Type:	Package
Title:	Import and Analyse Ego-Centered Network Data
Version:	1.24.2
Date:	2024-02-01
Description:	Tools for importing, analyzing and visualizing ego-centered network data. Supports several data formats, including the export formats of 'EgoNet', 'EgoWeb 2.0' and 'openeddi'. An interactive (shiny) app for the intuitive visualization of ego-centered networks is provided. Also included are procedures for creating and visualizing Clustered Graphs (Lerner 2008 <doi:10.1109/PACIFICVIS.2008.4475458>).
URL:	https://github.com/tilltnet/egor, https://egor.tillt.net/
BugReports:	https://github.com/tilltnet/egor/issues
License:	AGPL-3
Depends:	R (≥ 3.5.0), dplyr, tibble
Imports:	tidygraph, srvyr, tidyr, methods, utils, purrr, rlang, pillar
Suggests:	knitr, testthat (≥ 2.1.0), rmarkdown, survey, shiny, igraph, network, haven
VignetteBuilder:	knitr
RoxygenNote:	7.2.3
LazyData:	true
Encoding:	UTF-8
NeedsCompilation:	no
Packaged:	2024-02-02 03:52:06 UTC; till
Author:	Till Krenz [aut, cre], Pavel N. Krivitsky [aut], Raffaele Vacca [aut], Michal Bojanowski [aut], Markus Gamper [ctb], Andreas Herz [aut], Christopher McCarty [ctb]
Maintainer:	Till Krenz <egor@tillt.net>
Repository:	CRAN
Date/Publication:	2024-02-02 05:30:02 UTC

R Package for importing and analyzing ego-centered-network data

Description

Tools for importing, analyzing and visualizing ego-centered network data. Supports several data formats, including the export formats of 'EgoNet', 'EgoWeb 2.0' and 'openeddi'. An interactive (shiny) app for the intuitive visualization of ego-centered networks is provided. Also included are procedures for creating and visualizing Clustered Graphs (Lerner 2008 doi:10.1109/PACIFICVIS.2008.4475458).

Details

When analyzing ego-centered network data it is common to either include or exclude data on ego. By default egor excludes ego for most analytic and visual functions and offers to include ego with the argument include.ego. In order for egor to automatically detect which ego-level variables correspond to the alter-variables, the variables on the different levels need to be named exactly the same. Alternatively the functions that have an include.ego argument have additional arguments that allow to specify the name of the ego-level variable in question.

For further information see package webpage or GitHub.

Author(s)

Maintainer: Till Krenz egor@tillt.net

Authors:

• Pavel N. Krivitsky pavel@statnet.org

• Raffaele Vacca r.vacca@ufl.edu

• Michal Bojanowski michal2992@gmail.com (ORCID)

• Andreas Herz herzand@uni-hildesheim.de

Other contributors:

• Markus Gamper m.gamper@uni-koeln.de [contributor]

• Christopher McCarty ufchris@ufl.edu [contributor]

See Also

Useful links:

• https://github.com/tilltnet/egor

• https://egor.tillt.net/

• Report bugs at https://github.com/tilltnet/egor/issues

Calculate EI-Index of ego networks

Description

The EI-Index is the division of the surplus count intra-group edges over inter-group edges, divided by total count of all edges. This implementation uses the intra-group and inter-group density instead of edge counts, when rescale is set to TRUE (default). The EI-Index is calculated for the whole network and for subgroups. Alternatively, the EI index can be employed as a measurement for egos tendency to homo-/heterophily - use comp_ei(). for that variant of the EI-Index.

Usage

EI(object, alt.attr, include.ego = FALSE, ego.attr = alt.attr, rescale = TRUE)

Arguments

Details

The whole network EI is a metric indicating the tendency of a network to be clustered by the categories of a given factor variable (alt.attr). The EI value of a group describes the tendency of that group within a network to be connected (if between 0 and 1) or not connected (if between -1 and 0) to other groups. Differing group sizes can lead to a distortion of EI values i.e. the ability of a big group A to form relationships to much smaller group B is limited by the size of B. Even when all possible edges between A and B exist, the EI value for group A might still be negative, classifying it as homophile. The re-scaled EI-Index values provided by this implementation substitutes absolute edge counts by inter- and intra-group edge densities in order to avoid the distortion of the EI-Index values. These values express the extend of homo- or heterophily of the network and its subgroups, as made possible by subgroup sizes.

Value

Returns tibble with the following columns:

• ego ID (".egoID")

• network EI-Index ("ei")

• subgroup EI-Index values (named by value levels of alt.attr/ego.attr)

References

Krackhardt, D., Stern, R.N., 1988. Informal networks and organizational crises: an experimental simulation. Social Psychology Quarterly 51 (2), 123-140.

Everett, M. G., & Borgatti, S. P. (2012). Categorical attribute based centrality: E-I and G-F centrality. Social Networks, 34(4), 562-569.

See Also

comp_ei(), for an ego level homophily measure.

Examples

data("egor32")
EI(egor32, "sex")

32 sets of randomly created alter-alter ties belonging to ego-centered networks

Description

32 sets of randomly created alter-alter ties belonging to ego-centered networks

Usage

aaties32

Format

A data frame with 32 sets of alter-alter relations and 4 variables:

.EGOID

ego identifier

.SRCID

source alter ID

.TGTID

target alter ID

weight

weight of relation

Activate ego, alter or alter-alter tie data level of an egor dataset

Description

This function activates one of the data levels of an egor dataset, so that the dplyr verbs know which level to execute on.

