License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
Title:	Tools to Analyze Repertory Grid Data
LazyData:	yes
Type:	Package
LazyLoad:	yes
Description:	Analyze repertory grids, a qualitative-quantitative data collection technique devised by George A. Kelly in the 1950s. Today, grids are used across various domains ranging from clinical psychology to marketing. The package contains functions to quantitatively analyze and visualize repertory grid data (e.g. 'Fransella', 'Bell', & 'Bannister', 2004, ISBN: 978-0-470-09080-0). The package is part of the The package is part of the https://openrepgrid.org/ project.
Version:	0.1.18
Date:	2025-07-08
Encoding:	UTF-8
URL:	https://github.com/markheckmann/OpenRepGrid
Imports:	methods, graphics, grid, utils, stats, grDevices, crayon, plyr, stringr, abind, colorspace, psych, XML, pvclust, openxlsx, tidyr, dplyr, scales, igraph
Collate:	'bertin.r' 'calc.r' 'data-openrepgrid.r' 'dev-functions.r' 'distance.R' 'double-entry.R' 'export.r' 'globals.R' 'gmMain.r' 'gridlist.R' 'import.r' 'measures.r' 'onair.r' 'openrepgrid.r' 'perturbate.R' 'preferred_poles.R' 'repgrid.r' 'repgrid-basicops.r' 'repgrid-constructs.r' 'repgrid-elements.r' 'repgrid-output.r' 'repgrid-plots.r' 'repgrid-ratings.r' 'resampling.R' 'rgl-3d.r' 'settings.r' 'utils-import.r' 'utils.r' 'zzz.r'
RoxygenNote:	7.3.2
NeedsCompilation:	no
Suggests:	rgl, testthat (≥ 2.1.0), covr, styler, vdiffr, knitr, rmarkdown
Packaged:	2025-07-08 14:46:04 UTC; markheckmann
Author:	Mark Heckmann [aut, cre, cph], Alejandro García Gutiérrez [ctb], Diego Vitali [ctb]
Maintainer:	Mark Heckmann <heckmann.mark@gmail.com>
Depends:	R (≥ 3.5.0)
Repository:	CRAN
Date/Publication:	2025-07-08 15:10:02 UTC

OpenRepGrid: an R package for the analysis of repertory grids.

Description

The OpenRepGrid package provides tools for the analysis of repertory grid data. The repertory grid is a method devised by George Alexander Kelly in his seminal work "The Psychology of Personal Constructs" published in 1955. The repertory grid has been used in and outside the context of Personal Construct Psychology (PCP) in a broad range of fields. For an introduction into the technique see e.g. Fransella, Bell and Bannister (2003).

Note

To get started with OpenRepGrid visit the project's home under openrepgrid.org. On this site you will find tutorials, explanation about the theory, the analysis methods and the corresponding R code.

To see how to cite the OpenRepGrid package, type citation("OpenRepGrid") into the R console.

Author(s)

• Maintainer: Mark Heckmann (@markheckmann)

• Contributors: Richard C. Bell, Alejandro García Gutiérrez (@j4n7), Diego Vitali (@artoo-git), José Antonio González Del Puerto (@MindCartographer), Jonathan D. Raskin

• How to contribute: You can contribute in various ways. The OpenRepGrid code is hosted on GitHub, where you can issue bug reports or feature requests. You may email your request to the package maintainer.

References

Fransella, F., Bell, R. C., & Bannister, D. (2003). A Manual for Repertory Grid Technique (2. Ed.). Chichester: John Wiley & Sons.

Kelly, G. A. (1955). The psychology of personal constructs. Vol. I, II. New York: Norton, (2nd printing: 1991, Routledge, London, New York).

See Also

Useful links:

• https://github.com/markheckmann/OpenRepGrid

Concatenate repgrid objects.

Description

Simple concatenation of repgrid objects or list containing repgrid objects using the '+' operator.

Usage

## S4 method for signature 'repgrid,repgrid'
e1 + e2

## S4 method for signature 'list,repgrid'
e1 + e2

## S4 method for signature 'repgrid,list'
e1 + e2

## S4 method for signature 'repgrid,repgrid'
e1 / e2

Arguments

Examples

# join constructs
x <- bell2010
x + x
x + list(x, x)
list(x, x) + x

OpenRepGrid: Annotated overview of package functions.

Description

This documentation page contains an overview over the package functions ordered by topics. The best place to start learning OpenRepGrid will be the package website https://openrepgrid.org though.

Functions sorted by topic

Manipulating grids

Loading and saving data

Analyzing constructs

Descriptive statistics of constructs Construct correlations distance Root mean square of inter-construct correlations Somers' D Principal component analysis (PCA) of construct correlation matrix Cluster analysis of constructs

Analyzing elements

Visual representation

Index measures

Special features

Settings

OpenRepGrid uses several default settings e.g. to determine how many construct characters to display by default when displaying a grid. The function settings can be used to show and change these settings. Also it is possible to store the settings to a file and load the settings file to restore the settings.

Grid datasets

OpenRepGrid already contains some ready to use grid data sets. Most of the datasets are taken from the literature. To output the data simply type Type the name of the dataset to the console and press enter.

Single grids

Multiple grids

NOT YET AVAILABLE

Functions for developers

OpenRepGrid: internal functions overview for developers.

Below you find a guide for developers: these functions are usually not needed by the casual user. The internal functions have a twofold goal

1. to provide means for advanced numerical grid analysis and 2) to facilitate function development. The function for these purposes are internal, i.e. they are not visible in the package documentation. Nonetheless they do have a documentation that can be accesses in the same way as for other functions. More in the details section.

Functions for advanced grid analysis

The package provides functions to facilitate numerical research for grids. These comprise the generation of random data, permutation of grids etc. to facilitate Monte Carlo simulations, batch analysis of grids and other methods. With R as an underlying framework, the results of grid analysis easily lend themselves to further statistical processing and analysis within R. This is one of the central advantages for researchers compared to other standard grid software. The following table lists several functions for these purposes.

Modules for function development

Beside the advanced analysis feature the developer's functions comprise low-level modules to create new functions for grid analysis. Though the internal structure of a repgrid object in R is simple (type e.g. str(bell2010, 2) to get an impression), it is convenient to not have to deal with access on this level. Several function like e.g. getElementNames are convenient wrappers that perform standard tasks needed when implementing new functions. The following table lists several functions for these purposes.

Author(s)

Current members of the OpenRepGrid development team: Mark Heckmann. Everyone who is interested in developing the package is invited to join.

       The \pkg{OpenRepGrid} package development is hosted on github (<https://github.com/markheckmann/OpenRepGrid>).
       The github site provides information and allows to file bug reports or feature requests.
       Bug reports can also be emailed to the package maintainer or issued on
       <https://openrepgrid.org> under section *Suggestions/Issues*.
       The package maintainer is Mark Heckmann <heckmann(dot)mark(at)gmail(dot)com>.

See Also

Useful links:

• https://github.com/markheckmann/OpenRepGrid

Method for "<-" assignment of the repgrid ratings.

Description

It should be possible to use it for ratings on all layers.

Usage

## S4 replacement method for signature 'repgrid'
x[i, j, ...] <- value

Arguments

Examples

## Not run: 
x <- randomGrid()
x[1, 1] <- 2
x[1, ] <- 4
x[, 2] <- 3

# settings values outside defined rating scale
# range throws an error
x[1, 1] <- 999

# removing scale range allows arbitary values to be set
x <- setScale(x, min = NA, max = NA)
x[1, 1] <- 999

## End(Not run)

Extract parts of the repgrid object.

Description

Methods for "[", i.e., subsetting of repgrid objects.

Usage

## S4 method for signature 'repgrid'
x[i, j, ..., drop = TRUE]

Arguments

Examples

x <- boeker
x[1:4, ]
x[, 1:3]
x[1:4, 1:3]
x[1:4, c("self", "ideal self", "mother")]

subset method for openrepgridSettings class

Description

subset method for openrepgridSettings class

Print method for openrepgridSettings class

Usage

## S3 method for class 'openrepgridSettings'
x[i, ...]

## S3 method for class 'openrepgridSettings'
print(x, ...)

Add a new average element

Description

A new element is added to the grid. The scores are the arithmetic means across all selected elements.

Usage

addAvgElement(x, name = "avg", i, digits = Inf)

Arguments

Value

A repgrid object with additional new element.

Examples

addAvgElement(feixas2004, "others", i = 2:12)
addAvgElement(feixas2004, "others", i = 2:12, digits = 0) # integers

# exluding elements via negative indexes
addAvgElement(feixas2004, "others", i = c(-1, -13))

Add a new construct to an existing grid object.

Description

Add a new construct to an existing grid object.

Usage

addConstruct(
  x,
  l.name = NA,
  r.name = NA,
  scores = NA,
  l.preferred = NA,
  r.preferred = NA,
  l.emerged = NA,
  r.emerged = NA,
  position = NA,
  side = "pre"
)

Arguments

Value

repgrid object.

See Also

addElement()

Examples

## Not run: 

# show grid
bell2010
addConstruct(bell2010, "left pole", "pole right", c(3, 1, 3, 2, 5, 4, 6, 3, 7, 1))

## End(Not run)

Add an element to an existing grid.

Description

Add an element to an existing grid.

Usage

addElement(
  x,
  name = NA,
  scores = NA,
  abbreviation = NA,
  status = NA,
  position = NA,
  side = "pre"
)

Arguments

Value

repgrid object

See Also

addConstruct()

Examples

## Not run: 

bell2010
addElement(bell2010, "new element", c(1, 2, 5, 4, 3, 6, 5, 2, 7))

## End(Not run)

add index column for neater colnames

Description

add index column for neater colnames

Usage

addIndexColumnToMatrix(x)

Arguments

Examples

## Not run: 
x <- matrix(1:9, 3)
colnames(x) <- rep("Long names that occupiy too much space", 3)
rownames(x) <- rep("Some text", 3)
addIndexColumnToMatrix(x)

## End(Not run)

Adds the percentage of the sum-of-squares explained by each axis to the plot.

Description

Adds the percentage of the sum-of-squares explained by each axis to the plot.

Usage

addVarianceExplainedToBiplot2d(
  x,
  dim = c(1, 2, 3),
  var.cex = 0.7,
  var.show = TRUE,
  var.col = grey(0.1),
  axis.ext = 0.1,
  center = 1,
  normalize = 0,
  g = 0,
  h = 1 - g,
  col.active = NA,
  col.passive = NA,
  ...
)

Arguments

Align constructs using the ideal element to gain pole preferences.

Description

The direction of the constructs in a grid is arbitrary and a reflection of a scale does not affect the information contained in the grid. Nonetheless, the direction of a scale has an effect on inter-element correlations (Mackay, 1992) and on the spatial representation and clustering of the grid (Bell, 2010). Hence, it is desirable to follow a protocol to align constructs that will render unique results. A common approach is to align constructs by pole preference, i. e. aligning all positive and negative poles. This can e. g. be achieved using swapPoles(). If an ideal element is present, this element can be used to identify the positive and negative pole. The function alignByIdeal will align the constructs accordingly. Note that this approach does not always yield definite results as sometimes ratings do not show a clear preference for one pole (Winter, Bell & Watson, 2010). If a preference cannot be determined definitely, the construct direction remains unchanged (a warning is issued in that case).

Usage

alignByIdeal(x, ideal, high = TRUE)

Arguments

Value

repgrid object with aligned constructs.

References

Bell, R. C. (2010). A note on aligning constructs. Personal Construct Theory & Practice, 7, 42-48.

Mackay, N. (1992). Identification, Reflection, and Correlation: Problems in the bases of repertory grid measures. International Journal of Personal Construct Psychology, 5(1), 57-75.

Winter, D. A., Bell, R. C., & Watson, S. (2010). Midpoint ratings on personal constructs: Constriction or the middle way? Journal of Constructivist Psychology, 23(4), 337-356.

See Also

Aligning constructs alignByLoadings(), alignByPreferredPole()

Examples

feixas2004 # original grid
alignByIdeal(feixas2004, 13) # aligned with preference pole on the right

raeithel # original grid
alignByIdeal(raeithel, 3, high = FALSE) # aligned with preference pole on the left

Align constructs by loadings on first principal component.

Description

In case a construct loads negatively on the first principal component, the function alignByLoadings() will reverse it so that all constructs have positive loadings on the first principal component (see detail section for more).

Usage

alignByLoadings(x, trim = 20, index = TRUE)

Arguments

Details

The direction of the constructs in a grid is arbitrary and a reflection of a scale does not affect the information contained in the grid. Nonetheless, the direction of a scale has an effect on inter-element correlations (Mackay, 1992) and on the spatial representation and clustering of the grid (Bell, 2010). Hence, it is desirable to follow a protocol to align constructs that will render unique results. A common approach is to align constructs by pole preference, but this information is not always accessible. Bell (2010) proposed another solution for the problem of construct alignment. As a unique protocol he suggests to align constructs in a way so they all have positive loadings on the first component of a grid PCA.

Value

An object of class alignByLoadings containing a list of calculations with the following entries:

• cor.before: Construct correlation matrix before reversal

• loadings.before: Loadings on PCs before reversal

• reversed: Constructs that have been reversed

• cor.after: Construct correlation matrix after reversal

• loadings.after: Loadings on PCs after reversal

Note

Bell (2010) proposed a solution for the problem of construct alignment. As construct reversal has an effect on element correlation and thus on any measure that based on element correlation (Mackay, 1992), it is desirable to have a standard method for construct alignment independently from its semantics (preferred pole etc.). Bell (2010) proposes to align constructs in a way so they all have positive loadings on the first component of a grid PCA.

References

Bell, R. C. (2010). A note on aligning constructs. Personal Construct Theory & Practice, 7, 42-48.

Mackay, N. (1992). Identification, Reflection, and Correlation: Problems in the bases of repertory grid measures. International Journal of Personal Construct Psychology, 5(1), 57-75.

See Also

Aligning constructs alignByIdeal(), alignByPreferredPole()

Examples

# reproduction of the example in the Bell (2010)
# constructs aligned by loadings on PC 1
bell2010
alignByLoadings(bell2010)

# save results
a <- alignByLoadings(bell2010)

# modify printing of resukts
print(a, digits = 5)

# access results for further processing
names(a)
a$cor.before
a$loadings.before
a$reversed
a$cor.after
a$loadings.after

Align constructs by preferred pole

Description

The direction of the constructs in a grid is arbitrary. While their reversal (see reverse()) does not affect the information contained in the grid, it is often useful to align constructs for easier interpretation. One way of alignment is placing all positive poles on the same side. Note that this this is only possible if the preferred poles are defined (see preferredPoles()).

Usage

alignByPreferredPole(x, side_positive = "right")

Arguments

Value

A repgrid object with aligned constructs.

See Also

Aligning constructs alignByIdeal(), alignByLoadings()

Examples

x <- preferredPolesByIdeal(boeker, "ideal self")
x <- alignByPreferredPole(x)
x

Make indexes to order grid by angles in given dimensions.

Description

Reorder indexes for constructs and elements are calculated using the coordinates of the given dimensions.

Usage

angleOrderIndexes2d(x, dim = c(1, 2), clockwise = TRUE)

Arguments

Value

A list containing the indexes to reorder the grid. The first list element for the constructs, the second for the elements indexes.

Examples

## Not run: 

x <- randomGrid(15, 30) # make random grid
i <- angleOrderIndexes2d(x) # make indexes for ordering
x <- x[i[[1]], i[[2]]] # reorder constructs and elements
x # print grid

## End(Not run)

apply with a progress bar

Description

Can be used like standard base:::apply. The only thing it does is create an additional progress bar.

Usage

apply_pb(X, MARGIN, FUN, ...)

Arguments

Value

see ?apply

See Also

apply()

Examples

## Not run: 

apply_pb(anscombe, 2, sd, na.rm = TRUE)

# larger dataset
df <- data.frame(rnorm(30000), rnorm(30000))
head(apply_pb(df, 1, sd))

# performance comparison
df <- data.frame(rnorm(90000), rnorm(90000))
system.time(apply(df, 1, sd))
system.time(apply_pb(df, 1, sd))

## End(Not run)

Make Bertin display of grid data.

Description

One of the most popular ways of displaying grid data has been adopted from Bertin's (1974) graphical proposals, which have had an immense influence onto data visualization. One of the most appealing ideas presented by Bertin is the concept of the reorderable matrix. It is comprised of graphical displays for each cell, allowing to identify structures by eye-balling reordered versions of the data matrix (see Bertin, 1974). In the context of repertory grids, the display is made up of a simple colored rectangle where the color denotes the corresponding score. Bright values correspond to low, dark to high scores. For an example of how to analyze a Bertin display see e.g. Dick (2000) and Raeithel (1998).

