Strange results in parallel analysis -- weird output by rstudio but not R-Fiddle

Question

Major UPDATE based on discussion with Aleksandr Blekh's answer (thanks so much!):

This MRE would run with no problem in R-Fiddle

library(psych)

data(bock)

fa.parallel.poly(lsat6)

Output from R-Fiddle (Graph omitted as not relevant with error), no difference in no of factors suggested by the first and second line

 See the graphic output for a description of the results
Parallel analysis suggests that the number of factors =  3  and the number of components =  1 
Call: fa.parallel.poly(x = lsat6)
Parallel analysis suggests that the number of factors =  3  and the number of components =  1 

 Eigen Values of 
  Original factors Simulated data Original components simulated data
1             0.79           0.76                1.62           1.22
2             0.14           0.13                1.00           1.10
3             0.04           0.03                0.88           1.00

Output from R-Studio (Graph omitted as not relevant with error), note there are difference in no of factors suggested by the first (3) and second line (0), and the eigen Values section looks like messed up

 See the graphic output for a description of the results
Parallel analysis suggests that the number of factors =  3  and the number of components =  1 
Call: fa.parallel.poly(x = lsat6)
Parallel analysis suggests that the number of factors =  0  and the number of components =  1 

 Eigen Values of 
  Original factors Simulated data Original components simulated data   NA   NA   NA   NA   NA   NA   NA   NA    NA
1             0.79              1                  10           0.58 0.39 0.39 0.57 0.29 0.19 1.06 0.87 0.26 -2.02
    NA   NA NA NA   NA   NA   NA   NA   NA   NA   NA   NA   NA    NA   NA
1 0.12 1.62  1 10 1.17 0.04 1.15 1.16 0.03 1.11 1.24 0.14 0.58 -1.03 0.01

Here is the fresh results from R-studio together with sessionInfo() and options():

R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"
Copyright (C) 2013 The R Foundation for Statistical Computing
Platform: x86_64-apple-darwin10.8.0 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(psych)
> 
> data(bock)
> 
> fa.parallel.poly(lsat6)


  fa.parallel.poly |....................................................................................................| 100%

 See the graphic output for a description of the results
Parallel analysis suggests that the number of factors =  3  and the number of components =  1 
Call: fa.parallel.poly(x = lsat6)
Parallel analysis suggests that the number of factors =  0  and the number of components =  1 

 Eigen Values of 
  Original factors Simulated data Original components simulated data   NA   NA   NA   NA   NA   NA   NA   NA    NA
1             0.79              1                  10           0.58 0.39 0.39 0.57 0.29 0.19 1.06 0.87 0.26 -2.02
    NA   NA NA NA   NA   NA   NA   NA   NA   NA   NA   NA   NA    NA   NA
1 0.12 1.62  1 10 1.17 0.04 1.15 1.16 0.03 1.11 1.24 0.14 0.58 -1.03 0.01
> sessionInfo()
R version 3.0.2 (2013-09-25)
Platform: x86_64-apple-darwin10.8.0 (64-bit)

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] psych_1.5.1

loaded via a namespace (and not attached):
[1] mnormt_1.5-1   parallel_3.0.2 tools_3.0.2   
> options()
$add.smooth
[1] TRUE

$bitmapType
[1] "quartz"

$browser
function (url) 
{
    .Call("rs_browseURL", url)
}
<environment: 0x7fd21312cdc8>

$browserNLdisabled
[1] FALSE

$buildtools.check
    function (action) 
    {
        if (identical(.Platform$pkgType, "mac.binary.mavericks")) {
        .Call("rs_canBuildCpp")
    }
    else {
        if (!.Call("rs_canBuildCpp")) {
            .rs.installBuildTools(action)
            FALSE
        }
        else {
            TRUE
        }
    }
}
<environment: 0x7fd2131ecbc8>

$buildtools.with
function (code) 
{
    .rs.addRToolsToPath()
    on.exit(.rs.restorePreviousPath(), add = TRUE)
    force(code)
}
<environment: 0x7fd2131ecbc8>

$CBoundsCheck
[1] FALSE

$check.bounds
[1] FALSE

$citation.bibtex.max
[1] 1

$continue
[1] "+ "

$contrasts
        unordered           ordered 
"contr.treatment"      "contr.poly" 

$defaultPackages
[1] "datasets"  "utils"     "grDevices" "graphics"  "stats"     "methods"  

$demo.ask
[1] "default"

$deparse.cutoff
[1] 60

$device
[1] "RStudioGD"

$device.ask.default
[1] FALSE

$digits
[1] 7

$dvipscmd
[1] "dvips"

$echo
[1] TRUE

$editor
[1] "vi"

$encoding
[1] "native.enc"

$error
(function () 
{
    .rs.breakOnError(TRUE)
})()

$example.ask
[1] "default"

$expressions
[1] 5000

$ggvis.renderer
[1] "svg"

$help_type
[1] "html"

$help.search.types
[1] "vignette" "demo"     "help"    

$help.try.all.packages
[1] FALSE

$HTTPUserAgent
[1] "R (3.0.2 x86_64-apple-darwin10.8.0 x86_64 darwin10.8.0)"

$internet.info
[1] 2

$keep.source
[1] TRUE

$keep.source.pkgs
[1] FALSE

$locatorBell
[1] TRUE

$mailer
[1] "mailto"

$max.print
[1] 10000

$menu.graphics
[1] FALSE

$na.action
[1] "na.omit"

$nwarnings
[1] 50

$OutDec
[1] "."