Usage

## S3 method for class 'egor'
activate(.data, what)

Arguments

Examples

e <- make_egor(5,50)
e %>% 
   activate("aatie") %>% 
   mutate(weight2 = 2 + weight) %>% 
   activate("alter") %>% 
   mutate(age.years = age.years^3)

Simulated Allbus 2010 Data

Description

A dataset simulated based on the the original Allbus 2010 SPSS data. The dataset simulates 100 respondents and does not resemble any actual Allbus respondents. Each variable is randomly generated based on the range of the original variables, co-variations between variables are disregarded. The data’s purpose is purely to demonstrate how to technically work with the Allbus data using egor and R - no analytical assumptions should be made based on this data!

Usage

allbus_2010_simulated

Format

A tibble/ data.frame of 100 simulated respondents/ rows and 981 variables/ columns. Each variable is a ⁠labelled dbl⁠.

Details

The dataset contains (simulated!) answers to two ego-centered name generators.

Set and query the alter nomination design

Description

Extract, set, or update the alter nomination design associated with an ego-centered dataset.

Usage

alter_design(x, ...)

## S3 method for class 'egor'
alter_design(x, which, ...)

alter_design(x, ...) <- value

## S3 replacement method for class 'egor'
alter_design(x, which, ...) <- value

Arguments

32 sets of randomly created alters belonging to ego-centered networks

Description

32 sets of randomly created alters belonging to ego-centered networks

Usage

alters32

Format

A data frame with 32 sets of up to 32 alters per egoID and 7 variables:

.ALTID

alter identifier

.EGOID

ego identifier

age

age in categories

age.years

age in years

country

country

income

income

sex

gender

Calculate diversity measures on an egor object.

Description

alts_diversity_count() counts the categories of a variable present in the networks of an egor object. alts_diversity_entropy() calculates the Shannon entropy as a measurement for diversity of an alter attribute.

Usage

alts_diversity_count(object, alt.attr)

alts_diversity_entropy(object, alt.attr, base = 2)

Arguments

Value

A tibble with the ego ID and a numeric result vector.

Author(s)

Michał Bojanowski, michal2992@gmail.com

Till Krenz, egor@tillt.net

Examples

data("egor32")
alts_diversity_count(egor32, "age")
alts_diversity_entropy(egor32, "age")

Append rows/columns to ego, alter or aatie data

Description

These work like dplyr's bind_cols() and bind_rows(). The first argument has to be an egor object. Additional rows/columns are added bottom/RHS of the active data level (ego, alter, aatie).

Usage

append_rows(.egor, ..., .id = NULL)

append_cols(.egor, ...)

Arguments

Value

egor object containing the additional rows/ columns on the active level.

Examples

e <- make_egor(12, 15)

# Adding a column to the ego level
additional_ego_columns <-
  tibble(x = sample(1:3, 12, replace = TRUE))
  
append_cols(e, additional_ego_columns)

# Adding rows to the ego and alter level
additional_ego_rows <-
  list(
    .egoID = 13,
    sex = "w",
    age = factor("56 - 65"),
    age.years = 60,
    country = "Australia"
  ) %>%
  as_tibble()
  
additional_alter_rows <-
  list(
    .altID = 1:5,
    .egoID = rep(13, 5),
    sex = sample(c("f", "m"), 5, replace = TRUE)
  ) %>%
  as_tibble()
  
append_rows(e, additional_ego_rows) %>%
  activate(alter) %>%
  append_rows(additional_alter_rows)

egor - a data class for ego-centered network data.

Description

The function egor() is used to create an egor object from ego-centered network data. as.egor() converts a list of igraph/network objects or a nested_egor objects to an egor object.

Usage

as.egor(x, ...)

## S3 method for class 'nested_egor'
as.egor(
  x,
  ID.vars = list(ego = ".egoID", alter = ".alterID", source = ".Source", target =
    ".Target"),
  ...
)

## S3 method for class 'list'
as.egor(x, ego_name = NULL, ...)

egor(
  alters,
  egos = NULL,
  aaties = NULL,
  ID.vars = list(ego = "egoID", alter = "alterID", source = "Source", target = "Target"),
  ego_design = NULL,
  alter_design = list(max = Inf)
)

Arguments

Details

If parameters alters.df, egos.df, and aaties.df are data frames, they need to share a common ego ID variable, with corresponding values. If alters.df and aaties.df are lists of data frames, egoID is ignored and they are matched by position with the rows of egos.df. Of the three parameters only alters.df is necessary to create an egor object, and egos.df and aaties.df are optional.

Value

Returns an egor object, which is a ⁠named list⁠ with three ⁠tibble data.frames⁠: ego, alter and aatie (alter-alter ties). Each data set has an .egoID column, that groups the data belonging to one ego. Additionally the alter data has an .alterID column, that links to the columns .srcID and .tgtID in the alter-alter tie data.

In addition, egor has two attributes: ego_design, containing an object returned by srvyr::as_survey_design() specifying the sampling design by which the egos were selected and alter_design, a list containing specification of how the alters were nominated. See the argument above for currently implemented settings.

Functions

• as.egor(nested_egor): Can convert (legacy) nested_egor object to egor object.

Note

Column names .alts, .aaties, and .egoRow are reserved for internal use of egor and should not be used to store persistent data. Other .-led column names may be reserved in the future.

See Also

as_tibble() for extracting ego, alter, and alter–alter tables, as tibbles or as srvyr's tbl_svy surveys.

Examples

data("egos32")
data("alters32")
data("aaties32") 

egor(alters32, 
     egos32, 
     aaties32,
     ID.vars = list(ego = ".EGOID", 
                    alter = ".ALTID", 
                    source = ".SRCID",
                    target =  ".TGTID"))

Convert egor object to network or igraph objects

Description

These functions convert an egor object into a list of network or igraph objects. By default ego itself is not included in the created objects, there is a parameter (include.egor) that allows for including ego.