Usage

bertin(
  x,
  colors = c("white", "black"),
  showvalues = TRUE,
  xlim = c(0.2, 0.8),
  ylim = c(0, 0.6),
  margins = c(0, 1, 1),
  cex.elements = 0.7,
  cex.constructs = 0.7,
  cex.text = 0.6,
  col.text = NA,
  border = "white",
  lheight = 0.75,
  id = c(TRUE, TRUE),
  col.e = "black",
  col.c.left = "black",
  col.c.right = "black",
  col.e.lines = "black",
  cc = 0,
  cr = 0,
  cc.old = 0,
  cr.old = 0,
  col.mark.fill = "#FCF5A4",
  print = TRUE,
  ...
)

Arguments

Value

NULL just for the side effects, i.e. printing.

References

Bertin, J. (1974). Graphische Semiologie: Diagramme, Netze, Karten. Berlin, New York: de Gruyter.

Dick, M. (2000). The Use of Narrative Grid Interviews in Psychological Mobility Research. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(2).

Raeithel, A. (1998). Kooperative Modellproduktion von Professionellen und Klienten - erlauetert am Beispiel des Repertory Grid. Selbstorganisation, Kooperation, Zeichenprozess: Arbeiten zu einer kulturwissenschaftlichen, anwendungsbezogenen Psychologie (pp. 209-254). Opladen: Westdeutscher Verlag.

Examples

bertin(feixas2004)
bertin(feixas2004, c("white", "darkblue"))
bertin(feixas2004, showvalues = FALSE)
bertin(feixas2004, border = "grey")
bertin(feixas2004, cex.text = .9)
bertin(feixas2004, id = c(FALSE, FALSE))

bertin(feixas2004, cc = 3, cr = 4)
bertin(feixas2004, cc = 3, cr = 4, col.mark.fill = "#e6e6e6")

Bertin display with corresponding cluster analysis.

Description

Element columns and constructs rows are ordered according to cluster criterion. Various distance measures as well as cluster methods are supported.

Usage

bertinCluster(
  x,
  dmethod = c("euclidean", "euclidean"),
  cmethod = c("ward.D", "ward.D"),
  p = c(2, 2),
  align = TRUE,
  trim = NA,
  type = c("triangle"),
  xsegs = c(0, 0.2, 0.7, 0.9, 1),
  ysegs = c(0, 0.1, 0.7, 1),
  x.off = 0.01,
  y.off = 0.01,
  cex.axis = 0.6,
  col.axis = grey(0.4),
  draw.axis = TRUE,
  ...
)

Arguments

Value

A list of two hclust() object, for elements and constructs respectively.

See Also

cluster()

Examples

# default is euclidean distance and ward clustering
bertinCluster(bell2010)

### applying different distance measures and cluster methods

# euclidean distance and single linkage clustering
bertinCluster(bell2010, cmethod = "single")
# manhattan distance and single linkage clustering
bertinCluster(bell2010, dmethod = "manhattan", cm = "single")
# minkowksi distance with power of 2 = euclidean distance
bertinCluster(bell2010, dm = "mink", p = 2)

### using different methods for constructs and elements

# ward clustering for constructs, single linkage for elements
bertinCluster(bell2010, cmethod = c("ward.D", "single"))
# euclidean distance measure for constructs, manhatten
# distance for elements
bertinCluster(bell2010, dmethod = c("euclidean", "man"))
# minkowski metric with different powers for constructs and elements
bertinCluster(bell2010, dmethod = "mink", p = c(2, 1))

### clustering either constructs or elements only
# euclidean distance and ward clustering for constructs no
# clustering for elements
bertinCluster(bell2010, cmethod = c("ward.D", NA))
# euclidean distance and single linkage clustering for elements
# no clustering for constructs
bertinCluster(bell2010, cm = c(NA, "single"), align = FALSE)

### changing the appearance
# different dendrogram type
bertinCluster(bell2010, type = "rectangle")
# no axis drawn for dendrogram
bertinCluster(bell2010, draw.axis = FALSE)

### passing on arguments to bertin function via ...
# grey cell borders in bertin display
bertinCluster(bell2010, border = "grey")
# omit printing of grid scores, i.e. colors only
bertinCluster(bell2010, showvalues = FALSE)

### changing the layout
# making the vertical dendrogram bigger
bertinCluster(bell2010, xsegs = c(0, .2, .5, .7, 1))
# making the horizontal dendrogram bigger
bertinCluster(bell2010, ysegs = c(0, .3, .8, 1))

Concatenate the constructs of two grids.

Description

I.e. the constructs are combined to form one long grid. This function can be used in order to analyze multiple grids as one 'big grid' (eg. Slater, 1977, chap. 11).

Usage

bind(x, y, match = TRUE, index = TRUE)

Arguments

Value

repgrid object

References

Slater, P. (1977). The measurement of intrapersonal space by grid technique. London: Wiley.

Examples

## Not run: 

a <- randomGrid()
b <- randomGrid()
elements(b) <- rev(elements(a)) # reverse elements
bindConstructs(a, b)

bindConstructs(a, b, m = F) # no binding

## End(Not run)

Concatenate the constructs of two or more grids.

Description

I.e. the constructs are combined to form one long grid. The girds must have the same set of elements and an identical scale range. The order of the elements may differ.

Usage

bindConstructs(..., index = FALSE)

Arguments

Details

This function can be used in order to analyze multiple grids as one 'big grid' (eg. Slater, 1977, chap. 11).

Value

repgrid object with concatenated constructs.

References

Slater, P. (1977). The measurement of intrapersonal space by grid technique. London: Wiley.

Examples

a <- randomGrid()
b <- randomGrid()
elements(b) <- rev(elements(a)) # reverse elements
bindConstructs(a, b)
bindConstructs(a, b, a)

# using lists of repgrid objects
bindConstructs(a, list(a, b))

Draw a two-dimensional biplot.

Description

The biplot is the central way to create a joint plot of elements and constructs. Depending on the parameters chosen it contains information on the distances between elements and constructs. Also the relative values the elements have on a construct can be read off by projection the element onto the construct vector. A lot of parameters can be changed rendering different types of biplots (ESA, Slater's) and different looks (colors, text size). See the example section below to get started.

Usage

biplot2d(
  x,
  dim = c(1, 2),
  map.dim = 3,
  center = 1,
  normalize = 0,
  g = 0,
  h = 1 - g,
  col.active = NA,
  col.passive = NA,
  e.point.col = "black",
  e.point.cex = 0.9,
  e.label.col = "black",
  e.label.cex = 0.7,
  e.color.map = c(0.4, 1),
  c.point.col = "black",
  c.point.cex = 0,
  c.label.col = "black",
  c.label.col.left = NULL,
  c.label.col.right = NULL,
  c.label.cex = 0.7,
  c.color.map = c(0.4, 1),
  c.points.devangle = 91,
  c.labels.devangle = 91,
  c.points.show = TRUE,
  c.labels.show = TRUE,
  e.points.show = TRUE,
  e.labels.show = TRUE,
  inner.positioning = TRUE,
  outer.positioning = TRUE,
  c.labels.inside = FALSE,
  c.lines = TRUE,
  col.c.lines = grey(0.9),
  flipaxes = c(FALSE, FALSE),
  strokes.x = 0.1,
  strokes.y = 0.1,
  offsetting = TRUE,
  offset.labels = 0,
  offset.e = 1,
  axis.ext = 0.1,
  mai = c(0.2, 1.5, 0.2, 1.5),
  rect.margins = c(0.01, 0.01),
  srt = 45,
  cex.pos = 0.7,
  xpd = TRUE,
  unity = FALSE,
  unity3d = FALSE,
  scale.e = 0.9,
  zoom = 1,
  var.show = TRUE,
  var.cex = 0.7,
  var.col = grey(0.1),
  ...
)

Arguments

Details

For the construction of a biplot the grid matrix is first centered and normalized according to the prompted options.

Next, the matrix is decomposed by singular value decomposition (SVD) into

X = UDV^T

The biplot is made up of two matrices

X = GH^T

These matrices are construed on the basis of the SVD results.

\hat{X} = UD^gD^hV^T

Note that the grid matrix values are only recovered and the projection property is only given if g + h = 1

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

biplot2d(boeker) # biplot of boeker data
## Not run: 
biplot2d(boeker, c.lines = T) # add construct lines
biplot2d(boeker, center = 2) # with column centering
biplot2d(boeker, center = 4) # midpoint centering
biplot2d(boeker, normalize = 1) # normalization of constructs

biplot2d(boeker, dim = 2:3) # plot 2nd and 3rd dimension
biplot2d(boeker, dim = c(1, 4)) # plot 1st and 4th dimension

biplot2d(boeker, g = 1, h = 1) # assign singular values to con. & elem.
biplot2d(boeker, g = 1, h = 1, center = 1) # row centering (Slater)
biplot2d(boeker, g = 1, h = 1, center = 4) # midpoint centering (ESA)

biplot2d(boeker, e.color = "red", c.color = "blue") # change colors
biplot2d(boeker, c.color = c("white", "darkred")) # mapped onto color range

biplot2d(boeker, unity = T) # scale con. & elem. to equal length
biplot2d(boeker, unity = T, scale.e = .5) # scaling factor for element vectors

biplot2d(boeker, e.labels.show = F) # do not show element labels
biplot2d(boeker, e.labels.show = c(1, 2, 4)) # show labels for elements 1, 2 and 4
biplot2d(boeker, e.points.show = c(1, 2, 4)) # only show elements 1, 2 and 4
biplot2d(boeker, c.labels.show = c(1:4)) # show constructs labels 1 to 4
biplot2d(boeker, c.labels.show = c(1:4)) # show constructs labels except 1 to 4

biplot2d(boeker, e.cex.map = 1) # change size of texts for elements
biplot2d(boeker, c.cex.map = 1) # change size of texts for constructs

biplot2d(boeker, g = 1, h = 1, c.labels.inside = T) # constructs inside the plot
biplot2d(boeker,
  g = 1, h = 1, c.labels.inside = T, # different margins and elem. color
  mai = c(0, 0, 0, 0), e.color = "red"
)

biplot2d(boeker, strokes.x = .3, strokes.y = .05) # change length of strokes

biplot2d(boeker, flipaxes = c(T, F)) # flip x axis
biplot2d(boeker, flipaxes = c(T, T)) # flip x and y axis

biplot2d(boeker, outer.positioning = F) # no positioning of con.-labels

biplot2d(boeker, c.labels.devangle = 20) # only con. within 20 degree angle

## End(Not run)

Draw grid in rgl (3D device).

Description

The 3D biplot opens an interactive 3D device that can be rotated and zoomed using the mouse. A 3D device facilitates the exploration of grid data as significant proportions of the sum-of-squares are often represented beyond the first two dimensions. Also, in a lot of cases it may be of interest to explore the grid space from a certain angle, e.g. to gain an optimal view onto the set of elements under investigation (e.g. Raeithel, 1998).

Usage

biplot3d(
  x,
  dim = 1:3,
  labels.e = TRUE,
  labels.c = TRUE,
  lines.c = 2,
  lef = 1.3,
  center = 1,
  normalize = 0,
  g = 0,
  h = 1,
  col.active = NA,
  col.passive = NA,
  c.axis.show = TRUE,
  c.sphere.show = FALSE,
  c.sphere.col = grey(0.4),
  c.cex = 0.6,
  c.text.col = grey(0.4),
  e.sphere.show = TRUE,
  e.labels.show = TRUE,
  e.sphere.col = grey(0),
  e.cex = 0.6,
  e.text.col = grey(0),
  alpha.sphere = 0.05,
  col.sphere = "black",
  unity = FALSE,
  unity3d = FALSE,
  scale.e = 0.9,
  zoom = 1,
  ...
)

Arguments

References

Raeithel, A. (1998). Kooperative Modellproduktion von Professionellen und Klienten - erlauetert am Beispiel des Repertory Grid. Selbstorganisation, Kooperation, Zeichenprozess: Arbeiten zu einer kulturwissenschaftlichen, anwendungsbezogenen Psychologie (pp. 209-254). Opladen: Westdeutscher Verlag.

See Also

Unsophisticated biplot: biplotSimple();
2D biplots: biplot2d(), biplotEsa2d(), biplotSlater2d();
Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();
Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();
Function to set view in 3D: home().

Examples

## Not run: 
biplot3d(boeker)

biplot3d(boeker, e.sphere.show = 1:4)
biplot3d(boeker, e.sphere.show = 1:4, e.labels.show = 1:2)
biplot3d(boeker, c.axis.show = 1:2)

biplot3d(boeker, e.sphere.col = "red", c.text.col = "blue")
biplot3d(boeker, e.cex = 1)
biplot3d(boeker, col.sphere = "red")

biplot3d(boeker, g = 1, h = 1) # INGRID biplot
biplot3d(boeker, g = 1, h = 1, center = 4) # ESA biplot

## End(Not run)

biplotDraw is the workhorse doing the drawing of a 2D biplot.

Description

When the number of elements and constructs differs to a large extent, the absolute values of the coordinates for either constructs or elements will be much smaller or greater. This is an inherent property of the biplot. In the case it is not necessary to be able to read off the original data entries from the plot, the axes for elements and constructs can be scaled separately. The proportional projection values will stay unaffected. the absolute will change though. For grid interpretation the absolute values are usually oh no importance. Thus, there is an optional argument normalize which is FALSE as a default which rescales the axes so the longest construct and element vector will be set to the length of 1.

Usage

biplotDraw(
  x,
  inner.positioning = TRUE,
  outer.positioning = TRUE,
  c.labels.inside = FALSE,
  flipaxes = c(FALSE, FALSE),
  strokes.x = 0.1,
  strokes.y = 0.1,
  offsetting = TRUE,
  offset.labels = 0,
  offset.e = 1,
  axis.ext = 0.1,
  mai = c(0.2, 1.5, 0.2, 1.5),
  rect.margins = c(0.01, 0.01),
  srt = 45,
  cex.pos = 0.7,
  xpd = TRUE,
  c.lines = TRUE,
  col.c.lines = grey(0.9),
  zoom = 1,
  ...
)

Arguments

Value

Invisible return of dataframe used during construction of plot (useful for developers).

Plot an eigenstructure analysis (ESA) biplot in 2D.

Description

The ESA is a special type of biplot suggested by Raeithel (e.g. 1998). It uses midpoint centering as a default. Note that the eigenstructure analysis is just a special case of a biplot that can also be produced using the biplot2d() function with the arguments ⁠center=4, g=1, h=1⁠. Here, only the arguments that are modified for the ESA biplot are described. To see all the parameters that can be changed see biplot2d().

Usage

biplotEsa2d(x, center = 4, g = 1, h = 1, ...)

Arguments

References

Raeithel, A. (1998). Kooperative Modellproduktion von Professionellen und Klienten. Erlaeutert am Beispiel des Repertory Grid. In A. Raeithel (1998). Selbstorganisation, Kooperation, Zeichenprozess. Arbeiten zu einer kulturwissenschaftlichen, anwendungsbezogenen Psychologie (p. 209-254). Opladen: Westdeutscher Verlag.

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

## Not run: 
# See examples in [biplot2d()] as the same arguments
# can used for this function.

## End(Not run)

Draw the eigenstructure analysis (ESA) biplot in rgl (3D device).

Description

The 3D biplot opens an interactive 3D device that can be rotated and zoomed using the mouse. A 3D device facilitates the exploration of grid data as significant proportions of the sum-of-squares are often represented beyond the first two dimensions. Also, in a lot of cases it may be of interest to explore the grid space from a certain angle, e.g. to gain an optimal view onto the set of elements under investigation (e.g. Raeithel, 1998). Note that the eigenstructure analysis just a special case of a biplot that can also be produced using the biplot3d() function with the arguments ⁠center=4, g=1, h=1⁠.

Usage

biplotEsa3d(x, center = 1, g = 1, h = 1, ...)

Arguments

See Also

Unsophisticated biplot: biplotSimple();
2D biplots: biplot2d(), biplotEsa2d(), biplotSlater2d();
Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();
Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();
Function to set view in 3D: home().

Examples

## Not run: 

biplotEsa3d(boeker)
biplotEsa3d(boeker, unity3d = T)

biplotEsa3d(boeker,
  e.sphere.col = "red",
  c.text.col = "blue"
)
biplotEsa3d(boeker, e.cex = 1)
biplotEsa3d(boeker, col.sphere = "red")

## End(Not run)

Plot an eigenstructure analysis (ESA) in 2D grid with 3D impression (pseudo 3D).

