plot.psych {psych} | R Documentation |
Combines several plotting functions into one for objects of class “psych'. This can be used to plot the results of fa
, irt.fa
, VSS
, ICLUST
, omega
, factor.pa
, or principal
.
In R there are several ways to do exploratory factor and principal components analysis. Best reference, and developer of the ‘psych’ package: William Revelle, see links inside R in documentation on the ‘psych’ package. Ch 6 of his forthcoming book is highly recommended. Also very good, Michael Friendly’s page, not specific to R. A general purpose toolbox for personality, psychometric theory and experimental psychology. Functions are primarily for multivariate analysis and scale construction using factor analysis, principal component analysis, cluster analysis and reliability analysis, although others provide basic descriptive statistics. Item Response Theory is done using factor analysis of tetrachoric and polychoric. Rotations and transformations are either part of psych (Promax and cluster), of base R (varimax), or of GPArotation (simplimax, quartimax, oblimin, etc.). Of the various rotation/transformation options, varimax, Varimax, quartimax, bentlerT, geominT, and bifactor do orthogonal rotations. Since biplot.psych is looking for two elements of the x object: x$loadings and x$scores, you can create the appropriate object to plot. See the third example. Author(s) William Revelle. Fa, fa.poly, principal, fa.plot, pairs.panels.
x | The object to plot |
labels | Variable labels |
xlab | Label for the x axis – defaults to Latent Trait |
ylab | Label for the y axis |
ylim | Specify the limits for the y axis |
main | Main title for graph |
type | 'ICC' plots items, 'IIC' plots item information, 'test' plots test information, defaults to IIC. |
D | The discrimination parameter |
cut | Only plot item responses with discrimiantion greater than cut |
keys | Used in plotting irt results from irt.fa. |
y2lab | ylab for test reliability, defaults to 'reliability' |
qq | if TRUE, plot qq plot of residuals, otherwise plot a cor.plot of residuals |
... | other calls to plot |
Passes the appropriate values to plot. For plotting the results of irt.fa
, there are three options: type = 'IIC' (default) will plot the item characteristic respone function. type = 'IIC' will plot the item information function, and type= 'test' will plot the test information function.
Note that plotting an irt result will call either plot.irt or plot.poly depending upon the type of data that were used in the original irt.fa
call.
These are calls to the generic plot function that are intercepted for objects of type 'psych'. More precise plotting control is available in the separate plot functions. plot may be used for psych objects returned from fa
, irt.fa
, ICLUST
, omega
, as well as principal
A 'jiggle' parameter is available in the fa.plot function (called from plot.psych when the type is a factor or cluster. If jiggle=TRUE, then the points are jittered slightly (controlled by amount) before plotting. This option is useful when plotting items with identical factor loadings (e.g., when comparing hypothetical models).
Objects from irt.fa
are plotted according to 'type' (Item informations, item characteristics, or test information). In addition, plots for selected items may be done if using the keys matrix. Plots of irt information return three invisible objects, a summary of information for each item at levels of the trait, the average area under the curve (the average information) for each item as well as where the item is most informative.
If plotting multiple factor solutions in plot.poly, then main can be a vector of names, one for each factor. The default is to give main + the factor number.
It is also possible to create irt like plots based upon just a scoring key and item difficulties, or from a factor analysis and item difficulties. These are not true IRT type analyses, in that the parameters are not estimated from the data, but are rather indications of item location and discrimination for arbitrary sets of items. To do this, find irt.stats.like
and then plot the results.
Graphic output for factor analysis, cluster analysis and item response analysis.
More precise plotting control is available in the separate plot functions.
William Revelle
VSS.plot
and fa.plot
, cluster.plot
, fa
, irt.fa
, VSS
, ICLUST
, omega
, or principal
Hi,
I have a prcomp object (generated using the prcomp function) and I am trying to generate a biplot using ggbiplot, however I am confused about the different scaling options and their impact on the meaning of the plot.
scale = covariance biplot (scale = 1), form biplot (scale = 0). When scale = 1, the inner product between the variables approximates the covariance and the distance between the points approximates the Mahalanobis distance.
obs.scale = scale factor to apply to observations
var.scale = scale factor to apply to variables
I am not sure which of these scaling options I should choose for a prcomp object? I have seen lots of people set obs.scale = 1 and var.scale = 1, but I don't understand the reason for this.
What does the line length and angle actually correspond to in a biplot? I get the general idea that a high value on PC1 indicates that the variable has a strong influence on PC1 whilst a small value indicates a small influence. And that if the arrow is pointing to the right, then that variable has a positive impact on the PC.
Thanks for the help!!
Best wishes,
Lucy