Communication theories

In short

Communication theories are organised explanations of how messages are produced, carried, received and how they change what people know, feel and do. They give journalism and mass communication students the vocabulary — source, encoding, channel, noise, feedback, effect — used in every newsroom brief, campaign plan and media research study. They are examined in nearly every BJMC/BA(JMC) university paper (Introduction to Communication, Communication Theories, Media Effects), in UGC-NET Mass Communication and Journalism Paper II, and in entrance tests for IIMC, Jamia MCRC, Symbiosis (SIMC) and Xavier's. Employers test them indirectly: agenda-setting and framing in editorial and PR interviews, uses and gratifications in audience/OTT research, development communication in government and NGO campaign roles.

Explained

What a theory is, and the levels of communication

• Theory - a set of related statements that define concepts, explain relationships between them and predict outcomes; in mass communication theories are mostly *probabilistic* (tendencies), not deterministic laws.
• Levels: intrapersonal (within one person) → interpersonal (two people) → group → public/organisational → mass communication (one source, mediated channel, large, heterogeneous, anonymous, dispersed audience) → network/many-to-many (social media, which collapses these levels).
• Elements common to all models: source/communicator, encoding, message, channel/medium, decoding, receiver, feedback, noise (physical, semantic, psychological), context, effect.
• Three traditions of theory-building: (1) *normative* - what media ought to do; (2) *scientific/empirical* - measurable effects; (3) *critical/cultural* - ideology, power and meaning.
• Indian root: the Sadharanikaran model (introduced into communication studies by I.P. Tewari, later elaborated by J.S. Yadava and Nirmala Mani Adhikary) from Bharata's *Natyashastra* - communication as *simplification into commonness*; elements are *sahridayata* (commonality of heart), *sahridaya* (receiver), *bhava/rasa* (emotion and its relish), *sandesh* (message), *doshas* (noise). It is a consensus- and emotion-centred alternative to Western linear models and is routinely asked in Indian university papers.

Linear, interactive and transactional models

1. Aristotle (4th century BCE) - speaker, speech, occasion, audience, effect; *ethos, pathos, logos* as means of persuasion. Oldest model, purely speaker-centred.
2. Lasswell (1948) - "Who says What in Which Channel to Whom with What Effect"; maps directly to control analysis, content analysis, media analysis, audience analysis and effect analysis. Linear, no feedback, assumes an effect always occurs.
3. Shannon and Weaver (1949), the "mother of all models" - information source → transmitter (encoder) → channel → receiver (decoder) → destination, with a noise source acting on the channel; Weaver added three levels of problems - *technical* (accuracy of signal), *semantic* (meaning), *effectiveness* (behaviour). Engineering origin (Bell Labs); criticised as mechanistic.
4. Osgood-Schramm (1954) - circular: both parties continuously encode, interpret and decode; introduces the field of experience - communication works only where the fields overlap.
5. Berlo's SMCR (1960) - *Source* (communication skills, attitude, knowledge, social system, culture), *Message* (content, elements, treatment, structure, code), *Channel* (five senses), *Receiver* (same five attributes as source). No feedback element; stresses source-receiver similarity (homophily).
6. Newcomb's ABX (1953) - the first model to put communication in a *social* frame: A and B orient to each other and to object X; imbalance (disagreement) drives communication to restore symmetry.
7. Westley and MacLean (1957) - adds the C role, the gatekeeper/channel between advocate A and audience B, with feedback loops (fBA, fBC, fCA) - the classic model of mass media.
8. Gerbner (1956) - verbal and diagrammatic; perception → means and control → content; introduces that the communicator's perception of an event is itself selective.
9. Dance's helical model (1967) - communication as an upward spiral: never returns to the same point, each exchange builds on all previous ones.
10. Transactional view (Barnlund, 1970) - simultaneous sending and receiving, meaning created jointly; best fits social media.

