Written and maintained by CASRAI Editorial Board
Published · Last updated
On this page
Semiotic analysis is a qualitative method that studies how signs (words, images, gestures, objects, layouts) produce meaning, by examining what a sign stands for, to whom, and in relation to which other signs. It is an interpretive method: the result is a reasoned reading of a text, not a measured value, so there is no formula, test statistic or software routine to run.
This guide is for graduate and social, health and humanities researchers who need to choose the method, carry out an analysis in a defensible sequence, and describe it in a methods section. It uses the sign models of Saussure and Peirce and the analytic questions Daniel Chandler sets out in Semiotics for Beginners.
What is semiotic analysis?
Semiotic analysis treats a text (an advertisement, a road sign, a clinical leaflet, an interface screen, a photograph) as a set of signs and asks how those signs, in combination, support particular meanings. The analyst does not count features. The analyst identifies the signs, classifies how each relates to what it stands for, and argues for an interpretation using the text itself as evidence.
The method comes from two traditions of sign theory, set out below. Fields that use it include media and communication studies, cultural studies, literary studies and design research.
The two sign models: Saussure and Peirce
Saussure’s two-part sign
In the description Chandler gives in Semiotics for Beginners (Signs), Saussure defines the sign as the whole that results from associating a signifier (the form the sign takes) with a signified (the concept it represents). The sign is the pairing, not either half.
Peirce’s three-part sign
Peirce’s account, as summarised in the Stanford Encyclopedia of Philosophy entry by Albert Atkin, defines a sign as anything so determined by something else, its Object, that it produces an effect on a person, the interpretant. Peirce used representamen for the signifying element (the sign-vehicle). He also classified signs by how they relate to their object:
| Peirce type | Basis of the sign-object relation (SEP wording) | Example given in SEP |
|---|---|---|
| Icon | “a mere community in some quality” | Portraits |
| Index | “a correspondence in fact” | A weathercock |
| Symbol | “an imputed character” | Conventional signs such as words |
Side by side
| Saussure (dyadic) | Peirce (triadic) | |
|---|---|---|
| Parts | Signifier + signified | Representamen (sign-vehicle), Object, interpretant |
| What the model foregrounds | The form of the sign and the concept it evokes | The relation to the object and the effect on an interpreter |
| Useful when you want to | Compare signs against alternatives in the same system | Classify how each sign refers (icon, index, symbol) and track the reader’s interpretation |
The last row is an editorial summary of how analysts commonly use each model, not a quotation. Choose one model for the whole study and say which.
How to do a semiotic analysis, step by step
- Define the text and the question. State what is being analysed (for example a single sign, a set of advertisements, an interface) and what you want to explain. Justify why this text, since the analysis cannot generalise by sampling in the way a survey does.
- Identify the signs. List the signifiers present: words, images, colours, layout, typography, sound. Treat each as a candidate sign.
- Classify the relation to meaning. With Saussure, name the signifier and signified. With Peirce, name the object and the interpretant and decide whether the sign works as icon, index or symbol.
- Ask the paradigmatic questions. In Chandler’s description, these ask which set (paradigm) each signifier belongs to, why this choice was made over the alternatives, and which signifiers from the same set are noticeably absent.
- Ask the syntagmatic questions. Chandler describes this as identifying and describing structures in the text, such as narrative, argument or montage: how the signs are ordered and combined.
- Apply the commutation test. Chandler describes it as a test of whether a change at the level of the signifier leads to a change at the level of the signified. You imagine substituting one signifier for another that could occupy the same position and assess the effect. He notes that it has evolved into a rather more subjective form of textual analysis than its origin in the substitution tests of the Prague structuralists.
- Argue the interpretation. Write the reading as a claim supported by specific features of the text, and state what other readings you considered.
Worked example (illustrative, not a study)
The following is an illustrative composite constructed to show the procedure; it reports no data and describes no real study or audience. Take a generic pedestrian-crossing road sign showing a walking figure.
- Saussure. The signifier is the drawn figure on a coloured panel. The signified is the concept “pedestrians may cross here”.
