Cultivation Theory

Story

1968: A commission, one season of television

The year is 1968. America is in turmoil, with violence and unrest across the country, and everyone is asking, “where is all this violence coming from?” One of the suspects is television.

That year the National Commission on the Causes and Prevention of Violence is set up. The Commission does not just survey the existing research on violence on television; it also funds a single new study of its own: a content analysis of how much violence there was, and what kind, in the prime-time programmes of the 1967–68 season. The work is run at the University of Pennsylvania’s Annenberg School for Communication by George Gerbner.

A year later, in 1969, Congress sets aside 1 million dollars and establishes the Surgeon General’s (the US chief public health officer’s) Scientific Advisory Committee on Television and Social Behavior. A total of 23 projects are supported; Gerbner’s work is one of them.

Here is the interesting part: this apparently one-off count of violence is not dropped. It is repeated every year and grows into a long-running monitoring programme called Cultural Indicators. And the question slowly shifts, from “Do people who watch violence imitate it?” to “How does someone who has watched the same stories for years see the world?”

Myth check

“The message hits, the effect follows”?

What people assume: A media effect is a change you can see right after a message. Someone watches a violent scene and either imitates it or is unaffected; you measure it in the lab and you’re done.

What Gerbner and the theorists say: That framework defines “effect” only as change. But the media have no “before”; they are an environment we are born into. In the documentary, Michael Morgan relays Gerbner’s view this way. A “before and after” measurement is not much use for something that has no before.

The cultivation metaphor says exactly this: a single seed does not yield a crop in a day; it takes repetition, water and time. According to the theorists, the effect comes not from a downpour of single messages but from stable, repeated and pervasive patterns spread over years.

A note on history: according to Shanahan and Morgan, the “hypodermic needle” idea was never seriously defended as a theory by real researchers; it was a caricature that the limited-effects school argued against. Cultivation looks for a third way between “the media do nothing” and “the media hit like a bullet”. (This is the defenders’ reading of history; for the hypodermic model itself, see The Hypodermic Needle and Magic Bullet Model.)

Card

Who’s who? A team effort

Cultivation is often told as if it were one person’s invention, but it is really the work of a team that grew at the Annenberg School.

• George Gerbner: the name at the head of the project; dean of the school from 1964 to 1989.
• Larry Gross: ran Cultural Indicators together with Gerbner; co-author of the 1976 Violence Profile article.
• Michael Morgan and Nancy Signorielli: co-authors of the 1980 “mainstreaming” article; over the following decades they both defended and updated the theory. Signorielli was a member of the project’s original team and in 1974 published an early and frequently cited study of television characters.
• Critics: Paul Hirsch and Michael Hughes (1980–81), W. James Potter (1993, 2014).

A short timeline:

• 1967–68: first violence content analysis
• 1969: Surgeon General’s committee
• early 1970s: first national sample survey
• 1976: the Violence Profile
• 1980: “The ‘Mainstreaming’ of America”
• 1997, 2010, 2021: meta-analyses and state-of-the-field reports

Don’t mix them up: Gerbner looks at television’s cultural effect; agenda-setting (Agenda-Setting), which studies which issues come to be seen as important, is a separate line of research.

Lab

The one-sentence claim and a thought experiment

The theory in one sentence: Heavy television viewers, compared with demographically similar but lighter viewers, tend to perceive the real world in line with the most common and recurring messages of the on-screen world.

Three things are hidden in this sentence: (1) “heavy and long-term” viewing; (2) beliefs about the real world, not about programmes; (3) “compared with similar but lighter viewers”, that is, comparison and control.

A thought experiment (hypothetical). Imagine you have two relatives who are very much alike in age, education, income and where they live. One spends a few hours a week in front of a screen, the other hours every day. You ask both: “Do you think most people can be trusted?” What the theory predicts is that the heavy viewer’s answer will come out a little closer to the rules of the on-screen world. But a single question does not settle the “why”; researchers try to test this in samples of hundreds of people, holding other factors constant.

What it does not say: It does not say that everyone who watches will think the same thing, or that a single programme will transform anyone.

Why is it controversial? As Shanahan and Morgan also write, the hypothesis looks so reasonable at first glance that you may ask, “does that really deserve a book?” But after more than twenty years of testing and criticism it turned out to be neither this simple nor this obvious; it became one of the most debated approaches in communication research.

Card

Why “cultivation”? Why is television a “storyteller”?

