Patterns in Transportation, Income, and Poverty Across All 50 U.S. States: A 2015 Census-Based Analysis


Executive Summary

This report aims to analyze the US Census Demographic Data from 2015 to identify patterns into transportation quality, income, and poverty trends across the United States. Specifically, we aim to:

  1. Define and measure transportation efficiency.

  2. Contrast income and poverty levels across 50 states and different geographic locations.

  3. Explore regional, divisional, and state-level patterns across three key metrics: Income, Transportation Scores, and Poverty Rates


    Key Themes and Summary of Findings

Findings and Analysis


I - Identifying States with the Best Transportation


  1. What defines the "best transportation"?

    In the dataset provided, what I considered factors for calculating the transportation scores

    are:

    1. Positive Factors (Added)

      • Transit (weight 0.4): Strong public transit infrastructure is critical, so it's given the highest weight. High transit usage indicates effective public transport.

      • Walk (weight 0.2): High walkability shows good pedestrian infrastructure.

      • OtherTransp (weight 0.2): Includes bikes, rideshare, or other non-traditional modes.

      • Carpool (weight 0.2): Shared private transport is more efficient than single- occupancy driving.

      • Drive (weight 0.3): Recognizing its role in providing efficient transportation in areas where public transit is limited or impractical.


    2. Negative Factors (Subtracted)

      • MeanCommute (weight 0.5): Longer commute times are penalized as they indicate inefficiency in the transportation system. Shorter commute times indicate efficiency.

  2. Which states have the highest transportation scores?


    The five states with the highest transportation scores are North Dakota, South Dakota, Nebraska, Kansas, and Iowa.


  3. What is the average transit in each state?


    The state that has high transit usage is New York, which indicates that their public transportation is adequate. In contrast, Alabama and Mississippi are tied for having the worst transit usage.


  4. What is the average commute time in each state?


    New York, Maryland, and New Jersey are the top three states with the worst commute times. The states of South Dakota, North Dakota, and Montana have the shortest commute times.


  5. Are there outlier counties affecting the data?

    Upon checking the dataset in Jupyter Notebook, I found a total of 6.66% of outliers – kindly check the figure below. However, I did not do any cleaning or imputation since it is expected to have significant disparities in numbers in census data, especially for a country as big as the USA.


    Since the dataset is aggregated from county to state levels, individual county-level commute times and their impact on state-level averages cannot be directly analyzed. However, we can still draw meaningful insights based on state-level averages.

    New York stands out with exceptionally high transit usage (27.68%) but a relatively low transportation score (14.165), which may be influenced by densely populated metropolitan areas like New York, where commute times are known to be longer. Also, Maryland and New Jersey exhibit the most extended commute times despite moderate transit usage, indicating potential challenges in transportation infrastructure or system inefficiencies. These states mentioned can be considered to have outliers in these columns/metrics.

  6. How should you aggregate all the county data to represent the state effectively?


    The county-level data is transformed into state-level data by applying population- weighted means for numeric metrics. This ensures that states with larger populations have more influence on the aggregated values, accurately reflecting the population's characteristics.

    I did the aggregation in Jupyter Notebook using Python instead of Tableau's default aggregation. This way, not a single county is overrepresented or underrepresented. I can create calculated fields in Tableau, which will complicate the process as Tableau lacks complex calculations and formulas.

    As for the aggregation, I included columns - TotalPop, Transit, Walk, OtherTransp, Income, Poverty, Construction, ChildPoverty, and MeanCommute. I also added these columns: IsCoastal, TransportationScore, Division, and Region. I have included the formula I used in aggregating these columns and all the other processes I did with the dataset before loading it to Tableau for visualization in the Jupyter Notebook report.

  7. Which states have the best transportation? Is it the highest percentage of transit use? Is it the lowest mean commute times?



Findings based on these results are as follows:



States with robust transportation infrastructure tend to have lower commute times and minimal reliance on public transit, highlighting a direct link between infrastructure quality and commute efficiency. In contrast, highly urbanized states rely heavily on public transit, which often correlates with longer commute times due to overcrowding and urban congestion. This contrast underscores a rural vs. urban divide, where rural states generally show lower transit usage and shorter commute times. In contrast, urban states face the dual challenges of higher transit dependency and

extended commute durations. These findings emphasize the importance of urban planning and efficient public transit systems to address the unique challenges of densely populated areas. Targeted investments in improving urban transit systems and addressing infrastructure bottlenecks could significantly reduce commute times and enhance overall transportation efficiency in high- density regions.


Relationship Between Transit Usage and Transportation Score Using a Simple Linear Regression Model


As shown in the regression equation, for every unit increase in transit usage, the transportation score decreases by 0.168362, indicating a negative relationship between the two variables. The p-value is 0.0001268, which is much smaller than 0.05. This suggests that the relationship between Transit Usage and Transportation Score is statistically significant. The t-value for Transit Usage is -4.16665, which strongly indicates the variable's influence on the Transportation Score.

