Query about most efficient solution

[Bruce Miller]

Hi all,

I have working code that provides great data summaries from raw input data.  I save all data in R format *ROBJ so I can rerun things if need be.
Data are bat echolocation parameters used for identification of bats by their vocal signature.

Summary data is written so I can open in Excel and looks like this for each species.

Parameters	N	Min	Max	Mean	St.Dev	10%	25%	75%	90%
Dur	1972	0.24	5.05	1.76	0.65	0.93	1.30	2.20	2.57
Fmin	1972	63.24	69.87	65.24	1.28	63.75	64.26	66.12	67.20
Fmax	1972	68.38	89.89	76.27	5.37	70.48	72.07	80.00	84.66
Fmean	1972	64.52	73.62	67.68	1.74	65.48	66.33	68.90	70.07
Fk	1972	64.00	73.73	68.47	2.08	65.84	66.95	69.87	71.43
Fc	1972	63.24	69.87	65.63	1.43	63.75	64.52	66.67	67.80
Sc	1972	-493.66	774.11	78.53	83.90	11.53	28.32	103.02	166.00
Pmc	1972	0.40	41.60	16.24	8.09	8.10	9.80	21.60	28.70

The lines of code used once data has been read, are:

BatStats<- Deducer::descriptive.table (vars = d (Dur,Fmin,Fmax,Fmean,Fk,Fc,Sc,Pmc),data= BatStats,func.names =c("Valid N","Minimum","Maximum","Mean","St. Deviation","25th Percentile","75th Percentile"),func.additional= list(p10=function(x) quantile(x, c(0.10), na.rm=TRUE),p90=function(x) quantile(x, c(0.90), na.rm=TRUE)))

 

## Write the results Change output name and directory for stats.

write.csv(BatStats,file="C:\\=Bat data working\\Acoustic Parameters\\completed\\Myorip_Stats.csv")

I also have data sets that are size parameters that include only 2 size measurements of the same species.
I will try the merge from Data menu to see how well this works.

What I need to do is use plot Builder to plot the echolocation call parameters (each min, max and mean) by size for each species as dots with the 10 and 90th 
percentile boundaries as lines for upper and lower bounds.

My question is should I read in all of the summary tables (adding a column for Species ID) and merge with the size data or generate what is needed from the 
raw data tables anew?

Trying melt-cast of the summary tables did not seem to get me very far.

Thanks for any suggestions.

Bruce