| Title: | Quickly Explore Complex Survey Data |
|---|---|
| Description: | Visualize and tabulate single-choice, multiple-choice, matrix-style questions from survey data. Includes ability to group cross-tabulations, frequency distributions, and plots by categorical variables and to integrate survey weights. Ideal for quickly uncovering descriptive patterns in survey data. |
| Authors: | Liam Haller [aut, cre, cph] |
| Maintainer: | Liam Haller <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.2.0 |
| Built: | 2024-11-20 05:36:32 UTC |
| Source: | https://github.com/liamhaller/surveyexplorer |
A "survey" of bears in Berlin Report ...
berlinbearsberlinbears
berlinbearsA data frame with 500 rows and 22 columns describing bears and their preferences:
name of species
genus that the species belongs to
gender ofthe bear
age of the bear
survey questions on foods the bear will eat
example of likert questions
...
Base table for single & multiple choice questions
frequency_table(data.table, group_by)frequency_table(data.table, group_by)
data.table |
Output from either mutli or single summary |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
Gt table
Generate a grouped bar chart displaying the frequency distribution of
responses for a categorical variable. The function supports optional
subgrouping of data using the group_by variable, exclusion of specific
subgroups with 'subgroups_to_exclude,' and data weighting with the 'weights'
parameter. Users can also choose to exclude NA values from the questions
prior to analysis using the 'na.rm' parameter.
matrix_freq( dataset, question, response_order = NULL, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE, colors = NULL )matrix_freq( dataset, question, response_order = NULL, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE, colors = NULL )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
response_order |
An optional vector specifying the order of factor levels for the response categories. This parameter is particularly useful for ensuring that the response categories are presented in a specific, meaningful order when plotting. For instance, in surveys or questionnaires where responses range from strongly disagree to strongly agree, setting response_order allows the categories to be displayed in this logical sequence rather than an alphabetical or random order. |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
colors |
Optional vector specifying colors for each response category. |
A ggplot2 object representing a grouped bar chart displaying the frequency distribution of responses for the specified categorical variable. The chart supports grouping, weighting, and exclusion of subgroups.
Other matrix questions:
matrix_likert(),
matrix_mean(),
matrix_table()
#Array question (1-5) matrix_freq(berlinbears, dplyr::starts_with('p_')) #remove NA category matrix_freq(berlinbears, dplyr::starts_with('p_'), na.rm = TRUE) #Use `group_by` to partition the question into several groups matrix_freq(berlinbears, dplyr::starts_with('p_'), group_by = species, subgroups_to_exclude = c('panda bear', NA ), na.rm = TRUE) #Categorical input matrix_freq(berlinbears, dplyr::starts_with('c_'), group_by = is_parent, na.rm = TRUE)#Array question (1-5) matrix_freq(berlinbears, dplyr::starts_with('p_')) #remove NA category matrix_freq(berlinbears, dplyr::starts_with('p_'), na.rm = TRUE) #Use `group_by` to partition the question into several groups matrix_freq(berlinbears, dplyr::starts_with('p_'), group_by = species, subgroups_to_exclude = c('panda bear', NA ), na.rm = TRUE) #Categorical input matrix_freq(berlinbears, dplyr::starts_with('c_'), group_by = is_parent, na.rm = TRUE)
The function produces a visually appealing diverging stacked bar chart, allowing for easy interpretation of the distribution of responses to a specific Likert-scale question. The function supports customization of labels, colors, and weights, providing flexibility in data representation.
matrix_likert( dataset, question, labels = NULL, colors = NULL, weights = NULL, na.rm = TRUE )matrix_likert( dataset, question, labels = NULL, colors = NULL, weights = NULL, na.rm = TRUE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
labels |
Optional vector specifying labels for each response category. If not provided, it extracts labels from the original dataset. |
colors |
Optional vector specifying colors for each response category. Default colors are provided for 3 and 5 categories. If not specified, the function expects a vector of color codes. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A ggplot2 object representing a diverging stacked bar chart displaying the distribution of Likert-scale responses. The chart is customized based on the provided or extracted labels and colors.
Other matrix questions:
matrix_freq(),
matrix_mean(),
matrix_table()
This function creates a likert-style plot showing means and standard errors
for a specified numeric variable, question. Optionally, the plot can be
grouped by another variable, group_by, and subgroups can be excluded. If
survey weights are provided, the counts are adjusted accordingly. The plot is
flipped for better readability in likert-style format.
matrix_mean( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )matrix_mean( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A likert-style ggplot displaying means and standard errors. The plot is flipped for better readability, and if grouping is specified, different colors represent distinct subgroups.
