DedooseR is an R package that connects with Dedoose to support the analysis of qualitative data. It was built to help researchers streamline workflows and improve transparency and rigor in qualitative coding and analysis.
Key Features
DedooseR currently allows you to:
- Clean data exported from Dedoose
- Summarize code frquencies with one coder/media title in order of coder preference
- Plot raw frequencies or proportions of codes applied to total codes, define min frequency to visualize and/or exclude codes
- Set a saturation criteria and plot it
- Plot and play with saturation criteria sets
- Produce code co-occurrence heatmaps
Why This Package?
Ongoing challenges in qualitative research include defining what constitutes high-quality data and demonstrating transparency in how saturation is reached (Small & Calarco, 2022). Informed by guidelines for high-quality qualitative research (Hennink & Kaiser, 2022), DedooseR allows you to better understand your data with quality tags in Dedoose like:
- The concreteness of excerpts (priority code)
- The heterogeneity within codes (heterogeneity code)
By tagging these indicators in Dedoose and exploring them in R, this allows for gain greater confidence in both the depth and diversity of datasets.
Installation
You can install the development version of DedooseR from GitHub with:
# install.packages("pak")
pak::pak("abiraahmi/DedooseR")Example
This is a basic example which shows you how you may the package:
library(DedooseR)
# Load excerpts from Dedoose
excerpts <- read_xlsx("file path")
cooccurence <- read_xlsx("file path")
# Clean data
filepath <- "path/to/your/file"
preferred_coders <- c("s", "r", "l", "a")
excerpts <- clean_data(filepath = filepath, preferred_coders = preferred_coders)
# Summarize codes (1 coder/transcript according to listed coder preference)
df_all_summary <-summarize_codes(excerpts,
preferred_coders,
output_type = "datatable")
# Plot counts by raq frequency, proportion of total codes, set minimum count to
# visualize, and/or exclude codes
plot_counts(df_all_summary,
plot_proportion = FALSE,
min_count = 40,
exclude_codes = c("Priority excerpt", "Heterogeniety"))
# If you've been tagging excerpts by quality indicators, set them below and
# summarize code counts based on these selections
df_qual_summary <- quality_indicators(excerpts,
preferred_coders,
qual_indicators =
c("Priority excerpt", "Heterogeniety"))
df_all_summary <- summarize_codes(excerpts,
preferred_coders,
output_type = "tibble")
# Now call your plot function with these two datasets
# You can set a min code frequency to plot or plot by proportion of total counts
plot_saturation(
df_all_summary,
df_qual_summary,
qual_indicators = c("Priority excerpt", "Heterogeniety"),
min_counts = c("Priority excerpt" = 3, "Heterogeniety" = 3),
stacked = TRUE,
as_proportion = FALSE)
# Play around with the thresholds and see which codes reach saturation
# Define thresholds
thresholds_list <- list(
"Set 1" = list(
`Priority excerpt` = 2,
Heterogeniety = 3
),
"Set 2" = list(
`Priority excerpt` = 5,
Heterogeniety = 3
)
)
plot_saturation_comp(
summary_data = summary_data,
quality_summary = quality_summary,
thresholds_list = thresholds_list,
stacked = TRUE,
as_proportion = TRUE,
ncol = 2
)
# Load co-occurence datasets from Dedoose
cooccurence <- read_xlsx("file path")
# Produce code co-occurence heat map
cooccur(cooccurence, min_frequency = 10)References
Hennink, M., & Kaiser, B. N. (2022). Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Social science & medicine, 292, 114523.
Small, M. L., & Calarco, J. M. (2022). Qualitative Literacy: A Guide to Evaluating Ethnographic and Interview Research (1st ed.). University of California Press. https://doi.org/10.2307/j.ctv2vr9c4x