amended summary stats to include all data before keeping only UK

MTUS-W6-adult-survey-data-processing.Rmd

@@ -105,20 +105,27 @@ Things that are NOT fixed here:
 Loading `r sfile`.
 
 ```{r loadSurveyData}
-MTUSW6UKsurvey_DT <- as.data.table(read.spss(sfile))[countrya == "United Kingdom"] # load UK only from SPSS
+MTUSW6survey_DT <- as.data.table(read.spss(sfile)) # load from SPSS
 
-setkey(MTUSW6UKsurvey_DT, countrya, survey, swave, msamp, hldid, persid, id)
+setkey(MTUSW6survey_DT, countrya, survey, swave, msamp, hldid, persid, id)
 ```
 
-We have loaded `r length(MTUSW6UKsurvey_DT$id)` rows of data distributed as follows.
+We have loaded `r format(nrow(MTUSW6survey_DT), big.mark=",",scientific=FALSE)` rows of data for `r format(uniqueN(MTUSW6survey_DT$countrya), big.mark=",",scientific=FALSE)` countries.
 
 ```{r basicStats}
-kable(caption = "Number of cases per year",
-      table(MTUSW6UKsurvey_DT$survey, MTUSW6UKsurvey_DT$year, useNA = "always")
+kable(caption = "Number of diaries per year",
+      table(droplevels(MTUSW6survey_DT$countrya), # removes unused countries
+            MTUSW6survey_DT$survey)
       )
 ```
 
-# Process survey data
+```{r keepUKOnly}
+MTUSW6UKsurvey_DT <- subset(MTUSW6survey_DT, countrya == "United Kingdom")
+```
+
+We now delete the non-UK data leaving us with `r format(nrow(MTUSW6UKsurvey_DT), big.mark=",",scientific=FALSE)` rows of survey data.
+
+# Process UK survey data
 
 ```{r processSurveyData}
 print("-> Create uniq id for diaries (for matching) and persons")