Are more NC grandparents raising grandchildren because of the opioid epidemic? A look at the data.

The opioid epidemic has had devastating impacts in recent years, affecting the lives of those battling addiction as well as their family, friends, and colleagues. Children, too, are deeply affected; they may be separated from parents with substance abuse disorders if their caretaker becomes incarcerated, needs to enter a rehabilitation program, or becomes unable to care for them. First-hand accounts and interviews collected by non-profit groups and the media suggest that nearby family members typically step in first to take care of these children, followed by the foster care system.

Data from the American Community Survey and the Centers for Disease Control, as well as recent research from the Census Bureau supports these findings: the share of grandparents raising grandchildren was often higher in states with higher opioid prescribing rates.[1] (It is important to note that opioid prescriptions are not the only reason why grandparents are raising grandchildren, but are typically associated with higher rates of opioid prescriptions.)

Alabama, Arkansas, Louisiana, and Mississippi—all located in the Southeast—had some of the highest rates of both grandparents raising their grandchildren as well as people being prescribed opioids in 2016. More broadly, opioid prescribing rates are quite high across all southeastern states. Only the state of Virginia (63.4) had a rate lower than the national average of 66.5 prescriptions per 100 people.

Do these trends hold for North Carolina’s counties?

Carolina Demography sought to understand how these trends look at the county-level within North Carolina. Recent data has been released for 2017, resulting in slightly different results than those in the Census Bureau report linked above.

Nationally, the states that have the highest percentage of grandparents raising grandchildren are Mississippi (2.7%), Arkansas (2.2%), Louisiana (2.1%), Alabama (2.1%), compared to the nationwide average of 1.3%. In North Carolina, the statewide average of grandparents caring for their grandchildren is 1.6%.

In North Carolina, several counties exceed the national average of grandparents acting as primary caregivers for their grandchildren as well as the statewide rates listed above. Several of these also had high opioid prescription rates.

The Sandhills region—which is comprised of Bladen, Columbus, Cumberland, Hoke, Montgomery, Moore, Richmond, Robeson, Sampson, and Scotland Counties—had both the highest rates of grandparents as primary caregivers as well as the highest opioid prescription rates. Two Sandhills counties—Scotland and Columbus—exceeded the state and national averages on both measures.

Grandparent caregiver rates in North Carolina counties

The top ten North Carolina counties with the highest rates of grandparents acting as primary caregivers were:

  1. Anson County (5.4%)
  2. Clay County (4.6%)
  3. Graham County (3.9%)
  4. Scotland County (3.7%)
  5. Robeson County (3.6%)
  6. Columbus County (3.4%)
  7. Sampson County (2.9%)
  8. Person County (2.9%)
  9. Martin County (2.9%)
  10. Bladen County (2.8%)

Five of these ten counties are located in the Sandhills region, and all exceeded the state average of 1.6%. The share of grandparents raising their own grandchildren in these counties exceeded rates in the top five states: Mississippi (2.7%), Arkansas (2.2%), Louisiana (2.1%), Alabama (2.1%) and Kentucky (2.1%).

For the majority of these grandparents, caretaking has been a long-term job. In all counties except Person, at least 50% of grandparents have taken care of their grandchildren for three years or more.

Opioid prescribing rates in North Carolina counties[2]

North Carolina counties with the highest rates of prescriptions per 100 people were:

  1. Scotland County (168)
  2. Columbus County (140)
  3. Surry County (139)
  4. Richmond County (138)
  5. Beaufort County (131)
  6. Burke County (125)
  7. Mitchell County (124)
  8. Hertford County (123)
  9. Rockingham County (121)
  10. Caldwell County (119)

In each of these counties, the prescription rate exceeded 100, meaning that there were enough prescriptions dispensed that each person in the county could have one. This is compared to the state rate of 72 prescriptions per 100 people, which is higher than the national rate of 58.7.

Number one and two on this list—Scotland and Columbus counties in the Sandhills region—were also among those with the highest rates of grandparents acting as primary caregivers, at 3.7% (4th-highest) and 3.4% (6th-highest) respectively.

Counties with higher-than-average prescription rates also tended to have grandparent-caregiver rates exceeding the state average, except for Surry, Mitchell, and Hertford counties.

