Variable vectors
The Census datasets that the cancensus package provides
access to are rich in detail but they can be complex to navigate. There
are thousands of variable vectors, including separate vector indicators
for aggregations split by Total, Female, and
Male populations. As a result, the total number of vectors
per dataset is significant, ranging from 1,715 in the CA01
dataset to 6,623 in the CA16 one.
View available Census variable vectors
## # A tibble: 7,709 × 7
## vector type label units parent_vector aggregation details
## <chr> <fct> <chr> <fct> <chr> <chr> <chr>
## 1 v_CA21_1 Total Population, 2021 Numb… <NA> Additive CA 202…
## 2 v_CA21_2 Total Population, 2016 Numb… <NA> Additive CA 202…
## 3 v_CA21_3 Total Population percenta… Numb… <NA> Average of… CA 202…
## 4 v_CA21_4 Total Total private dwell… Numb… <NA> Additive CA 202…
## 5 v_CA21_5 Total Private dwellings o… Numb… v_CA21_4 Additive CA 202…
## 6 v_CA21_6 Total Population density … Ratio <NA> Average of… CA 202…
## 7 v_CA21_7 Total Land area in square… Numb… <NA> Additive CA 202…
## 8 v_CA21_8 Total Total - Age Numb… <NA> Additive CA 202…
## 9 v_CA21_9 Male Total - Age Numb… <NA> Additive CA 202…
## 10 v_CA21_10 Female Total - Age Numb… <NA> Additive CA 202…
## # ℹ 7,699 more rowslist_census_vectors(dataset) retrieves an index of all
available vectors for a given dataset from the CensusMapper API or local
cache if recently called. Each Census variable has a vector code
assigned to it with naming pattern that goes
v_{dataset}_{index}. This is the code by which vectors are
identified through the CensusMapper API. In addition the vector code,
there is additional information showing population type, aggregation
type, label and details, as well as variable hierarchy. This function
can also be used to show the variables for additional datasets made
accessible through the CensusMapper API.
Searching for Census variable vectors
Due to the large number of Census variables it can be hard to find
the right data. There is a function for searching through Census
variable metadata in a few different ways. There are three types of
searches possible using this function: exact search, which simply looks
for exact string matches for a given query against the vector dataset;
keyword search, which breaks vector metadata into unigram tokens and
then tries to find the vectors with the greatest number of unique
matches; and, semantic search which works better with search phrases and
has tolerance for inexact searches. Switching between search modes is
done using the query_type argument when calling
find_census_vectors() function.
Note that variable search is optimized for the Census variables in the main Census datasets. While searches generally work for variables in additional datasets such as cross-tabs and taxfiler data, they have not been extensively tested against these datasets.
Exact search
Exact search uses exact string matching. It is best used when you know exactly the term you are looking for, and is not robust to spelling errors.
## # A tibble: 4 × 4
## vector type label details
## <chr> <fct> <chr> <chr>
## 1 v_CA16_626 Total Oji-Cree Language; Total - Mother tongue for the total popu…
## 2 v_CA16_1433 Total Oji-Cree Language; Total - Language spoken most often at ho…
## 3 v_CA16_2676 Total Oji-Cree 25% Data; Total - Knowledge of languages for the p…
## 4 v_CA16_5930 Total Oji-Cree 25% Data; Work; Total - Language used most often a…This, on the other hand, will return a warning.
## Warning: No exact matches found. Please check spelling and try again or consider using semantic or keyword search.
## See ?find_census_vectors() for more details.
##
## Alternatively, you can launch the Censusmapper web API in a browser by calling explore_census_vectors(dataset)Unless otherwise specified, find_census_vectors() will
use exact search as the default option.
Keyword search
Keyword search is meant to be used when you know the approximate subject matter of the variable without knowing the exact variable you are looking for. Under the hood, keyword search splits the query and vector details into unique unigram tokens and then looks for the vectors with the greatest number of matches.
find_census_vectors('commute mode', dataset = 'CA16', type = 'female', query_type = 'keyword', interactive = FALSE)## # A tibble: 7 × 4
## vector type label details
## <chr> <fct> <chr> <chr>
## 1 v_CA16_5794 Female Total - Main mode of commuting for the employed la… 25% Da…
## 2 v_CA16_5797 Female Car, truck, van - as a driver 25% Da…
## 3 v_CA16_5800 Female Car, truck, van - as a passenger 25% Da…
## 4 v_CA16_5803 Female Public transit 25% Da…
## 5 v_CA16_5806 Female Walked 25% Da…
## 6 v_CA16_5809 Female Bicycle 25% Da…
## 7 v_CA16_5812 Female Other method 25% Da…Keyword search will show all results that have the highest number of
unique keyword matches. What happens if there are other keyword matches
that have fewer total matches? By default, the function argument is set
as interactive = TRUE, which will prompt the user with a
console menu option to see the rest of the matches or not. If using
find_census_vectors() in a script or reproducible
documentation, we recommend setting this argument to
interactive = FALSE.
Semantic search
Semantic search works best with phrases. Rather than the decontextualized unigrams of keyword search, semantic search splits queries and vector details into n-grams and finds matches using approximate string distances based on the generalized Levenshtein distance.
## # A tibble: 56 × 4
## vector type label details
## <chr> <fct> <chr> <chr>
## 1 v_CA16_2210 Total Number of after-tax income recipients aged 15 year… Income…
## 2 v_CA16_2213 Total Median after-tax income in 2015 among recipients (… Income…
## 3 v_CA16_2306 Total Percentage with after-tax income Income…
## 4 v_CA16_2297 Total Total - After-tax income groups in 2015 for the po… Income…
## 5 v_CA16_2300 Total Without after-tax income Income…
## 6 v_CA16_2303 Total With after-tax income Income…
## 7 v_CA16_2309 Total Under $10,000 (including loss) Income…
## 8 v_CA16_2312 Total $10,000 to $19,999 Income…
## 9 v_CA16_2315 Total $20,000 to $29,999 Income…
## 10 v_CA16_2318 Total $30,000 to $39,999 Income…
## # ℹ 46 more rowsSemantic search is more robust to spelling and punctuation issues that may come up with exact search. For example, while this throws a warning:
## Warning: No exact matches found. Please check spelling and try again or consider using semantic or keyword search.
## See ?find_census_vectors() for more details.
##
## Alternatively, you can launch the Censusmapper web API in a browser by calling explore_census_vectors(dataset)This will find the correct Census vector.
## Multiple possible matches. Results ordered by closeness.## # A tibble: 4 × 4
## vector type label details
## <chr> <fct> <chr> <chr>
## 1 v_CA16_626 Total Oji-Cree Language; Total - Mother tongue for the total popu…
## 2 v_CA16_1433 Total Oji-Cree Language; Total - Language spoken most often at ho…
## 3 v_CA16_2676 Total Oji-Cree 25% Data; Total - Knowledge of languages for the p…
## 4 v_CA16_5930 Total Oji-Cree 25% Data; Work; Total - Language used most often a…Results are ordered by string proximity if there are multiple possible matches.