Review the scholarly debate on your topic (not just framing theory — the substantive issue itself)
Situate your study within the framing literature: What do we already know about how this issue is framed? What gaps remain?
If using specific frames from prior research, introduce them here and explain their origins

Analytical Framework / Methodology

This is the chapter where you lay out the architecture of your analysis:

Approach. Explain whether you are using a deductive, inductive, or combined approach, and justify the choice
Corpus. Describe your source selection, time period, sampling strategy, and corpus size, with justification for each decision
Codebook. Present your frames with definitions, indicators, and examples. If deductive, explain where the frames come from. If inductive, describe the iterative process through which frames were developed
Coding procedure. Explain your unit of analysis, how you handle multi-frame texts, and any reliability measures
Limitations. Discuss methodological limitations honestly (single coder, language constraints, corpus boundaries)

Findings

Structure your findings around the patterns in your data. Common organizational strategies:

By frame. Dedicate a section to each frame: its frequency, where it appears, how it manifests in the text (with examples), and how it relates to the other frames. This works well when you have a manageable number of distinct frames.
By source or outlet. Compare framing across different media, speakers, or document types. This works well for comparative research designs.
By time period. Trace how framing evolves across key moments or phases. This works well for studies of framing during crises, campaigns, or policy processes.

In every case, combine quantitative presentation (tables showing frame frequencies and distributions) with qualitative analysis (representative excerpts demonstrating how each frame operates). The qualitative examples are what make framing analysis more than just counting categories.

Conclusion

Answer your research question directly
Summarize the main framing patterns you found
Discuss what these patterns mean — what do they tell us about how the issue is constructed in public discourse?
Connect your findings back to the literature: Do your results confirm, extend, or challenge prior research?
Acknowledge limitations and suggest directions for future research

Ask yourself

Does your analytical framework chapter clearly explain every methodological decision, so that another researcher could replicate your analysis?
Do your findings chapters present both the quantitative distribution of frames and qualitative examples of how each frame works?
In your conclusion, do you move beyond summarizing results to discussing what the framing patterns mean for your broader research question?

Example from the Literature

To see how framing analysis works in practice, consider Semetko and Valkenburg (2000), one of the most widely cited studies in the field. They analyzed 2,601 newspaper stories and 1,522 television news stories covering European politics — specifically, the Amsterdam meetings of European heads of state in 1997. Using a deductive approach, they tested for the presence of five generic news frames: conflict, human interest, economic consequences, morality, and responsibility. Each frame was operationalized through a set of yes/no indicator questions (e.g., “Does the story reflect disagreement between parties/individuals/groups/countries?” for the conflict frame), which were applied systematically to every item in the corpus.

The study found that the responsibility and conflict frames were the most prevalent in both press and television coverage, while the morality frame was the least common. It also revealed significant differences across media types: serious newspapers used the responsibility frame more frequently, while sensationalist outlets relied more heavily on the human interest frame. The study demonstrates several features worth emulating — a clearly justified deductive frame set, transparent operationalization through specific indicators, and systematic application across a large corpus that enables meaningful comparison across outlet types.

Common Pitfalls

These are the problems that come up most often in student framing analyses. Avoiding them will significantly strengthen your thesis.

1. Confusing frames with topics. This is the single most common mistake. A frame is not what a text is about — it is how the text presents and constructs the issue. “Climate change” is a topic. “Climate change as economic burden” or “climate change as existential threat” are frames. If your “frames” could serve as subject headings in a library catalogue, they are probably topics, not frames.

2. Defining frames too vaguely. If a frame definition is so broad that almost any text could be coded under it, it is not analytically useful. Each frame should have a clear definition and specific indicators. Test this: Could someone unfamiliar with your research apply your codebook and reach similar coding decisions?

3. Inconsistent coding. Applying different standards at different points in your analysis — coding more generously at the start, more strictly at the end, or shifting your interpretation of a frame mid-corpus. Pilot your codebook, code in a consistent order (or randomize), and revisit early coding decisions after you finish.

