dadams/texas-district-analysis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

double checked, updated narrative

Browse Source
This commit is contained in:
dadams-AU
2026-02-18 19:52:01 -08:00
parent 7e03e6abae
commit 64777d15e1
4 changed files with 45 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
analysis/draft_appendix.md
View File

@@ -5,7 +5,21 @@
### A1. Data integration
The analysis combines inspection and violation administrative records (2015-2025) into a district-year panel. Well-level linkage is done via `api_norm`.
The analysis combines inspection and violation administrative records (2015-2025) into a district-year panel. The estimation sample contains 143 district-year observations across 13 districts (52 pre-policy; 91 post-policy). Well-level linkage is done via `api_norm`.
### A1b. Pipeline volume and sample flow
| Stage | Count |
| :--- | ---: |
| Well records loaded (well universe table) | 1,010,432 |
| Inspection records loaded (all available years) | 1,878,764 |
| Violation records loaded (all available years) | 193,338 |
| Inspection records retained (2015-2025) | 1,867,859 |
| Violation records retained (2015-2025) | 191,762 |
| District-year panel observations | 143 |
| Districts represented | 13 |
These counts show that district-year inference is generated from very large underlying administrative record streams, with modest reductions due to the analytic time-window restriction.
### A2. Core variables
@@ -51,7 +65,7 @@ All models report district-clustered standard errors.
### B4. Spatial diagnostic (H4)
H4 is tested using permutation-based global Moran's I on estimated district treatment effects.
H4 is tested using permutation-based global Moran's I on estimated district treatment effects, using a manually specified district contiguity matrix and 5,000 random permutations for inference.
## Appendix C. Main Run Outputs
@@ -83,6 +97,13 @@ Substantively, this table supports the main-text conclusion that the policy effe
Pre-policy years are jointly non-significant in this decomposition.
The coefficient pattern reinforces parallel-pretrend credibility while showing that the post-policy effect strengthens in later years, consistent with delayed organizational adaptation.
### C2b. H2 omnibus heterogeneity test
- Wald chi-square (all district-by-post terms = 0): 0.670
- P-value: 0.4130
This omnibus test is not statistically significant in the current run, so district heterogeneity is interpreted primarily from the dispersion of district-specific estimates and mapped effect magnitudes.
### C3. Offshore differential annual effects (ref=2018)
| Year | Offshore differential coef | P-value |