The data suggests that states with higher transit usage tend to have slightly lower transportation scores. This might indicate challenges or limitations in transportation infrastructure in areas with high transit reliance. The statistically significant p-value confirms the reliability of this result. Further investigation might be needed to explore why transit usage negatively correlates with transportation score. Possible reasons include high dependency on public transit, overshadowing broader transportation infrastructure, or urban congestion challenges in high-transit areas.

Relationship Between Transit Usage and Transportation Score Using a Simple Linear Regression Model



As shown in the regression analysis above, it tells us that for every unit increase in mean commute time, the transportation score decreases by 0.38205. This indicates a strong negative relationship between mean commute time and transportation score. The p-value is < 0.0001, showing the relationship is highly statistically significant. There is strong evidence that mean commute time influences the transportation score. The negative coefficient, the Coefficient for Mean Commute (-0.38205), quantifies the decline in the transportation score as commute time increases. This suggests that longer average commute times are associated with worse transportation scores. The standard error for the mean commute coefficient is 0.0202635, indicating precise and reliable estimates. The t-value (-18.8541) for mean commute time is very large in magnitude, showing that it strongly affects the transportation score.

The analysis reveals a strong negative correlation between mean commute times and transportation scores. States with longer commute times tend to have lower transportation scores, likely reflecting greater transportation inefficiencies or challenges. For example, South Dakota (SD) has a mean commute of 16.65, and the transportation score is 18.334. This result is consistent with the regression model, indicating that shorter commute times are associated with higher transportation scores. These results imply that longer commutes may signify congestion,

inadequate road networks, or urban sprawl, which can lower transportation efficiency and user satisfaction. States with longer commute times may need to invest in better infrastructure, public transit options, or urban planning to improve transportation scores. However, this model only shows correlation, not causation. Other factors (e.g., public transit availability and urban density) could also influence the transportation score.


II - Income and Poverty Across America


  1. How do income and poverty rates vary across the United States?

    1. Income

      According to this bar chart, Maryland has the highest income of all states. On the other hand, Mississippi has the lowest income.


    2. Poverty Rates

      As for poverty rates, it appears that Mississippi has the worst poverty rate, and New Hampshire has the lowest poverty rate.



      Is there a relationship between poverty and income levels? The short answer is yes. The scatter plot above illustrates this relationship. There is a strong negative correlation between income and poverty rates. As income increases, poverty rates decrease significantly. States with lower incomes (closer to $40,000 or less) tend to have higher poverty rates (>15%). States with higher incomes ($60,000 and above) have lower poverty rates (<10%). The equation above tells us that for every $1,000 increase in income, poverty rates decrease by approximately 0.2855 percentage points. Additionally, the p-value for income (<0.0001) indicates that the relationship between income and poverty is highly statistically significant.

      The graph and regression model reveals a statistically significant and strong negative relationship between income and poverty rates. As income rises, poverty decreases, making income a critical factor in addressing poverty at the state level.


  2. Do counties with more construction experience more or less poverty?

    1. Although this question cannot be answered directly since the data is aggregated from county-level data to state-level data, we can still get insights from the state-level data.


      Based on this chart, Wyoming had the most construction activities when the census data was collected. On the right side, we can see that Massachusetts has the least construction activities.


      This simple regression line analysis illustrates the relationship between Construction Employment (x-axis) and Poverty Rates (y-axis). It shows a positive correlation between construction activities and poverty rates, meaning that as construction activities increase, poverty rates also tend to rise slightly. According to the model, the equation above shows that for every 1% increase in construction employment, poverty increases by 0.330371 units. However, the p- value for the construction coefficient is 0.176503, which is not statistically significant (typically, p < 0.05 is considered significant). This indicates that the observed relationship may be due to chance. The t-value for construction activities (1.36217) is low, indicating a weak relationship between these variables.

      With these said, it can be concluded that States with higher poverty rates are often associated with lower overall income levels and other factors, regardless of construction activity levels.


  3. Do counties near the coast experience more or less income?

This is another question that we cannot answer directly due to the limitation I have stated above. On a positive note, we can get valuable insights from the state-level data instead.



These charts tell us that the total average income for all fifty states is US$55,311, and the total income for coastal states surpassed the average income of those living in inland states. People in the inland states have an average income of US$52,066, while coastal dwellers have an average income of US$ 56,743 yearly. This could be because of their geographic advantages, which resulted in having several ports and being the first choice of most international companies.


III – Regional, Division, and State-Level Patterns


  1. Which region (North, South, East, West) consistently perform better regarding income, transportation, and poverty? Identify two states that bested the others in these metrics.




