Skip to contents

Example: Facebook Usage, Population Online, GDP, & Gender in Africa

Source: vignettes/example-pop-gdp-gender.Rmd
example-pop-gdp-gender.Rmd

How does Facebook usage relate to socio-economic indicators? This vignette illustates the extent that Facebook usage relates to internet connectivity and per capita GDP in sub-Saharan Africa—as well as how Facebook data can be used to examine the gender digital divide.

Setup

Below we load relevant packages and define Facebook keys.

## Load packages
library(rsocialwatcher)
library(dplyr)
library(tidyr)
library(ggplot2)
library(WDI)
library(janitor)
library(ggpubr)
library(knitr)
library(kableExtra)

## Set Facebook Keys
TOKEN        <- "TOKEN-HERE"
CREATION_ACT <- "CREATION-ACT-HERE"
VERSION      <- "VERSION-HERE"

Query data from WDI

We query data from the World Development Indicators (WDI). For each country, we query:

  • Total population (SP.POP.TOTL)
  • Male population (SP.POP.TOTL.MA.ZS)
  • Female population (SP.POP.TOTL.FE.ZS)
  • GDP per capita (NY.GDP.PCAP.CD)
  • Individuals using the Internet, % of population (IT.NET.USER.ZS)
wdi_df <- WDI(country = "all",
              indicator = c("SP.POP.TOTL",
                            "SP.POP.TOTL.MA.ZS",
                            "SP.POP.TOTL.FE.ZS",
                            "NY.GDP.PCAP.CD",
                            "IT.NET.USER.ZS"),
              start = 2021,
              end = 2021,
              extra = T)

wdi_df <- wdi_df %>%
  filter(region == "Sub-Saharan Africa")

Query data from Facebook

We separately query the number of male and female monthly active users on Facebook for each country.

fb_df <- query_fb_marketing_api(
  location_unit_type = "countries",
  location_keys      = map_param_vec(wdi_df$iso2c),
  gender             = map_param(1, 2),
  version            = VERSION, 
  creation_act       = CREATION_ACT, 
  token              = TOKEN)

Cleanup data

Here we merge together the WDI and Facebook datasets queried above and clean-up the dataset.

fb_clean_df <- fb_df %>%
  rename(iso2c = location_keys) %>%
  mutate(gender = case_when(
    gender == "1" ~ "fb_male",
    gender == "2" ~ "fb_female"
  )) %>%
  pivot_wider(id_cols = c(iso2c),
              names_from = gender,
              values_from = estimate_mau_upper_bound) %>%
  left_join(wdi_df, by = "iso2c") %>%
  clean_names() %>%
  mutate(fb_total = fb_female + fb_male,
         fb_per_female = fb_female/fb_total*100,
         wdi_per_female = sp_pop_totl_fe_zs,
         per_fb_pop = fb_total/sp_pop_totl*100) 