In counties with the lowest prescribing rates, rates were below the national average of 58.7 prescriptions per 100 people:

  1. Gates County (1)
  2. Caswell County (8)
  3. Northampton County (15)
  4. Currituck County (15)
  5. Alexander County (20)
  6. Bertie County (25)
  7. Greene County (29)
  8. Warren County (30)
  9. Perquimans County (30)
  10. Jones County (37)

Half of these counties had grandparent-caregiver rates above the state average (1.6%), while half fell below the state average.

In summary, counties with prescribing rates exceeding the state average also tended to have higher-than-average grandparent caregiver rates. This pattern was most apparent in the Sandhills region. We should note that grandparents care for their grandchildren for a variety of reasons—so the opioid crisis doesn’t completely explain this pattern—but there does appear to be a link between counties with higher rates of opioid perscriptions and grandparents taking on the role of primary caregivers for their grandchildren.

Families and communities in North Carolina are increasingly stretched thin as a result of this epidemic. When family members cannot take care of a neglected child, foster care and social services must step in, and many agencies in the state are seeing a spike in demand as a result of opioid abuse in homes.

Projects like ncIMPACT’s Opioid Response Project are working to meet these challenges in our state. They’re engaged in collaborative partnerships with various state and local agencies to help ten NC communities tackle their opioid crises.

[1] Prescribing rates measure the total number of opioid prescriptions dispensed in a given year and area divided by the total population, represented as a rate per 100 residents.

“For this database, a prescription is an initial or refill prescription dispensed at a retail pharmacy…and paid for by commercial insurance, Medicaid, Medicare, or cash or its equivalent. This database does not include mail order pharmacy data.” (CDC, 2017)

[2] Note: Opioid prescribing data was available for all counties except Camden County.

Posted in Health & Environment | Tagged , , , | Comments Off on Are more NC grandparents raising grandchildren because of the opioid epidemic? A look at the data.

A snapshot of single fathers in North Carolina: 2019

Father’s Day will be celebrated on Sunday, June 16, 2019. We’ve compiled some key stats about parenting and fatherhood in North Carolina.

A record number of households in North Carolina are headed by single dads.

There were 98,434 single father households in NC in 2017, an increase of more than 7,000 since 2016 and the highest number observed since 1960, when just 7,769 households with children were headed by single fathers.Rising number of single father headed households, 1960-2017

Single father households represented 8.1% of all NC households with children in 2017, the highest share of single father households since 1960, when it was just 1.1%.

The percentage of households in North Carolina with children under 18 is decreasing.

In 2017, there were 1.2 million households in North Carolina with at least one child under the age of 18, representing 30.6% all NC households.

This represents a historic low for the share of households with children. In 1960 and 1970, over half of NC households had children—60.2% and 52%, respectively.

This proportion has steadily declined, reflecting the combined impact of longer life expectancy (individuals are living for more years after their children have grown up) as well as declining fertility rates and delayed childbearing.

Most households with children in NC (89%) are headed by a biological, adoptive, or step parent.

Among the 1.2 million households with children:

  • 716,000 or 59.1% are headed by a married couple. This represents the highest rate of married-couple-headed households with children since 2009.
  • 269,396 or 22.2% are headed by a single mother*, the lowest number and lowest rate of single mother headed households since 2007 (272K and 22.8%).
  • 98,434 or 8.1% are headed by a single father*, the highest number and highest rate observed since 1960.

*Single mothers and fathers include individuals who are married but with spouse absent, as well as individuals who have an unmarried partner living in the household.8.1% of households with children headed by single father in 2017

The remaining 11% of households with children are headed by an individual who is not the mother or father of the child. They may be grandparents or other relatives, and the parents of the child may also be living in the house.

Defining “father” for this analysis

We are limiting our analysis to household heads age 15 and older who have at least one dependent child under the age of 18 living in the household. Children can be biological, adopted, or step sons or daughters.

Fathers who are not the head of the household (or married to the household head) and fathers who do not reside with their children are not included in this analysis.

Note: There were 128,286 NC households in 2017 with children under 18 living in the household headed by an individual who is not the mother or father of the child. They may be grandparents or other relatives, and the parents of the child may also be living in the house.

Defining “single father” for this analysis

Single fathers include fathers without a co-resident partner, as well as fathers who are married but their spouse is absent (meaning they live in a separate house) and fathers who are cohabiting with an unmarried partner.