Usage

as_igraph(
  x,
  directed = FALSE,
  include.ego = FALSE,
  ego.attrs = NULL,
  ego.alter.weights = NULL,
  graph.attrs = ".egoID"
)

## S3 method for class 'nested_egor'
as_igraph(
  x,
  directed = FALSE,
  include.ego = FALSE,
  ego.attrs = NULL,
  ego.alter.weights = NULL,
  graph.attrs = ".egoID"
)

## S3 method for class 'egor'
as.igraph(
  x,
  directed = FALSE,
  include.ego = FALSE,
  ego.attrs = NULL,
  ego.alter.weights = NULL,
  graph.attrs = ".egoID"
)

as_network(
  x,
  directed = FALSE,
  include.ego = FALSE,
  ego.attrs = NULL,
  ego.alter.weights = NULL,
  graph.attrs = ".egoID"
)

## S3 method for class 'egor'
as.network(
  x,
  directed = FALSE,
  include.ego = FALSE,
  ego.attrs = NULL,
  ego.alter.weights = NULL,
  graph.attrs = ".egoID"
)

Arguments

Details

The names of the variables specified in ego.attr and ego.alter.attr need to be the same as the names of corresponding alter attributes, in order for those variables to be merged successfully in the resulting network/ igraph object (see example).

Examples

e <- make_egor(3, 22)
as_igraph(e)

Extract ego, alter, and alter-alter tables from an egor object.

Description

Provided an egor object, these functions create a "global" tibble or srvyr's tbl_svy object containing egos, alter attributes, or alter-alter relations. The resulting tables are useful for advanced analysis procedures, e.g. multi-level regressions.

as_tibble() method for egor extracts the currently active component (ego, alter, or aaties) table, optionally joining it with the others, dropping any survey design information.

as_survey() method for egor instead returns a srvyr tbl_svy survey, taking into account any replication due to multiple alters or alter-alter ties incident on each ego. If no design is specified for the egos, then the default design (simple random sample with replacement) is assumed as the starting point.

as_egos_df(), as_alters_df(), as_aaties_df(), as_egos_survey(), as_alters_survey(), and as_aaties_survey() are convenience functions for the as_tibble() and as_survey() methods, activating the corresponding component of the egor object.

Usage

## S3 method for class 'egor'
as_tibble(x, ..., include.ego.vars = FALSE, include.alter.vars = FALSE)

## S3 method for class 'egor'
as_survey(.data, ..., include.ego.vars = FALSE, include.alter.vars = FALSE)

as_egos_df(object)

as_alters_df(object, include.ego.vars = FALSE)

as_aaties_df(object, include.ego.vars = FALSE, include.alter.vars = FALSE)

as_egos_survey(object, include.ego.vars = FALSE)

as_alters_survey(object, include.ego.vars = FALSE)

as_aaties_survey(object, include.ego.vars = FALSE, include.alter.vars = FALSE)

Arguments

Value

A tibble for the as_tibble and ⁠*_df⁠ functions and a tbl_svy for as_survey and the ⁠*_survey⁠ functions.

Examples

# Load example data
data(egor32)

as_tibble(egor32) # Ego table.

egor32 %>%
 activate("alter") %>%
 as_tibble(include.ego.vars=TRUE) # Alter table, but also with ego variables.

library(srvyr)
as_survey(egor32) # Ego table with survey design.

# Despite alter table being active, obtain the ego table.
(egor32 <- activate(egor32, "alter"))
as_egos_df(egor32)

# Create global alter table
as_alters_df(egor32)

# Create global alter-alter relations table
as_aaties_df(egor32)

# ... adding alter variables
as_aaties_df(egor32, include.alter.vars = TRUE)
as_egos_survey(egor32)
as_alters_survey(egor32) # Notice the resulting cluster design.

Cluster ego-centered networks by a grouping factor

Description

The idea of clustered graphs is to reduce the complexity of an ego-centered network graph by visualizing alters in clusters defined by a categorical variable (Lerner et al. 2008). clustered_graphs() calculates group sizes, inter and intra group tie densities and returns these informations in a list of igraph objects.

Usage

clustered_graphs(object, ..., clust.groups)

## S3 method for class 'list'
clustered_graphs(object, aaties, clust.groups, ...)

## S3 method for class 'egor'
clustered_graphs(object, clust.groups, ...)

## S3 method for class 'data.frame'
clustered_graphs(object, aaties, clust.groups, egoID = ".egoID", ...)

Arguments

Value

clustered_graphs returns a list of graph objects representing the clustered ego-centered network data;

References

Brandes, U., Lerner, J., Lubbers, M. J., McCarty, C., & Molina, J. L. (2008). Visual Statistics for Collections of Clustered Graphs. 2008 IEEE Pacific Visualization Symposium, 47-54.

See Also

vis_clustered_graphs for visualizing clustered graphs

Examples

data("egor32")

# Simplify networks to clustered graphs, stored as igraph objects
graphs <- clustered_graphs(egor32, "country") 

# Visualise
par(mfrow = c(2,3))
vis_clustered_graphs(
  graphs[1:5]
)
par(mfrow = c(1,1))

Obtain the index of a column in a data frame (or a list), producing an error if there is a problem.

Description

Obtain the index of a column in a data frame (or a list), producing an error if there is a problem.

Usage

col_idx(name, df)

Arguments

Value

An integer giving the column index of the named column.

Note

Numeric inputs for name are passed through, so this function is safe to use if the input is already a column index.

