Generating Subsets for testing: 1990¶

blocks, block groups parts, and blocks¶

From a national crosswalk:
Create target state-level subsets for NHGIS base crosswalks
Create target state-level subsets for NHGIS base tabular data
Record unit tests values for posterity

This is currently only intended for use with block-level data as base units.

James Gaboardi (jgaboardi@gmail.com), 2020-05

In [1]:

%load_ext watermark
%watermark

2020-06-13T12:08:18-04:00

CPython 3.7.6
IPython 7.15.0

compiler   : Clang 9.0.1 
system     : Darwin
release    : 19.5.0
machine    : x86_64
processor  : i386
CPU cores  : 8
interpreter: 64bit

In [2]:

import inspect
import nhgisxwalk
import numpy
import pandas

%load_ext autoreload
%autoreload 2
%watermark -w
%watermark -iv

watermark 2.0.2
numpy      1.18.5
nhgisxwalk 0.0.2
pandas     1.0.4

Set the state (for subsetting), source & target, and year & geography¶

In [3]:

subset_state = "10" # Delaware
#subset_state = "11" # DC
#subset_state = "15" # Hawaii
source_year, target_year = "1990", "2010"
gj_src, gj_trg = "GJOIN%s"%source_year, "GJOIN%s"%target_year

Set the base-level crosswalk file name¶

In [4]:

base_xwalk_name = "nhgis_blk%s_blk%s_gj.csv.zip" % (source_year, target_year)
base_xwalk_file = "../../crosswalks/%s" % base_xwalk_name
base_xwalk_file

Out[4]:

'../../crosswalks/nhgis_blk1990_blk2010_gj.csv.zip'

Set the base (source) summary file name¶

In [5]:

base_source_name = "%s_block/%s_block.csv" % (source_year, source_year)
base_source_file = "../../tabular_data/%s" % base_source_name
base_source_file

Out[5]:

'../../tabular_data/1990_block/1990_block.csv'

Set the supplementary summary file name¶

In [6]:

supp_source_name = "%s_blck_grp_598_103/%s_blck_grp_598_103.csv" % (
    source_year, source_year
)
supp_source_file = "../../tabular_data/%s" % supp_source_name
supp_source_file

Out[6]:

'../../tabular_data/1990_blck_grp_598_103/1990_blck_grp_598_103.csv'

Read in the national the base-level crosswalk¶

In [7]:

data_types = nhgisxwalk.str_types([gj_src, gj_trg])
base_xwalk = pandas.read_csv(base_xwalk_file, index_col=0, dtype=data_types)
base_xwalk.head()

/Users/the-gaboardi/miniconda3/envs/nhgis/lib/python3.7/site-packages/numpy/lib/arraysetops.py:569: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison
  mask |= (ar1 == a)

Out[7]:

	GJOIN1990	GJOIN2010
0	NaN	G01000300107014085
1	NaN	G01000300107014086
2	NaN	G01000300107014089
3	NaN	G01000300107014091
4	NaN	G01000300107014109

Create the state subset of the base-level crosswalk (for use in GH testing)¶

In [8]:

ss_base = base_xwalk[
    base_xwalk["GJOIN2010"].map(lambda x: x[1:3] == subset_state)
].copy()
ss_base.reset_index(drop=True, inplace=True)

Declare input variable¶

not needed for creating a subset perse, but should do regardless

In [9]:

input_vars = [
    nhgisxwalk.desc_code_1990["Persons"]["Total"],
    nhgisxwalk.desc_code_1990["Families"]["Total"],
    nhgisxwalk.desc_code_1990["Households"]["Total"],
    nhgisxwalk.desc_code_1990["Housing Units"]["Total"]
]
input_var_tags = ["pop", "fam", "hh", "hu"]

Generate the desired crosswalk and subset down to the target state¶

In [10]:

state_bgp1990trt2010 = nhgisxwalk.GeoCrossWalk(
    base_xwalk,
    source_year=source_year,
    target_year=target_year,
    source_geo="bgp",
    target_geo="trt",
    base_source_table=base_source_file,
    supp_source_table=supp_source_file,
    input_var=input_vars,
    weight_var=input_var_tags,
    keep_base=True,
    add_geoid=True,
    stfips=subset_state
)
del base_xwalk
state_bgp1990trt2010.xwalk