fb_clean_df %>% 
  kable() %>% 
  kable_styling() %>% 
  scroll_box(width = "100%",
             height = "300px")
iso2c fb_male fb_female x country iso3c year status lastupdated sp_pop_totl sp_pop_totl_ma_zs sp_pop_totl_fe_zs ny_gdp_pcap_cd it_net_user_zs region capital longitude latitude income lending fb_total fb_per_female wdi_per_female per_fb_pop
AO 3300000 2400000 1 Angola AGO 2021 NA 2024-03-28 34503774 49.41921 50.58079 1927.4741 32.602302 Sub-Saharan Africa Luanda 13.24200 -8.811550 Lower middle income IBRD 5700000 42.10526 50.58079 16.5199320
BJ 1600000 1000000 2 Benin BEN 2021 NA 2024-03-28 12996895 50.08666 49.91334 1360.9115 33.965331 Sub-Saharan Africa Porto-Novo 2.63230 6.477900 Lower middle income IDA 2600000 38.46154 49.91334 20.0047781
BW 690500 694000 3 Botswana BWA 2021 NA 2024-03-28 2588423 49.36345 50.63655 7238.7961 73.500000 Sub-Saharan Africa Gaborone 25.92010 -24.654400 Upper middle income IBRD 1384500 50.12640 50.63655 53.4881663
BF 2400000 1000000 4 Burkina Faso BFA 2021 NA 2024-03-28 22100683 49.82062 50.17938 888.7811 21.579827 Sub-Saharan Africa Ouagadougou -1.53395 12.360500 Low income IDA 3400000 29.41176 50.17938 15.3841399
BI 715700 412400 5 Burundi BDI 2021 NA 2024-03-28 12551213 49.65087 50.34913 221.1578 5.804922 Sub-Saharan Africa Bujumbura 29.36390 -3.378400 Low income IDA 1128100 36.55704 50.34913 8.9879759
CV 157700 159300 6 Cabo Verde CPV 2021 NA 2024-03-28 587925 49.72471 50.27512 3398.1654 69.762257 Sub-Saharan Africa Praia -23.50870 14.921800 Lower middle income Blend 317000 50.25237 50.27512 53.9184420
CM 3500000 2500000 7 Cameroon CMR 2021 NA 2024-03-28 27198628 49.87101 50.12899 1654.2570 45.602174 Sub-Saharan Africa Yaounde 11.51740 3.872100 Lower middle income Blend 6000000 41.66667 50.12899 22.0599363
CF 119800 61700 8 Central African Republic CAF 2021 NA 2024-03-28 5457154 49.99456 50.00546 461.1375 10.582640 Sub-Saharan Africa Bangui 21.64070 5.630560 Low income IDA 181500 33.99449 50.00546 3.3259094
TD 737800 218400 9 Chad TCD 2021 NA 2024-03-28 17179740 50.19593 49.80407 685.6903 17.868879 Sub-Saharan Africa N’Djamena 15.04450 12.104800 Low income IDA 956200 22.84041 49.80407 5.5658584
KM 162400 113400 10 Comoros COM 2021 NA 2024-03-28 821625 50.23009 49.77003 1577.4708 27.336943 Sub-Saharan Africa Moroni 43.24180 -11.698600 Lower middle income IDA 275800 41.11675 49.77003 33.5676251
CD 5000000 2800000 11 Congo, Dem. Rep.  COD 2021 NA 2024-03-28 95894118 49.61186 50.38815 576.9747 22.901585 Sub-Saharan Africa Kinshasa 15.32220 -4.325000 Low income IDA 7800000 35.89744 50.38815 8.1339713
CG 683100 501600 12 Congo, Rep.  COG 2021 NA 2024-03-28 5835806 49.94056 50.05944 2540.4732 NA Sub-Saharan Africa Brazzaville 15.26620 -4.276700 Lower middle income Blend 1184700 42.33983 50.05944 20.3005377
CI 5400000 3200000 13 Cote d’Ivoire CIV 2021 NA 2024-03-28 27478249 50.50436 49.49564 2613.3789 45.425533 Sub-Saharan Africa Yamoussoukro -4.03050 5.332000 Lower middle income IDA 8600000 37.20930 49.49564 31.2974819
GQ 113900 74900 14 Equatorial Guinea GNQ 2021 NA 2024-03-28 1634466 52.82306 47.17694 7406.0850 53.922402 Sub-Saharan Africa Malabo 8.77410 3.752300 Upper middle income IBRD 188800 39.67161 47.17694 11.5511733
