2. North Dakota's Mobility Profile & Concentrations of Need

Knowledge of the spatial distribution of individuals with mobility challenges is essential to the appropriate design of transportation policy. Furthermore, if one accepts the notion that the type and quality of services provided to those with greatest need serves as a measure of society itself, an understanding of these characteristics including the location of these persons is of paramount importance. A related approach would be to think of the location of these individuals as mobility demand nodes for governmental transportation facilities and services, in conjunction with the free market as the supply side of the mobility market.

This chapter is intended to describe various attributes of the North Dakota residents and to present a novel method of calculating the relative immobility of a particular region's inhabitants. The aim of the latter section is to aid in prioritizing the level of service that a region may need. We begin by presenting the mobility profile of North Dakota's residents.

2.1 Geographic Levels of Interest

To get an accurate picture of the mobility challenges facing North Dakota's residents, data is assembled at three geographic levels. These include North Dakota's eight planning regions, 53 counties, and 14 largest cities. Because their combined populations make Beulah-Hazen a large city, by North Dakota standards, these neighboring communities are considered one city for the purposes of this report.

The state is divided into eight super-county economic planning regions as shown in Figure 2.1. Those interested in additional regional level analysis of transportation issues should refer to a study by Hegland, Mielke, and Miller (2004).

Figure 2.1 North Dakota Economic Planning Regions and Hub Cities

Source: Enhancing Passenger Mobility Services in North Dakota through Increased Coordination, 2004

2.2 Urban Areas in North Dakota

The economic, political, and social structure of the State of North Dakota is relatively dichotomous in nature, with significant differences, both real and perceived, between rural and urban communities. In the context of this study, the definitions of these terms are quite different than those more commonly accepted. We do not define urban locales as metropolitan areas, but rather are referring to communities with 4,500 people or more. This is a smaller number of people than that used at the federal level by an order of magnitude, but to those familiar with North Dakota, the break is intuitive.

These cities provide the retail, medical, and governmental services that meet the regular needs of the vast majority of individuals including those with mobility challenges. As there are no cities in North Dakota with a population of 3,000 to 4,000 residents, this provides for a natural break for those cities selected and those omitted. These cities: Beulah-Hazen, Bismarck, Devils Lake, Dickinson, Fargo, Grafton, Grand Forks, Jamestown, Mandan, Minot, Valley City, Wahpeton, West Fargo, and Williston are displayed graphically in Figure 2.2.

Figure 2.2 Selected Cities

Source: Small Urban & Rural Transit Center, 2005

2.3 Population Mobility Characteristics

There are a number of possible attributes and data sources that describe pertinent aspects of a community's mobility needs. Much of this information is gathered by survey sampling which, if done correctly, produces results that describe the population. However, a great deal of this analysis, for example the Nationwide Personal Transportation Survey (NPTS) collected by the Federal Highway Administration (FHWA), is focused at either the regional or national level. This often results in little or no data being collected from within North Dakota, limiting the descriptive power of these analyses for the state or more local levels.

Making use of data available from the 2000 decennial census addresses this concern as it relies on actual population counts for the entire nation and is complemented by a more in-depth survey. This survey is uniformly completed by approximately one in six scientifically selected households across the nation. The use of this data also remedies challenges that will arise if the analysis is to be repeated in the future because there is consistency in questions from one census to the next.

In this study, six populations are of interest: the elderly, elementary and high school students, low-income individuals, the disabled, minorities, and individuals in households without vehicles. Data on these attributes, classified by city, county, and region, are presented in tables 2.1 to 2.8.

Population estimates for North Dakota arranged by region and county are presented in Table 2.1. Given its size of 70,704 square miles and population of 634,366, North Dakota is considered a rural state even though more than half its residents live in urban areas of 2,500 or more people. With just over half of its population, 322,347 individuals, located in Burleigh, Cass, Grand Forks, and Ward counties, there is clearly a urban-rural dichotomy in the state. Rural areas in the eastern part of the state tend to be more densely populated than in west.

Region V, which includes the North Dakota portion of the Fargo-Moorhead metropolitan area, is the largest region with an estimated population of 166,678. It is followed by Region VII, a 10-county area with 131,764 residents that includes the Bismarck-Mandan metropolitan area. The smallest region by population, northwest North Dakota, is Region I at 26,985. It is followed by sparsely populated Region VIII which is home to 36,961 residents in the southwestern part of the state.

The population of seniors, arranged by region and county, is presented in Table 2.2. Approximately 15 percent of all North Dakotans, or about 94,478 persons, are seniors. Cass County with 11,901 seniors has the largest elderly population of any county in the state. It is followed closely by Burleigh County with 8,640 and Ward County with 7,341 seniors respectively. The lowest senior populations are in the western part of the state where Billings County and Slope County, home to 142 and 137 seniors, are located. Sioux County with 226 seniors has the lowest percentage of elderly residents with only 6 percent age 65 or older. McIntosh County has the highest percentage of elderly citizens, 37 percent.

