Package 'ufs'

Description

This is simply 'in', but applies base::toupper() to both arguments, first.

Usage

find %IN% table

Arguments

Value

A logical vector.

Examples

letters[1:4] %IN% LETTERS

Description

Vargha & Delaney's A

Usage

A_VarghaDelaney(
  control,
  experimental,
  bootstrap = NULL,
  conf.level = 0.95,
  warn = FALSE
)

Arguments

Value

A numeric vector of length 1 with the A value, named 'A'.

Examples

ufs::A_VarghaDelaney(1:8, 3:12);

Description

Sample size for accuracy: d

Usage

aipedjmv(d = 0.5, w = 0.1, conf.level = 95)

Arguments

Value

A results object containing:

Description

Sample size for accuracy: r

Usage

aiperjmv(r = 0.3, w = 0.1, conf.level = 95)

Arguments

Value

A results object containing:

Description

This function by Josh O'Brien checks whether elements of a vector are valid colors. It has been copied from a Stack Exchange answer (see https://stackoverflow.com/questions/13289009/check-if-character-string-is-a-valid-color-representation).

Usage

areColors(x)

Arguments

Value

A logical vector.

Author(s)

Josh O'Brien

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

ufs::areColors(c(NA, "black", "blackk", "1", "#00", "#000000"));

Description

This is a function to conveniently and quickly compute the absolute relative risk (ARR) and its confidence interval.

Usage

arr(
  expPos,
  expN,
  conPos,
  conN,
  conf.level = 0.95,
  digits = 2,
  printAsPercentage = TRUE
)

## S3 method for class 'ufsARR'
print(x, digits = x$digits, printAsPercentage = x$printAsPercentage, ...)

Arguments

Value

An object with in estimate, the ARR, and in conf.int, the confidence interval.

Examples

ufs::arr(10, 60, 20, 60);

Description

associationMatrix produces a matrix with confidence intervals for effect sizes, point estimates for those effect sizes, and the p-values for the test of the hypothesis that the effect size is zero, corrected for multiple testing.

Usage

associationMatrix(
  dat = NULL,
  x = NULL,
  y = NULL,
  conf.level = 0.95,
  correction = "fdr",
  bootstrapV = FALSE,
  info = c("full", "ci", "es"),
  includeSampleSize = "depends",
  bootstrapV.samples = 5000,
  digits = 2,
  pValueDigits = digits + 1,
  colNames = FALSE,
  type = c("R", "html", "latex"),
  file = "",
  statistic = associationMatrixStatDefaults,
  effectSize = associationMatrixESDefaults,
  var.equal = TRUE
)

## S3 method for class 'associationMatrix'
print(x, type = x$input$type, info = x$input$info, file = x$input$file, ...)

## S3 method for class 'associationMatrix'
pander(x, info = x$input$info, file = x$input$file, ...)

Arguments

Value

An object with the input and several output variables, one of which is a dataframe with the association matrix in it. When this object is printed, the association matrix is printed to the screen. If the 'file' parameter is specified, a file with this matrix will also be written to disk.

Note

The 'statistic' and 'effectSize' parameter make it possible to use different functions to conduct null hypothesis significance testing and compute effect sizes. In both cases, the parameter needs to be a list containing four lists, named 'dichotomous', 'nominal', 'ordinal', and 'interval'. Each of these lists has to contain four elements, character vectors of length one (i.e. just one string value), again named 'dichotomous', 'nominal', 'ordinal', and 'interval'.

The combination of each of these names (e.g. 'dichotomous' and 'nominal', or 'ordinal' and 'interval', etc) determine which test should be done when computing the p-value to test the association between two variables of those types, or which effect sizes to compute. When called, associationMatrix determines the measurement levels of the relevant variables. It then uses these two levels (their string representation, e.g. 'dichotomous' etc) to find a string in the 'statistic' and 'effectSize' objects. Two functions with these names are then called from two lists, 'computeStatistic' and computeEffectSize. These lists list contain functions that have the same names as the strings in the 'statistic' list.

For example, when the default settings are used, the string (function name) found for two dichotomous variables when searching in associationMatrixStatDefaults is 'chisq', and the string found in associationMatrixESDefaults is 'v'. associationMatrix then calls computeStatistic[['chisq']] and computeEffectSize[['v']], providing the two variables as arguments, as well as passing the 'conf.level' argument. These two functions then each return an object that associationMatrix extracts the information from. Inspect the source code of these functions (by typing their names without parentheses in the R prompt) to learn how this object should look, if you want to write your own functions.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

### Generate a simple association matrix using all three variables in the
### Orange tree dataframe
associationMatrix(Orange);

### Or four variables from infert:
associationMatrix(infert, c("education", "parity",
                            "induced", "case"), colNames=TRUE);

### Use variable names in the columns and generate html
associationMatrix(Orange, colNames=TRUE, type='html');

Description

This function produces is a diamondplot that plots the confidence intervals for associations between a number of covariates and a criterion. It currently only supports the Pearson's r effect size metric; other effect sizes are converted to Pearson's r.

Usage

associationsDiamondPlot(
  dat,
  covariates,
  criteria,
  labels = NULL,
  criteriaLabels = NULL,
  decreasing = NULL,
  sortBy = NULL,
  conf.level = 0.95,
  criteriaColors = viridisPalette(length(criteria)),
  criterionColor = "black",
  returnLayerOnly = FALSE,
  esMetric = "r",
  multiAlpha = 0.33,
  singleAlpha = 1,
  showLegend = TRUE,
  xlab = "Effect size estimates",
  ylab = "",
  theme = ggplot2::theme_bw(),
  lineSize = 1,
  outputFile = NULL,
  outputWidth = 10,
  outputHeight = 10,
  ggsaveParams = ufs::opts$get("ggsaveParams"),
  ...
)

associationsToDiamondPlotDf(
  dat,
  covariates,
  criterion,
  labels = NULL,
  decreasing = NULL,
  conf.level = 0.95,
  esMetric = "r"
)

Arguments

Details

associationsToDiamondPlotDf is a helper function that produces the required dataframe.

This function can be used to quickly plot multiple confidence intervals.

Value

A plot.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

diamondPlot(), ggDiamondLayer()

Examples

### Simple diamond plot with correlations
### and their confidence intervals

associationsDiamondPlot(mtcars,
                        covariates=c('cyl', 'hp', 'drat', 'wt',
                                     'am', 'gear', 'vs', 'carb', 'qsec'),
                        criteria='mpg');

### Same diamond plot, but now with two criteria,
### and colouring the diamonds based on the
### correlation point estimates: a gradient
### is created where red is used for -1,
### green for 1 and blue for 0.

associationsDiamondPlot(mtcars,
                        covariates=c('cyl', 'hp', 'drat', 'wt',
                                     'am', 'gear', 'vs', 'carb', 'qsec'),
                        criteria=c('mpg', 'disp'),
                        generateColors=c("red", "blue", "green"),
                        fullColorRange=c(-1, 1));

Description

Measurement error (i.e. the complement of reliability) results in a downward bias of observed effect sizes. This attenuation can be emulated by this function.