ER 11500 6600 15 Eritrea ERI 2021 NA 2024-03-28 3620312 49.32835 50.67165 NA 21.730386 Sub-Saharan Africa Asmara 38.91830 15.331500 Low income IDA 18100 36.46409 50.67165 0.4999569
SZ 244200 236000 16 Eswatini SWZ 2021 NA 2024-03-28 1192271 49.63897 50.36112 4068.5738 58.912916 Sub-Saharan Africa Mbabane 31.46590 -26.522500 Lower middle income IBRD 480200 49.14619 50.36112 40.2760782
ET 6000000 3000000 17 Ethiopia ETH 2021 NA 2024-03-28 120283026 50.25098 49.74902 925.0007 16.698103 Sub-Saharan Africa Addis Ababa 38.74680 9.022740 Low income IDA 9000000 33.33333 49.74902 7.4823525
GA 505700 388100 18 Gabon GAB 2021 NA 2024-03-28 2341179 50.93207 49.06788 8635.7971 71.749406 Sub-Saharan Africa Libreville 9.45162 0.388320 Upper middle income IBRD 893800 43.42135 49.06788 38.1773457
GM 394000 195200 19 Gambia, The GMB 2021 NA 2024-03-28 2639916 49.74094 50.25902 772.1515 32.962290 Sub-Saharan Africa Banjul -16.58850 13.449500 Low income IDA 589200 33.12967 50.25902 22.3188920
GH 5500000 4100000 20 Ghana GHA 2021 NA 2024-03-28 32833031 49.87519 50.12481 2422.0859 68.200000 Sub-Saharan Africa Accra -0.20795 5.570450 Lower middle income IDA 9600000 42.70833 50.12481 29.2388479
GN 1700000 1200000 21 Guinea GIN 2021 NA 2024-03-28 13531906 49.40841 50.59159 1189.1760 34.680124 Sub-Saharan Africa Conakry -13.70000 9.516670 Low income IDA 2900000 41.37931 50.59159 21.4308317
GW 218600 139100 22 Guinea-Bissau GNB 2021 NA 2024-03-28 2060721 49.36214 50.63786 795.1186 35.154987 Sub-Saharan Africa Bissau -15.18040 11.803700 Low income IDA 357700 38.88734 50.63786 17.3580024
KE 9000000 7200000 23 Kenya KEN 2021 NA 2024-03-28 53005614 49.57811 50.42188 2069.6611 28.757528 Sub-Saharan Africa Nairobi 36.81260 -1.279750 Lower middle income Blend 16200000 44.44444 50.42188 30.5628004
LS 324800 361900 24 Lesotho LSO 2021 NA 2024-03-28 2281454 49.34125 50.65875 1029.5062 47.982999 Sub-Saharan Africa Maseru 27.71670 -29.520800 Lower middle income IDA 686700 52.70133 50.65875 30.0992262
LR 554800 451500 25 Liberia LBR 2021 NA 2024-03-28 5193416 49.77789 50.22211 675.6632 33.633159 Sub-Saharan Africa Monrovia -10.79570 6.300390 Low income IDA 1006300 44.86734 50.22211 19.3764567
MG 2300000 1900000 26 Madagascar MDG 2021 NA 2024-03-28 28915653 50.11450 49.88550 503.3521 19.730168 Sub-Saharan Africa Antananarivo 45.71670 -20.466700 Low income IDA 4200000 45.23810 49.88550 14.5250048
MW 985000 572800 27 Malawi MWI 2021 NA 2024-03-28 19889742 48.62205 51.37795 633.6097 24.405601 Sub-Saharan Africa Lilongwe 33.77030 -13.989900 Low income IDA 1557800 36.76980 51.37795 7.8321780
ML 1900000 617700 28 Mali MLI 2021 NA 2024-03-28 21904983 50.49419 49.50581 881.5101 34.490000 Sub-Saharan Africa Bamako -7.50034 13.566700 Low income IDA 2517700 24.53430 49.50581 11.4937318
MR 898500 505800 29 Mauritania MRT 2021 NA 2024-03-28 4614974 48.98381 51.01621 1998.3940 58.758190 Sub-Saharan Africa Nouakchott -15.98240 18.236700 Lower middle income IDA 1404300 36.01794 51.01621 30.4292072
MU 550600 510200 30 Mauritius MUS 2021 NA 2024-03-28 1266334 49.34472 50.65528 9068.9802 67.580000 Sub-Saharan Africa Port Louis 57.49770 -20.160500 Upper middle income IBRD 1060800 48.09578 50.65528 83.7693689
MZ 2200000 1600000 31 Mozambique MOZ 2021 NA 2024-03-28 32077072 49.06066 50.93934 504.0378 17.372837 Sub-Saharan Africa Maputo 32.57130 -25.966400 Low income IDA 3800000 42.10526 50.93934 11.8464678