4. Ignoring absent frames. If a frame that the literature suggests should be present is absent or marginal in your data, that is a significant finding. Do not only report what you found — discuss what you did not find and consider why.

5. Presenting only numbers without qualitative evidence. A table showing that “the conflict frame appeared in 43% of articles” is a starting point, not a finding. You must show how the conflict frame actually operates in your texts — what language, arguments, and narrative structures are used. The qualitative examples are what demonstrate that your coding was valid and that the frames are real patterns, not artifacts of your categories.

6. Failing to justify corpus boundaries. Why these sources and not others? Why this time period? If you cannot explain your selection decisions, reviewers will question whether different choices would produce different results. Every boundary should be justified in your methodology chapter.

7. Treating frames as mutually exclusive when they are not. Texts frequently contain multiple frames. If your coding scheme forces a single-frame assignment but your data is more complex, you will lose important information. Decide in advance how to handle multi-frame texts and document your decision.

Key Readings

These are the essential references for a framing analysis thesis. You do not need to cite all of them, but you should be familiar with the core works.

Foundational works

Entman, R. M. (1993). “Framing: Toward clarification of a fractured paradigm.” Journal of Communication, 43(4), 51-58. — The most widely cited definition of framing. Start here.
Gamson, W. A., & Modigliani, A. (1989). “Media discourse and public opinion on nuclear power.” American Journal of Sociology, 95(1), 1-37. — Introduces the concept of “media packages” and framing devices. A model for how to study frames in media discourse.
Iyengar, S. (1991). Is Anyone Responsible? How Television Frames Political Issues. University of Chicago Press. — The episodic/thematic distinction. Essential for anyone studying news framing.

Methodology and application

Semetko, H. A., & Valkenburg, P. M. (2000). “Framing European politics.” Journal of Communication, 50(2), 93-109. — Develops and applies the five generic news frames. The standard reference for deductive framing analysis.
Matthes, J., & Kohring, M. (2008). “The content analysis of media frames.” Journal of Communication, 58(2), 258-279. — A rigorous quantitative approach to identifying frames inductively. Useful for MA students considering cluster analysis.
D’Angelo, P., & Kuypers, J. A. (Eds.). (2010). Doing News Framing Analysis. Routledge. — An edited volume with practical examples of different approaches to framing analysis. Good for seeing the range of methods in practice.

Theory and review

Chong, D., & Druckman, J. N. (2007). “Framing theory.” Annual Review of Political Science, 10(1), 103-126. — A comprehensive review of framing theory from a political science perspective. Useful for understanding framing effects and the broader theoretical landscape.
Scheufele, D. A. (1999). “Framing as a theory of media effects.” Journal of Communication, 49(1), 103-122. — Distinguishes frame building from frame setting and connects framing to agenda-setting theory.
de Vreese, C. H. (2012). “New avenues for framing research.” American Behavioral Scientist, 56(3), 365-375. — Reviews the state of framing research and identifies directions for future work. Helpful for positioning your study.

How to use these readings. At minimum, read Entman (1993) and one of the methodology pieces (Semetko & Valkenburg for deductive work, Matthes & Kohring for inductive work). Then look for framing studies on your specific topic — they will show you how the method has been applied in your area and may provide the frames you build on.

Framing analysis often overlaps with or complements other methods. Depending on your research question, you may want to explore:

Discourse Analysis — Goes deeper into how language constructs meaning, identity, and power relations. Where framing analysis asks “how is this issue presented?”, discourse analysis asks “how does language constitute social reality?” If your interest is in the ideological or power dimensions of communication, discourse analysis may be more appropriate — or may complement framing analysis.
Building a Corpus — Every framing analysis depends on a well-constructed corpus. This guide covers source selection, database access, sampling strategies, and data management. Consult it before finalizing your research design.