Longest common prefix of a set of strings.

Description

Longest common prefix of a set of strings.

Usage

common_prefix(x)

Arguments

Value

A character vector that is the longest common substring at the start of each of the input vectors.

Calculate the EI-Indices of an egor object as a measurement of ego-alter homophily

Description

comp_ei() calculates the EI-Index values as a measurement for ego-alter homo-/heterophily.

Usage

comp_ei(object, alt.attr, ego.attr)

Arguments

Value

A tibble with the ego ID and a numeric result vector.

Examples

data("egor32")
comp_ei(egor32, "age", "age")

Calculate custom compositional measures on an egor object

Description

comp_ply() applies a function, that uses an alter attribute to calculate a compositional measurement, on all networks in an egor object and returns a ⁠numeric vector⁠.

Usage

comp_ply(object, alt.attr, .f, ..., ego.attr = NULL, result.name = "result")

Arguments

Details

When an ego attribute is used the .f is called like this: .f(alt.attr, ego.attr, ...). .f must return a single numeric value.

Value

A tibble with the ego ID and a numeric result vector.

Author(s)

Michał Bojanowski, michal2992@gmail.com

Till Krenz, egor@tillt.net

Examples

df <- make_egor(10, 32)
comp_ply(df, "age.years", sd, na.rm = TRUE)

Calculate the composition of alter attributes in an egor object

Description

composition() calculates the proportional or absolute composition of alters for a given attribute/variable.

Usage

composition(object, alt.attr, absolute = FALSE)

Arguments

Value

A tibble with the ego ID and values per category of alt.attr as numeric columns.

Examples

data("egor32")
composition(egor32, "sex")

Count attribute combinations of dyads in ego-centered networks

Description

count_dyads() counts the attribute combinations of alter-alter ties/ dyads in ego-centered networks. The results can be returned as a wide or long tibble/ data.frame.

Usage

count_dyads(
  object,
  alter_var_name,
  return_as = c("wide", "long"),
  prefix = NULL
)

Arguments

Value

Wide or long tibble/ data.frame.

Examples

data(egor32)
count_dyads(object = egor32,
            alter_var_name = "country")

# Return result as long tibble.
count_dyads(object = egor32,
            alter_var_name = "country",
            return_as = "long")

Transforms edge lists to alter-alter wide format data.

Description

Transforms edge lists to alter-alter wide format data.

Usage

edgelist_to_wide(edges, src_tgt_id = c(".SRCID", ".TGTID"), weight = "weight")

Arguments

Calculate Burt constraint for the egos of ego-centered networks

Description

This calculates Burt's network constraint for all egos in an egor object. It iterates over each network and applies igraph::constraint. A weight variable can be specified.

Usage

ego_constraint(object, weights = NULL, ego.alter.weights = weights)

Arguments

Details

The calculation of weighted network constraint only works, if the alter-alter tie weights are complemented by a alter level variable specifying the same weight for the ego-alter ties.

Value

Numeric vector with a constraint value for each ego.

References

Burt, R. (2004). Structural holes and good ideas. American Journal of Sociology, (110), 349–399.

Examples

data(egor32)
ego_constraint(egor32)

Calculate the relationship density in ego-centered networks

Description

This function uses an egor object and calculates the density of all the ego-centered networks listed in the 'egor' object. Instead of an egor object, alter and alter-alter data can be provided as lists or data.frames.

Usage

ego_density(object, ...)

## S3 method for class 'egor'
ego_density(object, weight = NULL, max.netsize = NULL, directed = FALSE, ...)

Arguments

Value

returns a vector of network density values.

Examples

data("egor32")
ego_density(egor32)

Set and query the ego sampling design

Description

Extract, set, remove, or update the survey design associated with an ego-centered dataset.

Usage

ego_design(x, ...)

## S3 method for class 'egor'
ego_design(x, ...)

## S3 method for class 'nested_egor'
ego_design(x, ...)

ego_design(x, ...) <- value

## S3 replacement method for class 'egor'
ego_design(x, ...) <- value

## S3 replacement method for class 'nested_egor'
ego_design(x, ...) <- value

has_ego_design(x)

## S3 method for class 'egor'
has_ego_design(x)

## S3 method for class 'nested_egor'
has_ego_design(x)

strip_ego_design(x)

Arguments

Note

This can be useful for adjusting or re-initializing the ego design information after the underlying ego attributes had been modified.

32 randomly created ego-centered networks stored as an egor object

Description

32 randomly created ego-centered networks stored as an egor object

Usage

egor32

Format

An egor object with 32 ego-centered networks (5 variables):

egoID

ego identifier

sex

ego's gender

age

ego's age

.alts

nested column/list containing alters

.aaties

nested column/list containing alter-alter relations

Display names and values of global egor options.

Description

Display names and values of global egor options.

Usage

egor_options()

Details

Currently egor makes use of the following global options. Use options() to change values.

egor.print.rows.active.level

• Numeric. Amount of rows to display when printing the active level of an egor object.

egor.print.rows.inactive.level

• Numeric. Amount of rows to display when printing the inactive levels of an egor object.

egor.print.switch.active.level.to.top

• Logical. When printing an egor object, should the active data-level always be printed first?

egor.return.results.with.design

• Logical. egor functions that return ego-level results (e.g. one value per ego) return a svy_tbl object containing the ego_design(), when this is set to TRUE.

egor Network Visualization App

Description

Launches an interactive Shiny Web App that creates a list of igraph objects from an 'egor' object and offers the user several graphical means of interacting with the visualization parameters for all networks in the egor object.