NA 460200 488600 32 Namibia NAM 2021 NA 2024-03-28 2530151 48.26455 51.73545 4919.1890 52.973479 Sub-Saharan Africa Windhoek 17.09310 -22.564800 Upper middle income IBRD 948800 51.49663 51.73545 37.4997382
NE 607100 154500 33 Niger NER 2021 NA 2024-03-28 25252722 50.72378 49.27622 590.6295 22.386139 Sub-Saharan Africa Niamey 2.10730 13.514000 Low income IDA 761600 20.28624 49.27622 3.0159125
NG 28700000 21000000 34 Nigeria NGA 2021 NA 2024-03-28 213401323 50.52781 49.47219 2065.7744 55.363714 Sub-Saharan Africa Abuja 7.48906 9.058040 Lower middle income Blend 49700000 42.25352 49.47219 23.2894526
RW 908900 530800 35 Rwanda RWA 2021 NA 2024-03-28 13461888 48.89536 51.10464 821.1714 30.461817 Sub-Saharan Africa Kigali 30.05870 -1.953250 Low income IDA 1439700 36.86879 51.10464 10.6946366
ST 38500 34200 36 Sao Tome and Principe STP 2021 NA 2024-03-28 223107 49.88862 50.11138 2350.4527 51.204092 Sub-Saharan Africa Sao Tome 6.60710 0.206180 Lower middle income IDA 72700 47.04264 50.11138 32.5852618
SN 2900000 1700000 37 Senegal SEN 2021 NA 2024-03-28 16876720 49.16273 50.83727 1633.5601 58.054819 Sub-Saharan Africa Dakar -17.47340 14.724700 Lower middle income IDA 4600000 36.95652 50.83727 27.2564811
SC 45700 40100 38 Seychelles SYC 2021 NA 2024-03-28 99258 52.69040 47.30866 12963.0590 81.593072 Sub-Saharan Africa Victoria 55.44660 -4.630900 High income IBRD 85800 46.73660 47.30866 86.4413952
SL 704800 549000 39 Sierra Leone SLE 2021 NA 2024-03-28 8420641 50.10119 49.89883 504.6213 NA Sub-Saharan Africa Freetown -13.21340 8.482100 Low income IDA 1253800 43.78689 49.89883 14.8896028
SO 1600000 1300000 40 Somalia SOM 2021 NA 2024-03-28 17065581 50.13382 49.86618 576.5237 NA Sub-Saharan Africa Mogadishu 45.32540 2.075150 Low income IDA 2900000 44.82759 49.86618 16.9932685
ZA 15300000 15700000 41 South Africa ZAF 2021 NA 2024-03-28 59392255 48.65046 51.34953 7073.6128 72.310492 Sub-Saharan Africa Pretoria 28.18710 -25.746000 Upper middle income IBRD 31000000 50.64516 51.34953 52.1953578
SS 608100 241400 42 South Sudan SSD 2021 NA 2024-03-28 10748272 49.50176 50.49824 NA NA Sub-Saharan Africa Juba 31.60000 4.850000 Low income IDA 849500 28.41672 50.49824 7.9035960
TZ 4800000 3400000 44 Tanzania TZA 2021 NA 2024-03-28 63588334 49.40789 50.59211 1146.0320 31.633131 Sub-Saharan Africa Dodoma 35.73820 -6.174860 Lower middle income IDA 8200000 41.46341 50.59211 12.8954471
TG 786300 354800 45 Togo TGO 2021 NA 2024-03-28 8644829 50.26448 49.73552 976.6662 34.984055 Sub-Saharan Africa Lome 1.22550 6.122800 Low income IDA 1141100 31.09281 49.73552 13.1997984
UG 1900000 1200000 46 Uganda UGA 2021 NA 2024-03-28 45853778 49.50658 50.49342 883.4657 10.342420 Sub-Saharan Africa Kampala 32.57290 0.314269 Low income IDA 3100000 38.70968 50.49342 6.7606207
ZM 2200000 1800000 47 Zambia ZMB 2021 NA 2024-03-28 19473125 49.34495 50.65505 1134.7135 21.231531 Sub-Saharan Africa Lusaka 28.29370 -15.398200 Low income IDA 4000000 45.00000 50.65505 20.5411304
ZW 1400000 1100000 48 Zimbabwe ZWE 2021 NA 2024-03-28 15993524 47.16715 52.83285 1773.9204 34.813971 Sub-Saharan Africa Harare 31.06720 -17.831200 Lower middle income Blend 2500000 44.00000 52.83285 15.6313268

