{smcl}
{June 2009}{...}
{hline}
help for {hi:sdyads}
{hline}

{title:Restructures dyadic data for use with scompute}

{p 8 15}{cmd:sdyads} CntrA CntrB year CntrA_Dyadic_Variable CntrB_Dyadic_Variable, newfile(filename)


{title:Description}

{p 0 4 4}{cmd:sdyads} Takes any dyadic data and restructures it so that
it is compatible with {cmd:scompute}. It restructures the data so as to 
 contain {hi: i vs i dyad} and {hi:j vs i dyad}; as well as, the original
 {hi: i vs j dyad}.


{title:Special Note}
{p 0 4 4}{cmd: sdyads} requires that the original data be non-directed dyads and 
that there be only one pair of non identifying variables that coincide 
with the members of a dyad. 
For example, if the data that is to be restructured
is non-directed dyadic data on trade then the original data must contain 
country A, country B, year, trade variable for Country A and trade variable for Country B. 
Nothing else!
Note that the trade variable could be exports, imports, total trade, etc.. So long as 
there is one for country A and one for country B. The program will not work if there 
are more than one pair of such variables. 
If the user has data that is non-directed dyads with
other pairs such as polity scores, capability scores, GDP scores, IGO membership, etc..
for each respective member of the dyad, the user must reuse {cmd: sdyads} for each
pair individually and then merge into a single file and then apply {cmd: scompute}


{p 0 4 4}{hi: i vs i dyads} the program {cmd:sdyads} does not fill in the values for 
the i vs i cell. The value for the i vs i cell has to be determined by the user. 
For theoretical suggestions on what possible values to use for the i vs i entry  
for different variables the user is directed to the articles by Signorino and Ritter (1999)
and Sweeney and Keshk (2005). 

{title:Options}

{p} {cmd: newfile} (filename) specifies the name of the file for the newly restructured data. 
The original data file is left in its original form.{p_end}

{title:Examples}

{p 8 12}{inp:. sdyads  ccode1 ccode2 year trade1 trade2, filename(TradeData)}{p_end}
{p 8 12}{inp:. sdyads  countryA countryB year gdpA gdpB, filename(GdpData)}{p_end}

{title:References}

Signorino, Curtis S., and Jeffrey M. Ritter. 1999. Tau-b or not tau-b: Measuring the similarity
of foreign policy positions. {it: International Studies Quarterly} 43:115-44.

Sweeney, Kevin and Omar M.G. Keshk. 2005. The Similarity of States: Using S to Compute Dyadic
Interest Similarity. {it: Conflict Management and Peace Science} 
22:165–187.


{title:Authors}

Omar M.G. Keshk 
The Ohio State University
keshk.1@osu.edu
June 2009