Usage

attenuate.d(d, reliability)

Arguments

Value

The attenuated value of Cohen's d

Author(s)

Gjalt-Jorn Peters & Stefan Gruijters

References

Bobko, P., Roth, P. L., & Bobko, C. (2001). Correcting the Effect Size of d for Range Restriction and Unreliability. Organizational Research Methods, 4(1), 46–61. doi:10.1177/109442810141003

Examples

attenuate.d(.5, .8);

Description

Attenuate a Pearson's r estimate for unreliability in the measurements

Usage

attenuate.r(r, reliability1, reliability2)

Arguments

Value

The attenuated value of Pearson's r

Examples

attenuate.r(.5, .8, .9);

Description

Bland-Altman Change plot

Usage

BAC_plot(
  data,
  cols = names(data),
  reliability = NULL,
  pointSize = 2,
  deterioratedColor = "#482576E6",
  unchangedColor = "#25848E80",
  improvedColor = "#7AD151E6",
  zeroLineColor = "black",
  zeroLineType = "dashed",
  ciLineColor = "red",
  ciLineType = "solid",
  conf.level = 0.95,
  theme = ggplot2::theme_minimal(),
  ignoreBias = FALSE,
  iccFromPsych = FALSE,
  iccFromPsychArgs = NULL
)

Arguments

Value

A ggplot2 plot.

Examples

### Create smaller dataset for example
dat <-
  ufs::testRetestSimData[
    1:25,
    c('t0_item1', 't1_item1')
  ];

ufs::BAC_plot(dat, reliability = .5);
ufs::BAC_plot(dat, reliability = .8);
ufs::BAC_plot(dat, reliability = .9);

Description

This is a dataset lifted from the psychTools package (which was originally in the psych package). For details, please check that help page (using "psychTools::bfi").

Usage

data(bfi)

Format

A data.frame with 2800 rows and 28 columns.

Examples

data(bfi);

Description

This is basically a meansDiamondPlot(), but extended to allow specifying subquestions and anchors at the left and right side. This is convenient for psychological questionnaires when the anchors or dimensions were different from item to item. This function is used to function the left panel of the CIBER plot in the behaviorchange package.

Usage

biAxisDiamondPlot(
  dat,
  items = NULL,
  leftAnchors = NULL,
  rightAnchors = NULL,
  subQuestions = NULL,
  decreasing = NULL,
  conf.level = 0.95,
  showData = TRUE,
  dataAlpha = 0.1,
  dataColor = "#444444",
  diamondColors = NULL,
  jitterWidth = 0.45,
  jitterHeight = 0.45,
  xbreaks = NULL,
  xLabels = NA,
  xAxisLab = paste0("Scores and ", round(100 * conf.level, 2), "% CIs"),
  drawPlot = TRUE,
  returnPlotOnly = TRUE,
  baseSize = 1,
  dotSize = baseSize,
  baseFontSize = 10 * baseSize,
  theme = ggplot2::theme_bw(base_size = baseFontSize),
  outputFile = NULL,
  outputWidth = 10,
  outputHeight = 10,
  ggsaveParams = ufs::opts$get("ggsaveParams"),
  ...
)

Arguments

Details

This is a diamondplot that can be used for items/questions where the anchors of the response scales could be different for every item. For the rest, it is very similar to meansDiamondPlot().

Value

Either just a plot (a gtable::gtable() object) or an object with all produced objects and that plot.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

CIBER() in the behaviorchange package, associationsDiamondPlot()

Examples

biAxisDiamondPlot(dat=mtcars,
                  items=c('cyl', 'wt'),
                  subQuestions=c('cylinders', 'weight'),
                  leftAnchors=c('few', 'light'),
                  rightAnchors=c('many', 'heavy'),
                  xbreaks=0:8);

Description

Create colours for a response scale for an item

Usage

biDimColors(start, mid, end, length, show = TRUE)

uniDimColors(start, end, length, show = TRUE)

Arguments

Value

The colours as hex codes.

Examples

uniDimColors("#000000", "#00BB00", length=5, show=FALSE);

Description

This function is a wrapper for the functions from the careless package. Normally, you'd probably call carelessReport which calls this function to generate a report of suspect participants.

Usage

carelessObject(
  data,
  items = names(data),
  flagUnivar = 0.99,
  flagMultivar = 0.95,
  irvSplit = 4,
  responseTime = NULL
)

Arguments

Value

An object of class carelessObject.

Examples

carelessObject(mtcars);

Description

This function wraps functions from the careless package to help inspect and diagnose careless participants. It is optimized for using in R Markdown files.

Usage

carelessReport(
  data,
  items = names(data),
  nFlags = 1,
  flagUnivar = 0.99,
  flagMultivar = 0.95,
  irvSplit = 4,
  headingLevel = 3,
  datasetName = NULL,
  responseTime = NULL,
  headingSuffix = " {.tabset}",
  digits = 2,
  missingSymbol = "Missing"
)

Arguments

Value

NULL, invisibly; and prints the report.

Examples

### Get the BFI data taken from the `psych` package
dat <- ufs::bfi;

### Get the variable names for the regular items
bfiVars <-
  setdiff(names(dat),
          c("gender", "education", "age"));

### Inspect suspect participants, very conservatively to
### limit the output (these are 2800 participants).
carelessReport(data = dat,
               items = bfiVars,
               nFlags = 5);

Description

The cat0 function is to cat what paste0 is to paste; it simply makes concatenating many strings without a separator easier.

Usage

cat0(..., sep = "")

Arguments

Value

Nothing (invisible NULL, like cat).

Examples

cat0("The first variable is '", names(mtcars)[1], "'.");

Description

This function is designed to make it easy to perform some data integrity checks, specifically checking for values that are impossible or unrealistic. These values can then be replaced by another value, or the offending cases can be deleted from the dataframe.

Usage

checkDataIntegrity(
  x,
  dat,
  newValue = NA,
  removeCases = FALSE,
  validValueSuffix = "_validValue",
  newValueSuffix = "_newValue",
  totalVarName = "numberOfInvalidValues",
  append = TRUE,
  replace = TRUE,
  silent = FALSE,
  rmarkdownOutput = FALSE,
  callingSelf = FALSE
)

Arguments

Value

The dataframe with the corrections, and the log stored in attribute checkDataIntegrity_log.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

### Default behavior: return dataframe with
### offending values replaced by NA

checkDataIntegrity(c('mpg', '<30'),
                   mtcars);

### Check two conditions, and instead of returning the
### dataframe with the results appended, only return the
### columns indicating which cases 'pass', what the new
### values would be, and how many invalid values were
### found for each case (to easily remove cases that
### provided many invalid values)

checkDataIntegrity(list(c('mpg', '<30'),
                        c('gear', '<5')),
                   mtcars,
                   append=FALSE);

Description

This function efficiently checks for the presence of a package without loading it (unlike library() or require(). This is useful to force yourself to use the package::function syntax for addressing functions; you can make sure required packages are installed, but their namespace won't attach to the search path.

Usage

checkPkgs(
  ...,
  install = FALSE,
  load = FALSE,
  repos = "https://cran.rstudio.com"
)

Arguments

Value

Invisibly, a vector of the available packages.

