bivan: Bivariate analysis measures computation
In Rsocialdata0: Handling survey data in R. (beta version, not for use)

Description Usage Arguments Details Value Author(s) References See Also Examples

bivan computes some main bivariate analysis measures.

bivan(
  formula,
  data,
  chi2 = T,
  phi = F,
  tschuprow = F,
  cramer.v = T,
  pearson.contingency = F,
  likelihood.ratio = F,
  gk.lambda = F,
  gk.tau = F,
  gk.tau.sqrt = T,
  theil.u = F,
  theil.u.sqrt = F,
  kendall.tau.a = F,
  kendall.tau.b = F,
  stuart.tau.c = F,
  gk.gamma = F,
  somers.d = T,
  wilson.e = F,
  calc.spearman.rho = F,
  std.res = T,
  quiet = F
  )

`formula`	a `formula` specifying the target and the predictors. Only first order formula can be supply.
`data`	a `Dataset` object.
`chi2`	if `TRUE` the Pearson's Khi-squared measure is computed.
`phi`	if `TRUE` the Phi measure is computed.
`tschuprow`	if `TRUE` the Tschuprow's coefficient is computed.
`cramer.v`	if `TRUE` the Cramer's V is computed.
`pearson.contingency`	if `TRUE` the Pearson contingency coefficient is computed.
`likelihood.ratio`	if `TRUE` the Khi-squared likelihood ratio is computed.
`gk.lambda`	if `TRUE` the Goodman and Kruskal's lambda is computed.
`gk.tau`	if `TRUE` the Goodman and Kruskal's tau is computed.
`gk.tau.sqrt`	if `TRUE` the square root of the Goodman and Kruskal's tau is computed.
`theil.u`	if `TRUE` the Theil's u is computed.
`theil.u.sqrt`	if `TRUE` the square root Theil's u is computed.
`kendall.tau.a`	if `TRUE` the Kendall's tau a is computed.
`kendall.tau.b`	if `TRUE` the Kendall's tau b is computed.
`stuart.tau.c`	if `TRUE` the Stuart's tau c is computed.
`gk.gamma`	if `TRUE` the Goodman and Kruskal's gamma is computed.
`somers.d`	if `TRUE` the Somers' D is computed.
`wilson.e`	if `TRUE` the Wilson's e is computed.
`calc.spearman.rho`	if `TRUE` the Spearman's rho is computed. (not available yet)
`std.res`	if `TRUE` the Pearson's Chi-squared standardize residuals are computed.
`quiet`	if `TRUE` console messages are turned off.

...

A Statdf object containing all the statistics asked and their approximate p-values (based on asymptotic variance approximation).

Emmanuel Rousseaux, Gilbert Ritschard.

Pearson, K. (1904). Mathematical contributions to the theory of evolution, XIII. On the theory of contingency and its relation to association and normal correlation.Draper's Co. Research Memoirs, Biometric Series, 1 (Reprinted in 1948 in: E. S. Pearson (ed.), Karl Pearson's Early Papers, Cambridge University Press, Cambridge.)

Tschuprow, A. A. (1918). On the mathematical expectation of moments of frequency distribution, Biometrika 12: 140-169.

Kendall, M. G. (1938). A new measure of rank correlation, Biometrika 30: 81-93.

Kendall, M. G. (1945). The treatment of ties in rank problems, Biometrika 33: 239-251.

Stuart, A. (1953). The estimation and comparison of strenghts of association in contingency tables, Biometrika 40: 105-110.

Goodman, L. A. & Kruskal, W. H. (1954). Measures of association for cross classifications, J. Amer. Statist. Ass. 37:54-115.

Somers, R. H. (1962). A new asymmetric measure for ordinal variables, Amer. Sociolog. Rev. 27: 799-811.

Theil, H. (1970). On the estimation of relationships involving qualitative variables, Amer. J. Sociol. 76: 103-154.

Cramer, H. (1971). Mathematical Methods of Statistics, Princeton University Press, Princeton.

Wilson, T. P. (1974). Measures of association for bivariate ordinal hypotheses, in: H. M. Blalock Jr. (ed.), Measurement in the Social Sciences, Aldine-Atherton, Chicago, pp. 327-342.

Ritschard, G. and al. (1996). Path analysis with partial association measures. International Journal of Methodology, vol. 30, number 1.

Statdf

# biv1 <- bivan(
#  formula = target ~ predictor1 + predictor2,
#  data = myData
# )
## print biv1
# biv1
## get global measures
# g <- globals(biv1)
## convert them to data.frame
# g.df <- sdf(g)