Who is a “single father”?

Among the 508,000 NC fathers who were the head were the head of household in 2017, nearly one in five—19%—were single fathers. Some fathers were more likely to be single fathers:

  • Young fathers (ages 15-29) were most likely to be single fathers (32%), followed by adults ages 30-39 (21%).
  • Black (35%) and Hispanic (26%) fathers were more likely to be single than white fathers (16%).
  • The probability that a father is a single father decreases as his educational attainment increases: more than 30% of fathers with a high school diploma or less were single fathers compared to 15% of fathers with an associate degree and 7% of fathers with a bachelor’s or higher.
  • Lower-income fathers are more likely to be single fathers: 42% of fathers in poverty were single fathers versus 13% of fathers in households with incomes at 200% of the federal poverty line or higher.Which dads are most likely to be single?

Single does not mean “without a partner”

The growth in “single fathers” has been driven by a rising number of fathers who live with an unmarried partner. Between 1990 and 2017, the number of single fathers increased from nearly 36,600 to more than 98,400, an increase of nearly 61,900 or 169% growth.

  • The number of single fathers without a partner in the household grew from 22,000 to 35,300, an increase of 13,300 or 60%.
  • The number of single fathers with a cohabiting partner grew from 14,500 to 63,100, an increase of 48,600 or 334%.

In 1990, 40% of single fathers were living with an unmarried partner; this proportion had increased 24 percentage points to 64% in 2017.Majority of single fathers live with a partner

Young fathers, Black and Latino fathers, fathers without a college degree, and fathers living in or near poverty are more likely to be living with an unmarried partner. As a result, these fathers are more likely to be classified as “single fathers.”

Please let us know if you have other questions that we can address in future blog posts by writing To receive monthly updates about demographic trends in North Carolina, subscribe to our monthly newsletter.

Posted in Carolina Demographics | Tagged , , , , | Comments Off on A snapshot of single fathers in North Carolina: 2019

Potential voters are fastest-growing segment of NC Hispanic population

North Carolina’s Hispanic or Latino population increased by nearly 116,000 residents between 2012 and 2017, an increase of 13.8% in just five years. The adult population grew faster than the child population over this period. And for both Latino adults and children, growth was due entirely to an increase in the citizen population. The non-citizen population of Latino children and adults decreased by 23.1% and 10.4%, respectively, between 2012 and 2017.

As a result of these changes, the voting eligible population—individuals who are both 18 or older and a citizen (native-born or naturalized)—is the fastest-growing segment of the state’s Hispanic population. This population increased by 105,514 or 49.4% over five years, more than twice the growth of the voting-eligible Latino population nationwide (22.1%). In 2012, just one in four NC Hispanic residents was voting eligible. As of 2017, this shared had increased to one in three.

The graph below shows the North Carolina Hispanic population by age and citizenship status based on data from the 2017 American Community Survey. The teal bars represent Hispanic individuals currently of voting age (18+): the dark shading represents citizens (voting eligible) while the light teal represents non-citizens. The blue bars represent North Carolina Hispanic or Latino children who are under 18. Similarly, the dark blue represents citizens (future voting eligible) and the light blue represents non-citizens.

In our post from 2014, we predicted that the size of the voting eligible Hispanic population would likely surpass 300,000 by 2020. According to the most recent ACS data, this has already occurred.

As of 2017, more than 319,000 of North Carolina’s 592,000 Hispanics age 18 and older were eligible to vote. This represents more than half of the state’s Latino adults (54%). An even greater percentage—88%—of the Hispanic residents who will turn 18 between 2017 and 2020 are citizens. These individuals will increase the size of the voting eligible Hispanic population by another 50,000 in 2020.

What does this mean?

These trends are not isolated to North Carolina, but they are more pronounced in the state due to the relatively recent impact of Hispanic immigrants. Compared to other states, North Carolina had the 16th largest population of voting-eligible Latinos in 2017. In contrast, North Carolina had the:

  • 2nd fastest-growing population of voting-eligible Latinos (49.4%), surpassed only by North Dakota (57.7%). Oregon (42.3%), Minnesota (40.9%), and South Carolina (39.9%) had the next highest increases.
  • 9th largest numeric increase in voting-eligible Latinos (105.5K). California (1.3M) had the largest numeric increase, followed by Texas (884K), Florida (657K), New York (270K), and Arizona (243K). These states already had large pools of potential Hispanic voters in 2012 due to a long history of immigration.