Other matrix questions:
matrix_freq(),
matrix_likert(),
matrix_table()
#basic plot matrix_mean(berlinbears, dplyr::starts_with('p_')) #with grouping and weights matrix_mean(berlinbears, dplyr::starts_with('p_'), group_by = species, subgroups_to_exclude = 'panda bear', weights = weights, na.rm = TRUE )#basic plot matrix_mean(berlinbears, dplyr::starts_with('p_')) #with grouping and weights matrix_mean(berlinbears, dplyr::starts_with('p_'), group_by = species, subgroups_to_exclude = 'panda bear', weights = weights, na.rm = TRUE )
This function creates a table showing percentages and counts for each
response option in a multiple-choice question, specified by question. If
grouping is provided with group_by, the table is extended to include
subgroups. Subgroups can be excluded, and survey weights are supported for
adjusted counts. The table is formatted for clarity and can be displayed in
wide format. When weights are used, counts are presented as percentages only,
and a note is added at the bottom of the table.
matrix_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE, column_order = NULL )matrix_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE, column_order = NULL )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
column_order |
reorder columns of final table with an argument to pass to |
A gt table summarizing percentages and counts for each response option in the specified multiple-choice question. If grouping is provided, the table includes subgroups and is formatted for clarity.
@examples #Array question (1-5) matrix_table(berlinbears, dplyr::starts_with('p_'))
#Use group_by to partition the question into several groups
matrix_table(berlinbears, dplyr::starts_with('p_'), group_by = species,
subgroups_to_exclude = 'panda bear' )
#Remove NA category matrix_table(berlinbears, dplyr::starts_with('p_'), group_by = species, subgroups_to_exclude = 'panda bear', na.rm = TRUE
#Categorical input matrix_table(berlinbears, dplyr::starts_with('c_'), group_by = is_parent)
Other matrix questions:
matrix_freq(),
matrix_likert(),
matrix_mean()
Visualize multiple-choice question responses with an upset plot, a visual
tool for exploring the overlap and distribution of multiple-choice question
responses. The function supports optional subgrouping of data using the
group_by variable, exclusion of specific subgroups with
'subgroups_to_exclude,' and data weighting with the 'weights' parameter.
Users can also choose to exclude NA values from the questions prior to
analysis using the 'na.rm' parameter.
multi_freq( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )multi_freq( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
An upset plot visualizing the distribution of responses to the multiple-choice question.
Other multiple-choice questions:
multi_summary(),
multi_table()
#Use dplyr to select questions library(dplyr) #Basic Upset plot #Use `group_by` to partition the question into several groups multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, subgroups_to_exclude = NA) #Specifiy survey weights with `weights` multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, weights = weights)#Use dplyr to select questions library(dplyr) #Basic Upset plot #Use `group_by` to partition the question into several groups multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, subgroups_to_exclude = NA) #Specifiy survey weights with `weights` multi_freq(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, weights = weights)
This function generates summary statistics, including frequencies, based on the provided question. It allows for optional grouping and weighting of data.
multi_summary( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )multi_summary( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A data frame containing summary statistics, including frequencies, for the specified question.
Other multiple-choice questions:
multi_freq(),
multi_table()
Generates a table presenting the distribution of responses for a specified
multiple-choice question. If a grouping variable, group_by, is provided,
the table extends to include row and column totals, along with additional count and
frequency columns for each level of group_by (excluding specified subgroups, if any).
When survey weights are specified with weights, the counts reflect the weighted values,
and a note is appended at the bottom of the table.
multi_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )multi_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The columns that contain each of the response options for a question, can be selected by using tidyselect semanatics or providing a vector of column names or numbers |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A gt table displaying frequencies and counts for the specified multiple-choice question. If a grouping variable is provided, the table includes subgroups for a comprehensive analysis. If survey weights are specified, the table notes that frequencies and counts are weighted.
Other multiple-choice questions:
multi_freq(),
multi_summary()
#Basic Table multi_table(berlinbears, question = dplyr::starts_with('will_eat')) #Use `group_by` to partition the question into several groups multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, subgroups_to_exclude = NA) #Specifiy survey weights with `weights` multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, weights = weights)#Basic Table multi_table(berlinbears, question = dplyr::starts_with('will_eat')) #Use `group_by` to partition the question into several groups multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, subgroups_to_exclude = NA) #Specifiy survey weights with `weights` multi_table(berlinbears, question = dplyr::starts_with('will_eat'), group_by = gender, weights = weights)
generates a bar chart of class ggplot illustrating how responses are
distributed for a specific single-choice question. If you provide a grouping
variable using group_by the chart includes facets for each subgroup.