Examples

ufs::checkPkgs('base');

### Require a specific version
ufs::checkPkgs(ufs = "0.3.1");

### This will show the error message
tryCatch(
  ufs::checkPkgs(
    base = "99",
    stats = "42.5",
    ufs = 20
  ),
  error = print
);

Description

Conceptual Independence Matrix

Usage

CIM(
  data,
  scales,
  conf.level = 0.95,
  colors = c("#440154FF", "#7AD151FF"),
  outputFile = NULL,
  outputWidth = 100,
  outputHeight = 100,
  outputUnits = "cm",
  faMethod = "minres",
  n.iter = 100,
  n.repeatOnWarning = 50,
  warningTolerance = 2,
  silentRepeatOnWarning = FALSE,
  showWarnings = FALSE,
  skipRegex = NULL,
  headingLevel = 2,
  printAbbreviations = TRUE,
  drawPlot = TRUE,
  returnPlotOnly = TRUE
)

CIM_partial(
  x,
  headingLevel = x$input$headingLevel,
  quiet = TRUE,
  echoPartial = FALSE,
  partialFile = NULL,
  ...
)

## S3 method for class 'CIM'
knit_print(
  x,
  headingLevel = x$input$headingLevel,
  quiet = TRUE,
  echoPartial = FALSE,
  partialFile = NULL,
  ...
)

Arguments

Value

A ggplot2::ggplot() plot.

Examples

### Load dataset `bfi`, originally from psychTools package
data(bfi, package= 'ufs');

### Specify scales
bfiScales <-
  list(Agreeableness     = paste0("Agreeableness_item_", 1:5),
       Conscientiousness = paste0("Conscientiousness_item_", 1:5),
       Extraversion      = paste0("Extraversion_item_", 1:5),
       Neuroticism       = paste0("Neuroticism_item_", 1:5),
       Openness          = paste0("Openness_item_", 1:5));

names(bfi) <- c(unlist(bfiScales),
                c('gender', 'education', 'age'));

### Only select first two and the first three items to
### keep it quick; just pass the full 'bfiScales'
### object to run for all five the full scales

CIM(bfi,
    scales=lapply(bfiScales, head, 3)[1:2],
    n.iter=10);

Description

These functions use some conversion to and from the t distribution to provide the Cohen's d distribution. There are four versions that act similar to the standard distribution functions (the d., p., q., and r. functions, and their longer aliases .Cohensd), three convenience functions (pdExtreme, pdMild, and pdInterval), a function to compute the confidence interval for a Cohen's d estimate cohensdCI, and a function to compute the sample size required to obtain a confidence interval around a Cohen's d estimate with a specified accuracy (pwr.cohensdCI and its alias pwr.confIntd).

Usage

cohensdCI(d, n, conf.level = 0.95, plot = FALSE, silent = TRUE)

dCohensd(
  x,
  df = NULL,
  populationD = 0,
  n = NULL,
  n1 = NULL,
  n2 = NULL,
  silent = FALSE
)

pCohensd(q, df, populationD = 0, lower.tail = TRUE)

qCohensd(p, df, populationD = 0, lower.tail = TRUE)

rCohensd(n, df, populationD = 0)

pdInterval(ds, n, populationD = 0)

pdExtreme(d, n, populationD = 0)

pdMild(d, n, populationD = 0)

pwr.cohensdCI(d, w = 0.1, conf.level = 0.95, extensive = FALSE, silent = TRUE)

Arguments

Details

The functions use convert.d.to.t() and convert.t.to.d() to provide the Cohen's d distribution.

The confidence interval functions, cohensdCI and pwr.cohensdCI, now use the same method as MBESS (a slightly adapted version of the MBESS function conf.limits.nct is used).

More details about cohensdCI and pwr.cohensdCI are provided in Peters & Crutzen (2017).

Value

dCohensd (or dd) gives the density, pCohensd (or pd) gives the distribution function, qCohensd (or qd) gives the quantile function, and rCohensd (or rd) generates random deviates.

pdExtreme returns the probability (or probabilities) of finding a Cohen's d equal to or more extreme than the specified value(s).

pdMild returns the probability (or probabilities) of finding a Cohen's d equal to or less extreme than the specified value(s).

pdInterval returns the probability of finding a Cohen's d that lies in between the two specified values of Cohen's d.

cohensdCI provides the confidence interval(s) for a given Cohen's d value.

pwr.cohensdCI provides the sample size required to obtain a confidence interval for Cohen's d with a desired width.

Author(s)

Gjalt-Jorn Peters (Open University of the Netherlands), with the exported MBESS function conf.limits.nct written by Ken Kelley (University of Notre Dame), and with an error noticed by Guy Prochilo (University of Melbourne).

Maintainer: Gjalt-Jorn Peters [email protected]

References

Peters, G. J. Y. & Crutzen, R. (2017) Knowing exactly how effective an intervention, treatment, or manipulation is and ensuring that a study replicates: accuracy in parameter estimation as a partial solution to the replication crisis. https://dx.doi.org/

Maxwell, S. E., Kelley, K., & Rausch, J. R. (2008). Sample size planning for statistical power and accuracy in parameter estimation. Annual Review of Psychology, 59, 537-63. https://doi.org/10.1146/annurev.psych.59.103006.093735

Cumming, G. (2013). The New Statistics: Why and How. Psychological Science, (November). https://doi.org/10.1177/0956797613504966

See Also

convert.d.to.t(), convert.t.to.d(), dt(), pt(), qt(), rt()

Examples

### Confidence interval for Cohen's d of .5
### from a sample of 200 participants, also
### showing this visually: this clearly shows
### how wildly our Cohen's d value can vary
### from sample to sample.
cohensdCI(.5, n=200, plot=TRUE);

### How many participants would we need if we
### would want a more accurate estimate, say
### with a maximum confidence interval width
### of .2?
pwr.cohensdCI(.5, w=.1);

### Show that 'sampling distribution':
cohensdCI(.5,
          n=pwr.cohensdCI(.5, w=.1),
          plot=TRUE);

### Generate 10 random Cohen's d values
rCohensd(10, 20, populationD = .5);

### Probability of findings a Cohen's d smaller than
### .5 if it's 0 in the population (i.e. under the
### null hypothesis)
pCohensd(.5, 64);

### Probability of findings a Cohen's d larger than
### .5 if it's 0 in the population (i.e. under the
### null hypothesis)
1 - pCohensd(.5, 64);

### Probability of findings a Cohen's d more extreme
### than .5 if it's 0 in the population (i.e. under
### the null hypothesis)
pdExtreme(.5, 64);

### Probability of findings a Cohen's d more extreme
### than .5 if it's 0.2 in the population.
pdExtreme(.5, 64, populationD = .2);

Description

These objects contain a number of settings and functions for associationMatrix.