The dynamics underlying these figures, especially when focus is placed on the proportion of elderly to total population, are of importance. The emigration of young residents seeking education and employment opportunities from rural areas and the emigration of old residents to adequate health care are notable. Areas with large Native American populations have higher birthrates which partially explains the lower percentages seen in Rolette and Sioux counties.

The number of students enrolled in K-12 institutions is presented in Table 2.3. Given the high correlation between the number of residents of school age and actual enrollees, the data presents a good proxy for individuals ages 5 to 18. Approximately 20 percent or 123,939 residents are enrolled in K-12 schools statewide. This ratio is relatively constant across the state. One exception is locations with large Native American populations, especially Sioux, Benson, and Rolette counties, which have a higher percentage of youth than other areas. The highest student populations are located in Cass, Burleigh, Grand Forks and Ward counties, having 20,881, 12,849, 11,757, and 11,203 students respectively.

There is a significant difference between the numbers reported by the Census and the State Department of Public Instruction which reports that there were 114,000 students in 1999. No reason is identified for the difference.

The low-income population arranged by region and county is presented in Table 2.4. About 11 percent of the state's population, 72,714 persons, reported having incomes below the poverty line. The numbers vary greatly across the state with the highest absolute numbers of low-income individuals in Cass, Grand Forks, Ward, and Burleigh counties. These same areas, except for Grand Forks County, also have some of the lowest proportions of low-income individuals.

Benson, Rolette, and Sioux Counties have relatively high proportions of low-income individuals with 29, 30, and 38 percent of their residents with incomes below the poverty line. A number of counties located in the central and south-central part of the state: Emmons, Grant, Kidder, and Sheridan also have relatively large low-income populations.

The number of individuals with go-outside-the-home disabilities is presented in Table 2.5. It is important to note that these individuals self identify and an individual who identifies themselves may have the same abilities as some who identifies themselves otherwise. Statewide there are 32,258 people, approximately 5 percent of state residents, who identify themselves as having such a disability. This percentage is relatively constant across the state.

Table 2.6 presents minority population counts by region and county. On a percentage basis, large Native American populations are present in McKenzie, Mountrail, Benson, Rolette, and Sioux Counties which have minority populations of 1,299; 2,255; 3,423; 10,239; and 3,464 respectively. Counties with urban centers are also home to large populations of minorities, many of whom are Native American though sizable African-American and Hispanic populations as well as new Americans and refugees are present. Cass, Burleigh, Ward, and Grand Forks counties have minority populations of 4,578; 3,210; 2,656; and 2,443 respectively.

The estimated number of individuals without vehicles in their households, arranged by region and county, are presented in Table 2.7. These numbers have been calculated using data from the U.S. Census by multiplying the number of households without vehicles times the average household size. Many of these households consist of elderly individuals whose household size is significantly smaller than the average. At the same time, many households without vehicles have low-incomes which traditionally have larger household sizes than the average. Data on actual personal automobile registration are presented in Chapter 4.

Statewide there are an estimated 42,739 individuals without vehicles; this is approximately 7 percent of the state's population. Low values are present in the western part of the state such as Slope County where every individual has access to a vehicle. Large numbers of individuals without vehicles are present in Cass County where 8,572 people are without access.

Data for the demographic categories presented thus far are presented for the selected cities in Table 2.8. Though demographic characteristics of these areas are relatively similar, there are some differences. Valley City, Devils Lake, and Grafton have relatively high proportions of seniors with 24, 22, and 20 percent of the total population over the age of 65. Only 6 percent of West Fargo's population is in that category. Sixteen percent of Devils Lake's population reported having low incomes while West Fargo, Beulah-Hazen, and Bismarck have low percentages of 6, 7, and 8 percent.

The size of disabled populations is similar across cities with the exception of Grafton, home to the State Developmental Center, where 11 percent of the population has a going-outside-the-home disability. Devils Lake has a relatively high number of Native Americans, 11 percent, while only 3 percent of Jamestown's and Valley City's populations are minority. Devils Lake also has a high proportion of individuals without vehicles, 14 percent, while West Fargo and Beulah have smaller ones. The relative size of the student population is quite similar with the exception of Beulah-Hazen where nearly 30 percent of the population consists of students enrolled in K-12.

2.4 Population Projections

The values presented thus far represent a snapshot of the state's people that is now more than 5 years old with the exception of the total population estimates which are for July 1, 2003. The dynamic nature of the mobility profile is important to policy makers as their goal is to address current needs not those of circumstances past. To understand the composition of the future population of the state, projections are made.

The basic concept behind population projections is that past behaviors, such as birthrate, mortality, and migration patterns, will continue in the future. This ignores the impact of public policy as well as the changes that would result from a change in economic climate.

Table 2.9 presents the projected population of North Dakota youth, those ages 5 to 19; seniors, those 65 and older; and the total population every five years from 2005 to 2025, in addition to the decennial census counts from the year 2000. The values are different than other projections that have been made, such as those by Rathge, Clemenson, Danielson (2002); however, this is not unexpected as such calculations often use different data and require subjective decisions regarding the model's formal construction. Of primary interest is the decrease in the number of youth while at the same time the number of seniors increases dramatically.