Voting-eligible Hispanics represent a contested piece of the electorate, though they have historically been less likely to register and vote. In future posts, we plan to explore:

  • how many eligible North Carolina Hispanic voters are registered to vote.
  • the percentage of North Carolina Hispanic voters who are affiliated with a political party and what that breakdown looks like.

Please let us know if you have other questions that we can address in future blog posts by writing To receive monthly updates about demographic trends in North Carolina, subscribe to our monthly newsletter.

Posted in Carolina Demographics, Elections & Voting | Tagged , , , | Comments Off on Potential voters are fastest-growing segment of NC Hispanic population

100 students start 9th grade in NC. What happens next?

Between 2014 and 2026, nearly 1.6 million students will enter the state’s K-12 public school system as ninth graders. Under current graduation rates, more than 213,000 of them are predicted to drop out or delay high school graduation.

Let’s break it down further.

For every 100 ninth graders that start public high school in NC:

  • 86 graduate high school within 4 years.
  • 14 dropout or take more than 4 years to complete high school.

In 12th grade, all public high school graduates in NC are asked in the spring what they intend to do after graduation. Of the 86 students that complete high school on time:

  • 1 has no clear path after graduation
  • 4 intend to join the military
  • 10 intend to go into the workforce
  • 71 intend to go on to post-secondary schooling

Of the 71 students who say they intend to go on to post-secondary schooling:

  • 28 say they intend to go to one of the 16 schools in the UNC system
  • 29 say they intend to go to one of the 58 NC community colleges across the state
  • 14 say they plan to go to private or out-of-state schools

But not everyone shows up to college in the fall, as our recent analysis of North Carolina’s public postsecondary pipeline shows.  Transition to college is the largest loss point in the postsecondary pipeline. Many high school graduates with postsecondary intentions do not matriculate in the fall, a process known as “summer melt.” Summer melt most often affects first-generation college students, low-income students, and students with intentions to enroll in a community college. Research finds that small-scale interventions, such as summertime college counseling or text messaging campaigns, are effective at reducing summer melt and can significantly increase first-year enrollment.

In NC, of the 71 students who say they intend to go to postsecondary education, only 49 of them enroll.

And of the 49 that enroll, 9 don’t return for a second year. Five more change schools between their first and second year.

And 15 return for a second year but don’t complete their degree or credential within 6 years.

Out of the 49 students who initially enroll in college, 25 receive a postsecondary degree or credential within 6 years. 24 more students have completed some college, but don’t have a degree. Out of our original 100 ninth graders:

  • 25 have a postsecondary degree or credential within 6 years
  • 24 have some college but no degree
  • 22 planned to go to college, but didn’t immediately transition
  • 15 entered work or the military
  • 14 dropped out of high school or took longer than 4 years to graduate

However, there are large differences in the predicted likelihood of postsecondary pipeline completion across demographic groups:

  • Female students are more likely to complete the postsecondary pipeline than male students. (31% versus 20%, respectively.)
  • Asian students (41%) are nearly twice as likely to complete the postsecondary pipeline on time than the state average of 16%.
  • Black (15%), Hispanic (14%) and American Indian (13%) students have predicted on-time pipeline completion rates lower than the stage average.
  • Disparities between men and women and between racial/ethnic group exist at all points in the pipeline: in four-year high school graduation rates, post-graduation intentions, immediate college enrollment, first-year persistence rates, and on-time graduation rates.

The challenge

North Carolina has set a goal for 2 million adults ages 25-44 to have postsecondary credential or degree by 2030; this represents about 67% of adults 25-44 with a postsecondary certificate, diploma, or associate degree or higher. But there are persistent inequities in educational outcomes between race and ethnic groups. Based on current projections, three underserved groups (Black, American Indian/Alaska Native, and Hispanic populations) will not meet the postsecondary attainment goal.

Of the 1.4 M children enrolled in public school in NC, 44% are Black, Hispanic, or American Indian. And these children are more likely to be first-generation college students, based on the current educational attainment of their parents.

