I was wondering if anyone could help me with some R code for extracting useful information from a nested list of dataframes.
The format of the nested dataframes are:
Level1: Five different ways of modelling data (one factor CFA, two factor CFA, ...)
Level2: The model applied to a set of 11 datasets
Level3: Various statistics that are useful for each dataframe
I have fitted the five models on the same eleven datasets.
An example of how this looks can be produced using the following script (sorry if the code is inefficient!)
### Example data that gets transformed
Testdfs<-list()
Testdfs$Summarydata1<-c(1:42)
Testdfs$Summarydata2<-c(43:84)
Testdfs$Summarydata3<-c(85:126)
### Data labels for my project
Datanames<-c("npar", "fmin", "chisq", "df", "pvalue", "baseline.chisq",
"baseline.df", "baseline.pvalue", "cfi", "tli", "nnfi", "rfi",
"nfi", "pnfi", "ifi", "rni", "logl", "unrestricted.logl", "aic",
"bic", "ntotal", "bic2", "rmsea", "rmsea.ci.lower", "rmsea.ci.upper",
"rmsea.pvalue", "rmr", "rmr_nomean", "srmr", "srmr_bentler",
"srmr_bentler_nomean", "crmr", "crmr_nomean", "srmr_mplus", "srmr_mplus_nomean",
"cn_05", "cn_01", "gfi", "agfi", "pgfi", "mfi", "ecvi")
names(Testdfs$Summarydata1)<-Datanames
names(Testdfs$Summarydata2)<-Datanames
names(Testdfs$Summarydata3)<-Datanames
### Creating the nested list from the three initial sets of example data
CombinedTest<-list()
Testfunction1<-function(X){X+1}
CombinedTest$OneFactor<-lapply(Testdfs,Testfunction1)
Testfunction2<-function(X){X+2}
CombinedTest$TwoFactor<-lapply(Testdfs,Testfunction2)
Testfunction3<-function(X){X+3}
CombinedTest$AlternativeTwoFactor<-lapply(Testdfs,Testfunction3)
Testfunction4<-function(X){X+4}
CombinedTest$TwoFactorOrthogonal<-lapply(Testdfs,Testfunction4)
Testfunction5<-function(X){X+5}
CombinedTest$SecondOrder<-lapply(Testdfs,Testfunction5)
remove(Testdfs,Datanames)
##Nested list contains 5 different "models" applied to the 3 dataframes (I have 11 in my actual database)
This produces a series of fit statistics for that dataset in that model i.e.
CombinedTest[["OneFactor"]][["Summarydata1"]]
npar fmin chisq df pvalue baseline.chisq baseline.df
2 3 4 5 6 7 8
baseline.pvalue cfi tli nnfi rfi nfi pnfi
9 10 11 12 13 14 15
ifi rni logl unrestricted.logl aic bic ntotal
16 17 18 19 20 21 22
bic2 rmsea rmsea.ci.lower rmsea.ci.upper rmsea.pvalue rmr rmr_nomean
23 24 25 26 27 28 29
srmr srmr_bentler srmr_bentler_nomean crmr crmr_nomean srmr_mplus srmr_mplus_nomean
30 31 32 33 34 35 36
cn_05 cn_01 gfi agfi pgfi mfi ecvi
37 38 39 40 41 42 43
What I would like to do, is create a table/dataframe for each of the initial dataframes (Summarydata1, summarydata2, summarydata3) that shows the fit statistics down the rows, and the five models that the dataframe has been applied to as columns. Please find an example below (without made up data).
Does anyone know how I could accomplish this? I'm new to R and know that using lists is important, but I find the logic around manipulating them difficult.
Thanks.