“Cultivation” means agriculture: sowing, watering, growing crops. This metaphor, chosen by Gerbner, says that influence forms not in one shot but through repetition and time. According to Shanahan and Morgan, the concept is a way of talking about “influencing” without saying “effect”. One of Gerbner’s often-quoted lines is: “Cultivation is what a culture does”.

According to Gerbner, human beings live by stories. Most of what we know we have not lived ourselves; we know it from stories. In the past it was the family, the teacher and the priest who told the stories. Theorists argue that today television tells most of the stories to most people most of the time.

Gerbner says television tells three kinds of stories:

1. Stories about how things work: fiction (series, films).
2. Stories about how things are: the news.
3. Stories about what to do and what is valuable: advertisements.

Together the three build the mediated culture we live in. This is why the theory asks not “what does a single programme do?” but “what does this story system, taken as a whole, teach?” The theorists themselves say this has a critical side: the power of storytelling is in the hands of a shrinking group of commercial institutions.

Step by step

A three-pronged design: who produces it, what does it show, what does it leave behind?

The Cultural Indicators project rests on three questions. Gerbner formulated this framework in 1973; each question calls for its own method.

1. Institutional process analysis: Under what pressures, constraints and processes are media messages selected, produced and distributed?
2. Message system analysis: Count the recurring patterns in the content: who appears (demographics), what they do, who does what to whom, which relationships and images keep repeating. message system analysis
3. Cultivation analysis: Is paying attention to this message system related to viewers’ conception of social reality? cultivation theory

The clever part of the design: each of the three questions has consequences for the other two. According to Gerbner, the “effect” of the media should be sought not in short-term attitude change but in the reciprocal relationship among the institutions that produce messages, the message systems, and the image structures of the culture.

Why “indicators”? Economists take the pulse of the economy with indicators such as inflation and unemployment; Gerbner wanted the cultural counterpart, a barometer of cultural issues. He also thought of the project as a relatively independent “Third Voice”, standing outside the clashes of political and commercial interests.

In Turkish studies these three components appear as “kurumsal süreç çözümlemesi, mesaj sistem çözümlemesi ve yetiştirme çözümlemesi” (institutional process analysis, message system analysis and cultivation analysis).

Card

Heavy viewers, light viewers and the “TV answer”

The interesting thing about cultivation surveys is that their questions are not about television. According to Morgan, the questions are about life: safety, values, attitudes. For example: “Would you be afraid to walk the streets of your neighbourhood at night?”.

Every question has two kinds of answer: the answer that fits the reality of the television world (the “TV answer”) and the answer that fits the official statistics. Then participants are grouped by how much television they watch: light, medium and heavy viewers. The cultivation differential is the difference in the percentage of TV answers between the two extreme groups. Variables such as gender, age, education and income are controlled heavy viewers.

How many hours count as “heavy” varies from study to study. For example, in Özer’s study of high-school students in Ankara, those who watched two hours or less were counted as “light” viewers and those who watched more than two hours as “heavy” viewers. Books may mention other thresholds too, such as “four hours”; instead of memorizing a number, focus on the logic: a relative comparison.

The aim is to find out whether heavy viewers give more “TV answers” than light viewers with similar demographic characteristics. A “TV answer” does not mean a wrong answer; it means the answer that fits the lesson of the on-screen world.

Remember: this does not say “people who watch television are aggressive”. What is being compared is beliefs about the world.

Story

Mean world syndrome: would you go out at night?

When Gerbner and his team split the survey results by amount of viewing, they saw the same pattern again and again: heavy viewers were more insecure, more fearful. Morgan sums it up this way: in almost every case, heavy viewers show a greater sense of insecurity. Gerbner calls this the mean world syndrome; viewers internalize a sense of danger, mistrust and meanness: “a sense of danger, of mistrust, of meanness in the world”. In Britannica’s definition: believing that the world is more violent and mean than it actually is.

Another of Gerbner’s terms is “happy violence”: violence that is fast, spectacular, witty and ends happily, in other words violence that is meant to be entertaining. Violence that is this “sweetened” on screen may normalize the danger of the world.

An ironic detail: Gerbner does not consider the pushing of people into real violence by television violence to be the main issue; he says the direct contribution is negligible. What really interests him is fear: in 1981 he argued in Congress that frightened people are more dependent and can be more easily steered and controlled.

In Turkish studies the concept appears as “kötü dünya sendromu” (e.g. Özkan 2017). One more caution: this is a finding of association; the debate on causality is in the critique section.

