Guidelines for using Racial and Ethnic Groups in Data Analyses
Updated: July 2003
Purpose
Background and Context
Definition of Minimum Categories
Guidelines
Implementation Schedule
Data Collection
Data Tabulation
Data Presentation
Additional Information
Special Issues
Estimates of "More than One Race" in Washington State
Bridging Ratios for Trend Analysis
Bridging Methods
Sample Use of Statistical Methods
Recommended Method
Additional Information
Converting Multiple to Single Race: Washington State Department of Health Discussion Paper
Guidelines
For Using Racial and Ethnic Groups in Data Analyses (Word Document)
<<<
Previous <<<
Washington State Department of Health discussion
paper developed December 2002 and updated April 2003
(continued)
Younger people are more likely to have been reported
as having more than one race in the 2000 US census.
Therefore, different methods of allocating from multiple
to single race for younger people will have a relatively
larger effect. Table 5 shows death rates for causes of
death that disproportionately affect younger people.
|
Ages 0 ? 14
|
Ages 15 - 24
|
|
Unintentional
Injury
|
|
|
White
|
Black
|
API2
|
AI/AN3
|
State rate
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
|
Single
|
8.5
|
18.9
|
9.6
|
40.8
|
9.8
|
38.4
|
21.3
|
25.3
|
96.2
|
37.9
|
|
Equal
|
8.3
|
15.4
|
7.8
|
30.4
|
9.2
|
37.6
|
19.1
|
22.5
|
73.8
|
36.5
|
|
Unequal
|
8.3
|
14.6
|
7.8
|
33.6
|
9.2
|
37.6
|
18.6
|
22.5
|
81.3
|
36.5
|
|
Maximum
|
8.0
|
12.8
|
6.6
|
23.5
|
8.7
|
36.9
|
17.1
|
20.2
|
58.7
|
35.2
|
|
% Range
|
5.9
|
32.5
|
31.3
|
42.4
|
10.8
|
3.8
|
19.6
|
19.9
|
39.0
|
7.1
|
|
|
|
|
Suicide
|
|
|
White
|
Black
|
API
|
AI/AN
|
State
rate
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
|
Single
|
0.6
|
0.0
|
0.0
|
0.0
|
0.5
|
12.1
|
9.1
|
8.4
|
18.0
|
11.8
|
|
Equal
|
0.6
|
0.0
|
0.0
|
0.0
|
0.5
|
11.8
|
8.2
|
7.5
|
13.8
|
11.4
|
|
Unequal
|
0.6
|
0.0
|
0.0
|
0.0
|
0.5
|
11.8
|
8.0
|
7.5
|
15.2
|
11.4
|
|
Maximum
|
0.5
|
0.0
|
0.0
|
0.0
|
0.5
|
11.6
|
7.3
|
6.7
|
11.0
|
11.0
|
|
%
Range
|
5.9
|
0.0
|
0.0
|
0.0
|
10.8
|
3.8
|
19.6
|
19.9
|
39.0
|
7.1
|
|
|
|
|
Homicide
|
|
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
|
Single
|
1.1
|
9.5
|
0.0
|
7.4
|
1.5
|
6.9
|
21.3
|
6.7
|
6.0
|
7.5
|
|
Equal
|
1.0
|
7.7
|
0.0
|
5.5
|
1.4
|
6.8
|
19.1
|
6.0
|
4.6
|
7.2
|
|
Unequal
|
1.0
|
7.3
|
0.0
|
6.1
|
1.4
|
6.8
|
18.6
|
6.0
|
5.1
|
7.2
|
|
Maximum
|
1.0
|
6.4
|
0.0
|
4.3
|
1.4
|
6.6
|
17.1
|
5.4
|
3.7
|
6.9
|
|
% Range
|
5.9
|
32.5
|
0.0
|
42.4
|
10.8
|
3.8
|
19.6
|
19.9
|
39.0
|
7.1
|
|
|
|
|
Motor
Vehicle
|
|
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
White
|
Black
|
API
|
AI/AN
|
State rate
|
|
Single
|
3.7
|
5.7
|
4.1
|
14.8
|
4.0
|
25.7
|
6.1
|
13.5
|
66.1
|
24.8
|
|
Equal
|
3.6
|
4.6
|
3.4
|
11.0
|
3.8
|
25.2
|
5.5
|
12.0
|
50.7
|
24.0
|
|
Unequal
|
3.6
|
4.4
|
3.4
|
12.2
|
3.8
|
25.2
|
5.3
|
12.0
|
55.9
|
24.0
|
|
Maximum
|
3.5
|
3.8
|
2.8
|
8.5
|
3.6
|
24.8
|
4.9
|
10.8
|
40.3
|
23.1
|
|
% Range
|
5.9
|
32.5
|
31.3
|
42.4
|
10.8
|
3.8
|
19.6
|
19.9
|
39.0
|
7.1
|
1 The percent range cannot be calculated from the data in this
table because of rounding. For example, the state rate
for suicide in the 0-14 age group is 0.50583 in the
?single? category and 0.45144 in the ?maximum?