Description

The ESA is a special type of biplot suggested by Raeithel (e.g. 1998). It uses midpoint centering as a default. Note that the eigenstructure analysis is just a special case of a biplot that can also be produced using the biplot2d() function with the arguments ⁠center=4, g=1, h=1⁠. Here, only the arguments that are modified for the ESA biplot are described. To see all the parameters that can be changed see biplot2d() and biplotPseudo3d().

Usage

biplotEsaPseudo3d(x, center = 4, g = 1, h = 1, ...)

Arguments

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

## Not run: 
# See examples in [biplotPseudo3d()] as the same arguments
# can used for this function.

## End(Not run)

Draws a biplot of the grid in 2D with depth impression (pseudo 3D).

Description

This version is basically a 2D biplot. It only modifies color and size of the symbols in order to create a 3D impression of the data points. This function will call the standard biplot2d() function with some modified arguments. For the whole set of arguments that can be used see biplot2d(). Here only the arguments special to biplotPseudo3d are outlined.

Usage

biplotPseudo3d(
  x,
  dim = 1:2,
  map.dim = 3,
  e.point.col = c("white", "black"),
  e.point.cex = c(0.6, 1.2),
  e.label.col = c("white", "black"),
  e.label.cex = c(0.6, 0.8),
  e.color.map = c(0.4, 1),
  c.point.col = c("white", "darkred"),
  c.point.cex = c(0.6, 1.2),
  c.label.col = c("white", "darkred"),
  c.label.cex = c(0.6, 0.8),
  c.color.map = c(0.4, 1),
  ...
)

Arguments

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

## Not run: 
# biplot with 3D impression
biplotPseudo3d(boeker)
# Slater's biplot with 3D impression
biplotPseudo3d(boeker, g = 1, h = 1, center = 1)

# show 2nd and 3rd dim. and map 4th
biplotPseudo3d(boeker, dim = 2:3, map.dim = 4)

# change elem. colors
biplotPseudo3d(boeker, e.color = c("white", "darkgreen"))
# change con. colors
biplotPseudo3d(boeker, c.color = c("white", "darkgreen"))
# change color mapping range
biplotPseudo3d(boeker, c.colors.map = c(0, 1))

# set uniform con. text size
biplotPseudo3d(boeker, c.cex = 1)
# change text size mapping range
biplotPseudo3d(boeker, c.cex = c(.4, 1.2))

## End(Not run)

A graphically unsophisticated version of a biplot.

Description

It will draw elements and constructs vectors using similar arguments as biplot2d(). It is a version for quick exploration used during development.

Usage

biplotSimple(
  x,
  dim = 1:2,
  center = 1,
  normalize = 0,
  g = 0,
  h = 1 - g,
  unity = T,
  col.active = NA,
  col.passive = NA,
  scale.e = 0.9,
  zoom = 1,
  e.point.col = "black",
  e.point.cex = 1,
  e.label.col = "black",
  e.label.cex = 0.7,
  c.point.col = grey(0.6),
  c.label.col = grey(0.6),
  c.label.cex = 0.6,
  ...
)

Arguments

Value

repgrid object.

See Also

Unsophisticated biplot: biplotSimple();
2D biplots: biplot2d(), biplotEsa2d(), biplotSlater2d();
Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();
Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();
Function to set view in 3D: home().

Examples

## Not run: 

biplotSimple(boeker)
biplotSimple(boeker, unity = F)

biplotSimple(boeker, g = 1, h = 1) # INGRID biplot
biplotSimple(boeker, g = 1, h = 1, center = 4) # ESA biplot

biplotSimple(boeker, zoom = .9) # zooming out
biplotSimple(boeker, scale.e = .6) # scale element vectors

biplotSimple(boeker, e.point.col = "brown") # change colors
biplotSimple(boeker,
  e.point.col = "brown",
  c.label.col = "darkblue"
)

## End(Not run)

Draws Slater's INGRID biplot in 2D.

Description

The default is to use row centering and no normalization. Note that Slater's biplot is just a special case of a biplot that can be produced using the biplot2d() function with the arguments ⁠center=1, g=1, h=1⁠. The arguments that can be used in this function are the same as in biplot2d(). Here, only the arguments that are set for Slater's biplot are described. To see all the parameters that can be changed see biplot2d().

Usage

biplotSlater2d(x, center = 1, g = 1, h = 1, ...)

Arguments

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

## Not run: 
# See examples in [biplot2d()] as the same arguments
# can used for this function.

## End(Not run)

Draw the Slater's INGRID biplot in rgl (3D device).

Description

The 3D biplot opens an interactive 3D device that can be rotated and zoomed using the mouse. A 3D device facilitates the exploration of grid data as significant proportions of the sum-of-squares are often represented beyond the first two dimensions. Also, in a lot of cases it may be of interest to explore the grid space from a certain angle, e.g. to gain an optimal view onto the set of elements under investigation (e.g. Raeithel, 1998). Note that Slater's biplot is just a special case of a biplot that can be produced using the biplot3d() function with the arguments ⁠center=1, g=1, h=1⁠.

Usage

biplotSlater3d(x, center = 1, g = 1, h = 1, ...)

Arguments

See Also

Unsophisticated biplot: biplotSimple();
2D biplots: biplot2d(), biplotEsa2d(), biplotSlater2d();
Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();
Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();
Function to set view in 3D: home().

Examples

## Not run: 

biplotSlater3d(boeker)
biplotSlater3d(boeker, unity3d = T)

biplotSlater3d(boeker,
  e.sphere.col = "red",
  c.text.col = "blue"
)
biplotSlater3d(boeker, e.cex = 1)
biplotSlater3d(boeker, col.sphere = "red")

## End(Not run)

Draws Slater's biplot in 2D with depth impression (pseudo 3D).

Description

The default is to use row centering and no normalization. Note that Slater's biplot is just a special case of a biplot that can be produced using the biplotPseudo3d() function with the arguments ⁠center=1, g=1, h=1⁠. Here, only the arguments that are modified for Slater's biplot are described. To see all the parameters that can be changed see biplot2d() and biplotPseudo3d().

Usage

biplotSlaterPseudo3d(x, center = 1, g = 1, h = 1, ...)

Arguments

See Also

• Unsophisticated biplot: biplotSimple();

• 2D biplots:biplot2d(), biplotEsa2d(), biplotSlater2d();

• Pseudo 3D biplots: biplotPseudo3d(), biplotEsaPseudo3d(), biplotSlaterPseudo3d();

• Interactive 3D biplots: biplot3d(), biplotEsa3d(), biplotSlater3d();

• Function to set view in 3D: home()

Examples

## Not run: 
# See examples in [biplotPseudo3d()] as the same arguments
# can used for this function.

## End(Not run)

Calculate angles for points in first two columns.

Description

The angles of the points given by the values in the first and second column of a matrix seen from the origin are calculated in degrees.

Usage

calcAngles(x, dim = c(1, 2), clockwise = TRUE)

Arguments

Value

vector. The angles of each row point with the origin as reference.

Examples

## Not run: 
m <- matrix(rnorm(9), 3)
calcAngles()

## End(Not run)

Calculate coordinates for biplot.

Description

Calculate coordinates for biplot.

Usage

calcBiplotCoords(x, g = 0, h = 1 - g, col.active = NA, col.passive = NA, ...)

Arguments

Value

a list.

Coordinates of a surrounding rectangle in direction of a given vector.

Description

An arbitrary numeric vector in 2D is to be extended so it will end on the borders of a surrounding rectangle of a given size. Currently the vector is supposed to start in the origin c(0,0).

Usage

calcCoordsBorders(x, y, xmax = 1, ymax = 1, cx = 0, cy = 0)

Arguments

Value

a dataframe containing the x and y coordinates for the extended vectors.

Examples

## Not run: 
calcCoordsBorders(1:10, 10:1)

x <- c(-100:0, 0:100, 100:0, 0:-100) / 10
y <- c(0:100, 100:0, -(0:100), -(100:0)) / 10
xy1 <- calcCoordsBorders(x, y)
xy2 <- calcCoordsBorders(x, y, xm = 1.2, ym = 1.2)
plot(xy2[, 1], xy2[, 2], type = "n")
segments(xy1[, 1], xy1[, 2], xy2[, 1], xy2[, 2])

## End(Not run)

make ascending and descending vector

Description

along a given length n make ascending indices until reaching the midpoint and descend afterwards again.

Usage

cascade(n, type = 1)

Arguments

Value

vector (type 1 or 3) or list (type 2)

Examples

## Not run: 
for (n in 1:10) {
  print(cascade(n))
}

## End(Not run)

Concatenate the elements of two grids

Description

Concatenate the elements of two grids

Usage

## S3 method for class 'repgrid'
cbind(..., .reorder = TRUE, .unique = FALSE)

Arguments

Value

repgrid with combined elements.

Examples

x <- boeker[, 1:2]
y <- boeker[, 5:7]
cbind(x, y)
x / y
y_reordered <- y[sample(nrow(y)), ]
cbind(x, y, y_reordered)

Centering of rows (constructs) and/or columns (elements).

Description

Centering of rows (constructs) and/or columns (elements).

Usage

center(x, center = 1, ...)

Arguments

Value

matrix containing the transformed values.

Note

If scale midpoint centering is applied no row or column centering can be applied simultaneously. TODO: After centering the standard representation mode does not work any more as it remains unclear what color values to attach to the centered values.

Examples

center(bell2010) # no centering
center(bell2010, rows = T) # row centering of grid
center(bell2010, cols = T) # column centering of grid
center(bell2010, rows = T, cols = T) # row and column centering

clear ratings

Description

set certain ratings in grid to NA (unknown)

Usage

clearRatings(x, rows = NA, cols = NA, layer = 1)

Arguments

Value

repgrid object

Examples

## Not run: 

####  TODO  ####

## End(Not run)

Cluster analysis (of constructs or elements).

Description

cluster is a preliminary implementation of a cluster function. It supports various distance measures as well as cluster methods. More is to come.

Usage

cluster(
  x,
  along = 0,
  dmethod = "euclidean",
  cmethod = "ward.D",
  p = 2,
  align = TRUE,
  trim = NA,
  main = NULL,
  mar = c(4, 2, 3, 15),
  cex = 0,
  lab.cex = 0.8,
  cex.main = 0.9,
  print = TRUE,
  ...
)

Arguments

Details

align: Aligning will reverse constructs if necessary to yield a maximal similarity between constructs. In a first step the constructs are clustered including both directions. In a second step the direction of a construct that yields smaller distances to the adjacent constructs is preserved and used for the final clustering. As a result, every construct is included once but with an orientation that guarantees optimal clustering. This approach is akin to the procedure used in FOCUS (Jankowicz & Thomas, 1982).

Value

Reordered repgrid object.

References

Jankowicz, D., & Thomas, L. (1982). An Algorithm for the Cluster Analysis of Repertory Grids in Human Resource Development. Personnel Review, 11(4), 15-22. doi:10.1108/eb055464.

See Also

bertinCluster()

Examples

cluster(bell2010)
cluster(bell2010, main = "My cluster analysis") # new title
cluster(bell2010, type = "t") # different drawing style
cluster(bell2010, dmethod = "manhattan") # using manhattan metric
cluster(bell2010, cmethod = "single") # do single linkage clustering
cluster(bell2010, cex = 1, lab.cex = 1) # change appearance
cluster(bell2010, lab.cex = .7, edgePar = list(lty = 1:2, col = 2:1)) # advanced appearance changes

Multiscale bootstrap cluster analysis.

Description

p-values are calculated for each branch of the cluster dendrogram to indicate the stability of a specific partition. clusterBoot will yield the same clusters as the cluster() function (i.e. standard hierarchical clustering) with additional p-values. Two kinds of p-values are reported: bootstrap probabilities (BP) and approximately unbiased (AU) probabilities (see Details section for more information).

Usage

clusterBoot(
  x,
  along = 1,
  align = TRUE,
  dmethod = "euclidean",
  cmethod = "ward.D",
  p = 2,
  nboot = 1000,
  r = seq(0.8, 1.4, by = 0.1),
  seed = NULL,
  trim = NA,
  ...
)

Arguments

Details

In standard (hierarchical) cluster analysis the question arises which of the identified structures are significant or just emerged by chance. Over the last decade several methods have been developed to test structures for robustness. One line of research in this area is based on resampling. The idea is to resample the rows or columns of the data matrix and to build the dendrogram for each bootstrap sample (Felsenstein, 1985). The p-values indicates the percentage of times a specific structure is identified across the bootstrap samples. It was shown that the p-value is biased (Hillis & Bull, 1993; Zharkikh & Li, 1995). In the literature several methods for bias correction have been proposed. In clusterBoot a method based on the multiscale bootstrap is used to derive corrected (approximately unbiased) p-values (Shimodaira, 2002, 2004). In conventional bootstrap analysis the size of the bootstrap sample is identical to the original sample size. Multiscale bootstrap varies the bootstrap sample size in order to infer a correction formula for the biased p-value on the basis of the variation of the results for the different sample sizes (Suzuki & Shimodaira, 2006).

align: Aligning will reverse constructs if necessary to yield a maximal similarity between constructs. In a first step the constructs are clustered including both directions. In a second step the direction of a construct that yields smaller distances to the adjacent constructs is preserved and used for the final clustering. As a result, every construct is included once but with an orientation that guarantees optimal clustering. This approach is akin to the procedure used in FOCUS (Jankowicz & Thomas, 1982).

Value

A pvclust object as returned by the function pvclust::pvclust()

References

Felsenstein, J. (1985). Confidence Limits on Phylogenies: An Approach Using the Bootstrap. Evolution, 39(4), 783. doi:10.2307/2408678

Hillis, D. M., & Bull, J. J. (1993). An Empirical Test of Bootstrapping as a Method for Assessing Confidence in Phylogenetic Analysis. Systematic Biology, 42(2), 182-192.

Jankowicz, D., & Thomas, L. (1982). An Algorithm for the Cluster Analysis of Repertory Grids in Human Resource Development. Personnel Review, 11(4), 15-22. doi:10.1108/eb055464.

Shimodaira, H. (2002) An approximately unbiased test of phylogenetic tree selection. Syst, Biol., 51, 492-508.

Shimodaira,H. (2004) Approximately unbiased tests of regions using multistep- multiscale bootstrap resampling. Ann. Stat., 32, 2616-2614.

Suzuki, R., & Shimodaira, H. (2006). Pvclust: an R package for assessing the uncertainty in hierarchical clustering. Bioinformatics, 22(12), 1540-1542. doi:10.1093/bioinformatics/btl117

Zharkikh, A., & Li, W.-H. (1995). Estimation of confidence in phylogeny: the complete-and-partial bootstrap technique. Molecular Phylogenetic Evolution, 4(1), 44-63.

Examples

## Not run: 

# pvclust must be loaded
library(pvclust)

# p-values for construct dendrogram
s <- clusterBoot(boeker)
plot(s)
pvrect(s, max.only = FALSE)

# p-values for element dendrogram
s <- clusterBoot(boeker, along = 2)
plot(s)
pvrect(s, max.only = FALSE)

## End(Not run)

Calculate correlations between constructs.

Description

Different types of correlations can be requested: PMC, Kendall tau rank correlation, Spearman rank correlation.

Usage

constructCor(
  x,
  method = c("pearson", "kendall", "spearman"),
  trim = 20,
  index = FALSE
)

Arguments

Value

Returns a matrix of construct correlations.

See Also

elementCor()

Examples

# three different types of correlations
constructCor(mackay1992)
constructCor(mackay1992, method = "kendall")
constructCor(mackay1992, method = "spearman")

# format output
constructCor(mackay1992, trim = 6)
constructCor(mackay1992, index = TRUE, trim = 6)

# save correlation matrix for further processing
r <- constructCor(mackay1992)
r
print(r, digits = 5)

# accessing the correlation matrix
r[1, 3]

Calculate Somers' d for the constructs.

Description

Somer's d is an asymmetric association measure as it depends on which variable is set as dependent and independent. The direction of dependency needs to be specified.

Usage

constructD(x, dependent = "columns", trim = 30, index = TRUE)

Arguments

Value

matrix of construct correlations.

Note

         Thanks to Marc Schwartz for supplying the code to calculate
               Somers' d.

References

Somers, R. H. (1962). A New Asymmetric Measure of Association for Ordinal Variables. American Sociological Review, 27(6), 799-811.

Examples

## Not run: 

constructD(fbb2003) # columns as dependent (default)
constructD(fbb2003, "c") # row as dependent
constructD(fbb2003, "s") # symmetrical index

# suppress printing
d <- constructD(fbb2003, out = 0, trim = 5)
d

# more digits
constructD(fbb2003, dig = 3)

# add index column, no trimming
constructD(fbb2003, col.index = TRUE, index = F, trim = NA)

## End(Not run)

Principal component analysis (PCA) of inter-construct correlations.