Usage

egor_vis_app(object = NULL, shiny_opts = list(launch.browser = TRUE))

Arguments

Note

This function requires shiny to be installed.

Examples

#if(interactive()){
#   data("egor32")
#   egor_vis_app(egor32)
#}

32 randomly created egos belonging to ego-centered networks

Description

32 randomly created egos belonging to ego-centered networks

Usage

egos32

Format

A data frame with 32 sets of alter-alter relations and 4 variables:

.EGOID

ego identifier

age

age in categories

age.years

age in years

country

country

income

income

sex

gender

This extracts egos from igraph/network data if they are named in ego_name and returns an egor object

Description

This extracts egos from igraph/network data if they are named in ego_name and returns an egor object

Usage

extract_egos_and_return(graph_attrs, alters, edges, ego_name = NULL)

Arguments

A selective subset of GSS 2004 data

Description

This is a selective subset of General Social Survey 2004 data containing variables from network questions. See Details for description how this particular subset was selected. The data has a near 0 research value, it is provided to illustrate the functions in egor package.

Usage

gss2004

Format

A tibble with 499 rows and the variables listed below. Data was imported from SPSS file and are labelled. Functions in the labelled package can be used to handle them.

Variables:

id

Case ID

vpsu, vstrat, wtssall

Design variables and weight

age

Ego's age in years

race

Ego's race. 1=white, 2=black, 3=other

sex

Ego's sex. 1=male, 2=female

marital

Ego's marital status. 1=married, 2=widowed, 3=divorced, 4=separated, 5=never married

numgiven

Number of alters mentioned

age[1-5]

Alter's age in years

race[1-5]

Alter's race. 1=asian, 2=black, 3=hispanic, 4=white, 5=other

sex[1-5]

Alter's sex. 1=male, 2=female

spouse[1-5]

Whether alter is a spouse of ego. 1=mentioned, 2=not mentioned

close[1-4][2-5]

How close are the two alters according to ego. 1=especially close, 2=know each other, 3=total strangers

Details

This dataset was created from original GSS 2004 data for illustrative purposes such that (1) it is small and (2) contains just enough variation in respondent's personal networks to illustrate various functions in the package. It is essentially a stratified sample from original data (1472 cases). Strata correspond to groups of cases created from unique combinations of values on the following ego variables: age (3 categories), race, sex, marital, numgiven. At most 2 cases were sampled from each stratum via simple random sampling with replacement.

Source

General Social Survey data at NORC: https://gss.norc.org/get-the-data

General helper functions

Description

Helper functions for ego centered network analysis

Usage

as_nested_egor(x)

alters_by_ego(x)

## S3 method for class 'egor'
alters_by_ego(x)

## S3 method for class 'nested_egor'
alters_by_ego(x)

aaties_by_ego(x)

## S3 method for class 'egor'
aaties_by_ego(x)

## S3 method for class 'nested_egor'
aaties_by_ego(x)

dyad.poss(max.alters, directed = FALSE)

sanitize.wide.edges(max.alters)

create_edge_names_wide(x)

dyads_possible_between_groups(x, y, geometric = TRUE)

din_page_dist(x)

Arguments

Functions

• as_nested_egor(): Converts an egor object to a "legacy" egor object with nested .alts and .aaties columns.

• alters_by_ego(): Splits the alter table into a list of tables (possibly 0-row) of alters associated with each ego, in the same order as the ego table.

• aaties_by_ego(): Splits the alter–alter ties table into a list of tables (possibly 0-row) of alter–alter associated with each ego, in the same order as the ego table.

• dyad.poss(): Returns the count of possible edges in an un-directed or directed, ego-centered network, based on the number of alters.

• sanitize.wide.edges(): Generates a data.frame marking possible dyads in a wide alter-alter relation data.frame. Row names corresponds to the network size. This is useful for sanitizing alter-alter relations in the wide format.

• create_edge_names_wide(): Creates a vector of names for variables containing data on alter-alter relations/ dyads in ego-centered networks.

• dyads_possible_between_groups(): Calculates the possible edges between members of different groups in an ego-centered network.

• din_page_dist(): Calculates the optimal distribution of a number of equally sized objects on a DIN-Norm DIN 476 (i.e. DIN A4) page in landscape view.

Create layout for an egogram

Description

This creates pairs of x and y coordinates for a egogram, accompanied by alter IDs for each coordinate pair.

Usage

layout_egogram(altID, venn_var, pie_var)

Arguments

Value

A dataframe with three columns: x, y and altID.

Generate a random edge list for one network.

Description

Generate a random edge list for one network.

Usage

make_edge_list(netsize)

Arguments

Generate random ego-centered-network data.

Description

This function generates random ego-centered-network data for a specified number of networks with a maximum network size. The network size of the generated networks is a normal distribution with sd=5.

Usage

make_egor(net.count, max.alters, netsize_fixed = FALSE, plot = FALSE)

Arguments

Import ego-centered network data from 'one file format'

Description

This function imports ego-centered network data, stored in a single file, providing ego, alter and edge data. This data format is used by the Allbus 2010 (GESIS) and similar social surveys.

Usage

onefile_to_egor(
  egos,
  netsize = NULL,
  ID.vars = list(ego = "egoID"),
  attr.start.col,
  attr.end.col,
  max.alters,
  aa.first.var,
  aa.regex = NULL,
  var.wise = FALSE,
  ...
)

Arguments

Value

An egor object is returned. It is a list of three data frames: (1) ego: dataframe of all egos and their attributes; (2) alter: dataframe of all alters; (3) aatie: dataframe of alter alter ties/ edges

References

Muller, C., Wellman, B., & Marin, A. (1999). How to Use SPSS to Study Ego-Centered Networks. Bulletin de Methodologie Sociologique, 64(1), 83-100.

