Readme.qmd

---
title: "H1B FOIA Data 2021-2023"
author: "Leonce Nshuti"
format: revealjs
editor: visual
---

#### [H-1B Visa Lottery Dashboard](https://leoncensh-h1b-visa-trends.hf.space/)

United States H-1B visa lottery system key metrics:

-   Employer

-   Salary

-   Job title

-   Country of origin

```{r, message=FALSE}
library(tidyverse)
library(arrow)
library(dbplyr, warn.conflicts = FALSE)
library(duckdb)
library(ggthemes)
```

## Data Overview

```{r}
trk_13139 <- open_dataset(
  sources = "data/TRK_13139_FY2021_2023.csv", 
  col_types = schema(ISBN = string()),
  format = "csv"
)

trk_13139 %>%
  as_tibble() %>%
  sample_n(150) %>%
  glimpse()
```

```{r}
honebs_sum_compensaton <- 
  trk_13139 |> 
  group_by(country_of_birth,lottery_year) |> 
  summarise(ben_comp_paid = sum(ben_comp_paid, na.rm = TRUE)) |> 
  arrange(desc(ben_comp_paid)) |> 
  collect() |>
  ungroup() |>
  mutate(lottery_year = factor(lottery_year)) |>
  mutate(ben_comp_paid=paste0('$', round(ben_comp_paid,0))) |>
  head(27)


honebs_sum_compensaton %>% 
  #filter(country_of_birth != "CHN") %>% 
  ggplot(data=., aes(x=lottery_year,
                     y=ben_comp_paid, 
                     group = country_of_birth,
                     colour=country_of_birth)) + 
  geom_point() + 
  geom_line() + 
  theme(panel.grid = element_blank(),
        panel.border = element_blank()) + 
  labs(
    title = "H1B salaries by Country of Origin and Lottery Year",
    subtitle = "",
    x = "Lottery Year", y = "Sum of Salary ($)",
    color = "country_of_birth", shape = "country_of_birth"
  ) +
  theme_tufte()

```

```{r}
trk_13139 |>
  as_tibble() |>
  filter(ed_level_definition == "MASTER'S DEGREE") |>
  group_by(country_of_nationality,lottery_year) |> 
  summarise(ben_comp_paid = sum(ben_comp_paid, na.rm = TRUE)) |> 
  arrange(desc(ben_comp_paid)) %>% 
  head(5)
```

### H1B salaries by Country of Origin and Lottery Year

| country_of_nationality | **lottery_year** | **ben_comp_paid** |
|------------------------|------------------|-------------------|
| CHN                    | 2021             | 1,176,384,333     |
| CHN                    | 2022             | 909,759,682       |
| CHN                    | 2023             | 251,258,543       |
| CAN                    | 2021             | 100,207,982       |
| TWN                    | 2021             | 82,747,461        |
| CAN                    | 2022             | 78,514,554        |
| TWN                    | 2022             | 75,308,303        |
| FRA                    | 2021             | 44,075,805        |
| KOR                    | 2021             | 36,157,220        |
| KOR                    | 2022             | 36,139,151        |

### Core Filters

1.  Fiscal year (2021, 2022, 2023)

2.  Employer name (e.g., Wipro, Amazon).

3.  Job title: (e.g., “Software Engineer”).

4.  Country of birth, Country of nationality.

5.  Salary: wage amount (e.g. \$50,000), wage unit(e.g. year).

6.  Worksite: work-site city, work-site state.

## Hong Kong H1Bs salary

![](images/clipboard-1910641578.png)

## Rwanda Aggregate salary

```{r}
trk_13139 %>% 
  as_tibble() %>%
  group_by(country_of_birth,lottery_year) %>% 
  summarise(ben_comp_paid = sum(ben_comp_paid, na.rm = TRUE)) %>% 
  arrange(desc(ben_comp_paid)) %>% 
  ungroup() %>%
  mutate(lottery_year = factor(lottery_year)) %>%
  mutate(ben_comp_paid=paste0('$', round(ben_comp_paid,0))) %>%
  filter(country_of_birth == "RWA") 
```