Usage

computeStatistic_t(var1, var2, conf.level = 0.95, var.equal = TRUE, ...)

computeStatistic_r(var1, var2, conf.level = 0.95, ...)

computeStatistic_f(var1, var2, conf.level = 0.95, ...)

computeStatistic_chisq(var1, var2, conf.level = 0.95, ...)

computeEffectSize_d(var1, var2, conf.level = 0.95, var.equal = TRUE, ...)

computeEffectSize_r(var1, var2, conf.level = 0.95, ...)

computeEffectSize_etasq(var1, var2, conf.level = 0.95, ...)

computeEffectSize_omegasq(var1, var2, conf.level = 0.95, ...)

computeEffectSize_v(
  var1,
  var2,
  conf.level = 0.95,
  bootstrap = FALSE,
  samples = 5000,
  ...
)

Arguments

Value

associationMatrixStatDefaults and associationMatrixESDefaults contain the default functions from computeStatistic and computeEffectSize that are called (see the help file for associationMatrix for more details).

The other functions return an object with the relevant statistic or effect size, with a confidence interval for the effect size.

For computeStatistic, this object always contains:

And in addition, it often contains (among other things, sometimes):

For computeEffectSize, this object always contains:

And in addition, it often contains (among other things, sometimes):

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

computeStatistic_f(Orange$Tree, Orange$circumference)
computeEffectSize_etasq(Orange$Tree, Orange$circumference)

Description

Effect Size Confidence Interval: Cohens's d

Usage

confintdjmv(d = 0.5, n = 128, conf.level = 95)

Arguments

Value

A results object containing:

Description

This function uses the MBESS functions conf.limits.ncf() (which has been copied into this package to avoid the dependency on MBESS) and convert.ncf.to.omegasq() to compute the point estimate and confidence interval for Omega Squared (which have been lifted out of MBESS to avoid importing the whole package)

Usage

confIntOmegaSq(var1, var2, conf.level = 0.95)

## S3 method for class 'confIntOmegaSq'
print(x, ..., digits = 2)

Arguments

Value

A confIntOmegaSq object is returned, with as elements:

Note

Formula 16 in Steiger (2004) is used for the conversion in convert.ncf.to.omegasq().

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

References

Steiger, J. H. (2004). Beyond the F test: Effect size confidence intervals and tests of close fit in the analysis of variance and contrast analysis. Psychological Methods, 9(2), 164-82. https://doi.org/10.1037/1082-989X.9.2.164

Examples

confIntOmegaSq(mtcars$mpg, mtcars$cyl);

Description

This function simply computes confidence intervals for proportions.

Usage

confIntProp(x, n, conf.level = 0.95, plot = FALSE)

Arguments

Details

This function is the adapted source code of binom.test(). Ir uses pbeta(), with some lines of code taken from the binom.test() source. Specifically, the count for the low category is specified as first 'shape argument' to pbeta(), and the total count (either the sum of the count for the low category and the count for the high category, or the total number of cases if compareHiToLo is FALSE) minus the count for the low category as the second 'shape argument'.

Value

The confidence interval bounds in a twodimensional matrix, with the first column containing the lower bound and the second column containing the upper bound.

Author(s)

Unknown (see binom.test(); adapted by Gjalt-Jorn Peters)

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

binom.test() and ggProportionPlot, the function for which this was written.

Examples

### Simple case
  confIntProp(84, 200);

  ### Using vectors
  confIntProp(c(2,3), c(10, 20), conf.level=c(.90, .95, .99));

Description

This function computes the confidence interval for a given correlation and its sample size. This is useful to obtain confidence intervals for correlations reported in papers when informing power analyses.

Usage

confIntR(r, N, conf.level = 0.95, plot = FALSE)

Arguments

Value

The confidence interval(s) in a matrix with two columns. The left column contains the lower bound, the right column the upper bound. The rownames() are the observed correlations, and the colnames() are 'lo' and 'hi'. The confidence level and sample size are stored as attributes. The results are returned like this to make it easy to access single correlation coefficients from the resulting object (see the examples).

Author(s)

Douglas Bonett (UC Santa Cruz, United States), with minor edits by Murray Moinester (Tel Aviv University, Israel) and Gjalt-Jorn Peters (Open University of the Netherlands, the Netherlands).

Maintainer: Gjalt-Jorn Peters [email protected]

References

Bonett, D. G., Wright, T. A. (2000). Sample size requirements for estimating Pearson, Kendall and Spearman correlations. Psychometrika, 65, 23-28.

Bonett, D. G. (2014). CIcorr.R and sizeCIcorr.R https://people.ucsc.edu/~dgbonett/psyc181.html

Moinester, M., & Gottfried, R. (2014). Sample size estimation for correlations with pre-specified confidence interval. The Quantitative Methods of Psychology, 10(2), 124-130. https://www.tqmp.org/RegularArticles/vol10-2/p124/p124.pdf

Peters, G. J. Y. & Crutzen, R. (forthcoming) An easy and foolproof method for establishing how effective an intervention or behavior change method is: required sample size for accurate parameter estimation in health psychology.

See Also

confIntR()

Examples

### To request confidence intervals for one correlation
  confIntR(.3, 100);

  ### The lower bound of a single correlation
  confIntR(.3, 100)[1];

  ### To request confidence intervals for multiple correlations:
  confIntR(c(.1, .3, .5), 250);

  ### The upper bound of the correlation of .5:
  confIntR(c(.1, .3, .5), 250)['0.5', 'hi'];

Description

Effect Size Confidence Interval: Pearson's r

Usage

confintrjmv(r = 0.3, N = 400, conf.level = 95)

Arguments

Value

A results object containing:

Description

This function is vectorized.

Usage

confIntSD(x, n = NULL, conf.level = 0.95)

Arguments

Value

A vector or matrix.

Examples

ufs::confIntSD(mtcars$mpg);
ufs::confIntSD(c(6, 7), c(32, 32));

Description

These are a number of functions to convert statistics and effect size measures from/to each other.

Arguments

Details

Note that by default, the behavior of convert.d.to.r depends on the sample size (see Bruce, Kromrey & Ferron, 1998).

Value

The converted value as a numeric value.

Author(s)

Gjalt-Jorn Peters and Peter Verboon

Maintainer: Gjalt-Jorn Peters [email protected]

References

Aaron, B. Kromrey J. D. & Ferron, J. (1998) Equating "r"-based and "d"-based Effect Size Indices: Problems with a Commonly Recommended Formula. Paper presented at the Annual Meeting of the Florida Educational Research Association (43rd, Orlando, FL, November 2-4, 1998).

Examples

convert.t.to.r(t=-6.46, n=200);
convert.r.to.t(r=-.41, n=200);

### Compute some p-values
convert.t.to.p(4.2, 197);
convert.chisq.to.p(5.2, 3);
convert.f.to.p(8.93, 3, 644);

### Convert d to r using both equations
convert.d.to.r(d=.2, n1=5, n2=5, akfEq8 = FALSE);
convert.d.to.r(d=.2, n1=5, n2=5, akfEq8 = TRUE);

Description

These functions are used by nnc() in the behaviorchange package to compute the Numbers Needed for Change, but are also available for manual use.