- Peirce. The figure is partly iconic, because it resembles a person, and partly symbolic, because the panel’s shape and colour are conventions a driver must have learned. Its interpretant is the driver’s understanding that they should watch for people crossing.
- Paradigmatic question. The designer chose a walking figure from the set of possible figures (a child, a person with a stick, a bicycle). What each alternative would signal is part of the reading, and the absence of the others is part of what the sign communicates.
- Syntagmatic question. The sign is read in sequence with the road markings, the kerb and other signs. The same figure in a different sequence (for example on a shop door) would not carry the same meaning.
- Commutation. Replace the figure with a different silhouette and ask whether the signified shifts. If it does, that signifier is carrying meaning at that position.
The output is a short argued reading, not a number. A real study would repeat these moves across a justified set of texts and cite the texts as evidence.
Reading the results
| If your analysis shows… | You can reasonably write… | You should not write… |
|---|---|---|
| Substituting a signifier changes the signified | That signifier contributes to the meaning at that position | That the change proves how audiences respond |
| A signifier is absent from an expected set | The absence is a notable choice worth interpreting | That the absence was intended, unless you have evidence of intent |
| Signs cluster as icons, indices or symbols | The text relies on that kind of reference | Frequencies or percentages, as if this were content analysis |
These rows are editorial guidance on the limits of interpretive claims. They follow from Chandler’s caution that the commutation test is subjective, not from a published standard.
Assumptions and common mistakes
- Treating a reading as a measurement. Semiotic analysis yields an interpretation. Another analyst may reach a different one, so make your reasoning inspectable.
- Mixing the models without saying so. Saussure’s dyad and Peirce’s triad define “sign” differently; state which you use.
- Claiming audience effects from the text alone. The analysis shows what the text makes available. Statements about how people actually read it need audience data.
- Skipping the alternatives. The paradigmatic and commutation steps depend on considering what could have been used instead.
- Unjustified selection of texts. Explain why these texts were chosen and what they can and cannot represent.
How to report a semiotic analysis
There is no standard reporting template or statistic for the method. In the methods section, state the theoretical model (Saussure, Peirce or both), the texts analysed and the basis for choosing them, the analytic steps you took (for example paradigmatic and syntagmatic questions and commutation), and how you handled competing interpretations. Cite the sign theory you rely on to its original or an authoritative source. For general structure, see the guide to the methods section in APA style, and for placing the method in a wider workflow, the data analysis process.
Because this is an interpretive method, no Excel, R or Python snippet applies.
FAQ
What is the difference between semiotics and semiotic analysis?
Semiotics is the study of signs and meaning-making. Semiotic analysis is the applied method that uses sign theory to interpret a particular text.
Is semiotic analysis qualitative or quantitative?
Qualitative. It rests on interpretation of specific signs and their relations, not on measurement or statistical testing.
Which model should I use, Saussure or Peirce?
Either is defensible. Saussure’s signifier and signified suit comparison of signs within a system. Peirce’s icon, index and symbol suit questions about how a sign refers to its object. Use one consistently and say why.
Is semiotic analysis reliable?
It is not reliable in the sense of a repeatable measurement. Chandler notes that the commutation test has become more subjective. Strengthen the work by making each step explicit and supporting claims with features of the text.
Sources
- Atkin, A. Peirce’s Theory of Signs. Stanford Encyclopedia of Philosophy (first published 2006-10-13; substantive revision 2022-08-04). plato.stanford.edu/entries/peirce-semiotics. Read 2026-11-08.
- Chandler, D. Semiotics for Beginners, “Signs” (Saussure’s signifier and signified). Read 2026-11-08 via a course mirror.
- Chandler, D. Semiotics for Beginners, “Paradigmatic Analysis” (commutation test, its Prague structuralist origin and its evolution into a more subjective form of textual analysis) and “D.I.Y. Semiotic Analysis” (paradigmatic and syntagmatic questions). Read 2026-11-08. See also Chandler, D., Semiotics: The Basics (Routledge).







