reciptest1: Social reciprocity measures and tests at different levels of...
In DLEIVA/DyaDA: Dyadic Data Analysis

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

Statistical significance for social reciprocity at different levels of analysis.

1	reciptest(X, pi, rep=9999, names=NULL, label=FALSE)

`X`	Original sociomatrix. Matrix X must be square.
`pi`	Matrix of probabilities Pij that specifies the null hypothesis regarding social reciprocity. Pi must be a square matrix
`rep`	Number of simulations for carrying out the randomization test. Number of simulations must be between 1 and 1000000.
`names`	(optional) Character array with the names of individuals. IF `label` TRUE. then this array should be specified.
`label`	(optional) Logical flag. TRUE. if `names` of individuals will be provided.

reciptest estimates statistical significance for several measures of social reciprocity at different levels of analysis: overall measures of social reciprocity as phi, see getphi; dyadic measures as ratiolambda, see getratiolambda; and individual measures as nuj, see getrationu. This procedure simulates a number of sociomatrices under specified null hypothesis regarding social reciprocity (e.g. complete reciprocation) by means of callings to a C routine included in the package, then computes the reciprocity measures at different levels. After rep simulations the sampling distributions for the statistics are estimated. Then statistical significance is computed as follows: p=NS+1/NOS+1 where NS is the number of times that simulated values is as great as or greater than the empirical values and NOS represents the number of simulated values when p-right value is computed. When estimating p-left value NS is the number of times that simulated values is as great as or lower than the empirical values.

`phi`	Overall measure of skew-symmetry in social interactions.
`psi`	Overall measure of symmetry in social interactions.
`dc`	Directional consistency in social interactions.
`delta`	Ratio skew-symmetry/symmetry.
`epsilon`	Generalized reciprocity index in social interactions.
`kappa`	Dyadic reciprocity index in social interactions.
`nuj`	Individuals' contributions to skew-symmetry.
`lambdaj`	Individuals' contributions to symmetry.
`rationu`	Dyadic contributions to skew-symmetry.
`ratiolambda`	Dyadic contributions to symmetry.
`omega`	Dyadic balanced reciprocity matrix.

David Leiva dleivaur@ub.edu.

Leiva, D., Solanas, A., & Salafranca, Ll. (2008). Testing reciprocity in social interactions: A comparison between the directional consistency and skew-symmetry statistics. Behavior Research Methods, 40, 626-634.

Solanas, A., Salafranca, Ll., Riba, C., Sierra, V., & Leiva, D. (2006). Quantifying social asymmetric structures. Behavior Research Methods, 38, 390-399.

van Hooff, J. A. R. A. M., & Wensing, J. A. B. (1987). Dominance and its behavioral measures in a captive wolf pack. In H. W. Frank (Ed.), Man and wolf (pp. 219-252). Dordrecht: Junk Publishers.

getphi, getdc, getnuj, getrationu, getomega

  X=matrix(c(0, 1, 2, 1, 0, 1, 3, 1, 0),nrow=3,ncol=3)
  pi=matrix(c(0, 0.5, 0.5, 0.5, 0, 0.5, 0.5, 0.5, 0),nrow=3,ncol=3)
  rep=99999
  names=c("Ind1","Ind2","Ind3")