## `summarise()` has grouped output by 'word'. You can override using the
## `.groups` argument.
## # A tibble: 7,351 × 3
## # Groups: word [5,697]
## word brewery n
## <chr> <chr> <int>
## 1 beer rhinegeist 552
## 2 beer madtree 466
## 3 pizza madtree 280
## 4 brewery rhinegeist 279
## 5 rhinegeist rhinegeist 249
## 6 madtree madtree 214
## 7 space rhinegeist 206
## 8 beers madtree 169
## 9 beers rhinegeist 165
## 10 bar rhinegeist 163
## # … with 7,341 more rows
I was curious about where out of state reviewers were from, there many from neighboring states, but there were also many from coastal states as well.
## `summarise()` has grouped output by 'word'. You can override using the
## `.groups` argument.
Reviews containing the word “people” were often rated higher, I though that that inclusion of the word “people” in reviews would be in a negative, “too many people” or something like that, but I was wrong. This pattern could also be related to how the brewerys have fewer negative ratings than positve ones.