Myth check

What does the theory assume, and what does it not?

What people assume: “Cultivation is the theory that says a child who watches a violent film will become aggressive.”

What the theorists say: Shanahan and Morgan write explicitly: cultivation does not study the effect of a news bulletin or an election ad on vote choice, of an advertisement on product sales, of aggression or nightmares after a violent programme, or of a series on fashion. The focus is long-term, total exposure.

The theory assumes the following:

• Television (especially fiction) offers consistent, repetitive, pervasive and unavoidable patterns of messages.
• It is assumed that messages are largely uniform and that viewing is not selective; critics debate precisely these two assumptions.
• Effects are not short-term but slow, indirect and cumulative.

An important subtlety: the theorists themselves do not say that the effect is a one-way flow from television to viewer; there is a continuing, dynamic interaction between message and context.

There is also a “hidden” assumption: Shanahan and Morgan regard the theory as a critical theory and write that they assume communication institutions are in the hands of social, cultural and above all economic elites, and that messages serve those elites’ purposes. They also say themselves that they have not tested these two assumptions directly and empirically.

Step by step

A cultivation study, step by step

Seeing how it runs on paper is the shortest way to understand the theory:

1. Recording and coding: Every year, one week of prime-time and weekend daytime fiction programmes is recorded; content analysis counts who does what.
2. The television world: The results yield a portrait of the on-screen world.
3. Comparison with reality: Where possible, this portrait is compared with official statistics.
4. Survey questions: A national sample is asked questions that reflect the “lesson” of the television world; each question has a “TV answer”.
5. Grouping: Participants are divided into light, medium and heavy viewers by how much television they watch.
6. Control: The comparison is balanced with variables such as gender, age, education, income and place of residence.
7. Cultivation differential: The percentage difference in “TV answers” between heavy and light viewers is calculated.

The first national sample survey was carried out in the early 1970s, funded by the National Institute of Mental Health (NIMH). So the order is: first the content, then the viewer.

In short, the flow is: content → real statistics → survey → comparison. If you skip a step (for example, if you only run a survey without a content analysis), what you have is no longer a full cultivation study.

Lab

Two fine-tunings: mainstreaming and resonance

In the early years the theory looked one-directional, like “heavy viewers are more afraid”. As criticisms came in, two concepts came to the fore; we will cover the details in the lesson Cultivation: Critiques and Later Findings, and here is the gist.

Mainstreaming. Heavy viewing erodes differences of opinion that stem from other factors. Because television is the shared storyteller of society, heavy viewers move closer to a common “television mainstream”. For example, on the question of political outlook, heavy viewers more often call themselves “moderate” and avoid saying “liberal” or “conservative”.

Resonance. If everyday experience resembles the messages on television, the effect gets a “double dose”. According to Özer’s account, the relationship between viewing and fear is strongest among people who live in city centres with high crime rates.

An example of mainstreaming from Özer’s account: looking by income, the heavy viewers in the middle and upper income groups come closer to each other; in terms of education, since the less educated already experience alienation, no additional effect of television viewing is seen (the “ceiling effect”).

Mini lab (hypothetical): Picture one person who lives in a high-crime neighbourhood and watches a lot of television, and another who lives in a safe district and also watches a lot. According to resonance, whose fear would be more strongly tied to television?

Story

The Violence Profile: the first big report (1976)

In 1976 Gerbner and Gross published “Living with Television: The Violence Profile” in the Journal of Communication. According to Shanahan and Morgan, this was the publication that contained the first results of cultivation analysis.

What did this profile show? According to Britannica, the Violence Profile aimed at continuously monitoring violence in prime time and examining who was harmed and who did the harming. Basing himself on these profiles, Gerbner showed Congress that the amount of prime-time violence changed very little from one season to the next. The demographics of the television world were also striking: Gerbner had noticed that the screen foregrounded young white men, and that women, older people and people of different skin colours rarely appeared as main characters.

On the viewer side, Morgan’s summary is that heavy viewers are more insecure in almost every case. A study of 1,280 children in Perth, Australia, repeated this relationship only among adolescents (details in the deep card).

I am deliberately not giving percentages: different figures circulate among the sources, and I could not verify them from the primary text.

Card

How big is the effect? Verified numbers

There are hundreds of studies on cultivation, so the most solid way is to look at meta-analyses. A meta-analysis converts the results of separate studies into a common measure and summarizes them together; this common measure is called the “effect size”.