    For this question, we have different results for each metric. They are as follows,

    1. The region with the best transportation score is the Midwest, and the two performing states are North Dakota and South Dakota. The average transportation score for all 50 states is 15.560. Midwest's average transportation score is 16.818. North Dakota has 18.430, and South Dakota has 18.334. The Midwest's higher average transportation scores likely result from efficient road networks and lower population density, reducing

      strain on transportation systems. The Northeast faces challenges in transportation infrastructure, potentially due to aging systems, higher urban congestion, and a reliance on public transit.

    2. As for the income, the average income is $55,311, with most states close to or slightly below this line. The Northeast shows the highest average income among all regions, with states like Connecticut and Massachusetts exceeding $75,000. However, the state with the highest income belongs to the South region, Maryland, with an average income of US$77,265. The second state is from the Northeast region, New Jersey, with an average income of US$73,166. Economic disparities among regions are apparent. The Northeast benefits from higher-paying industries like finance and technology, contributing to its higher average income. The South's lower income levels could reflect structural economic issues, such as reliance on lower-paying industries or limited access to high-paying jobs. One solution to this is targeted economic development and job training programs in underperforming states, particularly in the South, which could help bridge regional income gaps.

    3. Lastly, the poverty rates. The worst poverty rates belong to the southern region, with an average of 17.08, while the Northeast region has the lowest poverty rate, with an average of 12.31. All fifty states have an average poverty rate of 14.81. As for the states, the lowest poverty rate belongs to New Hampshire, which belongs to the Northeast region, and the other state is from the South region, which is Maryland, with an average of 10.02. Addressing poverty in Southern states should be a priority, with targeted interventions to improve education, job opportunities, and access to resources. The South shows the highest poverty rates among all regions, with many states exceeding the national average.

  2. Which division consistently performs better regarding income, transportation, and poverty? Identify two states in this division that bested the others in these metrics.




    We have different results for these metrics. I have outlined them below.

    1. The visualization above shows that West North Central (Midwest) outperformed the other divisions in transportation score. North Dakota and South Dakota are the two states that bested the different states in this division.

    2. Next, the income is analyzed according to divisions. The division that bested the other eight divisions is the Pacific, with an average income of US$64,057, and the top states from this region are Alaska and Hawaii. The chart above highlights significant regional income-level disparities, with the Northeast and Pacific Divisions leading and the East South-Central Division lagging. Addressing these disparities through economic development and infrastructure investment in low-income regions can help balance national income distribution.

    3. The last metric is poverty. East South Central (South) leads the country in poverty rates with an average of 16.52. Conversely, New England (Northeast) has the lowest poverty rate, averaging 11.76. This result highlights stark regional disparities in poverty rates, with the South lagging significantly behind other regions. The Northeast and Midwest lead in economic stability, while the West shows mixed performance. Addressing poverty in underperforming divisions through targeted interventions is critical to balancing national disparities.

Conclusion


The analysis of transportation, income, and poverty rates across regions, divisions, and states highlights significant geographic disparities in infrastructure and economic well-being. States in the Midwest, such as North Dakota, South Dakota, and Nebraska, consistently achieve the highest transportation scores due to well-maintained infrastructure and shorter commute times. On the other hand, urbanized states like New York exhibit high transit usage but lower transportation scores, indicating challenges related to congestion and aging infrastructure.

In terms of income, the Northeast leads the nation, with states such as Massachusetts and Connecticut benefiting from access to high-paying industries like finance and technology. Coastal states generally demonstrate stronger economic performance than inland states, as illustrated in the comparison of coastal and inland states. However, exceptions like South Carolina, which has lower income levels despite being coastal, suggest that additional factors contribute to these disparities.

Poverty rates are notably higher in the South, with states like Mississippi and Alabama ranking as the most economically challenged. In contrast, the Midwest maintains lower poverty rates, reflecting more substantial economic stability and robust infrastructure. The consistent underperformance of the South across all metrics highlights the region's need for targeted economic and infrastructural interventions to improve living conditions and reduce inequality.

The Midwest is the most balanced region, achieving high transportation scores, moderate income, and low poverty rates. Meanwhile, the Northeast excels in income but struggles with transportation challenges. The South, however, faces significant obstacles across all analyzed metrics, reinforcing the need for focused strategies to address systemic issues in this region.

These findings underline the importance of regional policies tailored to address specific needs. Improving transportation infrastructure in states with low scores and long commute times could enhance economic opportunities and reduce inefficiencies. Poverty alleviation efforts should focus on regions like the South, emphasizing economic diversification and job creation. The income gap between coastal and inland states further suggests the need for equitable economic policies to bridge this divide.


References:

Not Applicable