Era 1 - Powerful effects (1900-1940) and the limited-effects correction

• Hypodermic needle / magic bullet (1920s-30s): media inject uniform messages into a passive mass; evidence drawn from WWI propaganda, Nazi radio and Orson Welles's *War of the Worlds* (1938) panic. Now rejected as oversimplified.
• Propaganda analysis: Lasswell's categories, and the Institute for Propaganda Analysis's seven devices - name-calling, glittering generality, transfer, testimonial, plain folks, card stacking, bandwagon.
• Two-step flow (Lazarsfeld, Berelson, Gaudet, *The People's Choice*, 1944; Katz and Lazarsfeld, *Personal Influence*, 1955): ideas flow from media to opinion leaders and from them to the less active public; later generalised to multi-step flow. Explains why Indian political messaging works through local influencers, panchayat leaders and WhatsApp group admins.
• Individual differences / selective processes: selective exposure, selective attention, selective perception, selective retention - audiences avoid and reshape dissonant messages (Festinger's cognitive dissonance, 1957).
• Klapper's phenomenistic/limited effects (1960): media mainly *reinforce* existing attitudes; they rarely convert, and work through mediating factors.
• Diffusion of innovations (Rogers, 1962): knowledge → persuasion → decision → implementation → confirmation; adopter categories - innovators (2.5%), early adopters (13.5%), early majority (34%), late majority (34%), laggards (16%). Mass media create awareness; interpersonal channels persuade. Used in Indian agriculture extension and immunisation drives.

Era 2 - Moderate, cognitive and long-term effects

• Agenda setting (McCombs and Shaw, Chapel Hill, 1972): media do not tell us *what* to think but *what to think about*; correlation between media salience and public salience. First level - issue salience; second level (attribute agenda setting / framing) - which attributes of an issue are salient; third level (network agenda setting) - bundles of issues. Strength depends on the audience's need for orientation (relevance × uncertainty) and on agenda melding.
• Gatekeeping (Lewin, 1947; White's Mr Gates study, 1950; Shoemaker): selection and shaping of news at individual, routine, organisational, extra-media and ideological levels; news values (timeliness, proximity, conflict, prominence, human interest, consequence, oddity) are the working gates.
• Framing (Goffman, 1974; Entman, 1993): selecting some aspects of reality and making them salient - define problem, diagnose cause, make moral judgement, suggest remedy. Episodic vs thematic frames (Iyengar); priming - media coverage supplies the standards by which leaders are judged.
• Cultivation theory (Gerbner and Gross, 1970s): heavy television viewing *cultivates* a shared, television-flavoured worldview. Key constructs - mainstreaming, resonance, mean world syndrome; measured by the cultivation differential between heavy and light viewers using "TV answers". Cross-sectional, long-term, small but broad effect.
• Spiral of silence (Elisabeth Noelle-Neumann, 1974): people scan the climate of opinion with a *quasi-statistical sense*; those who feel in the minority fall silent, making the majority view appear even stronger. Hard core resists. Explains self-censorship in polarised online debate.
• Uses and gratifications (Katz, Blumler, Gurevitch, 1974): flips the question to "what do people do with media"; needs are cognitive, affective, personal integrative, social integrative and tension-release (surveillance, personal identity, personal relationships, diversion). Distinguishes gratifications sought from gratifications obtained. Criticised as too individualistic and for assuming a rational, active audience.
• Knowledge gap hypothesis (Tichenor, Donohue, Olien, 1970): as media information increases, better-educated segments gain faster, widening the gap - the basis of India's digital divide debates.
• Third-person effect (Davison, 1983): people believe media affect *others* more than themselves - the common justification for censorship demands.
• Social learning / social cognitive theory (Bandura, 1961 onward): behaviour is learned by observing models; attention, retention, reproduction, motivation. Basis of entertainment-education (Sabido method) - India's *Hum Log* (1984-85) and *Kalyani*, and radio serials like *Taru*.
• Persuasion routes: Elaboration Likelihood Model (Petty and Cacioppo) - central route (argument quality, high involvement) vs peripheral route (cues, source attractiveness); Hierarchy of effects/AIDA in advertising; Health Belief Model and Theory of Planned Behaviour in public-health campaigns.

