Analyzing NYCHA Housing Maintenance Code Violations

Laura Werner

Introduction

The NYCHA Housing Maintenance Code Violations dataset, available through NYC Open Data, provides detailed information about maintenance violations in public housing developments across New York City.

This dataset includes information such as the borough, development name, violation type, inspection dates, and status. Researchers, policymakers, and residents can use this data to better understand housing conditions, identify recurring maintenance issues, and explore patterns across boroughs.

Data

We begin by retrieving the dataset using the nycOpenData package:

violations <- nyc_pull_dataset("nyc_ha_violations")

violations |>
  dplyr::slice_head(n = 6)
#> # A tibble: 6 × 35
#>   viol_seq_no bldg_id  boro boro_nm  phn   low_hus_no high_hus_no str_nm     zip
#>         <dbl>   <dbl> <dbl> <chr>    <chr> <chr>      <chr>       <chr>    <dbl>
#> 1    18680111  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> 2    18680112  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> 3    18680113  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> 4    18680114  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> 5    18680115  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> 6    18680116  379363     3 BROOKLYN 1060  1060       1060        MYRTLE … 11206
#> # ℹ 26 more variables: development_name <chr>, tds_no <dbl>,
#> #   stairhall_no <dbl>, sctn_boro <dbl>, sctn_boro_nm <chr>, sctn_hus_no <chr>,
#> #   sctn_str_nm <chr>, sctn_zip <dbl>, actl_unit_insp <chr>, actl_stry <dbl>,
#> #   blk <dbl>, lot <dbl>, viol_desc <chr>, hzrd_clas <chr>, insp_dt <dttm>,
#> #   viol_appr_dt <dttm>, viol_typ_ordr_no <dbl>, issued_in_err <chr>,
#> #   latitude <dbl>, longitude <dbl>, community_board <dbl>,
#> #   council_district <dbl>, bin <dbl>, bbl <dbl>, census_tract <dbl>, …

The glimpse() function allows us to quickly inspect the structure of the dataset, including available variables and their data types.

Count violations by borough

Next, we summarize the number of violations recorded in each borough:

borough_counts <- violations |>
  count(boro_nm, sort = TRUE)

borough_counts
#> # A tibble: 5 × 2
#>   boro_nm           n
#>   <chr>         <int>
#> 1 MANHATTAN       998
#> 2 BROOKLYN        776
#> 3 QUEENS          251
#> 4 BRONX           249
#> 5 STATEN ISLAND    24

Visual

To better understand the distribution, we create a bar chart:

borough_counts |>
  ggplot(aes(x = reorder(boro_nm, n), y = n)) +
  geom_col(fill = "steelblue") +
  coord_flip() +
  labs(
    title = "Number of NYCHA Housing Maintenance Violations by Borough",
    x = "Borough",
    y = "Number of Violations"
  ) +
  theme_minimal()

This visualization highlights which boroughs have the highest number of recorded maintenance violations. Differences across boroughs may reflect variation in the number of developments, building age, reporting practices, or underlying housing conditions. Further analysis could investigate violation types, severity levels, or trends over time.