Out[10]:

	bgp1990gj	trt2010gj	trt2010ge	wt_pop	wt_fam	wt_hh	wt_hu
80076	G100001090444072500423009999999999921	G1000010043202	10001043202	1.0	1.0	1.0	1.0
80077	G100001090444444300422009999999999926	G1000010042202	10001042202	1.0	1.0	1.0	1.0
80078	G100001090444612650422009999999219011	G1000010041200	10001041200	0.0	0.0	0.0	0.0
80079	G100001090444612650422009999999219011	G1000010042201	10001042201	1.0	1.0	1.0	1.0
80080	G100001090444612650422009999999219012	G1000010042201	10001042201	1.0	1.0	1.0	1.0
...	...	...	...	...	...	...	...
81134	G100005093552999990515009999999999923	G1000050051500	10005051500	1.0	1.0	1.0	1.0
81135	G100005093552999990515009999999999924	G1000050051500	10005051500	1.0	1.0	1.0	1.0
81136	G100005093552999990516009999999999921	G1000050051702	10005051702	1.0	1.0	1.0	1.0
284765	G340033010610106000204029999999916014	G1000030990100	10003990100	0.0	0.0	0.0	0.0
507805	NaN	G1000050990000	10005990000	0.0	0.0	0.0	0.0

1063 rows × 7 columns

In [11]:

state_bgp1990trt2010.xwalk.drop_duplicates(subset=["bgp1990gj", "trt2010gj"])

Out[11]:

	bgp1990gj	trt2010gj	trt2010ge	wt_pop	wt_fam	wt_hh	wt_hu
80076	G100001090444072500423009999999999921	G1000010043202	10001043202	1.0	1.0	1.0	1.0
80077	G100001090444444300422009999999999926	G1000010042202	10001042202	1.0	1.0	1.0	1.0
80078	G100001090444612650422009999999219011	G1000010041200	10001041200	0.0	0.0	0.0	0.0
80079	G100001090444612650422009999999219011	G1000010042201	10001042201	1.0	1.0	1.0	1.0
80080	G100001090444612650422009999999219012	G1000010042201	10001042201	1.0	1.0	1.0	1.0
...	...	...	...	...	...	...	...
81134	G100005093552999990515009999999999923	G1000050051500	10005051500	1.0	1.0	1.0	1.0
81135	G100005093552999990515009999999999924	G1000050051500	10005051500	1.0	1.0	1.0	1.0
81136	G100005093552999990516009999999999921	G1000050051702	10005051702	1.0	1.0	1.0	1.0
284765	G340033010610106000204029999999916014	G1000030990100	10003990100	0.0	0.0	0.0	0.0
507805	NaN	G1000050990000	10005990000	0.0	0.0	0.0	0.0

1063 rows × 7 columns

In [12]:

state_bgp1990trt2010.xwalk["bgp1990gj"].nunique()

Out[12]:

In [13]:

state_bgp1990trt2010.xwalk["trt2010gj"].nunique()

Out[13]:

Write out the state subset of the base-level crosswalk (for use in GH testing)¶

In [14]:

ss_base.to_csv("../testing_data_subsets/%s" % base_xwalk_name)

In [15]:

ss_base

Out[15]:

	GJOIN1990	GJOIN2010	WEIGHT	PAREA_VIA_BLK00
0	NaN	G10000100432021078	0.0	0.0
1	NaN	G10000100432023014	0.0	0.0
2	NaN	G10000100432023015	0.0	0.0
3	NaN	G10000109900000011	0.0	0.0
4	NaN	G10000109900000012	0.0	0.0
...	...	...	...	...
38292	G10000500519289	G10000500519002125	1.0	1.0
38293	G34003300204401A	G10000309901000007	0.0	0.0
38294	G34003300204418	G10000309901000007	0.0	0.0
38295	G34003300204419	G10000309901000007	0.0	0.0
38296	G34003300204420	G10000309901000007	0.0	0.0

38297 rows × 4 columns

Record, subset, and write out the 1990 BLKs (sf1) needed to create this subset¶

In [16]:

blk1990 = ss_base[~ss_base["GJOIN1990"].isna()]["GJOIN1990"]
blk1990

Out[16]:

149       G10000100401101
150       G10000100401102
151       G10000100401102
152       G10000100401102
153       G10000100401103
               ...       
38292     G10000500519289
38293    G34003300204401A
38294     G34003300204418
38295     G34003300204419
38296     G34003300204420
Name: GJOIN1990, Length: 38148, dtype: object

In [17]:

blk1990.shape

Out[17]:

(38148,)

In [18]:

blk1990.nunique()

Out[18]:

In [19]:

# read in base source file
base_source_df = pandas.read_csv(base_source_file, dtype=str)
base_source_df

Out[19]:

	GISJOIN	YEAR	ANRCA	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	TRACTA	...	STATE	STATEA	URBRURALA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001
0	G01000100201101A	1990	99	9999	9999	9999	9	101A	1	0201	...	Alabama	01	1	5240	NaN	Block 101A	332	91	107	112
1	G01000100201101B	1990	99	9999	9999	9999	9	101B	1	0201	...	Alabama	01	2	9999	NaN	Block 101B	14	3	6	7
2	G01000100201102A	1990	99	9999	9999	9999	9	102A	1	0201	...	Alabama	01	1	5240	NaN	Block 102A	248	74	88	89
3	G01000100201103	1990	99	9999	9999	9999	9	103	1	0201	...	Alabama	01	1	5240	NaN	Block 103	49	15	15	16
4	G01000100201104	1990	99	9999	9999	9999	9	104	1	0201	...	Alabama	01	1	5240	NaN	Block 104	12	3	4	4
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
4934102	G56004509513577B	1990	99	9999	9999	9999	9	577B	5	9513	...	Wyoming	56	2	9999	NaN	Block 577B	5	1	3	3
4934103	G56004509513578	1990	99	9999	9999	9999	9	578	5	9513	...	Wyoming	56	1	9999	NaN	Block 578	59	20	21	22
4934104	G56004509513579	1990	99	9999	9999	9999	9	579	5	9513	...	Wyoming	56	1	9999	NaN	Block 579	30	8	10	11
4934105	G56004509513580	1990	99	9999	9999	9999	9	580	5	9513	...	Wyoming	56	2	9999	NaN	Block 580	35	9	11	11
4934106	G56004509513597	1990	99	9999	9999	9999	9	597	5	9513	...	Wyoming	56	2	9999	NaN	Block 597	39	10	13	13

4934107 rows × 30 columns

In [20]:

base_source_df = base_source_df[base_source_df["GISJOIN"].isin(blk1990.unique())]
base_source_df.reset_index(drop=True, inplace=True)
base_source_df

Out[20]:

	GISJOIN	YEAR	ANRCA	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	TRACTA	...	STATE	STATEA	URBRURALA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001
0	G10000100401101	1990	99	9999	9999	9999	9	101	1	0401	...	Delaware	10	2	9999	NaN	Block 101	24	6	8	9
1	G10000100401102	1990	99	9999	9999	9999	9	102	1	0401	...	Delaware	10	2	9999	NaN	Block 102	145	39	50	58
2	G10000100401103	1990	99	9999	9999	9999	9	103	1	0401	...	Delaware	10	2	9999	NaN	Block 103	75	23	27	29
3	G10000100401104	1990	99	9999	9999	9999	9	104	1	0401	...	Delaware	10	2	9999	NaN	Block 104	69	19	21	22
4	G10000100401105	1990	99	9999	9999	9999	9	105	1	0401	...	Delaware	10	2	9999	NaN	Block 105	2	0	2	3
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
11935	G10000500519289	1990	99	9999	9999	9999	9	289	2	0519	...	Delaware	10	2	9999	NaN	Block 289	8	1	6	6
11936	G34003300204401A	1990	99	9999	9999	9999	9	401A	4	0204	...	New Jersey	34	1	9160	NaN	Block 401A	122	31	51	52
11937	G34003300204418	1990	99	9999	9999	9999	9	418	4	0204	...	New Jersey	34	1	9160	NaN	Block 418	86	25	35	36
11938	G34003300204419	1990	99	9999	9999	9999	9	419	4	0204	...	New Jersey	34	1	9160	NaN	Block 419	207	50	114	123
11939	G34003300204420	1990	99	9999	9999	9999	9	420	4	0204	...	New Jersey	34	1	9160	NaN	Block 420	101	19	53	56

11940 rows × 30 columns

In [21]:

base_source_df.to_csv("../testing_data_subsets/%s_block.csv.zip" % source_year)

Record, subset, and write out the 1990 BGPs (sf1) needed to create this subset¶

In [22]:

bgp1990 = state_bgp1990trt2010.xwalk[
    ~state_bgp1990trt2010.xwalk["bgp1990gj"].isna()
]["bgp1990gj"]
bgp1990

Out[22]:

80076     G100001090444072500423009999999999921
80077     G100001090444444300422009999999999926
80078     G100001090444612650422009999999219011
80079     G100001090444612650422009999999219011
80080     G100001090444612650422009999999219012
                          ...                  
81133     G100005093552999990515009999999999922
81134     G100005093552999990515009999999999923
81135     G100005093552999990515009999999999924
81136     G100005093552999990516009999999999921
284765    G340033010610106000204029999999916014
Name: bgp1990gj, Length: 1062, dtype: object

In [23]:

# read in supplement source file
supp_source_df = pandas.read_csv(supp_source_file, dtype=str)
supp_source_df

Out[23]:

	GISJOIN	YEAR	ANRCA	AIANHH	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	...	STATEA	URBRURALA	URB_AREA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001
0	G0100010901710322002119999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	01	2	NaN	9999	NaN	BG 2 (pt.)	402	105	144	167
1	G0100010901710322002119999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	01	2	NaN	9999	NaN	BG 3 (pt.)	279	63	88	96
2	G0100010901719999902119999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	01	2	NaN	9999	NaN	BG 1	611	150	179	213
3	G0100010901719999902119999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	01	2	NaN	9999	NaN	BG 2 (pt.)	1244	318	410	444
4	G0100010901719999902119999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	01	2	NaN	9999	NaN	BG 3 (pt.)	447	126	166	197
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
367248	G5600450935209999995129999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	56	2	NaN	9999	NaN	BG 3	44	12	13	21
367249	G5600450935209999995129999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	56	2	NaN	9999	NaN	BG 4	42	12	18	22
367250	G5600450935209999995129999999999925	1990	99	NaN	9999	9999	9999	9	NaN	5	...	56	2	NaN	9999	NaN	BG 5	29	6	9	15
367251	G5600450935209999995129999999999926	1990	99	NaN	9999	9999	9999	9	NaN	6	...	56	2	NaN	9999	NaN	BG 6 (pt.)	30	8	10	15
367252	G5600450935209999995129999999999927	1990	99	NaN	9999	9999	9999	9	NaN	7	...	56	2	NaN	9999	NaN	BG 7 (pt.)	30	8	9	11

367253 rows × 34 columns

In [24]:

# GISJOIN ID correction
bgpidcols = nhgisxwalk.id_codes.code_cols("bgp", "1990")
supp_source_df = nhgisxwalk.id_codes.bgp_gj(supp_source_df, bgpidcols, cname="__GISJOIN")
supp_source_df

Out[24]:

	GISJOIN	YEAR	ANRCA	AIANHH	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	...	URBRURALA	URB_AREA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001	__GISJOIN
0	G0100010901710322002119999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	2	NaN	9999	NaN	BG 2 (pt.)	402	105	144	167	G010001090171032200211039999999999922
1	G0100010901710322002119999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	2	NaN	9999	NaN	BG 3 (pt.)	279	63	88	96	G010001090171032200211039999999999923
2	G0100010901719999902119999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	2	NaN	9999	NaN	BG 1	611	150	179	213	G010001090171999990211039999999999921
3	G0100010901719999902119999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	2	NaN	9999	NaN	BG 2 (pt.)	1244	318	410	444	G010001090171999990211039999999999922
4	G0100010901719999902119999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	2	NaN	9999	NaN	BG 3 (pt.)	447	126	166	197	G010001090171999990211039999999999923
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
367248	G5600450935209999995129999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	2	NaN	9999	NaN	BG 3	44	12	13	21	G560045093520999999512009999999999923
367249	G5600450935209999995129999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	2	NaN	9999	NaN	BG 4	42	12	18	22	G560045093520999999512009999999999924
367250	G5600450935209999995129999999999925	1990	99	NaN	9999	9999	9999	9	NaN	5	...	2	NaN	9999	NaN	BG 5	29	6	9	15	G560045093520999999512009999999999925
367251	G5600450935209999995129999999999926	1990	99	NaN	9999	9999	9999	9	NaN	6	...	2	NaN	9999	NaN	BG 6 (pt.)	30	8	10	15	G560045093520999999512009999999999926
367252	G5600450935209999995129999999999927	1990	99	NaN	9999	9999	9999	9	NaN	7	...	2	NaN	9999	NaN	BG 7 (pt.)	30	8	9	11	G560045093520999999512009999999999927

367253 rows × 35 columns

In [25]:

supp_source_df = supp_source_df[supp_source_df["__GISJOIN"].isin(bgp1990.unique())].copy()
supp_source_df

Out[25]:

