Managing an Identity Crisis: Forum Guide to Implementing New Federal Race and Ethnicity Categories
NFES 2008-802
October 2008

Case Study - North Dakota's Race/Ethnicity Conversion Matrix: A Homegrown Bridging Methodology

Case Study
case study icon

In late 2004, North Dakota took the step to implement some of the 1997 race and ethnicity data collection standards. The state increased the number of race categories to five groups and introduced the option to select more than one race.  Having taken this initiative, however, the state had to solve the problem of how to report its race and ethnicity data to EdFacts and the Department of Education, since both organizations still requested, and continue to request, race and ethnicity under the old standards.  To leap this hurdle, the North Dakota Department of Public Instruction (NDDPI) developed the matrix on the next page, which is essentially a bridging methodology used to convert race and ethnicity data collected under the new guidelines into data in line with the old standards, and therefore, in the form desired by the Federal government.

A diagram of the state's collection scheme follows this discussion. On the left side of the schematic are the new race and ethnicity categories.  Using the matrix, these codes pass through the diagram in the center from top to bottom in their conversion into the old categories on the right, which are reported to the federal government.  To convert multirace individuals, the first component race arrived at is the category to be used for conversion.

The matrix is simply a schematic representing the procedures followed by the state's online reporting system program and by programmers that process individual requests.  To illustrate these processes, let's walk through some examples:

If a student is Hispanic in the new collection, the matrix user moves from top to bottom and finds the diamond on the left side of the matrix labeled “Hispanic?”  He or she then moves to the right to find the category to which this new label should be converted.  In this case, the corresponding old category is “Hispanic” since that category is included in both the new and old schemes—no change in categorization is necessary. 

For another student who is Pacific Islander under NDDPI's new scheme, the system leads us to the sixth diamond down labeled “Pacific?”  From there, the conversion path leads right to the Asian category.  This is because in the old scheme, Asian encompasses the Pacific Islander group.  A direct conversion is therefore possible.

Finally, let us take the case of a multirace individual whose component races are Black and White.  Moving from top to bottom looking for either of those races, the first category we reach will be “Black,” and therefore, that is the race to be used for conversion.  We simply disregard the other component race – in this case, White.  Therefore, we find that this individual, who identified as Black/White under the new scheme, is now converted to Black under the old scheme.  So, North Dakota would report him or her as Black to the federal government.

Note that, in this matrix, “Hispanic” is included with the race categories rather than isolated as a separate demographic descriptor as is called for by the new 1997 standards.  An individual who identifies as “Hispanic” and one of the race categories, in this scheme, would be identified simply as “Hispanic” (see the multirace example below) according to the old guidelines.

The organization of the race and ethnicity categories in the matrix was guided by a set of criteria.  First, “Hispanic” was placed before the race categories because it is, in a sense, dominant in both the new and old schemes.  That is, a person who selects both “White” and “Hispanic,” for example, is categorized as “Hispanic” in both schemes.  So, placing Hispanic before the other races assures that all of those who select “Hispanic” are counted as such and not classified as belonging to one of the race groups.  Second, at the time of design, it was the preference of NDDPI to claim a high percentage of minorities.  So, the minority groups were placed above “White” in the matrix in an order that mirrors the sizes of those racial populations in the state from largest, American Indian, down to the smallest, Asian.  The one exception to this rule is the multirace group who select both Native American and Pacific Islander.  This is because it was the designing researchers' understanding that Native Hawaiians commonly select that race combination.  The decision was made to place that discrete group before the other races so that particular combination can be identified before the data system mistakenly lumps these individuals in as American Indian.

The matrix also includes a category called “Other.”  This category was added to the matrix for use in Adequate Yearly Progress reports at both the state and district levels.  “Other” has not been used for federal reporting.   Any individuals assigned to this category under the new scheme are converted to “White” under the currently used scheme.

North Dakotas Collection Scheme