• 1997: Morgan and Shanahan summarized twenty years of cultivation research with a meta-analysis; the average cultivation effect came out at .09.
• 2010: The same authors reported that more than 125 new studies had been published since 2000.
• 2021: Hermann, Morgan and Shanahan carried out a meta-analysis of cultivation studies from the 1970s to the present: 3,842 effect sizes from 406 independent samples, with an overall effect of .107. The effect stayed within a very narrow range over the decades; that is, although television changed a great deal, the relationship did not disappear.

These values are statistically small (this is our own general statistical reading). Two things are verified: the effect is not zero, and it has not visibly faded over the years. Critics, for their part, argue that the relationship may be explained by other factors; we will see this in the critique section.

A caution: these meta-analyses are based on different sets of studies; do not read .09, .107 and (for social media) .123 as a simple “growth story”. Each figure is a summary of its own set of samples.

Story

1980: Same data, different conclusion

In 1980 two researchers took another look at the Gerbner team’s findings using national survey data.

Michael Hughes reanalysed the alienation and fear-of-walking-near-home items in the 1975 and 1977 data of the General Social Survey (GSS). When he controlled variables such as age, sex, race, income, education, church attendance and association membership simultaneously, he reported that the effects the Gerbner team had found largely disappeared.

Paul Hirsch also pointed to inconsistencies in items, sampling and coding; he argued that when controls are applied simultaneously rather than one at a time, no linear relationship remains between amount of viewing and the “TV answer”, and concluded that the cultivation hypothesis lacks empirical support.

The Gerbner team did not stay silent: in Communication Research they published a reply titled “A Curious Journey into the Scary World of Paul Hirsch”. Shanahan and Morgan write that mainstreaming was one of the main responses to these first criticisms.

Our reading: this was mostly a methodological debate about control strategy; both sides were slicing the same data differently.

Card

Causality, uniform messages and the problem of measurement

The best-known branch of the criticism is causality. Doob and Macdonald (1979) ran a door-to-door survey in high-crime and low-crime neighbourhoods in Toronto. Pooling all the neighbourhoods, they repeated the earlier finding; but the relationships within neighbourhoods were non-significant. That is, once the actual crime rate was controlled, there was no general relationship between viewing and fear. A third factor (the real danger of the neighbourhood) could be explaining both viewing and fear. According to Özer, this finding became the basis for Gerbner developing the concept of resonance.

On the conceptual front, W. James Potter in 1993 examined the theory through assumptions such as uniform messages, non-selective viewing, linearity, causality, mainstreaming and resonance. In 2014 he argued that “cultivation” actually corresponds to three different things.

There are voices in the Turkish literature too. İrfan Erdoğan writes that the violence Gerbner deals with stays within a very narrow frame, and that there are much more important forms of violence. Yaylagül notes that the project’s method is traditional content analysis and survey, and that it can be considered administrative research lacking a historical-institutional context.

The defenders’ reply: cultivation is a reciprocal process, and some “causal” criticisms miss the main point.

Card

Cultivation in the age of streams and feeds: documented findings

In Gerbner’s world there were a few channels and everyone watched roughly the same stories. In today’s personalized feeds, this “shared message system” assumption becomes harder to hold. Morgan, Shanahan and Signorielli (2015) discuss these changing conditions. Documented findings:

• Television is still related: In the 2021 meta-analysis the effect remained stable for decades.
• Social media: The 2023 meta-analysis found 460 effect sizes from 66 independent samples, with an overall effect of .123. In studies of exposure to specific content the effect was higher than for general use; it was also higher in young samples and in Asian samples; when active use was measured, it was lower.
• Gaming: In a one-month online game experiment, participants’ perception of danger in the real world changed, but only for dangers seen in the game world.
• Facebook: Over 2010–2015 privacy attitudes relaxed; among heavy users perceived risk increased, while among light users it stayed constant.

Our interpretation (not documented): in a personalized feed the “same story for everyone” assumption weakens; on the other hand, the finding that exposure to a specific type of content gave a stronger effect (2023) may be a hint in exactly this direction. I have no verified study on streaming services such as Netflix, so I make no claim about them here.

Lab

Mini lab: the message system of your own feed

You can try the first prong of cultivation (the message system) at home. This is not a scientific study, but an exercise for getting a feel for the logic.

What to do (three days):

1. Each day, observe the first 30 items in your feed (news, video, social media, series). Only your own feed; do not note down other people’s accounts.
2. For each item choose just one category: danger/disaster/crime, everyday life, entertainment, other.
3. Over the three days, count how many were danger/disaster/crime.