Looking at the total population at the regional level provides a more local understanding of the projected population dynamics within the state over the next 20 years. Regions with larger urban areas, such as Region V, which includes the city of Fargo, and Region VII, in which Bismarck is located, are projected to experience an increase in population. The remaining regions are projected to have relatively stable or slightly declining populations. Those interested in more in-depth discussion and data on North Dakota population projections are referred to the study by Rathge, Clemenson, Danielson (2002) or a forthcoming paper that focuses on populations that face mobility challenges by Ripplinger.

2.5 Concentrations of Need – Mobility Needs Index

Developing an accurate, low-cost, reliable method of measuring mobility needs in a community is difficult. A generally accepted methodology does not currently exist, in part because any model adopted would likely be subjective in nature and vulnerable to some degree of criticism. An original theoretical model could be constructed; however, the cost of collecting, manipulating, and analyzing the data would almost certainly limit its pragmatic value. A method of ranking regions based on readily available data is used instead. This Mobility Needs Index, as it is referred to, is intended to provide an immediately intelligible measure of mobility needs.

It is important to note that this methodology is only an attempt to measure needs associated with identifiable demographic groups. This measurement does not suggest that all related needs are unmet. To the contrary, some cities may have systems and services in place that are satisfying many of their residents' mobility needs.

2.6 The Model

The Mobility Needs Index ranks the relative mobility needs of each of n predetermined geographic regions using a three step process that make use of m subjectively chosen variables that are related in some way to immobility. The first step scales the data value of a particular characteristic, v_ij, where i indexes the n regions and j indexes the m variables, by the physical area of the region, a_i, as described by (1). This data, arrayed in a n-row vector, _j, is then used to determine the quartile contained in the vector, _j, to which a particular region-variable combination belongs (2). A simple average of a region's m quartile values completes the three step process (3).

j = Q(j)

for j=1,2,...m

(2)

Ii =	qij	for i=1,2,...n	(3)
	m

The process thus ranks all regions on a scale of one to four with higher values identifying areas with greater mobility needs.

2.7 The Data

Determining what characteristics of a community affect its relative degree of mobility is an exercise in subjectivity. The cost of quality data collection and compilation is another concern. By making use of data collected by decennial census the latter apprehension is addressed. A side benefit is that this data is collected using a single instrument. This eliminates the need to accommodate the challenge of making use of data drawn from various samples taken at different times. The factors that were deemed important are those discussed previously in the chapter: the population of elderly, elementary and high school students, low-income individuals, the disabled, minorities, and individuals in households without vehicles.

Unfortunately, data on these populations are not readily available in cross tabulations so double counting does occur. However, by making the assumption that there is a proportional relationship for those with mobility challenges this may actually be a benefit. This would assert that, among other things, a low-income senior has twice the need of an individual in a household without an automobile. The model also requires that the marginal impact of an additional individual with a particular characteristic has the same impact on a community's mobility needs, which is clearly erroneous.

2.8 Mobility Needs Index Values

The values calculated for North Dakota's 53 counties are presented in Table 2.11 and Figure 2.3. The results are relatively intuitive as highly populated counties, such as Burleigh, Cass, Grand Forks, and Ward, have high values. Counties with relatively high minority populations have high values as is the case with Benson, Rolette, and Sioux.

As indicated earlier, this index is an attempt to measure concentrations of mobility needs. Resulting indices should not be interpreted to mean that all related needs are unmet; this is certainly not the case.

Figure 2.3 Mobility Needs Index Map I (including select cities)

Source: Small Urban & Rural Transit Center, 2005

2.9 Removing Selected Cities

The presence of cities is hypothesized to have a definitive, positive effect on a region's Mobility Needs Index Value. In order to isolate this effect, the 14 selected cities identified previously in the chapter were removed, and value for the state's 53 counties were re-tabulated, with the results shown in Table 2.12 and Figure 2.4.

Figure 2.4 Mobility Needs Index Map: North Dakota Counties Excluding City Data

Source: Small Urban & Rural Transit Center, 2005

A side by side comparison of Mobility Needs Index Values is presented in Table 2.13. Though there is a noticeable effect, the results were not a strong as expected. There appears to be a positive effect on the values in the counties with large Native American populations. Predominantly urban counties still had relatively high Mobility Needs Index Values while the opposite behavior was observed in rural counties, particularly those in western North Dakota. These results weakly support the hypothesis that the non-urban parts of a county with a large city are relatively similar to counties without ones. This is likely due to the impact urban areas have on adjacent, less urbanized areas.

References

• Hegland, Gary, Mielke, Jon, and Jim Miller, A Plan for Transportation Coordination in North Dakota, Small Urban & Rural Transit Center, Fargo, North Dakota, August 2004.
• Rathge, Richard, Clemenson, Mandy, and Ramona Danielson, North Dakota Population Projections: 2005 to 2010, North Dakota State Data Center, Fargo, North Dakota, September 2002.
• Ripplinger, David, The Demographics of Mobility in North Dakota, Small Urban & Rural Transit Center, Fargo, North Dakota, Forthcoming.