Examples

path_to_one_file_8 <- system.file("extdata", "one_file_8.csv", package = "egor")
egos_8 <- read.csv2(path_to_one_file_8)

onefile_to_egor(
  egos = egos_8, netsize = egos_8$netsize,
  attr.start.col = "alter.sex.1",
  attr.end.col = "alter.age.8",
  aa.first.var = "X1.to.2",
  max.alters = 8)

Order edge list columns with source and target at the beginning.

Description

Order edge list columns with source and target at the beginning.

Usage

order.edge.list.columns(edges, source_, target)

Plotting egor objects

Description

egor objects can be plotted as egographs or egograms. By default networks of the four first egos are plotted.

Usage

plot_egograms(
  x,
  ego_no = 1,
  x_dim = 1,
  y_dim = 1,
  venn_var = NULL,
  pie_var = NULL,
  ascending_inwards = TRUE,
  vertex_size_var = NULL,
  vertex_color_var = NULL,
  vertex_color_palette = "Heat Colors",
  vertex_color_legend_label = vertex_color_var,
  vertex_label_var = "name",
  edge_width_var = NULL,
  edge_color_var = NULL,
  edge_color_palette = "Heat Colors",
  highlight_box_col_var = NULL,
  highlight_box_col_palette = "Heat Colors",
  res_disp_vars = NULL,
  vertex_zoom = 1,
  edge_zoom = 2,
  font_size = 1,
  pie_colors = NULL,
  venn_gradient_reverse = FALSE,
  show_venn_labels = TRUE,
  include_ego = FALSE,
  ...
)

plot_ego_graphs(
  x,
  ego_no = 1,
  x_dim = 1,
  y_dim = 1,
  vertex_size_var = NULL,
  vertex_color_var = NULL,
  vertex_color_palette = "Heat Colors",
  vertex_color_legend_label = vertex_color_var,
  vertex_label_var = "name",
  edge_width_var = NULL,
  ego_alter_edge_width_var = if (!is.null(edge_width_var) & include_ego) edge_width_var,
  edge_color_var = NULL,
  ego_alter_edge_color_var = if (!is.null(edge_color_var) & include_ego) edge_color_var,
  edge_color_palette = "Heat Colors",
  highlight_box_col_var = NULL,
  highlight_box_col_palette = "Heat Colors",
  res_disp_vars = NULL,
  vertex_zoom = 1,
  edge_zoom = 3,
  font_size = 1,
  include_ego = FALSE,
  ...
)

plot_egor(
  x,
  ego_no = 1,
  x_dim = 2,
  y_dim = 2,
  ...,
  type = c("egograph", "egogram")
)

## S3 method for class 'egor'
plot(x, ...)

Arguments

Details

For type equals "egograph" ego networks are plotted with igraph's plotting engine. "egogram" uses a special layout that places the nodes on a map of (1) concentric circles with (2) subsections, that can be mapped to alter variables.

Available color palettes are:

• Heat Colors

• Yellow-Green

• Red-Yellow

• Blue-Red

• Black-White

• Greys

• Rainbow

• Topo Colors

Functions

• plot_egograms(): Plots an ego-socio-gram.

• plot_ego_graphs(): Plots an ego graph.

Examples

e <- make_egor(net.count = 5, max.alters = 12)
plot_egograms(x = e,
              ego_no = 2,
              venn_var = "sex",
              pie_var = "country",
              vertex_size_var = "age")
plot(e)

Read ego-centered network data exported with EgoNet software as an egor object

Description

This function imports ego-centered network data from folders with separate files for alters-level and edge data. It will run some basic checks upon the completeness of the data and inform the user of potential problems. This function can be used to import data exported from EgoNet (McCarty 2011).

Usage

read_egonet(
  egos.file,
  alter.folder,
  edge.folder,
  csv.sep = ",",
  ID.vars = list(ego = "egoID", alter = "alterID", source = "Source", target = "Target"),
  first.col.row.names = FALSE,
  ...
)

Arguments

Value

An egor object is returned. It is a list of three data frames: (1) ego: dataframe of all egos and their attributes; (2) alter: dataframe of all alters; (3) aatie: dataframe of alter alter ties/ edges

Examples

egos.file <-  system.file("extdata", "egos_32.csv", package = "egor")
alters.folder <- system.file("extdata", "alters_32", package = "egor")
edge.folder <-  system.file("extdata", "edges_32", package = "egor")

ef <- read_egonet(egos.file = egos.file, 
                          alter.folder = alters.folder, 
                          edge.folder = edge.folder, 
                          csv.sep = ";")

Objects exported from other packages

Description

These objects are imported from other packages. Follow the links below to see their documentation.

tidygraph

activate

Returns results inheriting srvyr design if the input egor object has a an ego_design and global option "egor.return.results.with.design" is TRUE or 'NULL'.

Description

Returns results inheriting srvyr design if the input egor object has a an ego_design and global option "egor.return.results.with.design" is TRUE or 'NULL'.

Usage

return_results(x, results)

Arguments

Rotate circle or star graph layout so that it 'stands' on two bottom nodes.

Description

Rotate circle or star graph layout so that it 'stands' on two bottom nodes.

Usage

rotate_to_equilibrium(xy_mat)

Arguments

Rotate a matrix of x and y coordinates by a radian

Description

Rotate a matrix of x and y coordinates by a radian

Usage

rotate_xy(xy_mat, rad, direction = 1)

Arguments

Convert a table to a list of rows

Description

A convenience function converting a data.frame() or a tibble().