Critical, cultural and medium theories

• Marxist base-superstructure; the Frankfurt School (Adorno, Horkheimer) - the culture industry produces standardised goods and mass deception; Marcuse, Habermas's public sphere.
• Hegemony (Gramsci) - dominance by consent manufactured in everyday culture, not force. Ideological State Apparatuses (Althusser).
• Political economy (Herman and Chomsky, 1988) - the propaganda model with five filters: ownership, advertising, sourcing, flak, and anti-communism (today, the dominant ideology). Relevant to Indian debates on media ownership and paid news.
• Encoding/decoding (Stuart Hall, 1973) - three reading positions: dominant-hegemonic, negotiated, oppositional; meaning is *polysemic*. Foundation of British Cultural Studies, with Morley's audience work and Fiske on popular culture.
• Medium theory (Marshall McLuhan) - "the medium is the message"; hot vs cool media; global village; the tetrad; Innis's bias of communication; Ong on orality and literacy.
• Semiotics - Saussure's signifier/signified, Peirce's icon-index-symbol, Barthes's denotation, connotation and myth.
• Feminist media theory - the male gaze (Mulvey), symbolic annihilation, representation audits; post-colonial theory (Said's *Orientalism*, Bhabha, subaltern studies) applied to the global flow of news.
• Public sphere and networked publics; Castells's mass self-communication and network society; Jenkins's convergence culture and participatory fandom.

Normative theories, development communication and the Indian framework

• Four Theories of the Press (Siebert, Peterson, Schramm, 1956): authoritarian (media serve the state), libertarian/free press (marketplace of ideas), social responsibility (freedom with obligations, from the Hutchins Commission 1947), Soviet-communist (media as party instrument).
• McQuail's additions: development media theory (media support nation-building, cultural autonomy; may accept restrictions) and democratic-participant theory (small-scale, horizontal, community media).
• Development communication: modernisation/dominant paradigm (Lerner, Schramm, Rogers) → dependency critique → participatory paradigm (Paulo Freire's dialogic *conscientização*, Beltrán, Servaes's multiplicity). MacBride Commission, *Many Voices One World* (1980) and the NWICO demand for a balanced flow of information.
• Indian milestones to cite: SITE (Satellite Instructional Television Experiment, 1975-76, six states, with ISRO and NASA ATS-6), Kheda Communication Project (1975-90), Jhabua development communication project, Krishi Darshan (1967), AIR's farm and school broadcasts, *Hum Log*, the Community Radio policy (open to NGOs since 2006; 500+ stations operating - numbers change, check MIB data).
• Indian legal and regulatory frame (current): Article 19(1)(a) freedom of speech with 19(2) reasonable restrictions; Press and Registration of Periodicals Act, 2023 (replaced the PRB Act, 1867; registration moved online via the Press Sewa portal); Press Council of India Act, 1978 and the PCI's Norms of Journalistic Conduct; Cable Television Networks (Regulation) Act, 1995 with the Programme and Advertising Codes and the three-tier grievance mechanism; IT Act, 2000 with the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 as amended, covering OTT self-classification (U, U/A 7+, 13+, 16+, A) and digital news publishers; Cinematograph Act, 1952 as amended in 2023 (CBFC certificates U, UA 7+/13+/16+, A, S); Bharatiya Nyaya Sanhita, 2023 replaced the IPC (defamation and related offences renumbered); Digital Personal Data Protection Act, 2023 (provisions being operationalised through rules - check the latest notification); self-regulation by NBDSA, ASCI's code for advertising and influencer disclosure, BARC India for TV ratings. A Broadcasting Services (Regulation) Bill has been under consultation and is not law as of November 2026 - verify current status before quoting it.