$pager
function (files, header, title, delete.file) 
{
    for (i in 1:length(files)) {
        if ((i > length(header)) || !nzchar(header[[i]])) 
            fileTitle <- title
        else fileTitle <- header[[i]]
        .Call("rs_showFile", fileTitle, files[[i]], delete.file)
    }
}
<environment: 0x7fd2131ecbc8>

$papersize
[1] "a4"

$pdfviewer
[1] "/usr/bin/open"

$pkgType
[1] "mac.binary"

$printcmd
[1] "lpr"

$prompt
[1] "> "

$repos
                     CRAN 
"http://cran.rstudio.com" 
attr(,"RStudio")
[1] TRUE

$restart
function (afterRestartCommand = "") 
{
    afterRestartCommand <- paste(as.character(afterRestartCommand), 
        collapse = "\n")
    .Call("rs_restartR", afterRestartCommand)
}
<environment: 0x7fd2131ecbc8>

$rl_word_breaks
[1] " \t\n\"\\'`><=%;,|&{()}"

$scipen
[1] 0

$show.coef.Pvalues
[1] TRUE

$show.error.messages
[1] TRUE

$show.signif.stars
[1] TRUE

$str
    $str$strict.width
[1] "no"

$str$digits.d
[1] 3

$str$vec.len
[1] 4


$str.dendrogram.last
[1] "`"

$stringsAsFactors
[1] TRUE

$texi2dvi
[1] "/usr/bin/texi2dvi"

$timeout
[1] 60

$ts.eps
[1] 1e-05

$ts.S.compat
[1] FALSE

$unzip
[1] "/usr/bin/unzip"

$useFancyQuotes
[1] TRUE

$verbose
[1] FALSE

$viewer
function (url, height = NULL) 
{
    if (!is.character(url) || (length(url) != 1)) 
        stop("url must be a single element character vector.", 
            call. = FALSE)
    if (identical(height, "maximize")) 
        height <- -1
    if (!is.null(height) && (!is.numeric(height) || (length(height) != 
        1))) 
        stop("height must be a single element numeric vector or 'maximize'.", 
            call. = FALSE)
    invisible(.Call("rs_viewer", url, height))
}
<environment: 0x7fd21312cdc8>

$warn
[1] 0

$warning.length
[1] 1000

$width
[1] 119

Answer 1

Pay attention to the plot's y-axis label. It says: "Eigen values of original and simulated factors and components" (emphasis mine). Parallel analysis (PA) produces separate sets of eigen values for both factors (factor analysis, FA) and components (principal component analysis, PCA). While FA and PCA seem to be similar, those methods are conceptually different. Here's how Professor William Revelle (well-known expert in psychometric theory, maintainer of the Personality Project and the author of psych R package) explains the difference (Revelle, 2015, p. 158):

Although on the surface, the component model and factor model appear to very similar (...), they are logically very different. In the components model, components are linear sums of the variables. In the factor model, on the other hand, factors are latent variables whose weighted sum accounts for the common part of the observed variables. In path analytic terms, for the component model, arrows go from the variables to the components, while in the factor model they go from the factors to the variables.

My statement above about the contents of the plot is intentionally not very accurate for pedagogical reasons (but, I hope still conceptually valid), as, in fact, fa.parallel() and fa.parallel.poly() functions plot "the eigenvalues for a principal components solution as well as the eigen values when the communalities are estimated by a one factor minres solution for a given data set as well as that of n (default value = 20) randomly generated parallel data sets of the same number of variables and subjects" (Revelle, 2015, p. 175-176) [bold emphasis mine].

UPDATE (based on the OP's clarification of the question):

I think that the double line output on number of factors/components is due to a second call to fa.parallel() or fa.parallel.poly() somewhere, because the standard output from fa.parallel.poly() contains only a single such line. See the MRE below.

Code:

library(psych)

data(bock)

fa.info <- fa.parallel.poly(lsat6)
summary(fa.info)

Output:

 See the graphic output for a description of the results
Parallel analysis suggests that the number of factors =  3  and the number of components =  1

References

Revelle, W. (2015). An introduction to psychometric theory with applications in R. [Website] Retrieved from http://www.personality-project.org/r/book

NOTE: Both my citations can be found in Chapter 6 of the referenced book, directly downloadable from http://www.personality-project.org/r/book/Chapter6.pdf.