Description

Various methods for rotation and methods for the calculation of the correlations are available. Note that the number of factors has to be specified. For more information on the PCA function itself type ?principal.

Usage

constructPca(
  x,
  nfactors = 3,
  rotate = "varimax",
  method = "pearson",
  trim = NA
)

Arguments

Value

Returns an object of class constructPca.

References

Fransella, F., Bell, R. & Bannister, D. (2003). A Manual for Repertory Grid Technique (2. Ed.). Chichester: John Wiley & Sons.

See Also

To extract the PCA loadings for further processing see constructPcaLoadings().

Examples

constructPca(bell2010)

# data from grid manual by Fransella et al. (2003, p. 87)
# note that the construct order is different
constructPca(fbb2003, nfactors = 2)

# no rotation
constructPca(fbb2003, rotate = "none")

# use a different type of correlation (Spearman)
constructPca(fbb2003, method = "spearman")

# save output to object
m <- constructPca(fbb2003, nfactors = 2)
m

# different printing options
print(m, digits = 5)
print(m, cutoff = .3)

Extract loadings from PCA of constructs.

Description

Extract loadings from PCA of constructs.

Usage

constructPcaLoadings(x)

Arguments

Value

A matrix containing the factor loadings.

Examples

p <- constructPca(bell2010)
l <- constructPcaLoadings(p)
l[1, ]
l[, 1]
l[1, 1]

Root mean square (RMS) of inter-construct correlations.

Description

The RMS is also known as 'quadratic mean' of the inter-construct correlations. The RMS serves as a simplification of the correlation table. It reflects the average relation of one construct to all other constructs. Note that as the correlations are squared during its calculation, the RMS is not affected by the sign of the correlation (cf. Fransella, Bell & Bannister, 2003, p. 86).

Usage

constructRmsCor(x, method = "pearson", trim = NA)

Arguments

Value

dataframe of the RMS of inter-construct correlations

References

Fransella, F., Bell, R. C., & Bannister, D. (2003). A Manual for Repertory Grid Technique (2. Ed.). Chichester: John Wiley & Sons.

See Also

elementRmsCor(), constructCor()

Examples

# data from grid manual by Fransella, Bell and Bannister
constructRmsCor(fbb2003)
constructRmsCor(fbb2003, trim = 20)

# modify output
r <- constructRmsCor(fbb2003)
print(r, digits = 5)
   # access calculation results
   r[2, 1]

Get or replace construct poles

Description

Allows to get and set construct poles. Replaces the older functions getConstructNames, getConstructNames2, and eNames which are deprecated.

Usage

constructs(x, collapse = FALSE, sep = " - ")

constructs(x, i, j) <- value

leftpoles(x)

leftpoles(x, position) <- value

rightpoles(x)

rightpoles(x, position) <- value

Arguments

Examples

# shorten object name
x <- boeker

## get construct poles
constructs(x) # both left and right poles
leftpoles(x) # left poles only
rightpoles(x)
constructs(x, collapse = TRUE)

## replace construct poles
constructs(x)[1, 1] <- "left pole 1"
constructs(x)[1, "leftpole"] <- "left pole 1" # alternative
constructs(x)[1:3, 2] <- paste("right pole", 1:3)
constructs(x)[1:3, "rightpole"] <- paste("right pole", 1:3) # alternative
constructs(x)[4, 1:2] <- c("left pole 4", "right pole 4")

l <- leftpoles(x)
leftpoles(x) <- sample(l) # brind poles into random order
leftpoles(x)[1] <- "new left pole 1" # replace name of first left pole

# replace left poles of constructs 1 and 3
leftpoles(x)[c(1, 3)] <- c("new left pole 1", "new left pole 3")

convertImportObjectToRepGridObject.

Description

Convert the returned object from an import function into a repgrid object. Works for all importXInternal functions (except scivesco).

Usage

convertImportObjectToRepGridObject(import)

Arguments

Convert the returned object from the sci:vesco import function into a repgrid object.

Description

Convert the returned object from the sci:vesco import function into a repgrid object.

Usage

convertScivescoImportObjectToRepGridObject(import)

Arguments

Grid data from Bell (2010).

Description

Grid data originated (but is not shown in the paper) from a study by Haritos, Gindinis, Doan and Bell (2004) on element role titles. It was used to demonstrate the effects of construct alignment in Bell (2010, p. 46).

References

Bell, R. C. (2010). A note on aligning constructs. Personal Construct Theory and Practice, 7, 43-48.

Haritos, A., Gindidis, A., Doan, C., & Bell, R. C. (2004). The effect of element role titles on construct structure and content. Journal of constructivist psychology, 17(3), 221-236.

Examples

bell2010

Grid data from Bell and McGorry (1992).

Description

The grid data set is used in Bell's technical report "Using SPSS to Analyse Repertory Grid Data" (1997, p. 6). Originally, the data comes from a study by Bell and McGorry (1992).

References

Bell, R. C. (1977). Using SPSS to Analyse Repertory Grid Data. Technical Report, University of Melbourne.

Bell, R. C., & McGorry, P. (1992). The analysis of repertory grids used to monitor the perceptions of recovering psychotic patients. In A. Thomson & P. Cummins (Eds.), European Perspectives in Personal Construct Psychology (p. 137-150). Lincoln, UK: European Personal Construct Association.

Examples

bellmcgorry1992

Grid data from Boeker (1996).

Description

Grid data from a schizophrenic patient undergoing psychoanalytically oriented psychotherapy. The data was taken during the last stage of therapy (Boeker, 1996, p. 163).

References

Boeker, H. (1996). The reconstruction of the self in the psychotherapy of chronic schizophrenia: a case study with the Repertory Grid Technique. In: Scheer, J. W., Catina, A. (Eds.): Empirical Constructivism in Europe - The Personal Construct Approach (p. 160-167). Giessen: Psychosozial-Verlag.

Examples

boeker

# data is also available as Excel file
path <- system.file("extdata", "boeker.xlsx", package = "OpenRepGrid")
x <- importExcel(path)

Grid data from Fransella, Bell and Bannister (2003).

Description

A dataset used throughout the book "A Manual for Repertory Grid Technique" (Fransella, Bell and Bannister, 2003, p. 60).

References

Fransella, F., Bell, R. & Bannister, D. (2003). A Manual for Repertory Grid Technique (2. Ed.). Chichester: John Wiley & Sons.

Examples

fbb2003

Grid data from Feixas and Saul (2004).

Description

A description by the authors: "When Teresa, 22 years old, was seen by the second author (LAS) at the psychological services of the University of Salamanca, she was in the final year of her studies in chemical sciences. Although Teresa proves to be an excellent student, she reveals serious doubts about her self worth. She cries frequently, and has great difficulty in meeting others, even though she has a boyfriend who is extremely supportive. Teresa is anxiously hesitant about accepting a new job which would involve moving to another city 600 Km away from home." (Feixas & Saul, 2004, p. 77).

References

Feixas, G., & Saul, L. A. (2004). The Multi-Center Dilemma Project: an investigation on the role of cognitive conflicts in health. The Spanish Journal of Psychology, 7(1), 69-78.

Examples

feixas2004

Pre- and post therapy dataset from Leach et al. (2001).

Description

Case as described by the authors: "Sarah, aged 32, was referred with problems of depression and sexual difficulties relating to childhood sexual abuse. She had three children and was living with her male partner. From the age of 9, her brother, an adult, had sexually abused Sarah. She attended a group for survivors of child sexual abuse and completed repertory grids prior to the group, immediately after the group and at 3- and 6-month follow-up." (Leach et al. 2001, p. 230).

Details

leach2001a is the pre-therapy, leach2001b is the post-therapy therapy dataset. The construct and elements are identical.

References

Leach, C., Freshwater, K., Aldridge, J., & Sunderland, J. (2001). Analysis of repertory grids in clinical practice. The British Journalof Clinical Psychology, 40, 225-248.

Examples

leach2001a
leach2001b

Grid data from Mackay (1992).

Description

Data set 'Grid C' used in Mackay's paper on inter-element correlation (1992, p. 65).

References

Mackay, N. (1992). Identification, reflection, and correlation: Problems in the bases of repertory grid measures. International Journal of Personal Construct Psychology, 5(1), 57-75.

Examples

mackay1992

Grid data from Raeithel (1998).

Description

Grid data to demonstrate the use of Bertin diagrams (Raeithel, 1998, p. 223). The context of its administration is unknown.

References

Raeithel, A. (1998). Kooperative Modellproduktion von Professionellen und Klienten. Erlaeutert am Beispiel des Repertory Grid. In A. Raeithel (1998). Selbstorganisation, Kooperation, Zeichenprozess. Arbeiten zu einer kulturwissenschaftlichen, anwendungsbezogenen Psychologie (p. 209-254). Opladen: Westdeutscher Verlag.

Examples

raeithel

Drug addict's grid data set from Slater (1977, p. 32).

Description

Drug addict's grid data set from Slater (1977, p. 32).

References

Slater, P. (1977). The measurement of intrapersonal space by grid technique. London: Wiley.

Examples

slater1977a

Grid data from Slater (1977).

Description

Grid data (ranked) from a seventeen year old female psychiatric patient (Slater, 1977, p. 110). She was depressed, anxious and took to cutting herself. The data was originally reported by Watson (1970).

References

Slater, P. (1977). The measurement of intrapersonal space by grid technique. London: Wiley.

Watson, J. P. (1970). The relationship between a self-mutilating patient and her doctor. Psychotherapy and Psychosomatics, 18(1), 67-73.

Examples

slater1977b

decouple a grid

Description

decouple a grid

Usage

decouple(x)

Arguments

Sample dataframe with grid data (constructs are columns)

Description

This dataframe can be converted into a repgrid object via importDataframe(). The columns names are elements followed by the constructs (left_pole_1:right_pole_1 to left_pole_3:right_pole_3). The poles are separated by a colon by default (change via arg pole_sep). The rows contain the elements' entries (element name and ratings). The min and max of the rating scale should be passed explicitly via the args rmin and rmax. See sample data df_construct_columns.

Details

See Also

importDataframe()

Other grid dataframes: df_element_columns, df_long

Examples

df_construct_columns
importDataframe(df_construct_columns, format = "construct_columns", rmin = 1, rmax = 5)

Sample dataframe with grid data (elements are columns)

Description

This dataframe can be converted into a repgrid object via importDataframe(). The dataframe column names are the minimum of the rating scale (1), the element names (element_1 to element_4), the maximum of the rating scale (5), and optionally a column indicating the preferred pole. Each row contains the constructs' entries (left pole, ratings, right pole, preferred pole). The preferred pole must be one of left, right, none, NA (see preferredPoles()). See sample data df_element_columns.

Details

See Also

importDataframe()

Other grid dataframes: df_construct_columns, df_long

Examples

df_element_columns
importDataframe(df_element_columns)

Sample dataframe with grid data (long)

Description

This dataframe can be converted into a repgrid object via importDataframe(). The long format has this name because it has few columns and many rows. It is a common format in data analytics. Here, each row contains a different element-construct combination and the corresponding rating value. The format looks like this:

Details

The columns element, left_pole, right_pole, and rating are mandatory, the columns preferred_pole, rmin, and rmax are optional. rmin and rmax contain the min and max of the rating scale. Alternatively, you may pass rmin and rmax as arguments in the function call.

See Also

importDataframe()

Other grid dataframes: df_construct_columns, df_element_columns

Examples

df_long
importDataframe(df_long, format = "long")

Number of constructs and elements all repgrids in gridlist

Description

Number of constructs and elements all repgrids in gridlist

Usage

## S3 method for class 'gridlist'
dim(x)

Arguments

Details

nrow and ncol are also covered by this approach as they access the dim method. The others are not generic and making them so introduces warnings, which I try to avoid.

Value

A list of length 2.

Return size of a grid.

Description

dim returns a numeric vector of length two containing the number of constructs and elements.

Usage

## S3 method for class 'repgrid'
dim(x)

Arguments

Value

Numeric vector of length two with the number of constructs and elements.

See Also

getNoOfConstructs(); getNoOfElements()

Examples

## Not run: 

dim(bell2010)

## End(Not run)

Calculate minimal and maximal possible distance

Description

While the minimal distance will usually be zero, the maximal distance can be used to normalize arbitrary distances.

Usage

dist_minmax(x, along = 1, dmethod = "euclidean", p = 2, max.only = FALSE)

Distance measures (between constructs or elements).

Description

Various distance measures between elements or constructs are calculated.

Usage

distance(
  x,
  along = 1,
  dmethod = "euclidean",
  p = 2,
  normalize = FALSE,
  trim = 20,
  index = TRUE,
  ...
)

Arguments

Value

matrix object.

Examples

# between constructs
distance(bell2010, along = 1)
distance(bell2010, along = 1, normalize = TRUE)

# between elements
distance(bell2010, along = 2)

# several distance methods
distance(bell2010, dm = "man") # manhattan distance
distance(bell2010, dm = "mink", p = 3) # minkowski metric to the power of 3

# to save the results without printing to the console
d <- distance(bell2010, trim = 7)
d

# some more options when printing the distance matrix
print(d, digits = 5)
print(d, col.index = FALSE)
print(d, upper = FALSE)

# accessing entries from the matrix
d[1, 3]

'Hartmann distance' (standardized Slater distances).

Description

Calculate Hartmann distance

Usage

distanceHartmann(
  x,
  method = "simulate",
  reps = 10000,
  prob = NULL,
  progress = TRUE,
  distributions = FALSE
)

Arguments

Details

Hartmann (1992) showed in a simulation study that Slater distances (see distanceSlater()) based on random grids, for which Slater coined the expression quasis, have a skewed distribution, a mean and a standard deviation depending on the number of constructs elicited. He suggested a linear transformation (z-transformation) which takes into account the estimated (or expected) mean and the standard deviation of the derived distribution to standardize Slater distance scores across different grid sizes. 'Hartmann distances' represent a more accurate version of 'Slater distances'. Note that Hartmann distances are multiplied by -1. Hence, negative Hartmann values represent dissimilarity, i.e. a big Slater distance.

There are two ways to use this function. Hartmann distances can either be calculated based on the reference values (i.e. means and standard deviations of Slater distance distributions) as given by Hartmann in his paper. The second option is to conduct an instant simulation for the supplied grid size for each calculation. The second option will be more accurate when a big number of quasis is used in the simulation.

It is also possible to return the quantiles of the sample distribution and only the element distances considered 'significant' according to the quantiles defined.

Value

A matrix containing Hartmann distances. In the attributes several additional parameters can be found:

• arguments: A list of several parameters including mean and sd of Slater distribution.

• quantiles: Quantiles for Slater and Hartmann distance distribution.

• distributions: List with values of the simulated distributions.

Calculation

The 'Hartmann distance' is calculated as follows (Hartmann 1992, p. 49).

D = -1 (\frac{D_{slater} - M_c}{sd_c})

Where D_{slater} denotes the Slater distances of the grid, M_c the sample distribution's mean value and sd_c the sample distribution's standard deviation.

References

Hartmann, A. (1992). Element comparisons in repertory grid technique: Results and consequences of a Monte Carlo study. International Journal of Personal Construct Psychology, 5(1), 41-56.

See Also

distanceSlater()

Examples

## Not run: 

### basics  ###

distanceHartmann(bell2010)
distanceHartmann(bell2010, method = "simulate")
h <- distanceHartmann(bell2010, method = "simulate")
h

# printing options
print(h)
print(h, digits = 6)
# 'significant' distances only
print(h, p = c(.05, .95))

# access cells of distance matrix
h[1, 2]

### advanced  ###

# histogram of Slater distances and indifference region
h <- distanceHartmann(bell2010, distributions = TRUE)
l <- attr(h, "distributions")
hist(l$slater, breaks = 100)
hist(l$hartmann, breaks = 100)

## End(Not run)

Calculate power-transformed Hartmann distances.

Description

Hartmann (1992) suggested a transformation of Slater (1977) distances to make them independent from the size of a grid. Hartmann distances are supposed to yield stable cutoff values used to determine 'significance' of inter-element distances. It can be shown that Hartmann distances are still affected by grid parameters like size and the range of the rating scale used (Heckmann, 2012). The function distanceNormalize applies a Box-Cox (1964) transformation to the Hartmann distances in order to remove the skew of the Hartmann distance distribution. The normalized values show to have more stable cutoffs (quantiles) and better properties for comparison across grids of different size and scale range.

Usage

distanceNormalized(
  x,
  reps = 1000,
  prob = NULL,
  progress = TRUE,
  distributions = TRUE
)

Arguments

Details

The function distanceNormalize can also return the quantiles of the sample distribution and only the element distances considered 'significant' according to the quantiles defined.