Key points

• Communication - a process of creating and sharing meaning, not mere transmission of signals.
• Noise - anything that distorts a message; physical, semantic (meaning) and psychological.
• Feedback - receiver's response; absent in Lasswell and Shannon-Weaver's original, central in Osgood-Schramm.
• Lasswell's formula - Who / What / Which channel / To whom / What effect; each term names a research area.
• Shannon-Weaver - technical, semantic and effectiveness levels of communication problems; source of "noise" and "redundancy".
• Berlo's SMCR - four blocks with five attributes each for source and receiver; no feedback.
• Sadharanikaran - Indian *Natyashastra*-based model built on *sahridayata*, commonality of heart, achieving *rasa*.
• Hypodermic needle - all-powerful media, passive audience; discredited but the historical starting point.
• Two-step flow - media → opinion leaders → public; explains interpersonal mediation.
• Agenda setting - media transfer issue *salience*; what to think about, not what to think.
• Framing and priming - how an issue is defined, and the standards used to evaluate actors.
• Gatekeeping - multi-level selection of news; news values are its practical criteria.
• Cultivation - heavy viewers absorb TV's worldview; mainstreaming, resonance, mean world syndrome.
• Spiral of silence - perceived minority opinion falls silent; quasi-statistical sense, hard core.
• Uses and gratifications - active audience selects media for surveillance, identity, relationships, diversion.
• Encoding/decoding - dominant, negotiated and oppositional readings of a polysemic text.
• Normative theories - authoritarian, libertarian, social responsibility, Soviet, plus development and democratic-participant.
• NWICO / MacBride - demand for a free and *balanced* flow of information against one-way news flow.

Formulas and examples

1. Entropy and redundancy of a headline alphabet. A headline generator uses 4 equally likely words.

• H_max = log₂ 4 = 2 bits per word.
• Actual use is skewed: p = 0.5, 0.25, 0.125, 0.125. H = 0.5(1) + 0.25(2) + 0.125(3) + 0.125(3) = 0.5 + 0.5 + 0.375 + 0.375 = 1.75 bits.
• Redundancy R = 1 − 1.75/2 = 0.125, i.e. 12.5% - that redundancy is what lets a reader still understand a partly garbled headline.

2. Channel capacity for a community radio link. B = 10 kHz, S/N = 1,023.

• C = 10,000 × log₂(1 + 1,023) = 10,000 × log₂ 1024 = 10,000 × 10 = 1,00,000 bits/s (100 kbit/s).
• Halving S/N to 511: log₂ 512 = 9 → C = 90 kbit/s. A 10% capacity loss for a doubling of noise power shows why engineers add power reluctantly.

3. Agenda-setting correlation. Five issues are ranked by number of news stories (media agenda) and by a public survey of "most important problem".

• Σd² = 4, n = 5 → rₛ = 1 − (6 × 4) / [5 × (25 − 1)] = 1 − 24/120 = 1 − 0.2 = +0.80.
• Interpretation: a strong positive rank correlation, the standard evidence McCombs and Shaw used - correlation, not proof of causation.

4. Cultivation differential. Of 400 heavy viewers (4+ h/day), 232 say "most people cannot be trusted"; of 600 light viewers (under 2 h/day), 216 say the same.

• Heavy = 232/400 = 58%; light = 216/600 = 36%.
• CD = 58 − 36 = +22 percentage points - consistent with mean world syndrome, though education, income and neighbourhood must be controlled before claiming cultivation.

5. Two-step flow reach. A Hindi daily reaches 8,00,000 readers; 5% act as opinion leaders and each talks to 12 others, of whom 60% recall the message.

• Opinion leaders = 0.05 × 8,00,000 = 40,000.
• Second-step contacts = 40,000 × 12 = 4,80,000; recalling = 0.60 × 4,80,000 = 2,88,000.
• Total informed ≈ 8,00,000 + 2,88,000 = 10,88,000, i.e. a 36% lift over direct readership - the quantitative sense of "personal influence".

Where it is used

• Newsrooms (The Hindu, Indian Express, NDTV, ANI, regional dailies): gatekeeping and news values in story selection; framing and priming in headline, visual and sequencing choices; inverted-pyramid redundancy as noise control.
• Advertising and PR (Ogilvy, Leo Burnett, Dentsu, Adfactors): ELM and hierarchy-of-effects in creative strategy; two-step flow as the formal basis of influencer and KOL campaigns; third-person effect in crisis messaging; ASCI code compliance.
• Government and development communication (MIB, CBC, DD and AIR, PIB, UNICEF India, SEWA, community radio): development media and participatory theory in campaigns such as Pulse Polio, Swachh Bharat, Beti Bachao Beti Padhao; diffusion of innovations in agriculture and health extension.
• Audience and market research (BARC India, Kantar, Nielsen, Ormax, OTT insights teams): uses and gratifications in binge-viewing and short-video studies; cultivation and content audits for representation.
• Media education, research and policy roles: content analysis, framing analysis and survey design for MA dissertations, UGC-NET, IIMC and PhD work; policy notes on the IT Rules, 2021 and self-regulation.
• Social-media and misinformation work (fact-check desks, platform trust and safety): spiral of silence and echo chambers in moderation policy; knowledge gap in digital-inclusion programmes.

