descriptives {tidySEM} | R Documentation |
Describe a dataset
Description
Provide descriptive statistics for a dataset.
Usage
descriptives(x, ...)
Arguments
x |
An object for which a method exists. |
... |
Additional arguments. |
Value
A data.frame
with descriptive statistics for x
.
Its elements are:
name | Character | Variable name |
type | character | Data type in R , as obtained by class(x)[1] |
n | Integer | Number of valid observations |
missing | Numeric | Proportion missing |
unique | Integer | Number of unique values |
mean | numeric | Mean value of non-missing entries, only defined for variables that can be coerced to numeric |
median | numeric | Median value of non-missing entries, only defined for numeric variables |
mode | Integer | For numeric variables: The mode value. For factors: The frequency of the mode value |
mode_value | Character | For factors: value of the mode |
sd | numeric | Standard deviation of non-missing entries, only defined for variables that can be coerced to numeric |
v | numeric | Variability coefficient V for factor variables (Agresti, 1990). V is the probability that two independent observations fall in different categories |
min | numeric | Minimum value for numeric variables |
max | numeric | Maximum value for numeric variables |
range | numeric | Range (distance between min and max) for numeric variables |
skew | numeric | Skewness. The normalized third central moment of a numeric variable, which reflects its skewness. A symmetric distribution has a skewness of zero |
skew_2se | numeric | Skewness, divided by two times its standard error. Values greater than one can be considered "significant" according to a Z-test with significance level of .05 |
kurt | numeric | Kurtosis. The normalized fourth central moment of a numeric variable, which reflects its peakedness. A heavy-tailed distribution has high kurtosis, a light-tailed distribution has low kurtosis (sometimes called platykurtic). |
kurt_2se | numeric | Kurtosis, divided by two times its standard error. Values greater than one can be considered "significant" according to a Z-test with significance level of .05 |
References
Agresti, A. (2012). Categorical data analysis (Vol. 792). John Wiley & Sons.
Examples
descriptives(iris)
[Package tidySEM version 0.2.7 Index]