Download Google Trends data on Covid-19

Downloads Google Trends data (https://trends.google.com/trends/) about the 2020 search activity for a given search term at global and country levels. The search term defaults to "coronavirus" to reflect the relative public attention to the Covid-19 pandemic.

download_google_trends_data(
  search_term = "coronavirus",
  type = "country_day",
  countries = NULL,
  low_search_volume = FALSE,
  pause = TRUE,
  silent = FALSE,
  cached = FALSE
)

Arguments

search_term	Defaults to "coronavirus".
type	The type of data that you want to retrieve. Can be any subset of "country": Relative search activity at the global level, reporting by country. "country_day": Relative search activity at the country level, reporting by country. "region": Relative search activity at the country level, reporting by region. "city": Relative search activity at the country level, reporting by city. Defaults to 'country_day'.
countries	A character vector of ISO3c country codes that you want to pull detailed data for. By default (countries = NULL) the code pulls detailed data for all countries the the global Google Trend request returns.
low_search_volume	Whether you want the include countries with low search volume. This increases the list of countries that are pulled considerably and also the risk that you hit a Google Trend search limit. Use with care. Defaults to FALSE.
pause	Whether you want the code to pause for a 2 to 5 seconds period between Google Trend API calls. As Google Trends has an unknown rate limit, this is probably a good idea and thus defaults to TRUE.
silent	Whether you want the function to send some status messages to the console. Might be informative as downloading will take some time and thus defaults to TRUE.
cached	Whether you want to download the cached version of the data from the tidycovid19 Github repository instead of retrieving the data from the authorative source. Downloading the cached version is faster and the cache is updated daily. Defaults to FALSE.

Value

If only one type was selected, a data frame containing the data. Otherwise, a list containing the desired data frames ordered as in type.

Details

Uses the gtrendsR package. Please note that Google Trends only reports relative search volume. For each data frame, the values are standardized so that the observations with the highest search volume gets a score of 100 and the other scores of the data frame are relative to that. This implies that comparisons across data frames are not feasible. When Google Trends reports a score of "<1" this is translated to 0.5 in the data.

Examples

df <- download_google_trends_data(type = "country", silent = TRUE, cached = TRUE)
df %>%
  dplyr::select(iso3c, gtrends_score) %>%
  dplyr::arrange(-gtrends_score)
#>    iso3c gtrends_score
#> 1    ITA           100
#> 2    ESP            82
#> 3    QAT            76
#> 4    GBR            74
#> 5    IRL            68
#> 6    FRA            62
#> 7    CHE            55
#> 8    URY            54
#> 9    ARE            52
#> 10   AUS            50
#> 11   ARG            49
#> 12   NPL            45
#> 13   USA            45
#> 14   PER            44
#> 15   ROU            43
#> 16   ZAF            42
#> 17   NZL            41
#> 18   AUT            40
#> 19   CAN            40
#> 20   PAN            39
#> 21   KWT            39
#> 22   COL            38
#> 23   SGP            36
#> 24   DEU            36
#> 25   BEL            36
#> 26   CHL            35
#> 27   PRY            33
#> 28   CRI            33
#> 29   IND            32
#> 30   BOL            32
#> 31   GHA            31
#> 32   ECU            30
#> 33   MEX            29
#> 34   GTM            29
#> 35   VEN            27
#> 36   PRT            26
#> 37   PAK            25
#> 38   MAR            24
#> 39   BGD            24
#> 40   DOM            24
#> 41   BRA            24
#> 42   KEN            23
#> 43   LKA            23
#> 44   DZA            20
#> 45   NLD            18
#> 46   MYS            15
#> 47   DNK            15
#> 48   SAU            14
#> 49   NGA            14
#> 50   SWE            14
#> 51   HKG            11
#> 52   ISR            11
#> 53   NOR            11
#> 54   PHL            10
#> 55   GRC             8
#> 56   EGY             6
#> 57   POL             4
#> 58   THA             3
#> 59   TUR             3
#> 60   IDN             2
#> 61   VNM             2
#> 62   RUS             2
#> 63   JPN             1

lst <- download_google_trends_data(type = c("region", "city"), silent = TRUE, cached = TRUE)
lst[[1]] %>%
  dplyr::filter(iso3c == "DEU") %>%
  dplyr::select(region, gtrends_score) %>%
  dplyr::arrange(-gtrends_score)
#>                    region gtrends_score
#> 1               Thuringia           100
#> 2                Saarland            94
#> 3    Rhineland-Palatinate            92
#> 4       Baden-Württemberg            91
#> 5           Saxony-Anhalt            88
#> 6                 Bavaria            88
#> 7             Brandenburg            87
#> 8                  Bremen            87
#> 9  North Rhine-Westphalia            87
#> 10           Lower Saxony            87
#> 11                 Berlin            86
#> 12 Mecklenburg-Vorpommern            85
#> 13     Schleswig-Holstein            83
#> 14                 Saxony            79
#> 15                Hamburg            77
#> 16                  Hesse            74

lst[[2]] %>%
  dplyr::filter(iso3c == "DEU") %>%
  dplyr::select(city, gtrends_score) %>%
  dplyr::arrange(-gtrends_score)
#>          city gtrends_score
#> 1 Nordstemmen           100
#> 2     Schlitz            83
#> 3   Gütersloh            83
#> 4    Sulzbach            81
#> 5  Angermünde            80
#> 6   Stuttgart            76
#> 7   Erftstadt            76
#> 8       Neuss            75
#> 9  Hildesheim            68