Then ask yourself: Is this higher than the frequency of danger I actually meet in daily life? Which stories repeat most often?

What is missing: A scientific message system analysis calls for a representative sample, consistent coding rules and more than one coder. Your exercise has none of these; so you cannot conclude from the result that “feeds produce a mean world”. You are only learning to ask the right question.

Bonus thought: With a friend, compare only the numbers (e.g. “how many out of 30 were danger?”); there is no need to share content or accounts. Why do the numbers come out different? How does this difference affect cultivation’s assumption that “everyone sees the same stories”?

Card

Cultivation studies in Turkey

Cultivation has been tested in Turkey as well. In Turkish the theory appears under the names “ekme” (“sowing”; Yaylagül, Erdoğan), “yetiştirme” (“raising”; Özer, Ercan and Demir) and “kültürel göstergeler” (“cultural indicators”).

• Ankara high schools (2005): In a field study with students from two high schools in Ankara, Turkey’s capital, Ömer Özer tested whether the perception of a dangerous world depends on living in violence-heavy places or on television, and took up the idea of resonance. The author reports that the results confirm the propositions of Gerbner and his colleagues.
• Anadolu University Faculty of Science: Ercan and Demir (2015) ran a survey that drew on the cultivation analysis component. In a study with students of the same faculty, Özer (2019) writes that heavy viewers regarded the police as necessary for this world and interpreted this as a sign that hegemony had been realized.
• Baku (2024): Özer and Jafarli first carried out a message system analysis of the prime time of Turkish channels (ATV, Kanal D, Star, Show TV, all national channels in Turkey): elements such as crime, violence, death, security forces and male dominance came to the fore. The survey questions were prepared according to these findings and applied to 500 university students (300 women, 200 men) in Baku, the capital of Azerbaijan. The authors concluded that television has a cultivating and mainstreaming role.
• Critical reading: Erdoğan (1998) evaluated Gerbner’s approach in terms of theory and method.

Note: these are small, local samples; do not read the results as national generalizations. The interpretations given belong to their authors.

Card

How it’s asked in exams

Frequent patterns

• “Who is the Cultural Indicators project and the cultivation (‘ekme’) theory associated with?” → George Gerbner (Annenberg, Pennsylvania).
• Concept matching: mainstreaming, resonance, mean world syndrome, heavy viewer.
• Theory comparison: cultivation (long-term perception of reality) versus agenda-setting (which issues are seen as important) and the hypodermic model (a single message, instant effect).

Tempting wrong answers

• “People who watch violence commit violence” → not cultivation’s claim; its claim is fear and perception of the world.
• “Hirsch supported the theory” → no, in 1980–81 he criticized it through a reanalysis.
• “Potter was Gerbner’s co-worker” → no, he was a conceptual critic.
• Years: Violence Profile 1976; mainstreaming article 1980.

Alternative names (in books and articles)

• Yaylagül (2006): “Kültürel Göstergeler ve Ekme Kuramı” (“Cultural Indicators and Cultivation Theory”); he also mentions the distinction between first- and second-level effects. In Polat’s textbook (unit 8) the title “Kültürel Göstergeler ve Ekme Modeli” (“Cultural Indicators and the Cultivation Model”) appears.
• Özer and others: “yetiştirme kuramı”, “yetiştirme çözümlemesi” (“cultivation theory”, “cultivation analysis”).
• Özkan (2017): “kötü dünya sendromu” (“mean world syndrome”).

In Turkish sources it may be written that Cultural Indicators began in the “mid-1960s” (“1960’ların ortası”); the first violence analysis, however, is the 1967–68 season.

Summary

Summary card

• 1967–69: In the context of the violence commission and the Surgeon General’s committee, Gerbner’s annual television content analysis begins; Cultural Indicators is born.
• Three prongs: institutional process analysis, message system analysis, cultivation analysis.
• Core claim: heavy viewers perceive the real world in line with the television world; the effect is spread over years.
• Concepts: heavy viewer, TV answer, mean world syndrome, mainstreaming, resonance.
• Evidence: meta-analyses; average effect .09 (1997) and .107 (2021), within a narrow range for decades.
• Critique: Hughes and Hirsch (1980–81), Doob and Macdonald (1979), Potter (1993, 2014); causality and measurement.
• Today: social media (.123), games and feeds; the specific type of content may be decisive.
• Exam: “perception of the world”, not “imitation”; years: 1976 and 1980.