category, both of which round to 0.5.
2 Asian and Pacific Islander
3 American Indian and Alaska Native
As can be seen from Table 5
For whites, blacks and API, the percent difference
between the highest and lowest rates are larger for
youth than for all ages combined.
- For whites,
- The percent difference between the highest and
lowest rate is about 6% for the younger age
group and 4% for the older age group.
- The absolute differences remain relatively
small.
- For blacks and APIs,
- The percent difference between the highest and
lowest rate is about 30% for the younger age
group and 20% for the older age group.
- There are not large differences between rates
based on unequal and equal fractional allocation
of people reporting more than one race.
- For most causes of death, the rates remain the
same relative to the rate for the state as a
whole. However, for some causes of death, the
rates for blacks or APIs change relative to the
state depending on which method of developing
rates is used. For example, for unintentional
injury, rates for APIs are similar to those for
the state as a whole using the
"single" method, but lower than the
state as a whole using the "maximum"
method.
- For AI/ANs,
- The differences between the highest and lowest
rates in these younger age groups are similar to
those for all ages combined.
- For some causes of death, the rate for AI/ANs
relative to the overall rate depends on which
method is used to allocate single to multiple
races.
In addition to larger variations for selected age
groups, there may also be a larger range of rates for
counties with higher proportions of people of more than
one race. For example, in King County, the percent
difference between the highest and lowest age-adjusted
death rates for AI/ANs for the causes of death shown in
Table 4 hovers around 50% versus 40% for the state as a
whole.
Conclusions and Recommendations
Based on this information, the Assessment Operations
Group (AOG) of the DOH recommended using unequal
fractional allocation as described above. Subsequent to
this decision, the NCHS bridged population counts became
available. As Table 6 illustrates, at the state level,
these counts were very close to the unequal fractional
allocation method recommended by the AOG.
|
Race
|
age group
|
NCHS
bridged counts
|
DOH
"unequal" counts
|
percent
difference
|
|
Race
|
age group
|
NCHS
bridged counts
|
DOH
"unequal" counts
|
percent
difference
|
|
AIAN1
|
0-4
|
9943
|
10077
|
-1.34%
|
|
Black
|
0-4
|
24012
|
23381
|
2.63%
|
|
AIAN
|
5-9
|
10902
|
11045
|
-1.32%
|
|
Black
|
5-9
|
24183
|
23508
|
2.79%
|
|
AIAN
|
10-14
|
11460
|
11640
|
-1.57%
|
|
Black
|
10-14
|
22364
|
21730
|
2.83%
|
|
AIAN
|
15-19
|
10612
|
10801
|
-1.79%
|
|
Black
|
15-19
|
20151
|
19585
|
2.81%
|
|
AIAN
|
20-24
|
8674
|
8885
|
-2.43%
|
|
Black
|
20-24
|
18478
|
18074
|
2.19%
|
|
AIAN
|
25-29
|
8365
|