Value

A matrix containing the standardized distances.
Further data is contained in the object's attributes:

Calculations

The 'power transformed Hartmann distance' are calculated as follows: The simulated Hartmann distribution is added a constant as the Box-Cox transformation can only be applied to positive values. Then a range of values for lambda in the Box-Cox transformation (Box & Cox, 1964) are tried out. The best lambda is the one maximizing the correlation of the quantiles with the standard normal distribution. The lambda value maximizing normality is used to transform Hartmann distances. As the resulting scale of the power transformation depends on lambda, the resulting values are z-transformed to derive a common scaling.

The code for the calculation of the optimal lambda was written by Ioannis Kosmidis.

References

Box, G. E. P., & Cox, D. R. (1964). An Analysis of Transformations. Journal of the Royal Statistical Society. Series B (Methodological), 26(2), 211-252.

Hartmann, A. (1992). Element comparisons in repertory grid technique: Results and consequences of a Monte Carlo study. International Journal of Personal Construct Psychology, 5(1), 41-56.

Heckmann, M. (2012). Standardizing inter-element distances in grids - A revision of Hartmann's distances, 11th Biennal Conference of the European Personal Construct Association (EPCA), Dublin, Ireland, Paper presentation, July 2012.

Slater, P. (1977). The measurement of intrapersonal space by Grid technique. London: Wiley.

See Also

distanceHartmann() and distanceSlater().

Examples

## Not run: 

### basics  ###

distanceNormalized(bell2010)
n <- distanceNormalized(bell2010)
n

# printing options
print(n)
print(n, digits = 4)
# 'significant' distances only
print(n, p = c(.05, .95))

# access cells of distance matrix
n[1, 2]

### advanced  ###

# histogram of Slater distances and indifference region
n <- distanceNormalized(bell2010, distributions = TRUE)
l <- attr(n, "distributions")
hist(l$bc, breaks = 100)

## End(Not run)

Slater distances (standardized Euclidean distances).

Description

The euclidean distance is often used as a measure of similarity between elements (see distance(). A drawback of this measure is that it depends on the range of the rating scale and the number of constructs used, i. e. on the size of a grid.
An approach to standardize the euclidean distance to make it independent from size and range of ratings and was proposed by Slater (1977, pp. 94). The 'Slater distance' is the Euclidean distance divided by the expected distance. Slater distances bigger than 1 are greater than expected, lesser than 1 are smaller than expected. The minimum value is 0 and values bigger than 2 are rarely found. Slater distances have been be used to compare inter-element distances between different grids, where the grids do not need to have the same constructs or elements. Hartmann (1992) showed that Slater distance is not independent of grid size. Also the distribution of the Slater distances is asymmetric. Hence, the upper and lower limit to infer 'significance' of distance is not symmetric. The practical relevance of Hartmann's findings have been demonstrated by Schoeneich and Klapp (1998). To calculate Hartmann's version of the standardized distances see distanceHartmann()

Usage

distanceSlater(x, trim = 20, index = TRUE)

Arguments

Value

A matrix with Slater distances.

Calculation

The Slater distance is calculated as follows. For a derivation see Slater (1977, p.94).
Let matrix D contain the row centered ratings. Then

P = D^TD

and

S = tr(P)

The expected 'unit of expected distance' results as

U = (2S/(m-1))^{1/2}

where m denotes the number of elements of the grid. The standardized Slater distances is the matrix of Euclidean distances E divided by the expected distance U.

E/U

References

Hartmann, A. (1992). Element comparisons in repertory grid technique: Results and consequences of a Monte Carlo study. International Journal of Personal Construct Psychology, 5(1), 41-56.

Schoeneich, F., & Klapp, B. F. (1998). Standardization of interelement distances in repertory grid technique and its consequences for psychological interpretation of self-identity plots: An empirical study. Journal of Constructivist Psychology, 11(1), 49-58.

Slater, P. (1977). The measurement of intrapersonal space by Grid technique. Vol. II. London: Wiley.

See Also

distanceHartmann()

Examples

distanceSlater(bell2010)
distanceSlater(bell2010, trim = 40)

d <- distanceSlater(bell2010)
print(d)
print(d, digits = 4)

# using Norris and Makhlouf-Norris (problematic) cutoffs
print(d, cutoffs = c(.8, 1.2))

Detect if two rectangles overlap.

Description

The overlap is assessed in x AND y.

Usage

doRectanglesOverlap(a, b, margin = 0)

Arguments

Value

logical. TRUE if rectangles overlap.

Examples

## Not run: 
# overlap in x and y
a <- c(0, 0, 2, 2)
b <- c(1, 1, 4, 3)
plot(c(a, b), c(a, b), type = "n")
rect(a[1], a[2], a[3], a[4])
rect(b[1], b[2], b[3], b[4])
doRectanglesOverlap(a, b)

# b contained in a vertically
a <- c(5, 0, 20, 20)
b <- c(0, 5, 15, 15)
plot(c(a, b), c(a, b), type = "n")
rect(a[1], a[2], a[3], a[4])
rect(b[1], b[2], b[3], b[4])
doRectanglesOverlap(a, b)

# overlap only in y
a <- c(0, 0, 2, 2)
b <- c(2.1, 1, 4, 3)
plot(c(a, b), c(a, b), type = "n")
rect(a[1], a[2], a[3], a[4])
rect(b[1], b[2], b[3], b[4])
doRectanglesOverlap(a, b)

## End(Not run)

Join the constructs of a grid with the same reversed constructs.

Description

Join the constructs of a grid with the same reversed constructs.

Usage

doubleEntry(x)

Arguments

Value

repgrid object

Examples

## Not run: 

data(bell2010)
doubleEntry(bell2010)

## End(Not run)

Calculate the correlations between elements.

Description

Note that simple element correlations as a measure of similarity are flawed as they are not invariant to construct reflection (Mackay, 1992; Bell, 2010). A correlation index invariant to construct reflection is Cohen's rc measure (1969), which can be calculated using the argument rc=TRUE which is the default option.

Usage

elementCor(x, rc = TRUE, method = "pearson", trim = 20, index = TRUE)

Arguments

Value

matrix of element correlations

References

Bell, R. C. (2010). A note on aligning constructs. Personal Construct Theory & Practice, (7), 42-48.

Cohen, J. (1969). rc: A profile similarity coefficient invariant over variable reflection. Psychological Bulletin, 71(4), 281-284.

Mackay, N. (1992). Identification, Reflection, and Correlation: Problems In The Bases Of Repertory Grid Measures. International Journal of Personal Construct Psychology, 5(1), 57-75.

See Also

constructCor()

Examples

elementCor(mackay1992) # Cohen's rc
elementCor(mackay1992, rc = FALSE) # PM correlation
elementCor(mackay1992, rc = FALSE, method = "spearman") # Spearman correlation

# format output
elementCor(mackay1992, trim = 6)
elementCor(mackay1992, index = FALSE, trim = 6)

# save as object for further processing
r <- elementCor(mackay1992)
r

# change output of object
print(r, digits = 5)
print(r, col.index = FALSE)
print(r, upper = FALSE)

# accessing elements of the correlation matrix
r[1, 3]

Root mean square (RMS) of inter-element correlations.

Description

The RMS is also known as 'quadratic mean' of the inter-element correlations. The RMS serves as a simplification of the correlation table. It reflects the average relation of one element with all other elements. Note that as the correlations are squared during its calculation, the RMS is not affected by the sign of the correlation (cf. Fransella, Bell & Bannister, 2003, p. 86).

Usage

elementRmsCor(x, rc = TRUE, method = "pearson", trim = NA)

Arguments

Details

Note that simple element correlations as a measure of similarity are flawed as they are not invariant to construct reflection (Mackay, 1992; Bell, 2010). A correlation index invariant to construct reflection is Cohen's rc measure (1969), which can be calculated using the argument rc=TRUE which is the default option in this function.

Value

dataframe of the RMS of inter-element correlations.

References

Fransella, F., Bell, R. C., & Bannister, D. (2003). A Manual for Repertory Grid Technique (2. Ed.). Chichester: John Wiley & Sons.

See Also

constructRmsCor(), elementCor()

Examples

# data from grid manual by Fransella, Bell and Bannister
elementRmsCor(fbb2003)
elementRmsCor(fbb2003, trim = 10)

# modify output
r <- elementRmsCor(fbb2003)
print(r, digits = 5)

# access second row of calculation results
r[2, "RMS"]

Get or replace element names

Description

Allows to get and set element names. Replaces the older functions getElementNames, getElementNames2, and eNames which are deprecated.

Usage

elements(x)

elements(x, position) <- value

Arguments

Examples

# copy Boeker grid to x
x <- boeker

## get element names
e <- elements(x)
e

## replace element names
elements(x) <- rev(e) # reverse all element names
elements(x)[1] <- "Hannes" # replace name of first element

# replace names of elements 1 and 3
elements(x)[c(1, 3)] <- c("element 1", "element 3")

Format a matrix and add index column.

Description

Format a matrix and add index column.

Usage

formatMatrix(
  x,
  rnames = rownames(x),
  pre.index = c(T, F),
  cnames = seq_len(ncol(x)),
  indexcol = F,
  digits = 2,
  diag = F,
  mode = 1
)

Arguments

Get construct names

Description

Get construct names

Usage

getConstructNames(x)

Arguments

Deprecated functions

getConstructNames(), and cNames() have been deprecated. Instead use constructs().

Retrieve construct names in needed format.

Description

Retrieves the construct names from a repgrid.

Usage

getConstructNames2(
  x,
  mode = 1,
  trim = 20,
  index = F,
  sep = " - ",
  pre = "(",
  post = ") "
)

Arguments

Details

Different features like trimming, indexing and choices of separators allow to return the kind of format that is needed.

Value

Vector with construct names.

Examples

## Not run: 

getConstructNames2(bell2010)
getConstructNames2(bell2010, mode = 2)
getConstructNames2(bell2010, index = T)
getConstructNames2(bell2010, index = T, mode = 3)

## End(Not run)

Retrieve element names of repgrid object.

Description

Function for convenience, so new users do not have to deal with object slots as they will typically not have knowledge about R object structures (S3, S4).

Usage

getElementNames(x)

Arguments

Value

vector Vector containing the names of the elements.

Deprecated functions

getElementNames(), getElementNames2(), and eNames() have been deprecated. Instead use elements().

Retrieve element names in needed format.

Description

Retrieves the element names from a repgrid.

Usage

getElementNames2(x, trim = 20, index = F, pre = "(", post = ") ")

Arguments

Details

Different features like trimming, indexing and choices of separators allow to return the kind of format that is needed.

Value

Vector with (trimmed) element names.

Examples

## Not run: 

getElementNames2(bell2010)
getElementNames2(bell2010, mode = 2)
getElementNames2(bell2010, index = T)
getElementNames2(bell2010, index = T, trim = 30)

## End(Not run)

Get number of constructs

Description

Get number of constructs

Usage

getNoOfConstructs(x)

Arguments

Value

numeric

Examples

## Not run: 

getNoOfConstructs(bell2010)

## End(Not run)

Get number of elements

Description

Get number of elements

Usage

getNoOfElements(x)

Arguments

Value

numeric

Examples

## Not run: 

getNoOfElements(bell2010)

## End(Not run)

get rating layer

Description

get rating layer

Usage

getRatingLayer(x, layer = 1, names = TRUE, trim = 10)

Arguments

Value

a matrix

Examples

## Not run: 

getRatingLayer(bell2010)

## End(Not run)

Get minimum and maximum scale value used in grid.

Description

The values are returned either as a vector or a list.

Usage

getScale(x, output = 1)

Arguments

Value

Vector or list (depends on output containing minimum and maximum scale value.

Get midpoint of the grid rating scale

Description

Get midpoint of the grid rating scale

Usage

getScaleMidpoint(x)

Arguments

Value

Midpoint of scale.

Examples

## Not run: 

getScaleMidpoint(bell2010)

## End(Not run)

Export a grid to dataframe with long format

Description

Export a grid to dataframe with long format

Usage

grid_to_df_long(x)

Arguments

Export a grid to dataframe with wide format

Description

Export a grid to dataframe with wide format

Usage

grid_to_df_wide(x)

Arguments

Add repgrids into a gridlist

Description

Add repgrids into a gridlist

Test or create object of class gridlist

Usage

gridlist(...)

is.gridlist(x)

as.gridlist(x)

Arguments

Resample constructs

Description

The goal of resampling is to build variations of a single grid. Two variants are implemented: The first is the leave-n-out approach which builds all possible grids when dropping n constructs. The second is a bootstrap approach, randomly drawing n constructs from the grid.

Usage

grids_leave_n_out(x, n = 0)

grids_bootstrap(x, n = nrow(x), reps = 100, replace = TRUE)

Arguments

Value

List of grids.

Examples

## All results for PVAFF index when one construct is left out 
p <- indexPvaff(boeker)
l <- grids_leave_n_out(boeker, n = 1)
pp <- sapply(l, indexPvaff)  # apply indexPvaff function to all grids
range(pp)   # min and max PVAFF
hist(pp, xlab = "PVAFF values")    # visualize
abline(v = p, col = "blue", lty = 2)

Rotate the interactive 3D device to default views.

Description

Rotate the interactive 3D device to a default viewpoint or to a position defined by theta and phi in Euler angles. Three default viewpoints are implemented rendering a view so that two axes span a plane and the third axis is pointing out of the screen.

Usage

home(view = 1, theta = NULL, phi = NULL)

Arguments

See Also

Interactive 3D biplots: biplot3d(), biplotSlater3d(), biplotEsa3d().

Examples

## Not run: 

biplot3d(boeker)
home(2)
home(3)
home(1)
home(theta = 45, phi = 45)

## End(Not run)

Convert a dataframe into a repgrid object.

Description

There are three different dataframe formats from which a repgrid object can be created: Columns are a) element_columns, b) construct_columns, c) long. Three corresponding sample dataframes, (df_element_columns, df_construct_columns, and df_long) are included in the package (see examples). See Detail section below for more info.

Usage

importDataframe(
  x,
  format = "element_columns",
  rmin = NULL,
  rmax = NULL,
  pole_sep = ":"
)

Arguments

Value

A 'repgrid“ object.

Format element_columns

In this format, each element has a separate column, and each row contains the ratings for one construct. It is a common way to represent grid data and looks like this.

The columns names contains the minimum of the rating scale (1), the names of the elements (element_1 to element_4), the maximum of the rating scale (5), and optionally the column preferred, indicating the preferred pole. Each row contains the constructs entries (left pole, ratings, right pole, preferred pole). The preferred pole must be one of left, right, none, NA (see preferredPoles()). See sample dataframe df_element_columns.

Format construct_columns

In this format, each construct has a separate column, and each row contains represents element. This format often results when summarising data (see examples). It looks like this:

The first column is named elements followed by the constructs. The construct poles are separated by a colon by default (see arg pole_sep). The rows below contain the elements' entries (element name, ratings). The min and max of the rating scale should be passed explicitly via the args rmin and rmax. See sample dataframe df_construct_columns.

Format long

The long format gets its name from the fact, that it has less columns, but many more rows. It is a common format in data analytics. Here, each row contains a different element-construct combination and the corresponding rating value. The format looks like this:

The columns element, left_pole, right_pole, and rating are mandatory, the columns preferred_pole, rmin, and rmax are optional. rmin and rmax contain the min and max of the rating scale. Alternatively, you may pass rmin and rmax as arguments in the function call. See sample dataframe df_long.

See Also

Import data importExcel(), importGridcor(), importGridstat(), importGridsuite(), importScivesco(), importTxt()

Examples

# dataframe with elements as columns (preferred way)
importDataframe(df_element_columns)

# dataframe with constructs as columns
importDataframe(df_construct_columns, format = "construct_columns", rmin = 1, rmax = 5)

# dataframe with long format
importDataframe(df_long, format = "long", rmin = 1, rmax = 5)

Import grid data from an Excel file.

Description

You can define a grid in a Microsoft Excel file (suffix .xlsx). The file must have one of two formats (wide or long, see format sections below).

Usage

importExcel(file, sheet = 1, format = "wide", rmin = NULL, rmax = NULL)

Arguments

Value

A repgrid object (one input file) or a named list with repgrid objects (several input files). List names are filename + sheet.

Format wide

In the wide format, each element has a separate column, and each row contains the ratzings for one construct. It is a common way to represent grid data and looks like this:

The header row contains the minimum of the rating scale (1), the names of the elements (element_1 to element_4), the maximum of the rating scale (5), and optionally the column preferred, indicating the preferred pole. Each row contains the constructs entries (left pole, ratings, right pole, preferred pole). The preferred pole must be one of left, right, none, NA (see preferredPoles()).

