Rationale

Cultivation theory is the idea that exposure to different television content can influence ones perception of reality. The study will explain whether or not the amount of television a person watches is in any way related to their perception of the commonality of certain occupations in the U.S. population.

Hypothesis

Participants who report more weekly hours of television viewing will believe a higher percentage of the U.S. population as being employed in law enforcement/criminal justice, medicine, or emergency response services.

Methods

The study included 400 participants. Participants average weekly television viewing was measured over a six-month period. Participants also estimated the percentage of the U.S. population employed in law enforcement/criminal justice, medicine, or emergency response services.

A bivariate linear regression was conducted in R. Weekly television viewing, “video” was the independent variable, and the estimated percentage of workers in the specified occupations marked as “pct” was the dependent variable.

Results

The regression analysis showed a positive relationship, quite significant at that, between weekly television viewing and the participants estimates of percentage of people working in law enforcement/criminal justice, medicine, or emergency response services. The regression coefficient for T.V. viewership was 0.844, with a p-value less than 2.2e-16.

The model had an R-squared value of 0.3111, meaning that weekly television viewing explained approximately 31.1% of the participants estimates.

Graph

Conclusion

The results support the hypothesis that participants who watched more television tended to estimate a higher percentage of the population as working in law enforcement/criminal justice, medicine, or emergency response services. The overwhelming positive correlation between television viewing and the estimates provided shows a direct consistency between T.V. exposure and the perceptions of social reality.

R Code


# Read the data from the web
FetchedData <- read.csv("https://github.com/drkblake/Data/raw/refs/heads/main/Cultivation.csv")

# Save the data on your computer
write.csv(FetchedData, "Cultivation.csv", row.names = FALSE)

# Remove the data from the environment
rm(FetchedData)

# Load the data
Cultivation <- read.csv("Cultivation.csv")

# View the first few rows
head(Cultivation)

# Run the bivariate regression
model <- lm(pct ~ video, data = Cultivation)

# View the regression results
summary(model)

# Load ggplot2
library(ggplot2)

# Create the graph
ggplot(Cultivation, aes(x = video, y = pct)) +
  geom_point() +
  geom_smooth(method = "lm", se = TRUE) +
  labs(
    x = "Weekly hours watching television",
    y = "Estimated percentage of law enforcement, medical, and emergency workers",
    title = "Television Viewing and Perceptions of Occupational Prevalence"
  )