Rationale

Cultivation theory suggests that individuals who are frequently exposed to television will have their perceptions of social reality infused by their viewing choices. An example being that heavy television viewers tend to overestimate the incidence of certain occupations in their respective communities, such as law enforcement, medicine, or emergency services, compared to the actual statistics. Thus, the current study grounds or tests the assumption that, television viewing will predict participants’ estimates of these professions workforce percentages.

Hypothesis

H₀ (null): Weekly hours of television watching do not predict participants’ estimates of workforce percentages.

H₁ (alternative): Participants who watch more television per week tend to overestimate workforce percentages.

Variables & method

Independent variable: video — average weekly hours spent watching television.

Dependent variable: pct — participants’ estimated percentage of workers in law enforcement, medicine, or emergency response.

Method: Bivariate linear regression, which models the relationship between TV viewing hours and perceived workforce percentages.

Results & discussion

The regression analysis investigates whether the number of hours viewing television predicts participants’ estimates of percentages in the workforce.

The regression coefficient (slope) = r round(coef(reg_model)[2], 2)

The intercept = r round(coef(reg_model)[1], 2)

R-squared = r round(summary(reg_model)$r.squared, 3)

The p-value = r signif(summary(reg_model)$coefficients[2,4], 3)

If the p-value is less than 0.05, we can conclude that viewing television is significantly associated with higher estimates of percentages in the workforce. This supports cultivation theory, as it suggests that those who view television more will see occupations as being more frequent than they actually are.

Code:

# Load dataset
FetchedData <- read.csv("https://github.com/drkblake/Data/raw/refs/heads/main/Cultivation.csv")
write.csv(FetchedData, "Cultivation.csv", row.names=FALSE)
data <- read.csv("Cultivation.csv")
# Quick data overview
head(data)
summary(data)
# Bivariate regression
reg_model <- lm(pct ~ video, data=data)
summary(reg_model)
# Scatterplot with regression line
library(ggplot2)
graph <- ggplot(data, aes(x=video, y=pct)) +
  geom_point(alpha=0.6) +
  geom_smooth(method="lm", se=TRUE, color="blue") +
  labs(x="Average Weekly TV Hours", y="Estimated Workforce Percentage",
       title="Cultivation Analysis: TV Viewing vs Workforce Estimates")
graph