8567
|
-2.41%
|
|
Black
|
25-29
|
18003
|
17670
|
1.85%
|
|
AIAN
|
30-34
|
8463
|
8697
|
-2.77%
|
|
Black
|
30-34
|
19026
|
18729
|
1.56%
|
|
AIAN
|
35-39
|
9081
|
9375
|
-3.24%
|
|
Black
|
35-39
|
19729
|
19437
|
1.48%
|
|
AIAN
|
40-44
|
8806
|
9141
|
-3.81%
|
|
Black
|
40-44
|
18119
|
17823
|
1.64%
|
|
AIAN
|
45-49
|
7472
|
7762
|
-3.89%
|
|
Black
|
45-49
|
13855
|
13624.5
|
1.66%
|
|
AIAN
|
50-54
|
5930
|
6197
|
-4.50%
|
|
Black
|
50-54
|
10157
|
9989.12
|
1.65%
|
|
AIAN
|
55-59
|
4104
|
4316
|
-5.17%
|
|
Black
|
55-59
|
6381
|
6283.58
|
1.53%
|
|
AIAN
|
60-64
|
2958
|
3110
|
-5.14%
|
|
Black
|
60-64
|
4608
|
4548.93
|
1.28%
|
|
AIAN
|
65-69
|
2053
|
2161
|
-5.27%
|
|
Black
|
65-69
|
3583
|
3545.15
|
1.06%
|
|
AIAN
|
70-74
|
1383
|
1464
|
-5.88%
|
|
Black
|
70-74
|
2878
|
2840.32
|
1.31%
|
|
AIAN
|
75-79
|
904
|
972
|
-7.58%
|
|
Black
|
75-79
|
2267
|
2240.6
|
1.16%
|
|
AIAN
|
80-84
|
499
|
538
|
-7.92%
|
|
Black
|
80-84
|
1327
|
1310.5
|
1.24%
|
|
AIAN
|
85+
|
397
|
431
|
-8.46%
|
|
Black
|
85+
|
975
|
965.133
|
1.01%
|
|
API2
|
0-4
|
27254
|
29522
|
-8.32%
|
|
White
|
0-4
|
333097
|
331327
|
0.53%
|
|
API
|
5-9
|
28000
|
29877
|
-6.70%
|
|
White
|
5-9
|
362824
|
361479
|
0.37%
|
|
API
|
10-14
|
28476
|
30081
|
-5.64%
|
|
White
|
10-14
|
372536
|
371385
|
0.31%
|
|
API
|
15-19
|
32048
|
33379
|
-4.15%
|
|
White
|
15-19
|
365157
|
364202
|
0.26%
|
|
API
|
20-24
|
32239
|
33163
|
-2.86%
|
|
White
|
20-24
|
330794
|
330064
|
0.22%
|
|
API
|
25-29
|
35766
|
36395
|
-1.76%
|
|
White
|
25-29
|
341518
|
341021
|
0.15%
|
|
API
|
30-34
|
33806
|
34291
|
-1.43%
|
|
White
|
30-34
|
376183
|
375760
|
0.11%
|
|
API
|
35-39
|
33341
|
33718
|
-1.13%
|
|
White
|
35-39
|
421799
|
421420
|
0.09%
|
|
API
|
40-44
|
31134
|
31452
|
-1.02%
|
|
White
|
40-44
|
433078
|
432721
|
0.08%
|
|
API
|
45-49
|
27546
|
27758
|
-0.77%
|
|
White
|
45-49
|
405350
|
405078
|
0.07%
|
|
API
|
50-54
|
21974
|
22081
|
-0.49%
|
|
White
|
50-54
|
353688
|
353482
|
0.06%
|
|
API
|
55-59
|
14768
|
14822
|
-0.37%
|
|
White
|
55-59
|
260252
|
260083
|
0.06%
|
|
API
|
60-64
|
11829
|
11862
|
-0.28%
|
|
White
|
60-64
|
191680
|
191554
|
0.07%
|
|
API
|
65-69
|
9321
|
9346
|
-0.27%
|
|
White
|
65-69
|
161268
|
161173
|
0.06%
|
|
API
|
70-74
|
7653
|
7676
|
-0.30%
|
|
White
|
70-74
|
149027
|
148960
|
0.04%
|
|
API
|
75-79
|
5228
|
5248
|
-0.38%
|
|
White
|
75-79
|
134309
|
134247
|
0.05%
|
|
API
|
80-84
|
2844
|
2863
|
-0.66%
|
|
White
|
80-84
|
93519
|
93477.3
|
0.04%
|
|
API
|
85+
|
1954
|
1963
|
-0.47%
|
|
White
|
85+
|
80759
|
80726.1
|
0.04%
|
1
Asian and Pacific Islander
2 American Indian and Alaska Native
The NCHS and DOH unequal allocation methods also
produced similar estimates at the county level. (Data
available upon request.) Therefore for consistency with
national standards, the AOG's final recommendation was
to use the NCHS counts as the agency standard.
|
