Tools for using Network Reporting Estimators

Functions useful for producing estimates from data that were collected using network reporting techniques like network scale-up, indirect sampling, network reporting, and sibling history.


The goal of the networkreporting package is to help people use network reporting methods.

In traditional survey methods, respondents report about themselves. In network report methods, survey respondents report about the people to whom they are socially connected. Thus, network reporting methods can be thought of a generalization of traditional surveys. Many existing methods rely on this kind of indirect reporting and are thus network reporting methods:

  • network scale-up method
  • sibling method
  • multiplicity sampling
  • network surivial method
  • generalized network scale-up method
  • confidant method
  • neighborhood method

Although many of the methods that rely on indirect reporting have been regarded as distinct in the past, there is great value in developing a unified statistical framework to deal with common problems that all of these methods share. This unified framework also enables us to develop new network reporting methods. This package will help you use both existing network reporting methods and any new network reporting methods that you develop.

For more information on network reporting methods see:

  • Feehan, Dennis (2015) "Network reporting methods", Ph.D. Dissertation. Office of Population Research, Princeton University.

The development of this software was supported by a grant from the National Institutes of Health (R01-HD075666).

You can install:

  • the latest released version from CRAN with

    install.packages("networkreporting")
  • the latest development version from github with

    install.packages("devtools")
    devtools::install_github("dfeehan/networkreporting")

The networkreporting package enables you to use several methods that many people currently think of as distinct. Here are some vignettes for how to use the package:

If you would like to suggest a feature or report a bug, please create an issue

If you use our package for your research, please cite it so that we can continue to develop it.

  • Feehan, Dennis M. and Salganik, Matthew J. (2014) "The networkreporting package." http://cran.r-project.org/package=networkreporting

News

networkreporting 0.1.0

  • very slight change to maintain compatability with upcoming version of dplyr

networkreporting 0.1.0

  • moved C++ code and other bootstrap code into a separate package, called surveybootstrap

  • some new unit tests and vignette

networkreporting 0.0.2

  • rescaled bootstrap code now written in C++ for increased speed

networkreporting 0.0.1

  • first release

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("networkreporting")

0.1.1 by Dennis M. Feehan, a year ago


Browse source code at https://github.com/cran/networkreporting


Authors: Dennis M. Feehan [aut, cre], Matthew J. Salganik [ths]


Documentation:   PDF Manual  


MIT + file LICENSE license


Imports functional, ggplot2, lazyeval, plyr, reshape2, stringr, dplyr, surveybootstrap

Suggests knitr, testthat

System requirements: GNU make


See at CRAN