Usage

rowlist(x)

Arguments

Value

A list() of length nrow(x), with each element itself a named list() containing the elements in the corresponding row.

Examples

library(tibble)
(df <- tibble(x=2:1, y=list(list(1:3), list(3:4))))
rowlist(df)

Filter and Subset Ego-centered Datasets

Description

Functions to index and take subsets of egor() objects: manipulate egos, alters, or alter-alter ties.

Usage

## S3 method for class 'egor'
subset(x, subset, ..., unit = attr(x, "active"))

## S3 method for class 'egor'
x[i, j, unit = attr(x, "active"), ...]

Arguments

Details

Removing or duplicating an ego will also remove or duplicate their alters and ties.

Value

An egor() object.

Examples

# Generate a small sample dataset
(e <- make_egor(5,4))

# First three egos in the dataset
e[1:3,]

# Using an external vector
# (though normally, we would use e[.keep,] here)
.keep <- rep(c(TRUE, FALSE), length.out=nrow(e$ego))
subset(e, .keep)

    # Filter egos
subset(x = egor32, subset = egor32$ego$variables$sex == "m", unit="ego")
subset(x = egor32, sex == "m")

# Filter alters
subset(x = egor32, sex == "m", unit = "alter")

# Filter aaties
subset(x = egor32, weight != 0, unit = "aatie")

# Filter egos by alter variables (keep only egos that have more than 13 alters)
subset(x = egor32, nrow(alter) > 13, unit = "alter")

# Filter alters by ego variables (keep only alters that have egos from Poland)
subset(x = egor32, ego$country == "Poland", unit = "ego")

# Filter edges by alter variables (keep only edges between alters where `sex == "m"`)
subset(x = egor32, all(alter$sex == "m"), unit = "aatie")

Methods to print and summarize egor objects

Description

Methods to print and summarize egor objects

Usage

## S3 method for class 'egor'
summary(object, ...)

## S3 method for class 'egor'
print(
  x,
  ...,
  n.active = getOption("egor.rows_active_level"),
  n.inactive = getOption("egor.rows_inactive_level")
)

Arguments

Read/ import ego-centered network data from the three files format, EgoWeb2.0 or openeddi.

Description

These functions read ego-centered network data from the three files format, EgoWeb2.0 or openeddi and transform it to an egoR object. The three files format consists of an ego file, on alters file and one file containing the edge data. EgoWeb2.0 and openeddi use variations of this format.

Usage

threefiles_to_egor(
  egos,
  alters.df,
  edges,
  ID.vars = list(ego = "egoID", alter = "alterID", source = "Source", target = "Target"),
  ego.vars = NULL,
  ...
)

read_egoweb(
  alter.file,
  edges.file,
  egos.file = NULL,
  ID.vars = list(ego = "EgoID", alter = "Alter.Number", source = "Alter.1.Number", target
    = "Alter.2.Number"),
  ego.vars = NULL,
  ...
)

read_openeddi(
  egos.file = NULL,
  alters.file = NULL,
  edges.file = NULL,
  ID.vars = list(ego = "puid", alter = "nameid", source = "nameid", target = "targetid"),
  ego.vars = NULL,
  ...
)

Arguments

Value

An egor object is returned. It is a list of three data frames: (1) ego: dataframe of all egos and their attributes; (2) alter: dataframe of all alters; (3) aatie: dataframe of alter alter ties/ edges

Functions

• read_egoweb(): This function reads in data from an EgoWeb 2.0 survey and transforms it to an egoR object. If no file name for the egos file is provided ego data is assumed to be merged with alters data and it will be extracted by read_egoweb. By default the standard ID variable names of EgoWeb are used, if you need to specify the ID variable names use the ID.vars parameter. Further Information: github.com/qualintitative/egoweb

• read_openeddi(): This function reads in data created by the openeddi survey software and transforms it to an egoR object. If no parameters are provided read_openeddi will try to find the adequate files in the working directory. By default the standard ID variable names of openeddi are used, if you need to specify the ID variable names use the ID.vars parameter. Further Information: www.openeddi.com

Examples

# The data for read.egonet.threefiles() needs to be loaded with read.csv(), 
# for it to be converted to an egoR object.
egos.file <-  system.file("extdata", "egos_32.csv", package = "egor")
alters.file <- system.file("extdata", "alters_32.csv", package = "egor")
edges.file <-  system.file("extdata", "edges_32.csv", package = "egor")

egos <- read.csv2(egos.file)
alters <- read.csv2(alters.file)
edges <- read.csv2(edges.file)

tf <- threefiles_to_egor(egos = egos, alters.df = alters, edges = edges)

# read_egoweb() and read_openeddi() read the files directly from the disk.

# Fetch current working directory
wd <- getwd()

setwd(system.file("extdata", "openeddi", package = "egor"))
oe <- read_openeddi()

setwd(system.file("extdata", "egoweb", package = "egor"))
ew <- read_egoweb(alter.file = "alters_32.csv", edges.file = "edges_32.csv", 
                  egos.file = "egos_32.csv")
                  
# Restore working directory                   
setwd(wd)

Transnational personal communities of social support of German migrants in Great Britain

Description

This is an egor object derived from a subset of the data of a personal network study on support relationships German migrants living in the UK maintain. The data was collected in 2010 using respondent driven sampling (snowball sampling). While the number of alters the respondents were allowed to enter was not limited, only a random subsample of up to eight alters were selected for the alter name interpreter and alter-alter tie questions. This data set contains the data for 50 of the originally 234 egos.

Usage

transnat

alter_df

ego_df

Format

transnat: an egor object of 50 egos.

alter_df: alter data.frame of the transnat dataset.

ego_df: ego data.frame of the transnat dataset.