Analysis

Setup

Below we create a theme that we will use across figures.

## Make theme for figures
p_theme <- theme(plot.title = element_text(face = "bold", size = 10),
                 plot.subtitle = element_text(face = "italic", size = 10),
                 axis.text = element_text(color = "black"),
                 axis.title = element_text(size = 10))

Facebook Usage, Internet Connectivity, and GDP

We create figures to show the distribution of the percent of population on Facebook and the correlation of this indicator with percent of population online and GDP as measured by WDI. Panel A shows notable variation in the percent of population on Facebook—values range from about 0% to over 75%. Both the percent of population online and per capita GDP appear positively correlated with the percent of population on Facebook.

p_1a <- fb_clean_df %>%
  ggplot() +
  geom_histogram(aes(x = per_fb_pop),
                 fill = "#4267B2",
                 color = "black") +
  labs(x = "% population on Facebook",
       y = "N countries",
       title = "A. % population on Facebook\nacross countries") +
  theme_classic2() +
  p_theme

p_1b <- fb_clean_df %>%
  ggplot() +
  geom_point(aes(x = per_fb_pop,
                 y = it_net_user_zs),
             fill = "#4267B2",
             pch = 21) +
  labs(x = "% population on Facebook",
       y = "% population using internet",
       title = "B. Internet connectivity vs\n% population on Facebook") +
  xlim(0, 100) +
  ylim(0, 100) +
  theme_classic2() +
  p_theme 

p_1c <- fb_clean_df %>%
  ggplot() +
  geom_point(aes(x = per_fb_pop,
                 y = ny_gdp_pcap_cd),
             fill = "#4267B2",
             pch = 21) +
  labs(x = "% population on Facebook",
       y = "GDP per capita",
       title = "C. Per capita GDP vs\n% population on Facebook") +
  theme_classic2() +
  p_theme

ggarrange(p_1a, p_1b, p_1c, nrow = 1)

Digital Gender Divide

Here we examine variation in the percent of Facebook users that are female across countries. Panel A shows notable variation in the percent of female Facebook users across countries, ranging from 20 to over 50%—while, as expected, the percent of female population as measured by WDI is about 50% for all countries.

p_2a <- fb_clean_df %>%
  ggplot() +
  geom_point(aes(x = fb_per_female,
                 y = wdi_per_female),
             pch = 21,
             size = 2,
             fill = "red") +
  xlim(20, 55) +
  ylim(20, 55) + 
  theme_classic2() +
  p_theme +
  labs(x = "% of Facebook users that are female",
       y = "% females in population (WDI)",
       title = "A. % females in population (WDI) vs\n% of Facebook users that are female")

p_2b <- fb_clean_df %>%
  mutate(income = income %>%
           factor(levels = c("Low income",
                             "Lower middle income",
                             "Upper middle income",
                             "High income"))) %>%
  ggplot(aes(x = fb_per_female,
             y = income)) +
  geom_boxplot(color = "gray50",
               outlier.size = 0) +
  geom_jitter(width = 0,
              height = 0.1,
              pch = 21,
              size = 2,
              fill = "red") +
  labs(x = "% of Facebook users that are female",
       y = NULL,
       title = "B. Income vs. % of Facebook users\nthat are female") + 
  theme_classic2() + 
  scale_x_continuous(breaks = seq(0, 60, 20),
                     limits = c(0, 60)) +
  p_theme

ggarrange(p_2a, p_2b, nrow = 1, widths = c(0.45, 0.55))