Effect sizes from one hundred years of social psychology

A total of 474 effect sizes from meta-analyses from social psychology during one hundred years extracted from the study by Richard et al. (2003) into a tab-separated CSV file.

Thanks to @richarddmorey we now have year as well.

Headers

• originalorder is the order of the effect size as in Richard et al. (2003).
• field is one of 18 research fields the effect sizes are grouped into, as in Richard et al. (2003).
• description is a short description of the effect size in question.
• k is the number of studies.
• r is the mean effect size (Pearson's r).
• sd is the standard deviation of the mean effect size. This is the only field with missing values, denoted by NA.
• documentnumber refers to the document number of the specific meta-analysis. See appendix in Richard et al. (2003).
• reference is the reference to the specific meta-analysis.
• year is when the meta-analysis was published (extracted from reference).

Fields

• Aggression
• Attitudes
• Attribution
• Expectancy effects
• Gender roles
• Group processes
• Health psychology
• Helping behavior
• Intergroup relations
• Law
• Leadership
• Methodology
• Motivation
• Nonverbal communication
• Personality
• Relationships
• Social cognition
• Social influence

Import into R

df <- read.csv("https://raw.githubusercontent.com/peterdalle/effectsizes/master/soc-psych.tsv",
               header=TRUE, sep="\t", stringsAsFactors=FALSE)

df$field <- factor(df$field)

# What is the mean effect size from all meta-analyses?
mean(df$r)

# What is the mean effect size from all meta-analyses in aggression research?
mean(subset(df$r, df$field == "Aggression"))

# Histogram of all effect sizes.
hist(df$r, breaks = 30)

library(tidyverse)

# Reproduce graph in Richard et al. (2003), but use density instead.
df %>%
  ggplot(aes(r)) +
  geom_density(fill="grey") + 
  scale_x_continuous(limits=c(0, 1), breaks=seq(0, 1, 0.1)) +
  theme_minimal() +
  labs(title="Magnitude of meta-analytic effect sizes in social psychology",
       x="Mean correlation coefficient",
       y="Density")

# Plot all effect sizes by field.
df %>% 
  ggplot(aes(year, r, color=factor(field))) +
  geom_point() + 
  theme_minimal() +
  labs(title="All effect sizes by year", color="Field", x="Field", y="Effect size (r)")

# Plot mean effect size by year.
df %>%
  group_by(year) %>%
  summarize(meanr = mean(r)) %>%
  ggplot(aes(year, meanr)) +
  geom_point() + 
  theme_minimal() +
  labs(title="Mean effect size by year", x="Year", y="Effect size (r)")

Import into Python

import pandas as pd
import numpy
import matplotlib.pyplot as plt

data = pd.read_csv("https://raw.githubusercontent.com/peterdalle/effectsizes/master/soc-psych.tsv", sep="\t")

# What is the mean effect size from all meta-analyses?
numpy.mean(data["r"])

# Histogram.
plt.hist(data["r"], bins="auto")
plt.title("Magnitude of meta-analytic effect sizes in social psychology")
plt.show()

See also

• Effect sizes from 60 years of communication research

Reference

Richard, F. D., Bond, C. F., & Stokes-Zoota, J. J. (2003). One Hundred Years of Social Psychology Quantitatively Described. Review of General Psychology, 7(4), 331–363. https://doi.org/10.1037/1089-2680.7.4.331