Rationale

Elaboration likelihood model suggests that enduring attitude change depends upon time spent processing information-rich, high-quality persuasive arguments. In this experiment, the dependent variable, Favor_1, measures whether a research participant favored (1) or opposed (0) legalizing recreational marijuana, while the independent variable, Time, measures how many seconds the participant spent looking at information-rich, high-quality persuasive arguments in support of legalizing recreational marijuana use.

Hypothesis

Based on ELM, I hypothesize the odds of supporting marijuana legalization will increase significantly as time spent reading supporting arguments increases.

Variables & Method

In this analysis, the dependent variable is “Favor_1”, while the independent variable is “Time”. Because the dependent variable is categorical with two categories and the independent variable is continuous, logistic regression offers a suitable statistical test of the hypothesis.

Results

The analysis found that the odds of favoring marijuana legalization rose significantly (z = 4.8, p < .05), supporting the hypothesis.

Code and output

# Read the data from the web
FetchedData <- read.csv("https://drkblake.com/wp-content/uploads/2023/10/ELM.csv")
# Save the data on your computer
write.csv(FetchedData, "ELM.csv", row.names=FALSE)
# remove the data from the environment
rm (FetchedData)

# Installing required packages
if (!require("tidyverse"))
  install.packages("tidyverse")
library(ggplot2)

# Read the data
mydata <- read.csv("ELM.csv") #Edit YOURFILENAME.csv

# Specify the DV and IV
mydata$DV <- mydata$Favor_1 #Edit YOURDVNAME
mydata$IV <- mydata$Time #Edit YOURIVNAME

# Look at the DV and IV
ggplot(mydata, aes(x = DV)) + geom_bar(color = "black", fill = "#1f78b4")

ggplot(mydata, aes(x = IV)) + geom_histogram(color = "black", fill = "#1f78b4")

# Logistic regression plot
ggplot(mydata, aes(x = IV,
                   y = DV))+
  geom_jitter(height = .03,
              alpha = .5) +
  geom_smooth(method = "glm",
              method.args = list(family = "binomial"),
              se = FALSE,
              color = "#1f78b4")

# Run the logistic regression and view the summary results
options(scipen = 999)
log.ed <- glm(DV ~ IV, data = mydata, family = "binomial")
summary(log.ed)
## 
## Call:
## glm(formula = DV ~ IV, family = "binomial", data = mydata)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.8855  -0.7788  -0.3463   0.7561   2.2548  
## 
## Coefficients:
##              Estimate Std. Error z value    Pr(>|z|)    
## (Intercept) -4.567169   0.920675  -4.961 0.000000702 ***
## IV           0.016209   0.003383   4.791 0.000001661 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 131.791  on 99  degrees of freedom
## Residual deviance:  94.123  on 98  degrees of freedom
## AIC: 98.123
## 
## Number of Fisher Scoring iterations: 5
p <- .50
Inflection_point <- (log(p/(1-p))-coef(log.ed)[1])/coef(log.ed)[2]
Inflection_point
## (Intercept) 
##    281.7672