	GISJOIN	YEAR	ANRCA	AIANHH	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	...	URBRURALA	URB_AREA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001	__GISJOIN
57044	G1000010904440725004239999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	2	NaN	9999	NaN	BG 1 (pt.)	179	48	83	168	G100001090444072500423009999999999921
57045	G1000010904444443004229999999999926	1990	99	NaN	9999	9999	9999	9	NaN	6	...	2	NaN	9999	NaN	BG 6 (pt.)	211	62	81	86	G100001090444444300422009999999999926
57046	G1000010904446126504229999999219011	1990	99	NaN	9999	9999	9999	9	NaN	1	...	1	Dover, DE	2190	NaN	BG 1 (pt.)	1848	490	615	694	G100001090444612650422009999999219011
57047	G1000010904446126504229999999219012	1990	99	NaN	9999	9999	9999	9	NaN	2	...	1	Dover, DE	2190	NaN	BG 2 (pt.)	67	17	21	22	G100001090444612650422009999999219012
57048	G1000010904446148004229999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	2	NaN	9999	NaN	BG 4 (pt.)	924	261	307	315	G100001090444614800422009999999999924
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
57816	G1000050935529999905159999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	2	NaN	9999	NaN	BG 2 (pt.)	1127	316	399	446	G100005093552999990515009999999999922
57817	G1000050935529999905159999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	2	NaN	9999	NaN	BG 3 (pt.)	228	74	99	108	G100005093552999990515009999999999923
57818	G1000050935529999905159999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	2	NaN	9999	NaN	BG 4	808	214	269	310	G100005093552999990515009999999999924
57819	G1000050935529999905169999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	2	NaN	9999	NaN	BG 1	1100	323	401	445	G100005093552999990516009999999999921
213548	G3400330106101060002049999999916014	1990	99	NaN	9999	9999	9999	9	NaN	4	...	1	Wilmington, DE--NJ--MD--PA	9160	NaN	BG 4 (pt.)	1049	215	380	400	G340033010610106000204029999999916014

777 rows × 35 columns

In [26]:

supp_source_df.drop(columns=["__GISJOIN"], inplace=True)
supp_source_df.reset_index(drop=True, inplace=True)
supp_source_df

Out[26]:

	GISJOIN	YEAR	ANRCA	AIANHH	AIANHHA	RES_ONLYA	TRUSTA	RES_TRSTA	BLOCKA	BLCK_GRPA	...	STATEA	URBRURALA	URB_AREA	URB_AREAA	CD103A	ANPSADPI	ET1001	EUD001	EUO001	ESA001
0	G1000010904440725004239999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	10	2	NaN	9999	NaN	BG 1 (pt.)	179	48	83	168
1	G1000010904444443004229999999999926	1990	99	NaN	9999	9999	9999	9	NaN	6	...	10	2	NaN	9999	NaN	BG 6 (pt.)	211	62	81	86
2	G1000010904446126504229999999219011	1990	99	NaN	9999	9999	9999	9	NaN	1	...	10	1	Dover, DE	2190	NaN	BG 1 (pt.)	1848	490	615	694
3	G1000010904446126504229999999219012	1990	99	NaN	9999	9999	9999	9	NaN	2	...	10	1	Dover, DE	2190	NaN	BG 2 (pt.)	67	17	21	22
4	G1000010904446148004229999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	10	2	NaN	9999	NaN	BG 4 (pt.)	924	261	307	315
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
772	G1000050935529999905159999999999922	1990	99	NaN	9999	9999	9999	9	NaN	2	...	10	2	NaN	9999	NaN	BG 2 (pt.)	1127	316	399	446
773	G1000050935529999905159999999999923	1990	99	NaN	9999	9999	9999	9	NaN	3	...	10	2	NaN	9999	NaN	BG 3 (pt.)	228	74	99	108
774	G1000050935529999905159999999999924	1990	99	NaN	9999	9999	9999	9	NaN	4	...	10	2	NaN	9999	NaN	BG 4	808	214	269	310
775	G1000050935529999905169999999999921	1990	99	NaN	9999	9999	9999	9	NaN	1	...	10	2	NaN	9999	NaN	BG 1	1100	323	401	445
776	G3400330106101060002049999999916014	1990	99	NaN	9999	9999	9999	9	NaN	4	...	34	1	Wilmington, DE--NJ--MD--PA	9160	NaN	BG 4 (pt.)	1049	215	380	400

777 rows × 34 columns

In [27]:

supp_source_df.to_csv(
    "../testing_data_subsets/%s_blck_grp_598_103.csv.zip" % source_year
)