R: Lack of cronbach's alpha information

Question

I am trying to calculate Cronbach alpha for an index of 5 different survey question. They are scaled identically, 0-10. I have created a matrix with the variable

I have 47000 observations, so the problem is not due to a lack of observations. When I use the psych package, I get less information than I would expect.

trust_alpha <- alpha(x = trust_ma)
trust_alpha

the output is:

Number of categories should be increased in order to count frequencies. 
Reliability analysis   
Call: alpha(x = trust_ma)

 
 lower alpha upper     95% confidence boundaries
0.89 0.9 0.9 

 Reliability if an item is dropped:

 Item statistics

I can't figure out what "Number of categories should be increased in order to count frequencies." means, and why I don't get the reliability if an item is dropped or item statistics. Is there a way, I can get this information?

I really appreciate any help you can provide.

EDIT

overview over the data with dput(head(trust))

structure(list(idno = c(27L, 137L, 194L, 208L, 220L, 254L), trstplt = c(5L, 
3L, 5L, 3L, 7L, 5L), trstprt = c(5L, 4L, 5L, 3L, 7L, 5L), trstprl = c(5L, 
7L, 6L, 0L, 7L, 6L), trstplc = c(10L, 8L, 8L, 8L, 8L, 7L), trstlgl = c(10L, 
8L, 8L, 5L, 8L, 5L), trust_norris = c(7, 6, 6.4, 3.8, 7.4, 5.6
), trust_norris_na = c(0L, 0L, 0L, 0L, 0L, 0L)), row.names = c(NA, 
-6L), groups = structure(list(.rows = structure(list(1L, 2L, 
    3L, 4L, 5L, 6L), ptype = integer(0), class = c("vctrs_list_of", 
"vctrs_vctr", "list"))), row.names = c(NA, -6L), class = c("tbl_df", 
"tbl", "data.frame")), class = c("rowwise_df", "tbl_df", "tbl", 
"data.frame"))

with head(trust)

 head(trust)
# A tibble: 6 x 8
# Rowwise: 
   idno trstplt trstprt trstprl trstplc trstlgl trust_norris trust_norris_na
  <int>   <int>   <int>   <int>   <int>   <int>        <dbl>           <int>
1    27       5       5       5      10      10          7                 0
2   137       3       4       7       8       8          6                 0
3   194       5       5       6       8       8          6.4               0
4   208       3       3       0       8       5          3.8               0
5   220       7       7       7       8       8          7.4               0
6   254       5       5       6       7       5          5.6               0

idno is my ID variable, as it subset of a larger dataset, trust_norris is a reflexis index, that is the mean of trstplt, trstprt, trstprl, trstplc and trstlgl. trust_norris_na is a variable that describes how many NA there are in trstplt, trstprt, trstprl, trstplc and trstlgl pr. row.

I created a matrix of the relevant variables

trust_ma <- data.matrix(subset(trust, select = c(trstplt, trstprt, trstprl, trstplc, trstlgl)))

 

Answer 1

I found out that the reason for the lack of output was due to I was using Rstudio and I wrote the code in Rmarkdown. When written in the console I got the full results.

Answer 2

This is absolutely idiotic behavior. This was working 6 Months ago. I found out now that putting the assignment operator at the end of the command, will still produce the error in RMarkdown, but now at least it will also compute the object. I don't know what is going on, but it broke an entire paper i was working on.

ICC.o <- df %>% dplyr::select(O1:O6) %>% psych::alpha() #does not work
df %>% dplyr::select(O1:O6) %>% psych::alpha() -> ICC.o #works with warning