Usage

convert.cer.to.d(
  cer,
  eer,
  eventDesirable = TRUE,
  eventIfHigher = TRUE,
  dist = "norm",
  distArgs = NULL,
  distNS = "stats"
)

convert.d.to.eer(
  d,
  cer,
  eventDesirable = TRUE,
  eventIfHigher = TRUE,
  dist = "norm",
  distArgs = list(),
  distNS = "stats"
)

convert.d.to.nnc(d, cer, r = 1, eventDesirable = TRUE, eventIfHigher = TRUE)

convert.eer.to.d(
  eer,
  cer,
  eventDesirable = TRUE,
  eventIfHigher = TRUE,
  dist = "norm",
  distArgs = NULL,
  distNS = "stats"
)

Arguments

Value

The converted value.

Author(s)

Gjalt-Jorn Peters & Stefan Gruijters

Maintainer: Gjalt-Jorn Peters [email protected]

References

Gruijters, S. L., & Peters, G. Y. (2019). Gauging the impact of behavior change interventions: A tutorial on the Numbers Needed to Treat. PsyArXiv. doi:10.31234/osf.io/2bau7

See Also

nnc() in the behaviorchange package.

Examples

convert.d.to.eer(d=.5, cer=.25);
convert.d.to.nnc(d=.5, cer=.25);

Description

This function simply returns the result of pnorm() for Cohen's d.

Usage

convert.d.to.U3(d)

Arguments

Value

An unnames numeric vector with the U3 values.

Examples

convert.d.to.U3(.5);

Description

Tries to 'smartly' convert factor and character vectors to numeric.

Usage

convertToNumeric(vector, byFactorLabel = FALSE)

Arguments

Value

The converted vector.

Examples

ufs::convertToNumeric(as.character(1:8));

Description

These functions compute the point estimate and confidence interval for Cramer's V.

Usage

cramersV(x, y = NULL, digits = 2)

## S3 method for class 'CramersV'
print(x, digits = x$input$digits, ...)

confIntV(
  x,
  y = NULL,
  conf.level = 0.95,
  samples = 500,
  digits = 2,
  method = c("bootstrap", "fisher"),
  storeBootstrappingData = FALSE
)

## S3 method for class 'confIntV'
print(x, digits = x$input$digits, ...)

Arguments

Value

A point estimate or a confidence interval for Cramer's V, an effect size to describe the association between two categorical variables.

Examples

### Get confidence interval for Cramer's V
### Note that by using 'table', and so removing the raw data, inhibits
### bootstrapping, which could otherwise take a while.
confIntV(table(infert$education, infert$induced));

Description

normalityAssessment can be used to assess whether a variable and the sampling distribution of its mean have an approximately normal distribution.

Usage

dataShape(
  sampleVector,
  na.rm = TRUE,
  type = 2,
  digits = 2,
  conf.level = 0.95,
  plots = TRUE,
  xLabs = NA,
  yLabs = NA,
  qqCI = TRUE,
  labelOutliers = TRUE,
  sampleSizeOverride = NULL
)

## S3 method for class 'dataShape'
print(x, digits = x$input$digits, extraNotification = TRUE, ...)

## S3 method for class 'dataShape'
pander(x, digits = x$input$digits, extraNotification = TRUE, ...)

normalityAssessment(
  sampleVector,
  samples = 10000,
  digits = 2,
  samplingDistColor = "#2222CC",
  normalColor = "#00CC00",
  samplingDistLineSize = 2,
  normalLineSize = 1,
  xLabel.sampleDist = NULL,
  yLabel.sampleDist = NULL,
  xLabel.samplingDist = NULL,
  yLabel.samplingDist = NULL,
  sampleSizeOverride = TRUE
)

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

## S3 method for class 'normalityAssessment'
pander(x, headerPrefix = "#####", suppressPlot = FALSE, ...)

samplingDistribution(
  popValues = c(0, 1),
  popFrequencies = c(50, 50),
  sampleSize = NULL,
  sampleFromPop = FALSE,
  ...
)

Arguments

Details

samplingDistribution is a convenient wrapper for normalityAssessment that makes it easy to quickly generate a sample and sampling distribution from frequencies (or proportions).

dataShape computes the skewness and kurtosis.

normalityAssessment provides a number of normality tests and draws histograms of the sample data and the sampling distribution of the mean (most statistical tests assume the latter is normal, rather than the first; normality of the sample data guarantees normality of the sampling distribution of the mean, but if the sample size is sufficiently large, the sampling distribution of the mean is approximately normal even when the sample data are not normally distributed). Note that for the sampling distribution, the degrees of freedom are usually so huge that the normality tests, negligible deviations from normality will already result in very small p-values.

samplingDistribution makes it easy to quickly assess the distribution of a variables based on frequencies or proportions, and dataShape computes skewness and kurtosis.

Value

An object with several results, the most notably of which are:

Examples

### Note: the 'not run' is simply because running takes a lot of time,
###       but these examples are all safe to run!
## Not run: 

normalityAssessment(rnorm(35));

### Create a distribution of three possible values and
### show the sampling distribution for the mean
popValues <- c(1, 2, 3);
popFrequencies <- c(20, 50, 30);
sampleSize <- 100;
samplingDistribution(popValues = popValues,
                     popFrequencies = popFrequencies,
                     sampleSize = sampleSize);

### Create a very skewed distribution of ten possible values
popValues <- c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10);
popFrequencies <- c(2, 4, 8, 6, 10, 15, 12, 200, 350, 400);
samplingDistribution(popValues = popValues,
                     popFrequencies = popFrequencies,
                     sampleSize = sampleSize, digits=5);

## End(Not run)

Description

This function provides a number of descriptives about your data, similar to what SPSS's DESCRIPTIVES (often called with DESCR) does.

Usage

descr(
  x,
  digits = 4,
  errorOnFactor = FALSE,
  include = c("central tendency", "spread", "range", "distribution shape", "sample size"),
  maxModes = 1,
  t = FALSE,
  conf.level = 0.95,
  quantileType = 2
)

## Default S3 method:
descr(
  x,
  digits = 4,
  errorOnFactor = FALSE,
  include = c("central tendency", "spread", "range", "distribution shape", "sample size"),
  maxModes = 1,
  t = FALSE,
  conf.level = 0.95,
  quantileType = 2
)

## S3 method for class 'descr'
print(
  x,
  digits = attr(x, "digits"),
  t = attr(x, "transpose"),
  row.names = FALSE,
  ...
)

## S3 method for class 'descr'
pander(x, headerPrefix = "", headerStyle = "**", ...)

## S3 method for class 'descr'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)

## S3 method for class 'data.frame'
descr(x, ...)

Arguments

Details

Note that R (of course) has many similar functions, such as summary, psych::describe() in the excellent psych::psych package.

The Hartigans' Dip Test may be unfamiliar to users; it is a measure of uni- vs. multidimensionality, computed by diptest::dip.test() from the dip.test package. Depending on the sample size, values over .025 can be seen as mildly indicative of multimodality, while values over .05 probably warrant closer inspection (the p-value can be obtained using diptest::dip.test(); also see Table 1 of Hartigan & Hartigan (1985) for an indication as to critical values).