Format long

The long format has this name because it has few columns and many rows. It is a common format in data analytics. Here, each row contains a different element-construct combination and the corresponding rating value. The format looks like this:

The columns element, left_pole, right_pole, and rating are mandatory, the columns preferred_pole, rmin, and rmax are optional. rmin and rmax contain the min and max of the rating scale. Alternatively, you may pass rmin and rmax as arguments in the function call.

See Also

Import data importDataframe(), importGridcor(), importGridstat(), importGridsuite(), importScivesco(), importTxt()

Examples

file <- system.file("extdata", "grid_01.xlsx", package = "OpenRepGrid")
rg_1 <- importExcel(file, sheet = "wide")
rg_2 <- importExcel(file, sheet = "long", format = "long")

# Open sample file to inspect it (requires Excel to be installed).
## Not run: 
browseURL(file) # may not work on all systems 
## End(Not run)

# Import more than one Excel file
files <- system.file("extdata", c("grid_01.xlsx", "grid_02.xlsx"), package = "OpenRepGrid")
rgs <- importExcel(files) # returns a list of grids

# Impoort from several sheets at once (all must have same format)
file <- system.file("extdata", "grid_03.xlsx", package = "OpenRepGrid")
rgs <- importExcel(file, sheet = 1:3) # by index
rgs <- importExcel(file, sheet = c("grid 1", "grid 2", "grid 3")) # or by name

Import GRIDCOR data files.

Description

Reads the file format that is used by the grid program GRIDCOR (Feixas & Cornejo, 2002).

Usage

importGridcor(file, dir = NULL)

Arguments

Value

a single repgrid object in case one file and a list of repgrid objects in case multiple files are imported.

Note

Note that the GRIDCOR data sets the minimum ratings scale range to 1. The maximum value can differ and is defined in the data file.

Also note that both Gridcor and Gridstat data files do have the same suffix .dat. Make sure not to mix them up.

References

Feixas, G., & Cornejo, J. M. (2002). GRIDCOR: Correspondence Analysis for Grid Data (version 4.0). Barcelona: Centro de Terapia Cognitiva. Retrieved from https://repertorygrid.net/en/.

See Also

Import data importDataframe(), importExcel(), importGridstat(), importGridsuite(), importScivesco(), importTxt()

Examples

## Not run: 

# supposing that the data file gridcor.dat is in the current directory
file <- "gridcor.dat"
rg <- importGridcor(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
rg <- importGridcor(file, dir)

# using a full path
rg <- importGridcor("/Users/markheckmann/data/gridcor.dat")

## End(Not run)

Internal parser for GRIDCOR data files.

Description

Parse the file format that is used by the grid program GRIDCOR (Feixas & Cornejo).

Usage

importGridcorInternal(file, dir = NULL)

Arguments

Note

     Note that the GRIDCOR data sets the minimum ratings scale range to 1.
           The maximum value can differ and is defined in the data file.

References

https://repertorygrid.net/en/

Examples

## Not run: 

# supposing that the data file gridcor.dat is in the current directory
file <- "gridcor.dat"
imp <- importGridcorInternal(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
imp <- importGridcorInternal(file, dir)

# using a full path
imp <- importGridcorInternal("/Users/markheckmann/data/gridcor.dat")

## End(Not run)

Import Gridstat data files.

Description

Reads the file format that is used by the latest version of the grid program gridstat (Bell, 1998).

Usage

importGridstat(file, dir = NULL, min = NULL, max = NULL)

Arguments

Value

A single repgrid object in case one file and a list of repgrid objects in case multiple files are imported.

Note

Note that the gridstat data format does not contain explicit information about the range of the rating scale used (minimum and maximum). By default the range is inferred by scanning the ratings and picking the minimal and maximal values as rating range. You can set the minimal and maximal value by hand using the min and max arguments or by using the setScale() function. Note that if the rating range is not set, it may cause several functions to not work properly. A warning will be issued if the range is not set explicitly when using the importing function.

The function only reads data from the latest GridStat version. The latest version allows the separation of the left and right pole by using on of the following symbols ⁠/:-⁠ (hyphen, colon and dash). Older versions may not separate the left and right pole. This will cause all labels to be assigned to the left pole only when importing. You may fix this by simply entering one of the construct separator symbols into the GridStat file between each left and right construct pole.

The third line of a GridStat file may contain a no labels statement (i.e. a line containing any string of 'NOLA', 'NO L', 'NoLa', 'No L', 'Nola', 'No l', 'nola' or 'no l'). In this case only ratings are supplied, hence, default names are assigned to elements and constructs.

References

Bell, R. C. (1998) GRIDSTAT: A program for analyzing the data of a repertory grid. Melbourne: Author.

See Also

Import data importDataframe(), importExcel(), importGridcor(), importGridsuite(), importScivesco(), importTxt()

Examples

## Not run: 

# supposing that the data file gridstat.dat is in the current working directory
file <- "gridstat.dat"
rg <- importGridstat(file)

# specifying a directory (example)
dir <- "/Users/markheckmann/data"
rg <- importGridstat(file, dir)

# using a full path (example)
rg <- importGridstat("/Users/markheckmann/data/gridstat.dat")

# setting rating scale range
rg <- importGridstat(file, dir, min = 1, max = 6)

## End(Not run)

Parser for Gridstat data files.

Description

Parse the file format that is used by the latest version of grid program gridstat (Bell, 1998).

Usage

importGridstatInternal(file, dir = NULL, min = NULL, max = NULL)

Arguments

Value

a list with imported parameters

Note

     Note that the gridstat data format does not contain explicit
           information about the range of the rating scale (minimum and
           maximum). By default the range is inferred by scanning
           the ratings and picking the minimal and maximal values as rating
           range. You can set the minimal and maximal value by hand using the `min` and
           `max` arguments or by using the `setScale()` function.
           Note that if the rating range is not set, it may cause several
           functions to not work properly. A warning will be issued if the range is
           not set explicitly when using the importing function.

           The function only reads data from the latest GridStat version.
           The latest version allows the separation of the left and right pole
           by using on of the following symbols `/:-` (hyphen, colon and dash). Older versions may not
           separate                              the left and right pole. This will cause all labels to be assigned to
           the left pole only when importing. You may fix this by simply entering
           one of the construct separator symbols into the GridStat file between each
           left and right construct pole.

           The third line of a GridStat file may contain a no labels statement (i.e. a
           line containing any string of 'NOLA', 'NO L', 'NoLa', 'No L', 'Nola', 'No l',
           'nola' or 'no l'). In this case only ratings are supplied, hence, default
           names are assigned to elements and constructs.

     Email from Richard: The gridstat file has a fixed format with a title line, number
     of constructs and elements on second line. The third line can say No labels
     (actually it looks at the first 4 characters which can be any of 'NOLA','NO L',
     'NoLa','No L','Nola','No l','nola','no l') in which case it skips to the data and
     creates dummy labels for elements and constructs, otherwise it reads the construct
     labels then the element labels, then the data. Construct labels were originally
     stored as one, hence it didn't matter what the separator between left and right
     pole labels was, but in the latest version where constructs can be reversed, it
     looks for a fixed separator - one of slash(/), dash(-), or colon(:). Some of my
     old data files might not conform.

References

Bell, R. C. (1998) GRIDSTAT: A program for analyzing the data of a repertory grid. Melbourne: Author.

Examples

## Not run: 

# supposing that the data file gridstat.dat is in the current working directory
file <- "gridstat.dat"
imp <- importGridstatInternal(file)

# specifying a directory (example)
dir <- "/Users/markheckmann/data"
imp <- importGridstatInternal(file, dir)

# using a full path (example)
imp <- importGridstatInternal("/Users/markheckmann/data/gridstat.dat")

# setting rating scale range
imp <- importGridstatInternal(file, dir, min = 1, max = 6)

## End(Not run)

Import Gridsuite data files.

Description

Import Gridsuite data files.

Usage

importGridsuite(file, dir = NULL)

Arguments

Value

A single repgrid object in case one file and a list of repgrid objects in case multiple files are imported.

Note

The developers of Gridsuite have proposed to use an XML scheme as a standard exchange format for repertory grid data (Walter, Bacher & Fromm, 2004).

TODO: The element and construct IDs are not used yet. Thus, if the output should be in different order the current mechanism will cause false assignments.

References

http://www.gridsuite.de/

Walter, O. B., Bacher, A., & Fromm, M. (2004). A proposal for a common data exchange format for repertory grid data.Journal of Constructivist Psychology, 17(3), 247. doi:10.1080/10720530490447167

See Also

Import data importDataframe(), importExcel(), importGridcor(), importGridstat(), importScivesco(), importTxt()

Examples

## Not run: 

# supposing that the data file gridsuite.xml is in the current directory
file <- "gridsuite.xml"
rg <- importGridsuite(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
rg <- importGridsuite(file, dir)

# using a full path
rg <- importGridsuite("/Users/markheckmann/data/gridsuite.xml")

## End(Not run)

Internal parser for Gridsuite data files

Description

Internal parser for Gridsuite data files

Usage

importGridsuiteInternal(file, dir = NULL)

Arguments

Note

     The developers of Gridsuite have proposed to use an XML scheme as
           a standard exchange format for repertory grid data (Walter,
           Bacher & Fromm, 2004). This approach is also embraced by the
           `OpenRepGrid` package.

     TODO: The element and construct IDs are not used yet. Thus,
           if the output should be in different order the current mechanism
           will cause false assignments.

References

http://www.gridsuite.de/

           Walter, O. B., Bacher, A., & Fromm, M. (2004). A proposal
           for a common data exchange format for repertory grid data.
           *Journal of Constructivist Psychology, 17*(3), 247.
           doi:10.1080/10720530490447167

Examples

## Not run: 

# supposing that the data file gridsuite.xml is in the current directory
file <- "gridsuite.xml"
imp <- importGridsuite(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
imp <- importGridsuite(file, dir)

# using a full path
imp <- importGridsuite("/Users/markheckmann/data/gridsuite.xml")

## End(Not run)

Import sci:vesco data files.

Description

Import sci:vesco data files.

Usage

importScivesco(file, dir = NULL)

Arguments

Value

A single repgrid object in case one file and a list of repgrid objects in case multiple files are imported.

Note

Sci:Vesco offers the options to rate the construct poles separately or using a bipolar scale. The separated rating is done using the "tetralemma" field. The field is a bivariate plane on which each of the four (tetra) corners has a different meaning in terms of rating. Using this approach also allows ratings like: "both poles apply", "none of the poles apply" and all intermediate ratings can be chosen. This relaxes the bipolarity assumption often assumed in grid theory and allows for deviation from a strict bipolar rating if the constructs are not applied in a bipolar way. Using the tetralemma field for rating requires to analyze each construct separately though. This means we get a double entry grid where the emergent and contrast pole ratings might not simply be a reflection of on another. The tetralemma field is not yet supported and importing will fail. Currently only bipolar ratings are supported.

If a tetralemma field has been used for rating, OpenRepGrid will offer the option to transform the scores into "normal" grid ratings (i.e. restricted to bipolarity) by projecting the ratings from the bivariate tetralemma field onto the diagonal of the tetralemma field and thus forcing a bipolar rating type. This option is not recommended due to the fact that the conversion is susceptible to error when both ratings are near to zero.

TODO: For developers: The element IDs are not used yet. This might cause wrong assignments.

References

Menzel, F., Rosenberger, M., Buve, J. (2007). Emotionale, intuitive und rationale Konstrukte verstehen. Personalfuehrung, 4(7), 91-99.

See Also

Import data importDataframe(), importExcel(), importGridcor(), importGridstat(), importGridsuite(), importTxt()

Examples

## Not run: 

# supposing that the data file scivesco.scires is in the current directory
file <- "scivesco.scires"
rg <- importScivesco(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
rg <- importScivesco(file, dir)

# using a full path
rg <- importScivesco("/Users/markheckmann/data/scivesco.scires")

## End(Not run)

Internal parser for sci:vesco files (suffix scires).

Description

Internal parser for sci:vesco files (suffix scires).

Usage

importScivescoInternal(file, dir = NULL)

Arguments

Value

a list with extracted parameters.

Note

     Sci:Vesco offers the options to rate the construct poles separately or using
           a bipolar scale. The separated rating is done using the "tetralemma" field.
           The field is a bivariate plane on which each of the four (tetra) corners
           has a different meaning in terms of rating. Using this approach also allows ratings
           like: "both poles apply", "none of the poles apply" and all intermediate ratings
           can be chosen. This relaxes the bipolarity assumption often assumed in grid theory and
           allows for deviation from a strict bipolar rating if the constructs are not applied
           in a bipolar way. Using the tetralemma field for rating requires to analyze
           each construct separately though. This means we get a double entry grid where the
           emergent and contrast pole ratings might not simply be a reflection of on another.
           If a tetralemma field has been used for rating, `OpenRepGrid` offers the option
           to transform the scores into "normal" grid ratings (i.e. restricted to bipolarity)
           by projecting the ratings from the bivariate tetralemma field onto the diagonal
           of the tetralemma field and thus forcing a bipolar rating type. This option is
           not recommended due to the fact that the conversion is susceptible to error
           when both ratings are near to zero.

     TODO: The element IDs are not used yet. This might cause wrong assignments.

Examples

## Not run: 

# supposing that the data file scivesco.scires is in the current directory
file <- "scivesco.scires"
imp <- importScivescoInternal(file)

# specifying a directory (arbitrary example directory)
dir <- "/Users/markheckmann/data"
imp <- importScivescoInternal(file, dir)

# using a full path
imp <- importScivescoInternal("/Users/markheckmann/data/scivesco.scires")

## End(Not run)

Import grid data from a text file.

Description

You can define a grid using a standard text editor and saving it as a .txt file. The Details section describes the required format of the .txt file. However, you may also consider using the Excel format instead, as it has a more intuitive format (see importExcel()).

Usage

importTxt(file, dir = NULL, min = NULL, max = NULL)

Arguments

Details

The .txt file has to be in a fixed format. There are three mandatory blocks each starting and ending with a predefined tag in uppercase letters. The first block starts with ELEMENTS and ends with ⁠END ELEMENTS⁠. It contains one element per line. The other mandatory blocks are CONSTRUCTS and RATINGS (see below). In the block containing the constructs the left and right pole are separated by a colon (:). To define missing values use NA. The block PREFERRED is optional. Each line indicated the preferred construct pole. Allowed values are left, right, none (no pole preferred), and NA (unknown). The block RANGE is optional but recommended. It gives the rating scale range defined by two numbers. The order of the blocks is arbitrary. All text oustide the blocks is discarded and can be used for comments.

The content of a sample .txt file is shown below. The package also contains a sample file (see Examples).

---------------- sample .txt file -------------------

    Note: anything outside the tag pairs is discarded

ELEMENTS
element 1
element 2
element 3
END ELEMENTS

CONSTRUCTS
left pole 1 : right pole 1
left pole 2 : right pole 2
left pole 3 : right pole 3
left pole 4 : right pole 4
END CONSTRUCTS

PREFERRED
left
left
right
none
END PREFERRED

RATINGS
1 3 2
4 1 1
1 4 4
3 1 1
END RATINGS

RANGE
1 4
END RANGE

------------------ end of file ------------------

Note that the maximum and minimum value has to be defined using the min and max arguments if no RANGE block is contained in the data file. Otherwise the scaling range is inferred from the available data and a warning is issued as the range may be erroneous. This may effect other functions that depend on knowing the correct range and it is thus strongly recommended to set the scale range correctly.

Value

   A single `repgrid` object in case one file and
           a list of `repgrid` objects in case multiple files are imported.

See Also

Import data importDataframe(), importExcel(), importGridcor(), importGridstat(), importGridsuite(), importScivesco()

Examples

# Import a .txt file delivered along with the package
file <- system.file("extdata", "grid_01.txt", package = "OpenRepGrid")
rg <- importTxt(file)

# To see the structure of the file, try opening it as follows.
# (may not work on all systems)
## Not run: 
file.show(file)

## End(Not run)

# Import more than one .txt file
files <- system.file("extdata", c("grid_01.txt", "grid_02.txt"), package = "OpenRepGrid")
rgs <- importTxt(files)

Calculate 'bias' of grid as defined by Slater (1977).

Description

"Bias records a tendency for responses to accumulate at one end of the grading scale" (Slater, 1977, p.88).

Usage

indexBias(x, min = NULL, max = NULL, digits = 2)

Arguments

Value

Numeric.

Note

STATUS: Working and checked against example in Slater, 1977, p. 87.

References

Slater, P. (1977). The measurement of intrapersonal space by Grid technique. London: Wiley.