Interview questions

• Q: What is the difference between a model and a theory of communication? A: A model is a simplified representation of the process that shows elements and their relations; a theory explains and predicts why those relations produce particular effects.
• Q: State Lasswell's formula and its research value. A: "Who says what in which channel to whom with what effect" - the five terms correspond to control, content, media, audience and effects research.
• Q: What did Shannon and Weaver add that earlier models lacked? A: The explicit concepts of encoder/decoder and noise, plus Weaver's three levels of problems - technical, semantic and effectiveness.
• Q: Does agenda setting say the media tell us what to think? A: No - it says they tell us what to think *about*, transferring issue salience; the second level (attribute agenda setting) comes closer to influencing *how* we think.
• Q: How is framing different from agenda setting? A: Agenda setting is about which issues are prominent; framing is about which aspects of an issue are selected and made salient so that a particular definition, cause and remedy seem natural.
• Q: Why is the hypodermic needle model rejected? A: It assumes a uniform passive mass and ignores selective exposure, interpersonal influence and audience differences, which empirical work from the 1940s onward disproved.
• Q: Explain cultivation theory with one Indian example. A: Heavy television viewing gradually produces a TV-like worldview; heavy viewers of crime-based news channels typically overestimate crime rates in their city, the mean world syndrome.
• Q: What does the spiral of silence predict about online debate? A: Those who sense their view is in the minority stay quiet, so the majority view looks stronger than it is and the minority grows quieter - a self-reinforcing spiral, resisted only by a hard core.
• Q: How would you use uses and gratifications to plan an OTT content slate? A: Segment viewers by gratifications sought - surveillance, diversion, identity, social connection - then test gratifications obtained after launch and reallocate commissioning to the gaps.
• Q: Which normative theory best describes Indian broadcasting policy? A: Mainly social responsibility with strong development-media elements - freedom under Article 19(1)(a) restricted by 19(2), the Programme and Advertising Codes, the IT Rules, 2021 and layered self-regulation.
• Q: What is the Sadharanikaran model and why does it matter? A: An Indian model from the *Natyashastra* treating communication as achieving commonness and shared *rasa* through *sahridayata*; it foregrounds emotion and consensus rather than linear transmission and persuasion.
• Q: What are Herman and Chomsky's propaganda model filters? A: Ownership, advertising, sourcing from official channels, flak, and an overarching ideology - together they shape what counts as news without explicit censorship.
• Q: Give one limitation of uses and gratifications. A: It over-assumes a rational, self-aware, active audience and underplays how ownership, availability and ideology limit what people can choose.

Common mistakes

• Writing "media tell us what to think" for agenda setting - the examiner looks for "what to think *about*", and for the distinction from framing.
• Mixing up Berlo's SMCR with Shannon-Weaver, or claiming SMCR has feedback - it does not.
• Treating the hypodermic needle as a current theory instead of a historical, discredited position.
• Confusing cultivation (long-term worldview from heavy viewing) with social learning (imitation of observed behaviour).
• Naming only Siebert's four normative theories and forgetting McQuail's development and democratic-participant additions - a frequent two-mark loss.
• Quoting repealed law: the PRB Act, 1867 (now the Press and Registration of Periodicals Act, 2023) and the IPC (now the Bharatiya Nyaya Sanhita, 2023); and treating the pending broadcasting bill as enacted.
• Describing theories without any example, diagram or Indian case - most university schemes award marks for a labelled model diagram and one applied illustration.

Related topics

Media industry in India · Film studies basics · Media laws and ethics · Development and rural communication · Advertising and public relations principles · Media research methods and content analysis