Data Access - PeekbankR

The peekbankr package provides easy access to the peekbank database from R without writing SQL queries.

Installation

devtools is required to install peekbankr, so install it using

install.packages("devtools")

Some setups require a manual installation of RMariaDB, install is using:

install.packages("RMariaDB")

You can then install peekbankr using

devtools::install_github("peekbank/peekbankr")

Basic Usage

peekbankr offers a range of get_ functions to pull data into your environment, for example:

library(peekbankr)

datasets <- get_datasets()
head(datasets)

range_admins <- get_administrations(age = c(24, 36))
head(range_admins)

pomper_admins <- get_administrations(dataset_name = "pomper_saffran_2016")

Refer to the Data Schema to learn more about the shape of the resulting data.

Quick Analysis Example

library(dplyr)

aoi_data <- get_aoi_timepoints(dataset_name = "pomper_saffran_2016")

target_looks <- aoi_data %>%
  group_by(t_norm) %>%
  summarise(prop_target = mean(aoi == "target", na.rm = TRUE), .groups = "drop")

library(ggplot2)
ggplot(target_looks, aes(x = t_norm, y = prop_target)) +
  geom_line() +
  labs(x = "Time (ms)", y = "Proportion target", title = "Target looking")

Full documentation