0

I have the following problem with the describe function of psych: I want to describe selected variables of a dataframe and then remove a few of the results with subset and select. That seemingly only works with a dataframe but I get a describe class. For me it seems that sometimes it works and sometimes it doesn't which I assume is actually impossible. However, indeed it worked a few times and I already could save the output, nicely arranged, exactly as I wanted it to look like. But now again it returns the error that class describe cannot be converted into dataframe. I see the problem probably is that I get a list of lists (at least the environment says so). Since I'm a complete newbie in programming I just can't solve this problem, even after searching how to convert this class, I just don't get it.

Descriptives = describe(NumericData[5:44], na.rm = TRUE, interp = FALSE, 
                        skew = TRUE, ranges = TRUE, trim = .1, type = 3, 
                        check = TRUE, fast = NULL, quant = c(.25, .50, .75), 
                        IQR = FALSE)
Descriptives = as.data.frame(Descriptives) 
Descriptives = subset(Descriptives, select = -c(vars, median, trimmed, mad, range))
colnames(Descriptives) = c("N", "MEAN", "SD", "MIN", "MAX", "SKEW", "KURTOSIS", "SE", "Q1", "MEDIAN", "Q3")
Descriptives = round(Descriptives, digits = 4)
options(max.print = 1000)
print(as.data.frame(Descriptives))
write.table(Descriptives, file = "Descriptives.txt", sep = ",")
camille
  • 16,432
  • 18
  • 38
  • 60
Tamara Traub
  • 11
  • 2
  • 1
    Without a [reproducible](https://stackoverflow.com/q/5963269/5325862) sample of your data, we can't run your code or see any output. Where exactly does the error occur? Run the code line by line to start debugging it. Also, if the issue is with converting to a data frame, the last 4 lines are most likely irrelevant to the question—you could remove them from the question & focus on identifying the problem – camille Jan 13 '22 at 00:12
  • On second look, `psych::describe` and `psych::describeBy` both take a data frame or matrix and return a data frame. Were you trying to convert to data frame to drop the `describe` class? If I swap out your data for one of the example datasets from the package docs I don't get any error – camille Jan 13 '22 at 00:22

2 Answers2

0

Okay, so the error now occurs in every line (first block = line 1), the describe function actually works fine, but then I can't select the subset of resulting variables (line 3), here it says argument subset is missing and or name the columns left (line 4) which tell me tried to set columnnames for an object with less than two dimensions. Additionally round (line 5) returns an error that says can't use numerical argument for non mathematical function. The error "cannot coerce class ‘"describe"’ to a data.frame" which occurred previously in line 2 appears now when I want to print it. It worked one time and lately at least without choosing subset of resulting variables, but now nothing works and I don't understand why..The code remains the same.

2 rows of the Dataset I use:

structure(list(Age = c(24, 23, 44, 48, 35, 56, 64, 29, 20, 62, 
35, 31, 32, 60, 57, 66, 46, 18, 52, 63, 64, 35, 54, 58, 61, 52, 
52, 33, 49, 28, 22, 27, 40, 53, 18, 19, 43, 44, 26, 28, 38, 18, 
50, 45, 23, 38, 50, 36, 72, 62, 33, 28, 29, 42, 48, 42, 29, 70, 
27, 33, 22, 62, 67, 20, 32, 22, 32, 67, 29, 55, 49, 19, 52, 20, 
30, 24, 18, 24, 23, 22, 19, 20, 29, 22, 20, 19, 21, 18, 22, 22, 
18, 24, 22, 24, 19, 25, 24, 25, 20, 21, 23, 39, 60, 53, 47, 48, 
40, 29, 24, 27, 21, 21, 27, 22, 20, 23, 36, 22, 25, 27, 66, 54, 
54, 64, 49, 40), FTND = c(5, 7, 0, 6, 0, 6, 0, NA, 3, 4, 0, 7, 
NA, 0, 4, 3, 4, 1, 0, 6, 0, 5, 0, NA, NA, 3, 0, 2, NA, 0, 0, 
0, NA, NA, NA, NA, NA, 4, 0, 10, NA, NA, 8, NA, 3, 7, 0, 0, 5, 
2, 0, 6, 7, 0, 4, 2, 0, NA, 0, 0, 0, 0, 0, 0, 4, 0, 0, NA, 3, 
NA, NA, NA, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, NA, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)), row.names = c(NA, 
-126L), class = "data.frame")

The dataset I get with describe is a list of lists, which does not contain the resulting variables, but all variables of my original dataset. Also by using (dput(Descriptives[5:6]) it should have printed variables age and FTND instead of EmQ (actually variable/row 9):

structure(list(EmQ = structure(list(descript = "EmQ", units = NULL, 
    format = NULL, counts = c(n = "97", missing = "29", distinct = "39", 
    Info = "0.998", Mean = "44.04", Gmd = "12.33", `.05` = "23.6", 
    `.10` = "27.6", `.25` = "38.0", `.50` = "45.0", `.75` = "50.0", 
    `.90` = "58.4", `.95` = "60.2"), values = list(value = c(19, 
    21, 22, 24, 25, 27, 28, 30, 32, 33, 34, 36, 37, 38, 39, 40, 
    41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 
    56, 57, 58, 59, 60, 61, 64, 66), frequency = structure(c(2, 
    2, 1, 2, 2, 1, 1, 2, 2, 2, 3, 1, 3, 1, 2, 2, 3, 4, 3, 4, 
    8, 7, 3, 6, 2, 4, 3, 1, 1, 1, 1, 4, 2, 1, 2, 3, 2, 2, 1), .Dim = 39L)), 
    extremes = c(L1 = 19, L2 = 21, L3 = 22, L4 = 24, L5 = 25, 
    H5 = 59, H4 = 60, H3 = 61, H2 = 64, H1 = 66)), class = "describe"), 
    EmQ10 = structure(list(descript = "EmQ10", units = NULL, 
        format = NULL, counts = c(n = "108", missing = "18", 
        distinct = "19", Info = "0.993", Mean = "10.16", Gmd = "4.346", 
        `.05` = "4.0", `.10` = "5.7", `.25` = "7.0", `.50` = "10.0", 
        `.75` = "13.0", `.90` = "15.0", `.95` = "16.0"), values = list(
            value = c(2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 
            14, 15, 16, 17, 18, 19, 20), frequency = structure(c(1, 
            3, 4, 3, 5, 13, 10, 11, 10, 11, 5, 11, 5, 6, 5, 2, 
            1, 1, 1), .Dim = 19L)), extremes = c(L1 = 2, L2 = 3, 
        L3 = 4, L4 = 5, L5 = 6, H5 = 16, H4 = 17, H3 = 18, H2 = 19, 
        H1 = 20)), class = "describe")), descript = "NumericData[5:44]", dimensions = c(126L, 
2L), class = "describe")

The data I get with describeBy. Also a list, with just 2 groups in list 1(?) namely controls and patients, also contains needed result-variables like trimmed, median as attributes I guess):

structure(list(NULL, NULL), .Dim = 2L, .Dimnames = list(Group = c(NA_character_, 
NA_character_)))

I'm sorry for the very long post, I don't know how to put it better..

Tamara Traub
  • 11
  • 2
0

This is what I once got and want to get again:


      N      MEAN    SD     MIN MAX  SKEW   KURTOSIS  SE     Q1    MEDIAN   Q3
Age  126    36.254  15.6578 18  72  0.6067  -0.9925  1.3949 22.25   30.5    49
FTND 107    1.2617  2.3121   0  10  1.7475  2.0378   0.2235   0       0     1.5
Tamara Traub
  • 11
  • 2