Value

A list of dataframes with the requested values.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

References

Hartigan, J. A.; Hartigan, P. M. The Dip Test of Unimodality. Ann. Statist. 13 (1985), no. 1, 70–84. doi:10.1214/aos/1176346577. https://projecteuclid.org/euclid.aos/1176346577.

See Also

summary, psych::describe()

Examples

descr(mtcars$mpg);

Description

These functions are used by diamondPlot() to construct a diamond plot. It's normally not necessary to call this function directly: instead, use meansDiamondPlot(), meanSDtoDiamondPlot(), and factorLoadingDiamondCIplot().

Usage

diamondCoordinates(
  values,
  otherAxisValue = 1,
  direction = "horizontal",
  autoSize = NULL,
  fixedSize = 0.15
)

ggDiamondLayer(
  data,
  ciCols = 1:3,
  colorCol = NULL,
  generateColors = NULL,
  fullColorRange = NULL,
  color = "black",
  lineColor = NA,
  otherAxisCol = 1:nrow(data),
  autoSize = NULL,
  fixedSize = 0.15,
  direction = "horizontal",
  ...
)

rawDataDiamondLayer(
  dat,
  items = NULL,
  itemOrder = 1:length(items),
  dataAlpha = 0.1,
  dataColor = "#444444",
  jitterWidth = 0.5,
  jitterHeight = 0.4,
  size = 3,
  ...
)

varsToDiamondPlotDf(
  dat,
  items = NULL,
  labels = NULL,
  decreasing = NULL,
  conf.level = 0.95
)

Arguments

Value

ggDiamondLayer returns a ggplot() geom_polygon() object, which can then be used in ggplot() plots (as diamondPlot() does).

diamondCoordinates returns a set of four coordinates that together specify a diamond.

varsToDiamondPlotDf returns a dataframe of diamondCoordinates.

rawDataDiamondLayer returns a geom_jitter() object.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

meansDiamondPlot(), meanSDtoDiamondPlot(), factorLoadingDiamondCIplot(), diamondPlot()

Examples

## Not run: 
### (Don't run this example as a test, because we
###  need the ggplot function which isn't part of
###  this package.)

### The coordinates for a simple diamond
diamondCoordinates(values = c(1,2,3));

### Plot this diamond
ggplot() + ggDiamondLayer(data.frame(1,2,3));

## End(Not run)

Description

This function constructs a diamond plot using ggDiamondLayer(). It's normally not necessary to call this function directly: instead, use meansDiamondPlot() meanSDtoDiamondPlot(), and factorLoadingDiamondCIplot().

Usage

diamondPlot(
  data,
  ciCols = 1:3,
  colorCol = NULL,
  otherAxisCol = NULL,
  yValues = NULL,
  yLabels = NULL,
  ylab = NULL,
  autoSize = NULL,
  fixedSize = 0.15,
  xlab = "Effect Size Estimate",
  theme = ggplot2::theme_bw(),
  color = "black",
  returnLayerOnly = FALSE,
  outputFile = NULL,
  outputWidth = 10,
  outputHeight = 10,
  ggsaveParams = ufs::opts$get("ggsaveParams"),
  ...
)

Arguments

Value

A ggplot2::ggplot() plot with a ggDiamondLayer() is returned.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

meansDiamondPlot(), meanSDtoDiamondPlot(), ggDiamondLayer(), factorLoadingDiamondCIplot()

Examples

tmpDf <- data.frame(lo = c(1, 2, 3),
                    mean = c(1.5, 3, 5),
                    hi = c(2, 4, 10),
                    color = c('green', 'red', 'blue'));

### A simple diamond plot
diamondPlot(tmpDf);

### A diamond plot using the specified colours
diamondPlot(tmpDf, colorCol = 4);

### A diamond plot using automatically generated colours
### using a gradient
diamondPlot(tmpDf, generateColors=c('green', 'red'));

### A diamond plot using automatically generated colours
### using a gradient, specifying the minimum and maximum
### possible values that can be attained
diamondPlot(tmpDf, generateColors=c('green', 'red'),
            fullColorRange=c(1, 10));

Description

Measurement error (i.e. the complement of reliability) results in a downward bias of observed effect sizes. This attenuation can be reversed by disattenuation.

Usage

disattenuate.d(d, reliability)

Arguments

Value

The disattenuated value of Cohen's d

Author(s)

Gjalt-Jorn Peters & Stefan Gruijters

References

Bobko, P., Roth, P. L., & Bobko, C. (2001). Correcting the Effect Size of d for Range Restriction and Unreliability. Organizational Research Methods, 4(1), 46–61. doi:10.1177/109442810141003

Examples

disattenuate.d(.5, .8);

Description

Disattentuate a Pearson's r estimate for unreliability

Usage

disattenuate.r(r, reliability1, reliability2)

Arguments

Value

The disattenuated value of Pearson's r

Examples

disattenuate.r(.5, .8, .9);

Description

These are two diamond plot functions to conveniently make diamond plots to compare subgroups or different samples. They are both based on a univariate diamond plot where colors are used to distinguish the data points and diamonds of each subgroup or sample. The means comparison diamond plot produces only this plot, while the duo comparison diamond plot combines it with a diamond plot visualising the effect sizes of the associations. The latter currently only works for two subgroups or samples, while the simple meansComparisonDiamondPlot also works when comparing more than two sets of datapoints. These functions are explained more in detail in Peters (2017).

Usage

duoComparisonDiamondPlot(
  dat,
  items = NULL,
  compareBy = NULL,
  labels = NULL,
  compareByLabels = NULL,
  decreasing = NULL,
  conf.level = c(0.95, 0.95),
  showData = TRUE,
  dataAlpha = 0.1,
  dataSize = 3,
  comparisonColors = viridisPalette(length(unique(dat[, compareBy]))),
  associationsColor = "grey",
  alpha = 0.33,
  jitterWidth = 0.5,
  jitterHeight = 0.4,
  xlab = c("Scores and means", "Effect size estimates"),
  ylab = c(NULL, NULL),
  plotTitle = NULL,
  theme = ggplot2::theme_bw(),
  showLegend = TRUE,
  legend.position = "top",
  lineSize = 1,
  drawPlot = TRUE,
  xbreaks = "auto",
  outputFile = NULL,
  outputWidth = 10,
  outputHeight = 10,
  ggsaveParams = ufs::opts$get("ggsaveParams"),
  ...
)

meansComparisonDiamondPlot(
  dat,
  items = NULL,
  compareBy = NULL,
  labels = NULL,
  compareByLabels = NULL,
  decreasing = NULL,
  sortBy = NULL,
  conf.level = 0.95,
  showData = TRUE,
  dataAlpha = 0.1,
  dataSize = 3,
  comparisonColors = viridisPalette(length(unique(dat[, compareBy]))),
  alpha = 0.33,
  jitterWidth = 0.5,
  jitterHeight = 0.4,
  xlab = "Scores and means",
  ylab = NULL,
  plotTitle = NULL,
  theme = ggplot2::theme_bw(),
  showLegend = TRUE,
  legend.position = "top",
  lineSize = 1,
  xbreaks = "auto",
  outputFile = NULL,
  outputWidth = 10,
  outputHeight = 10,
  ggsaveParams = ufs::opts$get("ggsaveParams"),
  ...
)

Arguments

Details

These functions are explained in Peters (2017).

