compute_name_index: Compute multiple features of surnames and given names.

In psychbruce/ChineseNames: Chinese Name Database 1930-2008

Compute multiple features of surnames and given names.

Description

Compute all available name features (indices) based on familyname and givenname. You can either input a data frame with a variable of Chinese full names (and a variable of birth years, if necessary) or just input a vector of full names (and a vector of birth years, if necessary).

• Usage 1: Input a single value or a vector of name [and birth, if necessary].

• Usage 2: Input a data frame of data and the variable name of var.fullname (or var.surname and/or var.givenname) [and var.birthyear, if necessary].

Caution. Name-character uniqueness (NU) for birth year >= 2010 is estimated by forecasting and thereby may not be accurate.

Usage

compute_name_index(
  data = NULL,
  var.fullname = NULL,
  var.surname = NULL,
  var.givenname = NULL,
  var.birthyear = NULL,
  name = NA,
  birth = NA,
  index = c("NLen", "SNU", "SNI", "NU", "CCU", "NG", "NV", "NW", "NC"),
  NU.approx = TRUE,
  digits = 4,
  return.namechar = TRUE,
  return.all = FALSE
)

Arguments

data	Data frame.
var.fullname	Variable name of Chinese full names (e.g., "name").
var.surname	Variable name of Chinese surnames (e.g., "surname").
var.givenname	Variable name of Chinese given names (e.g., "givenname").
var.birthyear	Variable name of birth year (e.g., "birth").
name	If no data, you can just input a vector of full name(s).
birth	If no data, you can just input a vector of birth year(s).
index	Which indices to compute? By default, it computes all available name indices: NLen: full-name length (2~4). SNU: surname uniqueness (1~6). SNI: surname initial (1~26). NU: name-character uniqueness (1~6). CCU: character-corpus uniqueness (1~6). NG: name gender (-1~1). NV: name valence (1~5). NW: name warmth (1~5). NC: name competence (1~5). For details, see https://psychbruce.github.io/ChineseNames/
NU.approx	Whether to approximately compute name-character uniqueness (NU) using the nearest two birth cohorts with relative weights (which would be more precise than just using a single birth cohort). Default is TRUE.
digits	Number of decimal places. Default is 4.
return.namechar	Whether to return separate name characters. Default is TRUE.
return.all	Whether to return all temporary variables in the computation of the final variables. Default is FALSE.

Value

A new data frame (of class data.table) with name indices appended. Full names are split into name0 (surnames, with compound surnames automatically detected), name1, name2, and name3 (given-name characters).

Note

For details and examples, see https://psychbruce.github.io/ChineseNames/

Examples

## Prepare ##
sn = familyname$surname[1:12]
gn = c(top100name.year$name.all.1960[1:6],
       top100name.year$name.all.2000[1:6],
       top100name.year$name.all.1960[95:100],
       top100name.year$name.all.2000[95:100])
demodata = data.frame(name=paste0(sn, gn),
                      birth=c(1960:1965, 2000:2005,
                              1960:1965, 2000:2005))
demodata

## Compute ##
newdata = compute_name_index(demodata,
                             var.fullname="name",
                             var.birthyear="birth")
newdata

psychbruce/ChineseNames documentation built on Oct. 1, 2023, 3:04 p.m.