Details

The questionnaire used seven name generators:

1. From time to time, people rely on other people's advice and opinions to help them find their way in life better. In the last 12 months, who have you sought advice from when it came to important decisions, for example about your family or work? (emotional)

2. In the last 12 months who has done little jobs and favors for you or helped you, for example in filling in forms or moving home? (instrumental)

3. In the past year, who have you turned to when you felt down and wanted someone to talk to? (emotional)

4. In the last 12 months, who have you borrowed money from? (instrumental)

5. In the past year, who have you spent your free time with or shared a hobby? ( social companionship)

6. In the past year who have you had disagreements or arguments with (e.g. about everyday affairs, money or property)? (conflict)

7. Who has let you know that you can rely on them (e.g. that they will always be there for you if you need help)? (emotional).

References

• Herz, A. (2015). Relational constitution of social support in migrants' transnational personal communities. Social Networks, 40 (1), S. 64-74.

• Herz, A. (2012). Strukturen transnationaler sozialer Unterstützung. Springer Fachmedien Wiesbaden.

Trims alter-alter ties of alters that are missing/ deleted from alters data.

Description

This is used in the background by dplyr methods, to maintain the alter-alter ties according to changes made to the ego and alter data levels.

Usage

trim_aaties(object)

Arguments

Value

An egor object with trimmed alter-alter ties (.aaties).

Trims alters that are missing/ deleted from ego data.

Description

This is used in the background by dplyr methods, to maintain the alter ties according to changes made to the ego data level.

Usage

trim_alters(object)

Arguments

Value

An egor object with trimmed alter-alter ties (.aaties).

Import ego-centered network data from two file format

Description

This function imports ego-centered network data, stored in two files, where one file contains the ego attributes and the edge information and the other file contains the alters data. This form of data storage for ego-centered network data is proposed by Muller, Wellman and Marin (1999).

Usage

twofiles_to_egor(
  egos,
  alters,
  ID.vars = list(ego = "egoID", alter = "alterID", source = "Source", target = "Target"),
  max.alters,
  aa.first.var,
  selection = NULL,
  ...
)

Arguments

Value

An egor object is returned. It is a list of three data frames: (1) ego: dataframe of all egos and their attributes; (2) alter: dataframe of all alters; (3) aatie: dataframe of alter alter ties/ edges

Examples

path_to_alters_8.csv <- system.file("extdata", "alters_8.csv", package = "egor")
path_to_one_file_8 <- system.file("extdata", "one_file_8.csv", package = "egor")

# read data from disk
egos_8 <- read.csv2(path_to_one_file_8)
alters_8 <- read.csv2(path_to_alters_8.csv)

# convert to egor object
  twofiles_to_egor(
    egos = egos_8,
    alters = alters_8,
    max.alters = 8,
    aa.first.var = "X1.to.2")

Visualize clustered graphs

Description

vis_clustered_graphs visualizes clustered_graphs using a list of clustered graphs created with clustered_graphs.

Usage

vis_clustered_graphs(
  graphs,
  node.size.multiplier = 1,
  node.min.size = 0,
  node.max.size = 200,
  normalise.node.sizes = TRUE,
  edge.width.multiplier = 1,
  center = 1,
  label.size = 0.8,
  labels = FALSE,
  legend.node.size = 45,
  pdf.name = NULL,
  ...
)

Arguments

Value

vis_clustered_graphs plots a list of igraph objects created by the clustered_graphs function.

clustered_graphs returns a list of graph objects representing the clustered ego-centered network data;

References

Brandes, U., Lerner, J., Lubbers, M. J., McCarty, C., & Molina, J. L. (2008). Visual Statistics for Collections of Clustered Graphs. 2008 IEEE Pacific Visualization Symposium, 47-54.

See Also

clustered_graphs for creating clustered graphs objects

Examples

data("egor32")

# Simplify networks to clustered graphs, stored as igraph objects
graphs <- clustered_graphs(egor32, "country") 

# Visualise
par(mfrow = c(2,3))
vis_clustered_graphs(
  graphs[1:5]
)
par(mfrow = c(1,1))

weights.egor() extracts the (relative) sampling weights of each ego in the dataset.

Description

weights.egor() extracts the (relative) sampling weights of each ego in the dataset.

Usage

## S3 method for class 'egor'
weights(object, ...)

Arguments

See Also

weights.survey.design

Transform wide alter-alter data to an edge list. When alter-alter for numerous networks is stored in one file/object it is common use the 'wide' data format. This function transforms such data to an edge lists.

Description

Transform wide alter-alter data to an edge list. When alter-alter for numerous networks is stored in one file/object it is common use the 'wide' data format. This function transforms such data to an edge lists.

Usage

wide.dyads.to.edgelist(
  e.wide,
  first.var,
  max.alters,
  alters.list = NULL,
  selection = NULL
)

Arguments

Transform wide alter-alter data to an edge list. A regex based implementation to convert a wide list to an edgelist.

Description

Transform wide alter-alter data to an edge list. A regex based implementation to convert a wide list to an edgelist.

Usage

wide.dyads.to.edgelist.regex(e.wide, aa.regex, netsize)

Arguments

Transform 'wide' alter-level data to the 'long'-format

Description

A function to transform a wide-format dataframe of ego-centered network data into a long-format data-frame, where every row represents one alter/dyad. In the created dataframe numerous networks can be distinguished by a network ID (egoID).

Usage

wide.to.long(
  wide,
  egoID = "egoID",
  max.alters,
  start.col,
  end.col,
  var.wise = FALSE
)

Arguments