Value

A Diamond plots: a ggplot2::ggplot() plot meansComparisonDiamondPlot, and a gtable() by duoComparisonDiamondPlot.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

References

Peters, G.-J. Y. (2017). Diamond Plots: a tutorial to introduce a visualisation tool that facilitates interpretation and comparison of multiple sample estimates while respecting their inaccuracy. PsyArXiv. http://doi.org/10.17605/OSF.IO/9W8YV

See Also

diamondPlot(), meansDiamondPlot(), the CIBER() function in the behaviorchange package

Examples

meansComparisonDiamondPlot(mtcars,
                           items=c('disp', 'hp'),
                           compareBy='vs',
                           xbreaks=c(100,200, 300, 400));
meansComparisonDiamondPlot(chickwts,
                           items='weight',
                           compareBy='feed',
                           xbreaks=c(100,200,300,400),
                           showData=FALSE);
duoComparisonDiamondPlot(mtcars,
                         items=c('disp', 'hp'),
                         compareBy='vs',
                         xbreaks=c(100,200, 300, 400));

Description

Escapes any characters that would have special meaning in a reqular expression.

Usage

escapeRegex(string)

Arguments

Details

escapeRegex will escape any characters that would have special meaning in a reqular expression. For any string grep(regexpEscape(string), string) will always be true.

Value

The value of the string with any characters that would have special meaning in a reqular expression escaped.

Note

Note that this function was copied literally from the Hmisc package (to prevent importing the entire package for one line of code).

Author(s)

Charles Dupont
Department of Biostatistics
Vanderbilt University

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

grep, Hmisc, https://hbiostat.org/R/Hmisc/, https://github.com/harrelfe/Hmisc

Examples

string <- "this\\(system) {is} [full]."
escapeRegex(string)

Description

This function can be used to detect exceptionally high or low scores in a vector.

Usage

exceptionalScore(
  x,
  prob = 0.025,
  both = TRUE,
  silent = FALSE,
  quantileCorrection = 1e-04,
  quantileType = 8
)

Arguments

Details

Note that of course, by definition, prob or 2 * prob percent of the values is exceptional, so it is usually not a wise idea to remove scores based on their 'exceptionalness'. Instead, use exceptionalScores(), which calls this function, to see how often participants answered exceptionally, and remove them based on that.

Value

A logical vector, indicating for each value in the supplied vector whether it is exceptional.

Examples

exceptionalScore(
  c(1,1,2,2,2,3,3,3,4,4,4,5,5,5,5,6,6,7,8,20),
  prob=.05
);

Description

A function to detect participants that consistently respond exceptionally.

Usage

exceptionalScores(
  dat,
  items = NULL,
  exception = 0.025,
  totalOnly = TRUE,
  append = TRUE,
  both = TRUE,
  silent = FALSE,
  suffix = "_isExceptional",
  totalVarName = "exceptionalScores"
)

Arguments

Value

Either a vector containing the number of exceptional values, a dataset containing, for each inspected variable, which values are exceptional, or the provided dataset where either the total or the exceptional values for each variable are appended.

Examples

exceptionalScores(mtcars);

Description

This function exports data frames or matrices to HTML, sending output to one or more of the console, viewer, and one or more files.

Usage

exportToHTML(
  input,
  output = ufs::opts$get("tableOutput"),
  tableOutputCSS = ufs::opts$get("tableOutputCSS")
)

Arguments

Value

Invisibly, the (potentially concatenated) input as character vector.

Examples

exportToHTML(mtcars[1:5, 1:5]);

Description

Functions often get passed variables from within dataframes or other lists. However, printing these names with all their dollar signs isn't very userfriendly. This function simply uses a regular expression to extract the actual name.

Usage

extractVarName(x)

Arguments

Value

The actual variables name, with all containing objectes stripped off.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

extractVarName('mtcars$mpg');

Description

Do factor-analysis, logging warnings and errors

Usage

fa_failsafe(
  ...,
  n.repeatOnWarning = 50,
  warningTolerance = 2,
  silentRepeatOnWarning = FALSE,
  showWarnings = TRUE
)

Arguments

Value

A list with the fa object and a warnings and an errors object.

Description

This function contains some code from a function in psych::psych-package that's not exported print.psych.fa.ci but useful nonetheless. It basically takes the outcomes of a factor analysis and extracted the confidence intervals.

Usage

faConfInt(fa)

Arguments

Details

THis function extract confidence interval bounds and combines them with factor loadings using the code from the print.psych.fa.ci in psych::psych-package.

Value

A list of dataframes, one for each extracted factor, with in each dataframe three variables:

Author(s)

William Revelle (extracted by Gjalt-Jorn Peters)

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

## Not run: 
### Not run because it takes too long to run to test it,
### and may produce warnings, both because of the bootstrapping
### required to generate the confidence intervals in fa
faConfInt(psych::fa(Thurstone.33, 2, n.iter=100, n.obs=100));

## End(Not run)

Description

This function uses the diamondPlot() to visualise the results of a factor analyses. Because the factor loadings computed in factor analysis are point estimates, they may vary from sample to sample. The factor loadings for any given sample are usually not relevant; samples are but means to study populations, and so, researchers are usually interested in population values for the factor loadings. However, tables with lots of loadings can quickly become confusing and intimidating. This function aims to facilitate working with and interpreting factor analysis based on confidence intervals by visualising the factor loadings and their confidence intervals.

Usage

factorLoadingDiamondCIplot(
  fa,
  xlab = "Factor Loading",
  colors = viridisPalette(max(2, fa$factors)),
  labels = NULL,
  theme = ggplot2::theme_bw(),
  sortAlphabetically = FALSE,
  ...
)

Arguments

Value

A ggplot2::ggplot() plot with several ggDiamondLayer()s is returned.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

psych::fa()ss, meansDiamondPlot(), meanSDtoDiamondPlot(), diamondPlot(), ggDiamondLayer()

Examples

## Not run: 
### (Not run during testing because it takes too long and
###  may generate warnings because of the bootstrapping of
###  the confidence intervals)

factorLoadingDiamondCIplot(psych::fa(psych::Bechtoldt,
                                     nfactors=2,
                                     n.iter=50,
                                     n.obs=200));

### And using a lower alpha value for the diamonds to
### make them more transparent

factorLoadingDiamondCIplot(psych::fa(psych::Bechtoldt,
                                     nfactors=2,
                                     n.iter=50,
                                     n.obs=200),
                           alpha=.5,
                           size=1);

## End(Not run)

Description

This function uses the diamondPlot() to visualise the results of a factor analyses. Because the factor loadings computed in factor analysis are point estimates, they may vary from sample to sample. The factor loadings for any given sample are usually not relevant; samples are but means to study populations, and so, researchers are usually interested in population values for the factor loadings. However, tables with lots of loadings can quickly become confusing and intimidating. This function aims to facilitate working with and interpreting factor analysis based on confidence intervals by visualising the factor loadings and their confidence intervals.

