使用Lavaan的多个数据帧的R循环

Question

I'm currently working on data with 8 waves, each wave is for now stored in it's own dataframe. I've done most of the data cleaning with a lot of repetition, since I couldn't figure out how to make R run the same script over different dataframes. Now I want to do a CFA using Lavaan, and I notice the script is getting very messy, with CFA's being done for multiple variables over all 8 waves. I would love to find a way to simplify my script, so that it's clear for both me and others. Underneath you can find a short example of what the 'long' version of the code would look like, but I'm hoping to get some help in shortening it!

I've tried using a for-loop, but I could not get it to work.

If you have any tips, please let me know so I can stop copy-pasting the same code and replacing the number in the df!

# Example with 2 waves in 2 df
model_ADI_aff <- "aff =~ bds89 + bds39 + bds50 + bds29 + bds84 + bds49 + bds70 + bds88 + bds11 + bds28
                "
fit_ADI_aff_1 <- cfa(model_ADI_aff,
                 data = bds_1,
                 missing = "fiml",
                 estimator = "MLR",
                 se = "robust.huber.white",
                 test = "yuan.bentler")
summary(fit_ADI_aff_1, standardized = TRUE, fit.measures = TRUE)
modindices(fit_ADI_aff_1, sort.=TRUE, minimum.value=3)

fit_ADI_aff_2 <- cfa(model_ADI_aff,
                     data = bds_2,
                     missing = "fiml",
                     estimator = "MLR",
                     se = "robust.huber.white",
                     test = "yuan.bentler")
summary(fit_ADI_aff_2, standardized = TRUE, fit.measures = TRUE)
modindices(fit_ADI_aff_2, sort.=TRUE, minimum.value=3)

...

Answer 1

One way to make it simpler would be putting all your bds_x data.frames into a list and then running your model in all of them with lapply()

model_ADI_aff <- "aff =~ bds89 + bds39 + bds50 + bds29 + bds84 + bds49 + bds70 + bds88 + bds11 + bds28"

fun = function(bds){
 fit = cfa(model_ADI_aff,
           data = bds,
           missing = "fiml",
           estimator = "MLR",
           se = "robust.huber.white",
           test = "yuan.bentler")
 summ = summary(fit)
 mo = modindices(fit_ADI_aff_2, sort.=TRUE, minimum.value=3)

 list(fit = fit,summary = summ, modindices = mo)
}

df_list = "list containing all bds"
results = lapply(df_list,fun)

results will be a list of lists, each containing the fit, summary and modindices for each data.frame

Answer 2

@Fino

I made a few adjustments, and it seems to have done the trick, thank you so much!

fun = function(bds){
  fit = cfa(model_ADI_aff,
            data = bds,
            missing = "fiml",
            estimator = "MLR",
            se = "robust.huber.white",
            test = "yuan.bentler")
  summ = summary(fit, standardized = TRUE, fit.measures = TRUE)
  mo = modindices(fit, sort.=TRUE, minimum.value=3)

  list(fit = fit,summary = summ, modindices = mo)
}

df_list <- list (bds_1, bds_2, bds_3, bds_4, bds_5, bds_6, bds_7, bds_8)
results = lapply(df_list,fun)