Additionally, if you specify survey weights with weights the chart reflects
weighted response frequencies.
single_freq( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )single_freq( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The categorical variable of interest for which frequencies and counts will be calculated, can be selected by using tidyselect semantics |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A ggplot2 object with a bar chart displaying response frequencies. If "group_by" is provided, facets show subgroup details. If "weights" are specified, the chart displays weighted frequencies.
Other single-choice questions:
single_summary(),
single_table()
#Simple barchart single_freq(berlinbears, question = income) #Use `group_by` to facet the graph into several groups single_freq(berlinbears, question = income, group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` single_freq(berlinbears, question = income, group_by = species, subgroups_to_exclude = c('black bear', NA)) #Specify survey weights with `weights` single_freq(berlinbears, question = h_winter, group_by = gender, weights = weights) #to ignore NA values in the responses to `question`, set na.rm = TRUE single_freq(berlinbears, question = h_winter, na.rm = TRUE)#Simple barchart single_freq(berlinbears, question = income) #Use `group_by` to facet the graph into several groups single_freq(berlinbears, question = income, group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` single_freq(berlinbears, question = income, group_by = species, subgroups_to_exclude = c('black bear', NA)) #Specify survey weights with `weights` single_freq(berlinbears, question = h_winter, group_by = gender, weights = weights) #to ignore NA values in the responses to `question`, set na.rm = TRUE single_freq(berlinbears, question = h_winter, na.rm = TRUE)
This function analyzes a specified categorical variable, question,
optionally grouping by another variable, group_by. Counts and frequencies
are computed, taking into account provided survey weights. Subgroups can be
excluded, and NAs can be removed if necessary.
single_summary( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm )single_summary( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The categorical variable of interest for which frequencies and counts will be calculated, can be selected by using tidyselect semantics |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A tabled data frame with counts and frequencies for the specified variable and optional grouping variable. The output is pre-processed, considering subgroup exclusions, NA removal, and survey weights if provided.
Other single-choice questions:
single_freq(),
single_table()
Generates a detailed table summarizing the frequencies and counts for each level
of the specified variable, question. If a grouping variable, group_by, is provided,
the table extends to include row and column totals, along with additional count and
frequency columns for each level of group_by (excluding specified subgroups, if any).
When survey weights are specified with weights, the counts reflect the weighted values,
and a note is appended at the bottom of the table.
single_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )single_table( dataset, question, group_by = NULL, subgroups_to_exclude = NULL, weights = NULL, na.rm = FALSE )
dataset |
The input dataframe (or tibble) of survey questions |
question |
The categorical variable of interest for which frequencies and counts will be calculated, can be selected by using tidyselect semantics |
group_by |
Optional variable to group the analysis. If provided, the frequencies and counts will be calculated within each subgroup. |
subgroups_to_exclude |
Optional vector specifying subgroups to exclude from the analysis. |
weights |
Optional variable containing survey weights. If provided, frequencies and counts will be weighted accordingly. |
na.rm |
Logical indicating whether to remove NA values from |
A gt table summarizing frequencies and counts based on the specified
parameters. If the optional group_by parameter is provided, the output
will be a grouped gt table, displaying frequencies and counts for each
subgroup as well as row and column totals.
Other single-choice questions:
single_freq(),
single_summary()
#Simple table single_table(berlinbears, question = income) #Use `group_by` to partition the question into several groups single_table(berlinbears, question = income, group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` single_table(berlinbears, question = income, group_by = species, subgroups_to_exclude = c('black bear', NA)) #Specifiy survey weights with `weights` single_table(berlinbears, question = h_winter, group_by = gender, weights = weights) #to ignore NA values in the responses to `question`, set na.rm = TRUE single_table(berlinbears, question = h_winter, na.rm = TRUE)#Simple table single_table(berlinbears, question = income) #Use `group_by` to partition the question into several groups single_table(berlinbears, question = income, group_by = gender) #to ignore a subgroup, use `subgroups_to_exclude` single_table(berlinbears, question = income, group_by = species, subgroups_to_exclude = c('black bear', NA)) #Specifiy survey weights with `weights` single_table(berlinbears, question = h_winter, group_by = gender, weights = weights) #to ignore NA values in the responses to `question`, set na.rm = TRUE single_table(berlinbears, question = h_winter, na.rm = TRUE)