Usage

factorLoadingHeatmap(
  fa,
  xlab = "Factor Loading",
  colors = viridisPalette(max(2, fa$factors)),
  labels = NULL,
  showLoadings = FALSE,
  heatmap = FALSE,
  theme = ggplot2::theme_minimal(),
  sortAlphabetically = FALSE,
  digits = 2,
  labs = list(title = NULL, x = NULL, y = NULL),
  themeArgs = list(panel.grid = ggplot2::element_blank(), legend.position = "none",
    axis.text.x = ggplot2::element_blank()),
  ...
)

Arguments

Value

A ggplot2::ggplot() plot with several ggDiamondLayer()s is returned.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

psych::fa()ss, meansDiamondPlot(), meanSDtoDiamondPlot(), diamondPlot(), ggDiamondLayer()

Examples

## Not run: 
### (Not run during testing because it takes too long and
###  may generate warnings because of the bootstrapping of
###  the confidence intervals)

factorLoadingHeatmap(psych::fa(psych::Bechtoldt,
                               nfactors=2,
                               n.iter=50,
                               n.obs=200));

### And using a lower alpha value for the diamonds to
### make them more transparent

factorLoadingHeatmap(psych::fa(psych::Bechtoldt,
                               nfactors=2,
                               n.iter=50,
                               n.obs=200),
                     alpha=.5,
                     size=1);

## End(Not run)

Description

This function takes a numeric vector, sorts it, and then finds the shortest interval and returns its length.

Usage

findShortestInterval(x)

Arguments

Value

The length of the shortest interval.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

Examples

findShortestInterval(c(1, 2, 4, 7, 20, 10, 15));

Description

Pretty formatting of confidence intervals

Usage

formatCI(
  ci,
  sep = "; ",
  prefix = "[",
  suffix = "]",
  digits = 2,
  noZero = FALSE
)

Arguments

Value

A character vector of one element.

See Also

noZero(), formatR(), formatPvalue()

Examples

### With leading zero ...
formatCI(c(0.55, 0.021));

### ... and without
formatCI(c(0.55, 0.021), noZero=TRUE);

Description

Pretty formatting of p values

Usage

formatPvalue(values, digits = 3, spaces = TRUE, includeP = TRUE)

Arguments

Value

A formatted P value, roughly according to APA style guidelines. This means that the noZero function is used to remove the zero preceding the decimal point, and p values that would round to zero given the requested number of digits are shown as e.g. p<.001.

See Also

formatCI(), formatR(), noZero()

Examples

formatPvalue(cor.test(mtcars$mpg,
                      mtcars$disp)$p.value);
formatPvalue(cor.test(mtcars$drat,
                      mtcars$qsec)$p.value);

Description

Pretty formatting of correlation coefficients

Usage

formatR(r, digits = 2)

Arguments

Value

The formatted correlation.

See Also

noZero(), formatCI(), formatPvalue()

Examples

formatR(cor(mtcars$mpg, mtcars$disp));

Description

getData() and getDat() provide an easy way to load SPSS datafiles.

Usage

getData(
  filename = NULL,
  file = NULL,
  errorMessage = "[defaultErrorMessage]",
  applyRioLabels = TRUE,
  use.value.labels = FALSE,
  to.data.frame = TRUE,
  stringsAsFactors = FALSE,
  silent = FALSE,
  ...
)

getDat(..., dfName = "dat", backup = TRUE)

Arguments

Value

getData returns the imported dataframe, with the filename from which it was read stored in the 'filename' attribute.

getDat is a simple wrapper for getData() which creates a dataframe in the parent environment, by default with the name 'dat'. Therefore, calling getDat() in the console will allow the user to select a file, and the data from the file will then be read and be available as 'dat'. If an object with dfName (i.e. 'dat' by default) already exists, it will be backed up with a warning. getDat() also invisibly returns the data.frame.

Note

getData() currently can't read from LibreOffice or OpenOffice files. There doesn't seem to be a platform-independent package that allows this. Non-CRAN package ROpenOffice from OmegaHat should be able to do the trick, but fails to install (manual download and installation using https://www.omegahat.org produces "ERROR: dependency 'Rcompression' is not available for package 'ROpenOffice'" - and manual download and installation of RCompression produces "Please define LIB_ZLIB; ERROR: configuration failed for package 'Rcompression'"). If you have any suggestions, please let me know!

Examples

## Not run: 
### Open a dialogue to read an SPSS file
getData();

## End(Not run)

Description

This function provides a simple interface to create a ggplot2::ggplot() bar chart.

Usage

ggBarChart(vector, plotTheme = ggplot2::theme_bw(), ...)

Arguments

Value

A ggplot2::ggplot() plot is returned.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

ggplot2::geom_bar()

Examples

ggBarChart(mtcars$cyl);

Description

This function provides a simple interface to create a ggplot box plot, organising different boxplots by levels of a factor is desired, and showing row numbers of outliers.

Usage

ggBoxplot(
  dat,
  y = NULL,
  x = NULL,
  labelOutliers = TRUE,
  outlierColor = "red",
  theme = ggplot2::theme_bw(),
  ...
)

Arguments

Details

This function is based on JasonAizkalns' answer to a question on Stack Exchange (Cross Validated; see https://stackoverflow.com/questions/33524669/labeling-outliers-of-boxplots-in-r).

Value

A ggplot plot is returned.

Author(s)

Jason Aizkalns; implemented in this package (and tweaked a bit) by Gjalt-Jorn Peters.

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

geom_boxplot

Examples

### A box plot for miles per gallon in the mtcars dataset:
ggBoxplot(mtcars$mpg);

### And separate for each level of 'cyl' (number of cylinder):
ggBoxplot(mtcars, y='mpg', x='cyl');

Description

These are convenience functions to quickly generate plots for multiple variables, with the variables in the y axis.

Usage

ggEasyBar(
  data,
  items = NULL,
  labels = NULL,
  sortByMean = TRUE,
  xlab = NULL,
  ylab = NULL,
  scale_fill_function = NULL,
  fontColor = "white",
  fontSize = 2,
  labelMinPercentage = 1,
  showInLegend = "both",
  legendRows = 2,
  legendValueLabels = NULL,
  biAxisLabels = NULL
)

ggEasyRidge(
  data,
  items = NULL,
  labels = NULL,
  sortByMean = TRUE,
  xlab = NULL,
  ylab = NULL
)

Arguments

Value

A ggplot() plot is returned.

Author(s)

Gjalt-Jorn Peters

Maintainer: Gjalt-Jorn Peters [email protected]

See Also

geom_ridgeline(), geom_bar()

Examples

ggEasyBar(mtcars, c('gear', 'carb'));
ggEasyRidge(mtcars, c('disp', 'hp'));

### When plotting single questions, if you want to show the anchors:
ggEasyBar(mtcars, c('gear'),
          biAxisLabels=list(leftAnchors="Fewer",
                            rightAnchors="More"));

### Or for multiple questions (for e.g. semantic differentials):
ggEasyBar(mtcars, c('gear', 'carb'),
          biAxisLabels=list(leftAnchors=c("Fewer", "Lesser"),
                            rightAnchors=c("More", "Greater")));