lensModel: Lens Model Regressions

In multicon: Multivariate Constructs

Description

A function for computing key statistics from a Lens Model (Brunswick, 1952) analysis.

Usage

lensModel(inSet, exSet, cueSet)

Arguments

inSet	A data.frame containing the variables on the validity side of the lens model. It must have the same dimensions as exSet and columns corresponding to the columns in exSet.
exSet	A data.frame containing the variables on the utilization side of the lens model. It must have the same dimensions as inSet and columns corresponding to the columns in inSet.
cueSet	A data.frame containing the cues to use in the lens analysis.

Details

This function is designed to perform so-called Lens Model analyses. If a set of targets has known criterion values on some dimensions (e.g., self-reports of personality) a set of judges may make judgments of those targets (e.g., other reports of personality) based on some information (i.e., Cues) presented to the judges (e.g., some behavioral acts). A lens model analyses examines (a) the achievement of the judges (i.e., accuracy) for each dimension being judged as the correlation between the judgments and the criterion, (b) the validities of the cues for each dimension as the linear regression coefficients predicting the criterion from all of the cues, and (c) the cue utilization of the judges for each dimension as the linear regression coefficients predicting the judgments from the cues. This function computes all of this and much more in one step.

Value

Returns a list containing the following

Lens Stats

A data.frame containing the following statistics for each variable in inSet:

1. Validity SaturationThe multiple R for the Validity side of the Lens Model

2. Utilization SaturationThe multiple R for the Utilization side of the Lens Model

3. Coefficient CorrelationThe correlation between the Cue Validities and the Cue Utilizations (not including the intercept)

4. AchievementThe correlation between inSet and exSet

5. Linear KnowledgeThe correlation between the fitted values from the validity and utilization sides of the model

6. Unmodeled KnowledgeThe correlation between the residuals from the validity and utilization sides of the model

Cue Validities	A data.frame of size ncol(cueSet)+1 X ncol(inSet) containing the regression coefficients (including intercept) for the validity side of the lens model.
Cue Utilizations	A data.frame of size ncol(cueSet)+1 X ncol(exSet) containing the regression coefficients (including intercept) for the utilization side of the lens model.

Author(s)

Ryne A. Sherman

References

Brunswik, E. (1952). The conceptual framework of psychology. Chicago: University of Chicago Press

See Also

lensDetect print.lensMod

Examples

data(lensData)
DIAMONDS.in <- lensData[,32:39] # Self-ratings on 8 Situation Characteristics
DIAMONDS.ex <- lensData[,40:47] # Coder-ratings on 8 Situation Characteristics
CUES <- lensData[,3:31]         # Coded Situation Cues

mod <- lensModel(DIAMONDS.in, DIAMONDS.ex, CUES) # Get the lens statistics
mod$'Lens Stats' # View the overall stats
print(mod)       # View the individual coefficients and p-values

multicon documentation built on May 2, 2019, 3:18 a.m.