| Type: | Package |
| Title: | Simple Method for Purging Mediation Effects among Independent Variables |
| Version: | 1.0.0 |
| Author: | Philip D. Waggoner <philip.waggoner@gmail.com> |
| Maintainer: | Philip D. Waggoner <philip.waggoner@gmail.com> |
| Description: | Simple method of purging independent variables of mediating effects. First, regress the direct variable on the indirect variable. Then, used the stored residuals as the new purged (direct) variable in the updated specification. This purging process allows for use of a new direct variable uncorrelated with the indirect variable. Please cite the method and/or package using Waggoner, Philip D. (2018) <doi:10.1177/1532673X18759644>. |
| License: | MIT + file LICENSE |
| Imports: | MASS |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 6.0.1 |
| NeedsCompilation: | no |
| Packaged: | 2018-03-20 13:51:35 UTC; bpwaggo |
| Repository: | CRAN |
| Date/Publication: | 2018-03-21 09:59:03 UTC |
Purges mediator effects between two independent variables (linear)
Description
Purges mediator effects between two independent variables, where selection (direct) variable is continuous, and returns new "purged" direct variable to be used in multivariate specification.
Usage
purge.lm(x, "direct", "indirect")
Arguments
x |
Represents data frame, though usage requires the data.frame name |
direct |
Represents "direct", or mediator variable, though usage requires column's name |
indirect |
Represents "indirect", or mediated variable, though usage requires column's name |
Details
Purging of mediator effects between two independent variables in two steps. First, the function regresses the direct (mediator) variable on the indirect (mediated) variable. Second, it stores and uses the residuals from the bivariate specification as the new "purged" variable to be used in place of the original "direct" variable in multivariate analyses. Regarding syntax, the function is built with placeholder objects to calculate the quantities of interest. Then, the usage allows placing the real objects' names from working datasets (including, data frame, direct variable name in quotes, and indirect variable name in quotes) for intuitive usage.
Value
purged
Examples
df <- data.frame(A = 1:10, B = 2:11) # linear/continuous example
purge.lm(df, "A", "B")
Purges mediator effects between two independent variables (logit link function)
Description
Purges mediator effects between two independent variables, where selection (direct) variable is binary, and returns new "purged" direct variable to be used in multivariate specification.
Usage
purge.logit(x, "direct", "indirect")
Arguments
x |
Represents data frame, though usage requires the data.frame name |
direct |
Represents "direct", or mediator variable, though usage requires column's name |
indirect |
Represents "indirect", or mediated variable, though usage requires column's name |
Details
Purging of mediator effects between two independent variables in two steps. First, the function regresses the direct (mediator) variable on the indirect (mediated) variable. Second, it stores and uses the residuals from the bivariate specification as the new "purged" variable to be used in place of the original "direct" variable in multivariate analyses. Regarding syntax, the function is built with placeholder objects to calculate the quantities of interest. Then, the usage allows placing the real objects' names from working datasets (including, data frame, direct variable name in quotes, and indirect variable name in quotes) for intuitive usage.
Value
purged
Examples
df <- data.frame(A = rep(0:1, 20), B = 1:20) # logit/binary example
purge.logit(df, "A", "B")
Purges mediator effects between two independent variables (negative binomial)
Description
Purges mediator effects between two independent variables, where selection (direct) variable is an event count, and returns new "purged" direct variable to be used in multivariate specification.
Usage
purge.negbin(x, "direct", "indirect")
Arguments
x |
Represents data frame, though usage requires the data.frame name |
direct |
Represents "direct", or mediator variable, though usage requires column's name |
indirect |
Represents "indirect", or mediated variable, though usage requires column's name |
Details
Purging of mediator effects between two independent variables in two steps. First, the function regresses the direct (mediator) variable on the indirect (mediated) variable. Second, it stores and uses the residuals from the bivariate specification as the new "purged" variable to be used in place of the original "direct" variable in multivariate analyses. Regarding syntax, the function is built with placeholder objects to calculate the quantities of interest. Then, the usage allows placing the real objects' names from working datasets (including, data frame, direct variable name in quotes, and indirect variable name in quotes) for intuitive usage.
Value
purged
Examples
df <- data.frame(A = c(1,1,1,1,2,2,3,3,4,5), B = 1:10) # negative binomial/counts example
purge.negbin(df, "A", "B")
Purges mediator effects between two independent variables (Poisson)
Description
Purges mediator effects between two independent variables, where selection (direct) variable is an event count, and returns new "purged" direct variable to be used in multivariate specification.
Usage
purge.poisson(x, "direct", "indirect")
Arguments
x |
Represents data frame, though usage requires the data.frame name |
direct |
Represents "direct", or mediator variable, though usage requires column's name |
indirect |
Represents "indirect", or mediated variable, though usage requires column's name |
Details
Purging of mediator effects between two independent variables in two steps. First, the function regresses the direct (mediator) variable on the indirect (mediated) variable. Second, it stores and uses the residuals from the bivariate specification as the new "purged" variable to be used in place of the original "direct" variable in multivariate analyses. Regarding syntax, the function is built with placeholder objects to calculate the quantities of interest. Then, the usage allows placing the real objects' names from working datasets (including, data frame, direct variable name in quotes, and indirect variable name in quotes) for intuitive usage.
Value
purged
Examples
df <- data.frame(A = c(1,1,1,1,1,2,2,2,3,4), B = 1:10) # Poisson/counts example
purge.poisson(df, "A", "B")
Purges mediator effects between two independent variables (probit link function)
Description
Purges mediator effects between two independent variables, where selection (direct) variable is binary, and returns new "purged" direct variable to be used in multivariate specification.
Usage
purge.probit(x, "direct", "indirect")
Arguments
x |
Represents data frame, though usage requires the data.frame name |
direct |
Represents "direct", or mediator variable, though usage requires column's name |
indirect |
Represents "indirect", or mediated variable, though usage requires column's name |
Details
Purging of mediator effects between two independent variables in two steps. First, the function regresses the direct (mediator) variable on the indirect (mediated) variable. Second, it stores and uses the residuals from the bivariate specification as the new "purged" variable to be used in place of the original "direct" variable in multivariate analyses. Regarding syntax, the function is built with placeholder objects to calculate the quantities of interest. Then, the usage allows placing the real objects' names from working datasets (including, data frame, direct variable name in quotes, and indirect variable name in quotes) for intuitive usage.
Value
purged
Examples
df <- data.frame(A = rep(0:1, 20), B = 2:21) # probit/binary example
purge.probit(df, "A", "B")