Best How To :

If you want separate bars for each gear, then you should add fill=gear to the aes in geom_bar:

ggplot(cdata[cdata$year==2012 & cdata$sitecode==678490,],
       aes(x = factor(month), y = totalvalue, fill=gear)) + 
  geom_bar(stat = "identity", position="dodge") +
  labs(x = "Month", y = "Total value")

this gives:

When you want to make a plot per site, per year, showing the 'total value' for each gear, for each month, as a bar, you can use facet_grid. For example:

ggplot(cdata, aes(x = factor(month), y = totalvalue, fill=gear)) + 
  geom_bar(stat = "identity", position="dodge") +
  labs(x = "Month", y = "Total value") +
  facet_grid(sitecode ~ year)

this gives:

Some additional comments:

• It's probably better not to use spaces in your column names (in the code above I removed the spaces)
• Add an example to your question which illustrative for the problem you are facing. In this case, it's better to give an example dataset that includes several sitecodes and several years.

I therefore made up some data:

df1 <- read.table(text="sitecode  year month gear totalvalue
678490     2012 3     GL   13882
678490     2012 4     GL   50942
678490     2012 5     GL   54973
678490     2012 6     GL   63938
678490     2012 7     GL   23825
678490     2012 8     GL   8195
678490     2012 9     GL   14859
678490     2012 9     RT   3225
678490     2012 10    GL   981
678490     2012 10    RT   19074
678490     2012 11    SD   106384
678490     2012 11    RT   2828
678490     2012 12    GL   107167
678490     2012 12    RT   4514", header= TRUE)

df2 <- df1
df2$sitecode <- 7849
df2$year <- 2013
df3 <- df1
df3$sitecode <- 7849
df4 <- df1
df4$year <- 2013

cdata <- rbind(df1,df2,df3,df4)