See Also

indexVariability()

Examples

indexBias(boeker)

Bieri's index of cognitive complexity

Description

The index builds on the number of rating matches between pairs of constructs. It is the relation between the total number of matches and the possible number of matches.

Usage

indexBieri(x, deviation = 0)

Arguments

Details

CAVEAT: The Bieri index will change when constructs are reversed.

Value

List of class indexBieri:

• grid: The grid used to calculate the index

• deviation The deviation parameter.

• matches_max Maximum possible number of matches across constructs.

• matches Total number of matches across constructs.

• constructs: Matrix with no. of matches for constructs.

• bieri: Bieri index (= matches / matches_max)

Examples

m <- indexBieri(boeker)

# several output options
print(m)
print(m, output = "IC")  # construct matches

# extract the matrix of matches
m$constructs

# CAVEAT: Bieri's index changes when constructs are reversed
nr <- nrow(boeker)
l <- replicate(1000, swapPoles(boeker, sample(nr, sample(nr, 1))))
bieri <- sapply(l, function(x) indexBieri(x)$bieri)
hist(bieri, breaks = 50)
abline(v = mean(bieri), col = "red", lty = 2)

Conflict measure for grids (Slade & Sheehan, 1979) based on correlations.

Description

Conflict measure as proposed by Slade and Sheehan (1979)

Usage

indexConflict1(x)

Arguments

Details

The first approach to mathematically derive a conflict measure based on grid data was presented by Slade and Sheehan (1979). Their operationalization is based on an approach by Lauterbach (1975) who applied the balance theory (Heider, 1958) for a quantitative assessment of psychological conflict. It is based on a count of balanced and imbalanced triads of construct correlations. A triad is imbalanced if one or all three of the correlations are negative, i. e. leading to contrary implications. This approach was shown by Winter (1982) to be flawed. An improved version was proposed by Bassler et al. (1992) and has been implemented in the function indexConflict2.

The table below shows when a triad made up of the constructs A, B, and C is balanced and imbalanced:

Value

A list with the following elements:

• total: Total number of triads

• imbalanced: Number of imbalanced triads

• prop.balanced: Proportion of balanced triads

• prop.imbalanced: Proportion of imbalanced triads

References

Bassler, M., Krauthauser, H., & Hoffmann, S. O. (1992). A new approach to the identification of cognitive conflicts in the repertory grid: An illustrative case study. Journal of Constructivist Psychology, 5(1), 95-111.

Heider, F. (1958). The Psychology of Interpersonal Relation. John Wiley & Sons.

Lauterbach, W. (1975). Assessing psychological conflict. The British Journal of Social and Clinical Psychology, 14(1), 43-47.

Slade, P. D., & Sheehan, M. J. (1979). The measurement of 'conflict' in repertory grids. British Journal of Psychology, 70(4), 519-524.

Winter, D. A. (1982). Construct relationships, psychological disorder and therapeutic change. The British Journal of Medical Psychology, 55 (Pt 3), 257-269.

See Also

indexConflict2() for an improved version of this measure; see indexConflict3() for a measure based on distances.

Examples

indexConflict1(feixas2004)
indexConflict1(boeker)

Conflict measure for grids (Bassler et al., 1992) based on correlations.

Description

The function calculates the conflict measure as devised by Bassler et al. (1992). It is an improved version of the ideas by Slade and Sheehan (1979) that have been implemented in the function indexConflict1(). The new approach also takes into account the magnitude of the correlations in a trait to assess whether it is balanced or imbalanced. As a result, small correlations that are psychologically meaningless are considered accordingly. Also, correlations with a small magnitude, i. e. near zero, which may be positive or negative due to chance alone will no longer distort the measure (Bassler et al., 1992).

Usage

indexConflict2(x, crit = 0.03)

Arguments

Details

Description of the balance / imbalance assessment:

1. Order correlations of the triad by absolute magnitude, so that r_{max} > r_{mdn} > r_{min}, r_{max} > r_{mdn} > r_{min}.

2. Apply Fisher's Z-transformation and division by 3 to yield values between 1 and -1 ( Z_{max} > Z_{mdn} > Z_{min}, Z_{max} > Z_{mdn} > Z_{min}).

3. Check whether the triad is balanced by assessing if the following relation holds:

   • If Z_{max} Z_{mdn} > 0, Z_{max} x Z_{mdn} > 0, the triad is balanced if Z_{max} Z_{mdn} - Z_{min} <= crit, Z_{max} x Z_{mdn} - Z_{min} <= crit.

   • If Z_{max} Z_{mdn} < 0, Z_{max} x Z_{mdn} < 0, the triad is balanced if Z_{min} - Z_{max} Z_{mdn} <= crit, Z_{min} - Z_{max} x Z_{mdn} <= crit.

Personal remarks (MH)

I am a bit suspicious about step 2 from above. To devide by 3 appears pretty arbitrary. The r for a z-values of 3 is 0.9950548 and not 1. The r for 4 is 0.9993293. Hence, why not a value of 4, 5, or 6? Denoting the value to devide by with a, the relation for the first case translates into a Z_{max} Z_{mdn} <= \frac{crit}{a} + Z_{min}, a x Z_{max} x Z_{mdn} =< crit/a + Z_{min}. This shows that a bigger value of a will make it more improbable that the relation will hold.

References

Bassler, M., Krauthauser, H., & Hoffmann, S. O. (1992). A new approach to the identification of cognitive conflicts in the repertory grid: An illustrative case study. Journal of Constructivist Psychology, 5(1), 95-111.

Slade, P. D., & Sheehan, M. J. (1979). The measurement of 'conflict' in repertory grids. British Journal of Psychology, 70(4), 519-524.

See Also

See indexConflict1() for the older version of this measure; see indexConflict3() for a measure based on distances instead of correlations.

Examples

indexConflict2(bell2010)

x <- indexConflict2(bell2010)
print(x)

# show conflictive triads
print(x, output = 2)

# accessing the calculations for further use
x$total
x$imbalanced
x$prop.balanced
x$prop.imbalanced
x$triads.imbalanced

Conflict or inconsistency measure for grids (Bell, 2004) based on distances.

Description

Measure of conflict or inconsistency as proposed by Bell (2004). The identification of conflict is based on distances rather than correlations as in other measures of conflict indexConflict1() and indexConflict2(). It assesses if the distances between all components of a triad, made up of one element and two constructs, satisfies the "triangle inequality" (cf. Bell, 2004). If not, a triad is regarded as conflictive. An advantage of the measure is that it can be interpreted not only as a global measure for a grid but also on an element, construct, and element by construct level making it valuable for detailed feedback. Also, differences in conflict can be submitted to statistical testing procedures.

Usage

indexConflict3(
  x,
  p = 2,
  e.out = NA,
  e.threshold = NA,
  c.out = NA,
  c.threshold = NA,
  trim = 20
)

Arguments

Details

Status: working; output for euclidean and manhattan distance checked against Gridstat output.
TODO: standardization and z-test for discrepancies; Index of Conflict Variation.

Value

A list (invisibly) containing:

• potential: number of potential conflicts

• actual: count of actual conflicts

• overall: percentage of conflictive relations

• e.count: number of involvements of each element in conflictive relations

• e.perc: percentage of involvement of each element in total of conflictive relations

• c.count: number of involvements of each construct in conflictive relation

• c.perc: percentage of involvement of each construct in total of conflictive relations

• e.stats: detailed statistics for prompted elements

• c.stats: detailed statistics for prompted constructs

• e.threshold: threshold percentage. Used by print method

• c.threshold: threshold percentage. Used by print method

• enames: trimmed element names. Used by print method

• cnames: trimmed construct names. Used by print method

output

For further control over the output see print.indexConflict3().

References

Bell, R. C. (2004). A new approach to measuring inconsistency or conflict in grids. Personal Construct Theory & Practice, (1), 53-59.

See Also

See indexConflict1() and indexConflict2() for conflict measures based on triads of correlations.

Examples

# calculate conflicts
indexConflict3(bell2010)

# show additional stats for elements 1 to 3
indexConflict3(bell2010, e.out = 1:3)

# show additional stats for constructs 1 and 5
indexConflict3(bell2010, c.out = c(1, 5))

# finetune output
## change number of digits
x <- indexConflict3(bell2010)
print(x, digits = 4)

## omit discrepancy matrices for constructs
x <- indexConflict3(bell2010, c.out = 5:6)
print(x, discrepancies = FALSE)

Dispersion of dependency index (DDI)

Description

Measures the degree of dispersion of dependency in a situation-resource grid (dependency grid), i.e. the degree to which a person dispersed critical situations over resource persons (Walker et al., 1988, p. 66). The index is a renamed adoption of the ⁠diversity index⁠ from the field of ecology where it is used to measure the diversity of species in a sample. Both are computationally identical. The index is applicable to dependency grids (e.g., situation-resource) only, i.e., all grid ratings must be 0 or 1.

Usage

indexDDI(x, ds)

Arguments

Details

Caveat: The DDI depends on the chosen sample size ds. Also, its measurement range is not normalized between 0 and 1, allowing only comparison between similarly sized grids (see Bell, 2001).

Theoretical Background: Dispersion of Dependency: Kelly (1969) proposed that it is problematic to view people as either independent or dependent because everyone is, to greater or lesser degrees, dependent upon others in life. What Kelly felt was important was how well people disperse their dependencies across different people. Whereas young children tend to have their dependencies concentrated on a small number of people (typically parents), adults are more likely to spread their dependencies across a variety of others. Dispersing one's dependencies is generally considered more psychologically adjusted for adults (Walker et al., 1988).

References

Bell, R. C. (2001). Some new Measures of the Dispersion of Dependency in a Situation-Resource Grid. Journal of Constructivist Psychology, 14(3), 227-234, doi:10.1080/713840106.

Kelly, G. A. (1962). In whom confide: On whom depend for what. In Maher, B. (Ed.). Clinical psychology and personality: The selected papers of George Kelly, 189-206. New York Krieger.

Walker, B. M., Ramsey, F. L., & Bell, R. (1988). Dispersed and Undispersed Dependency. International Journal of Personal Construct Psychology, 1(1), 63-80, doi:10.1080/10720538808412765.

See Also

indexUncertainty

Examples

# sample grid from Walker et al. (1988), p. 67
file <- system.file("extdata", "dep_grid_walker_1988_2.xlsx" , package = "OpenRepGrid")
x <- importExcel(file)

indexDDI(x, ds = 2:5) 

# using named vector
ds = c("2"=2, "3"=3, "4"=4, "5"=5)  
indexDDI(x, ds)

Implicative Dilemmas

Description

Implicative dilemmas are closely related to the notion of conflict. An implicative dilemma arises when a desired change on one construct is associated with an undesired implication on another construct. E. g. a timid subject may want to become more socially skilled but associates being socially skilled with different negative characteristics (selfish, insensitive etc.). Hence, he may anticipate that becoming less timid will also make him more selfish (cf. Winter, 1982). As a consequence, the subject will resist to the change if the negative presumed implications will threaten the patients identity and the predictive power of his construct system. From this stance the resistance to change is a logical consequence coherent with the subjects construct system (Feixas, Saul, & Sanchez, 2000). The investigation of the role of cognitive dilemma in different disorders in the context of PCP is a current field of research (e.g. Feixas & Saul, 2004, Dorough et al. 2007).

Usage

indexDilemma(
  x,
  self = 1,
  ideal = ncol(x),
  diff.mode = 1,
  diff.congruent = NA,
  diff.discrepant = NA,
  diff.poles = 1,
  r.min = 0.35,
  exclude = FALSE,
  digits = 2,
  show = FALSE,
  output = 1,
  index = TRUE,
  trim = 20
)

Arguments

Details

The detection of implicative dilemmas happens in two steps. First the constructs are classified as being 'congruent' or 'discrepant'. Secondly, the correlation between a congruent and discrepant construct pair is assessed if it is big enough to indicate an implication.

Classifying the construct

To detect implicit dilemmas the construct pairs are first identified as 'congruent' or 'discrepant'. The assessment is based on the rating differences between the elements 'self' and 'ideal self'. A construct is 'congruent' if the construction of the 'self' and the preferred state (i.e. ideal self) are the same or similar. A construct is discrepant if the construction of the 'self' and the 'ideal' is dissimilar.

There are two popular accepted methods to identify congruent and discrepant constructs:

1. "Scale Midpoint criterion" (cf. Grice 2008)

2. "Minimal and maximal score difference" (cf. Feixas & Saul, 2004)

"Scale Midpoint criterion" (cf. Grice 2008)

As reported in the Idiogrid (v. 2.4) manual: "... The Scale Midpoint uses the scales as the 'dividing line' for discrepancies; for example, if the actual element is rated above the midpoint, then the discrepancy exists (and vice versa). If the two selves are the same as the actual side of the scale, then a discrepancy does not exist". As an example:

Assuming a scoring range of 1-7, the midpoint score will be 4, we then look at self and ideal-self scoring on any given construct and we proceed as follow:

• If the scoring of Self AND Ideal Self are both < 4: construct is "Congruent"

• If the scoring of Self AND Ideal Self are both > 4: construct is "Congruent"

• If the scoring of Self is < 4 AND Ideal Self is > 4 (OR vice versa): construct is "discrepant"

• If scoring Self OR Ideal Self = 4 then the construct is NOT Discrepant and it is "Undifferentiated"

Minimal and maximal score difference criterion (cf. Feixas & Saul, 2004)

This other method is more conservative and it is designed to minimize Type I errors by a) setting a default minimum correlation between constructs of r=.34; b) discarding cases where the ideal Self and self are neither congruent or discrepant; c) discarding cases where ideal self is "not oriented", i.e. scored at the midpoint.

E.g. suppose the element 'self' is rated 2 and 'ideal self' 5 on a scale from 1 to 6. The ratings differences are 5-2 = 3. If this difference is smaller than e.g. 1 the construct is 'congruent', if it is bigger than 3 it is 'discrepant'.

The values used to classify the constructs 'congruent' or 'discrepant' can be determined in several ways (cf. Bell, 2009):

1. They are set 'a priori'.

2. They are implicitly derived by taking into account the rating differences to the other constructs. (Not yet implemented)

The value mode is determined via the argument diff.mode.

If no 'a priori' criteria to determine whether the construct is congruent or discrepant is supplied as an argument, the values are chosen according to the range of the rating scale used. For the following scales the defaults are chosen as:

Defining the correlations

As the implications between constructs cannot be derived from a rating grid directly, the correlation between two constructs is used as an indicator for implication. A large correlation means that one construct pole implies the other. A small correlation indicates a lack of implication. The minimum criterion for a correlation to indicate implication is set to .35 (cf. Feixas & Saul, 2004). The user may also choose another value. To get a an impression of the distribution of correlations in the grid, a visualization can be prompted via the argument show. When calculating the correlation used to assess if an implication is given or not, the elements under consideration (i. e. self and ideal self) can be included (default) or excluded. The options will cause different correlations (see argument exclude).

Example of an implicative dilemma

A depressive person considers herself as 'timid' and wished to change to the opposite pole she defines as 'extraverted'. This construct is called discrepant as the construction of the 'self' and the desired state (e.g. described by the 'ideal self') on this construct differ. The person also considers herself as 'sensitive' (preferred pole) for which the opposite pole is 'selfish'. This construct is congruent, as the person construes herself as she would like to be. If the person now changed on the discrepant construct from the undesired to the desired pole, i.e. from timid to extraverted, the question can be asked what consequences such a change has. If the person construes being timid and being sensitive as related and that someone who is extraverted will not be timid, a change on the first construct will imply a change on the congruent construct as well. Hence, the positive shift from timid to extraverted is presumed to have a undesired effect in moving from sensitive towards selfish. This relation is called an implicative dilemma. As the implications of change on a construct cannot be derived from a rating grid directly, the correlation between two constructs is used as an indicator of implication.

Value

List object of class indexDilemma, containing the result from the calculations.

Author(s)

Mark Heckmann, Alejandro García, Diego Vitali

References

Bell, R. C. (2009). Gridstat version 5 - A Program for Analyzing the Data of A Repertory Grid (manual). University of Melbourne, Australia: Department of Psychology.

Dorough, S., Grice, J. W., & Parker, J. (2007). Implicative dilemmas and psychological well-being. Personal Construct Theory & Practice, (4), 83-101.

Feixas, G., & Saul, L. A. (2004). The Multi-Center Dilemma Project: an investigation on the role of cognitive conflicts in health. The Spanish Journal of Psychology, 7(1), 69-78.

Feixas, G., Saul, L. A., & Sanchez, V. (2000). Detection and analysis of implicative dilemmas: implications for the therapeutic process. In J. W. Scheer (Ed.), The Person in Society: Challenges to a Constructivist Theory. Giessen: Psychosozial-Verlag.

Winter, D. A. (1982). Construct relationships, psychological disorder and therapeutic change. British Journal of Medical Psychology, 55 (Pt 3), 257-269.

Grice, J. W. (2008). Idiogrid: Idiographic Analysis with Repertory Grids (Version 2.4). Oklahoma: Oklahoma State University.

See Also

print.indexDilemma(), plot.indexDilemma()

Examples

id <- indexDilemma(boeker, self = 1, ideal = 2)
id

# adjust minimal correlation
indexDilemma(boeker, self = 1, ideal = 2, r.min = .5)

# adjust congruence and discrepance ranges
indexDilemma(boeker, self = 1, ideal = 2, diff.congruent = 0, diff.discrepant = 4)

# print options (see ?print.indexDilemma for help)
print(id, output = "D")   # dilemmas only
print(id, output = "OD")  # overview and dilemmas

# plot dilemmas as network graph (see ?plot.indexDilemma for help)
# set a seed for reproducibility
plot(id, layout = "rows")
plot(id, layout = "circle")
plot(id, layout = "star")

Dilemmatic constructs

Description

A Dilemmatic Construct (DC) is one where the ideal element is rated on the scale midpoint. This means, the person cannot decide which of the poles is preferable. Such constructs are called "dilemmatic". For example, on a rating scale from 1 to 7, a rating of 4 on the ideal element means that the construct is dilemmatic. By definition, DCs can only emerge in scales with an uneven number of rating options, i.e. 5-point scale, 7-point scale etc. However, the function makes it possible to allow for a deviation from the midpoint, to still count as dilemmatic. This is useful if the grid uses a large rating scale, e.g. from 0 to 100 or a visual analog scale, as some grid administration programs do. In this case you may want to set ratings, for example, between 45 and 55 as close enough to the midpoint to indicate that both poles are equally desirable.

Usage

indexDilemmatic(x, ideal, deviation = 0, warn = TRUE)

Arguments

Value

List of class indexDilemmatic:

• ideal: Name of the ideal element.

• n_constructs Number of grid's constructs.

• scale: Minimum and maximum of grid rating scale.

• midpoint: Midpoint of rating scale.

• ⁠lower,upper⁠: Lower and upper value to for a rating to be considered in the midpoint range.

• midpoint_range: Midpoint range as interval.

• n_dilemmatic: Number of dilemmatic constructs.

• perc_dilemmatic: Percentage of constructs which are dilemmatic.

• i_dilemmatic: Index of dilemmatic constructs.

• dilemmatic_constructs: Labels of dilemmatic constructs.

• summary: Summary dataframe.

Examples

dc <- indexDilemmatic(feixas2004, ideal = 13)
dc

# control the output
print(dc, output = "S")  # Summary 
print(dc, output = "D")  # Details

Intensity index

Description

Calculate intensity index.

Usage

indexIntensity(x, rc = FALSE, trim = 30)

Arguments

Details

The Intensity index has been suggested by Bannister (1960) as a measure of the amount of construct linkage. Bannister suggested that the score reflects the degree of organization of the construct system under investigation (Bannister & Mair, 1968). The index resulted from his and his colleagues work on construction systems of patient suffering schizophrenic thought disorder. The concept of intensity has a theoretical connection to the notion of "tight" and "loose" construing as proposed by Kelly (1991). While tight constructs lead to unvarying prediction, loose constructs allow for varying predictions. Bannister hypothesized that schizophrenic thought disorder is liked to a process of extremely loose construing leading to a loss of predictive power of the subject's construct system. The Intensity score as a structural measure is thought to reflect this type of system disintegration (Bannister, 1960).

Implementation as in the Gridcor program and explained on the correspoding help pages: "... the sum of the squared values of the correlations of each construct with the rest of the constructs, averaged by the total number of constructs minus one. This process is repeated with each element, and the overall Intensity is calculated by averaging the intensity scores of constructs and elements." (Gridcor manual). Currently the total is calculated as the unweighted average of all single scores (for elements and construct).

Value

An object of class indexIntensity containing a list with the following elements:

c.int: Intensity scores by construct. e.int: Intensity scores by element. c.int.mean: Average intensity score for constructs. e.int.mean: Average intensity score for elements. total.int: Total intensity score.

Development

TODO: Results have not been tested against other programs' results.

References

Bannister, D. (1960). Conceptual structure in thought-disordered schizophrenics. The Journal of mental science, 106, 1230-49.

Examples

indexIntensity(bell2010)
indexIntensity(bell2010, trim = NA)

# using Cohen's rc for element correlations
indexIntensity(bell2010, rc = TRUE)

# save output
x <- indexIntensity(bell2010)
x

# printing options
print(x, digits = 4)

# accessing the objects' content
x$c.int
x$e.int
x$c.int.mean
x$e.int.mean
x$total.int

Polarization (percentage of extreme ratings)

Description

Polarization is the percentage of extreme ratings, e.g. the values 1 and 7 for a grid with a 7-point ratings scale.

Usage

indexPolarization(x, deviation = 0)

Arguments

Value

List of class indexPolarization:

• scale: Minimum and maximum of grid rating scale.

• ⁠lower,upper⁠ Lower and upper value to decide which ratings are considered extreme.

• polarization_total: Grid's overall polarization.

• polarization_constructs: Polarization per construct.

• polarization_elements: Polarization per element.

Examples

p <- indexPolarization(boeker)
p

# control the output
print(p, output = "T")  # total polarization
print(p, output = "C")  # construct polarization
print(p, output = "E")  # element polarization

Percentage of Variance Accounted for by the First Factor (PVAFF)

Description

The PVAFF is used as a measure of cognitive complexity. It was introduced in an unpublished PhD thesis by Jones (1954, cit. Bonarius, 1965). To calculate it, the 'first factor' two different methods may be used. One applies principal component analysis (PCA) to the construct centered raw data (default), the second applies SVD to the construct correlation matrix. The PVAFF reflects the amount of variation that is accounted for by a single linear component. If a single latent component is able to explain the variation in the grid, the cognitive complexity is said to be low. In this case the construct system is regarded as 'simple' (Bell, 2003).

Usage

indexPvaff(x, method = 1)

Arguments

References

Bell, R. C. (2003). An evaluation of indices used to represent construct structure. In G. Chiari & M. L. Nuzzo (Eds.), Psychological Constructivism and the Social World (pp. 297-305). Milan: FrancoAngeli.

Bonarius, J. C. J. (1965). Research in the personal construct theory of George A. Kelly: role construct repertory test and basic theory. In B. A. Maher (Ed.), Progress in experimental personality research (Vol. 2). New York: Academic Press.

James, R. E. (1954). Identification in terms of personal constructs (Unpublished doctoral thesis). Ohio State University, Columbus, OH.

Examples

indexPvaff(bell2010)

Self construction profile

Description

TBD

Usage

indexSelfConstruction(
  x,
  self,
  ideal,
  others = c(-self, -ideal),
  method = "euclidean",
  p = 2,
  normalize = TRUE,
  round = FALSE
)

Arguments

Value

List object of class indexSelfConstruction, containing the results from the calculations:

• grid: Reduced grid with self, ideal and others

• method_type: method type (correlation or distance)

• method: correlation or distance method used

• self_element: name of the self element

• ideal_element: name of the ideal element

• other_elements: name(s) of other elements

• self_ideal: measure between self and ideal

• self_others: measure between self and others

• ideal_others: measure betwen ideal and others

References

TBD

Examples

# using distance measures
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "euclidean")
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "manhattan")
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "minkowski", p = 3)

# using correlation measures
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "pearson")
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "spearman")

# using not-normalized distances
indexSelfConstruction(boeker, 1, 2, c(3:11), method = "euclidean", normalize = FALSE)

# printing the results (biplot only works with)
cp <- indexSelfConstruction(boeker, 1, 2, c(3:11))
cp$grid  # grid with self, ideal and others
biplot2d(cp$grid, center = 4)  # midopoint centering

Uncertainty index

Description

A measure for the degree of dispersion of dependency in a dependency grid (Bell, 2001). It is normalized measure with a value range between 0 and 1. The index is applicable to dependency grids (e.g., situation-resource) only, i.e., all grid ratings must be 0 or 1.

Usage

indexUncertainty(x)

Arguments

Details

Theoretical Background: Dispersion of Dependency: Kelly (1969) proposed that it is problematic to view people as either independent or dependent because everyone is, to greater or lesser degrees, dependent upon others in life. What Kelly felt was important was how well people disperse their dependencies across different people. Whereas young children tend to have their dependencies concentrated on a small number of people (typically parents), adults are more likely to spread their dependencies across a variety of others. Dispersing one's dependencies is generally considered more psychologically adjusted for adults (Walker et al., 1988).

References

Bell, R. C. (2001). Some new Measures of the Dispersion of Dependency in a Situation-Resource Grid. Journal of Constructivist Psychology, 14(3), 227-234, doi:10.1080/713840106.

See Also

indexDDI

Examples

# sample grid from Bell (2001, p.231)
file <- system.file("extdata", "dep_grid_bell_2001.xlsx" , package = "OpenRepGrid")
x <- importExcel(file)

indexUncertainty(x) 

Calculate 'variability' of a grid as defined by Slater (1977).

Description

Variability records a tendency for the responses to gravitate towards both end of the gradings scale. (Slater, 1977, p.88).

Usage

indexVariability(x, min = NULL, max = NULL, digits = 2)

Arguments

Value

Numeric.

Note

STATUS: working and checked against example in Slater, 1977 , p.88.

References

Slater, P. (1977). The measurement of intrapersonal space by Grid technique. London: Wiley.

See Also

indexBias()

Examples

indexVariability(boeker)

Test if object has class repgrid

Description

Test if object has class repgrid

Usage

is.repgrid(x)

Arguments

lapply with a progress bar

Description

Can be used like standard base:::lapply. The only thing it does is create an additional progress bar.

Usage

lapply_pb(X, FUN, ...)

Arguments

Value

list see ?lapply

See Also

lapply()

Examples

## Not run: 

l <- sapply(1:20000, function(x) list(rnorm(1000)))
lapply_pb(l, mean)

## End(Not run)

Move construct or element in grid to the left, right, up or down.

Description

Move element in grid to the right.

Usage

left(x, pos = 0)

right(x, pos = 0)

up(x, pos = 0)

down(x, pos = 0)

Arguments

Value

repgrid object.

repgrid object

repgrid object

repgrid object

Examples

## Not run: 
x <- randomGrid()
left(x, 2) # 2nd element to the left
right(x, 1) # 1st element to the right
up(x, 2) # 2nd construct upwards
down(x, 1) # 1st construct downwards

## End(Not run)

Constructor for repgrid class

Description

Constructor for repgrid class

Usage

makeEmptyRepgrid()

Value

repgrid object

Make a new repgrid object.

Description

The function creates a repgrid object from scratch. A number of parameters have to be defined in order to make a new grid (see parameters).

Usage

makeRepgrid(args)

Arguments

Value

NULL

Examples

## Not run: 

# make list object containing the arguments
args <- list(
  name = c("element_1", "element_2", "element_3", "element_4"),
  l.name = c("left_1", "left_2", "left_3"),
  r.name = c("right_1", "right_2", "right_3"),
  scores = c(
    1, 0, 1, 0,
    1, 1, 1, 0,
    1, 0, 1, 0
  )
)
# make grid object
x <- makeRepgrid(args)
x

## End(Not run)

map a value onto others

Description

map a value onto others

Usage

map(x, each)

Arguments

Value

vector

Examples

## Not run: 
map(1:10, 3)

m <- matrix(1:12, ncol = 4)
for (i in 1:12) {
  print(m[ring(i, 3), map(i, 4)])
}

## End(Not run)

Determine color values according to a given range of values.

Description

From a given numeric vector z the range is determined and the values are linearly mapped onto the interval given by val.range. Then a color ramp using the colors given by color is created and the mapped values are transformed into hex color values.

Usage

mapCoordinatesToColor(z, colors = c("white", "black"), val.range = c(0.2, 0.8))

Arguments

Value

numeric vector

Map arbitrary numeric vector to a given range of values.

Description

From a given numeric vector z the range is determined and the values are linearly mapped onto the interval given by val.range. This function can be used in order to map arbitrary vectors to a given range of values.

Usage

mapCoordinatesToValue(z, val.range = c(0.5, 1))

Arguments

Value

numeric vector

Number of matches in ratings

Description

Count the number of matches, i.e. (near) identical ratings between two elements or constructs. Matches are used as the basis for the calculation of grid indexes.

Usage

matches(x, deviation = 0, diag.na = TRUE)

Arguments

Value

A list of class org.matches with:

• grid: The grid used to calculate the matches.

• deviation The deviation parameter.

• max_constructs Maximum possible number of matches across constructs.

• max_elements Maximum possible number of matches across elements.

• total_constructs Total number of matches across constructs.

• total_elements Total number of matches across elements.

• constructs: Matrix with no. of matches for constructs.

• elements: Matrix with no. of matches for elements.

Examples

m <- matches(boeker)

# several output options
print(m, index = FALSE, names = FALSE, upper = FALSE)
print(m, output = "C")  # construct matches
print(m, output = "E")  # element matches

# extract the matrices
m$constructs
m$elements

Midpoint of the grid rating scale

Description

Midpoint of the grid rating scale

Usage

midpoint(x)

Arguments

Value

Midpoint of scale.

Examples

midpoint(bell2010)

modify a construct

Description

change the attributes of a construct

Usage

modifyConstruct(
  x,
  pos,
  l.name = NA,
  l.preferred = NA,
  l.emerged = NA,
  r.name = NA,
  r.preferred = NA,
  r.emerged = NA,
  replace = FALSE
)

Arguments

Value

repgrid object

Examples

## Not run: 

####  TODO  ####

## End(Not run)

modifyElement

Description

change the attributes of an element i.e. name, abbreviation, status etc.

Usage

modifyElement(
  x,
  pos,
  name = NA,
  abbreviation = NA,
  status = NA,
  replace = FALSE
)

Arguments

Value

repgrid object

Examples

## Not run: 

####  TODO  ####

## End(Not run)

Converts a Gridstat multigrid file into temporary single grid files and returns their path

Description

The format for a multigrid file resembles the single Gridstat data file. The lines of the single files are simply placed below each other without any blank lines in between. The function reads in a file and tests if it is a multigrid file. Multigrid files are separated into single Gridstat temp files. The file path for the temp files is returned. If the file is a single grid files the path is left unaltered.

Usage

multigridFileToSinglegridFiles(file)

Arguments

Value

A vector containing the paths to the temp files

Normalize rows or columns by its standard deviation.

Description

Normalize rows or columns by its standard deviation.

Usage

normalize(x, normalize = 0, ...)

Arguments

Value

Not yet defined TODO!

Examples

x <- matrix(sample(1:5, 20, rep = TRUE), 4)
normalize(x, 1) # normalizing rows
normalize(x, 2) # normalizing columns

order one vector by another

Description

small wrapper to order one vector by another, hardly worth a function

Usage

orderBy(x, y)

Arguments

Value

vector

find the order of a string vector so it will match the order of another

Description

find the order of a string vector so it will match the order of another

Usage

orderByString(x, y)

Arguments

Value

a vector of strings

Examples

## Not run: 
a <- c("c", "a", "b")
b <- c("b", "c", "a")
index <- orderByString(a, b) # to order b like a needs what indexes?
index
b[index]

## End(Not run)

Generate a list with all possible construct reflections of a grid.

Description

Generate a list with all possible construct reflections of a grid.

Usage

permuteConstructs(x, progress = TRUE)

Arguments

Value

A list of repgrid objects with all possible permutations of the grid.

Examples

## Not run: 

l <- permuteConstructs(mackay1992)
l

## End(Not run)

Permute rows, columns or whole grid matrix.

Description

Generate one or many permutations of the grid by shuffling the rows, the columns or the whole grid matrix.

Usage

permuteGrid(x, along = 1, n = 1)

Arguments

Value

A repgrid object if n = 1 or a list of repgrid objects if n > 1.

Examples

## Not run: 

# permute grid
permuteGrid(bell2010)
permuteGrid(bell2010)
permuteGrid(bell2010)

# generate a list of permuted grids
permuteGrid(bell2010, n = 5)

## End(Not run)

Perturbate grid ratings

Description

Randomly subtract or add an amount to a proportion of the grid ratings. This emulates randomness during the rating process, producing a grid which might also have resulted.

Usage

perturbate(x, prop = 0.1, amount = c(-1, 1), prob = c(0.5, 0.5))

grids_perturbate(x, n = 10, prop = 0.1, amount = c(-1, 1), prob = c(0.5, 0.5))

Arguments