Programs that reach, engage, and successfully enroll, retain, and graduate our state’s growing population of first-generation and minority students will be vital to ensure North Carolina’s progress towards any statewide attainment goal. Failure to improve these outcomes is not only detrimental to the future economic well-being of these children and their families: it will ensure that our state’s primary path to future attainment growth will be the continued reliance on in-migration of highly educated individuals from other states and countries.


Posted in Education | Tagged , , , , , | Comments Off on 100 students start 9th grade in NC. What happens next?

Over the past decade, returns to postsecondary degrees have diminished nationwide and in NC

Key takeaways: A college degree is associated with several positive life outcomes, ranging from greater financial security to better health. However, over the past decade, returns on postsecondary degrees have diminished both nationwide and in North Carolina.

We reported recently that the level of educational attainment among North Carolina’s degree-earners continues to rise. This is related, in part, to a shifting job market favoring higher-skilled workers. However, as more individuals seek out greater levels of postsecondary education, the question remains as to whether these advanced degrees continue to provide larger financial returns over the course of an individual’s life.

A publication from Georgetown University’s Center on Education and the Workforce sought to answer this question, and determined that there are four general rules to earnings and education:

Rule 1: Degree level matters.

In general, higher degrees are associated with higher earnings.

Rule 2: Occupational choice can trump degree level.

Individuals with less education may earn more in higher-paying fields, such as STEM careers, than those with more education in lower-paying fields.

Rule 3: Although occupation can trump education, degree level still matters most within individual occupations.

Individuals with more advanced education in their career field earn more than those that with lower levels of education.

Rule 4: Race/ethnicity and gender are wild cards.

These demographic factors can play a larger role in future earnings than even education or occupation.

Georgetown’s analysis used national-level data, so we decided to investigate how well these rules hold up for workers in North Carolina. In this blog post, we examine the first rule—whether degree level matters—on degree level attainment and earnings advantages over time. We will look at the other rules in future posts.

Our findings at the state level closely matched Georgetown’s national findings.

Calculating Synthetic Work-Life Earnings

For this analysis, we conducted a synthetic work-life earnings estimate. A “synthetic” estimate means that these are not the actual dollar amounts earned by real people over the course of their working lives (which would require an extensive amount of sensitive, longitudinal data). Instead, these are estimates derived from a point-in-time cross-sectional survey—the American Community Survey, in this case.

We focused on individuals aged to 25 to 64 because these are our prime working ages. We also chose to focus the analysis on full-time, year-round workers—defined as being employed at least 35 hours per week and working at least 50 weeks per year—because we believe that they are a better representation of wage returns to education in the labor market. This means that we are excluding workers who work seasonally, who work part time, or who drop out or stop out temporarily from the labor force.

Using wage and employment data, we calculated the median earnings for each 5-year age group between ages 25 and 64 for full-time, year-round workers in North Carolina by educational attainment. (So, for example, we looked at the median earnings for 25-29 year-olds with a high school diploma, an associate degree, a bachelor’s degree, etc.) We then calculated each educational group’s predicted lifetime earnings by multiplying each age category by 5 to obtain their estimated total salary for each age range. Finally, we summed the estimated total salary for each age range from 25-29 to 60-64 to obtain the total synthetic work-life earnings estimate.

For example: Let’s calculate the median earnings for a full-time, year-round employee with a bachelor’s degree

Median earning between 25-29: $42,000
              Total salary for age range 25-29 = $42,000/year * 5 years = $210,000
Median earning between 30-34: $48,000
                Total salary for age range 30-34 = $48,000/year * 5 years = $240,000
Median earning between 35-39: $56,000
                Total salary for age range 35-39 = $56,000/year * 5 years = $280,000
Median earning between 40-44: $62,000
                Total salary for age range 40-44 = $62,000/year * 5 years = $310,000
Median earning between 45-49: $66,000
                Total salary for age range 45-49 = $66,000/year * 5 years = $330,000
Median earning between 50-54: $65,000
                Total salary for age range 50-54 = $65,000/year * 5 years = $325,000
Median earning between 55-59: $65,000
                Total salary for age range 55-59 = $65,000/year * 5 years = $325,000
Median earning between 60-64: $63,000
                Total salary for age range 60-64 = $63,000/year * 5 years = $315,000

Lifetime earnings = sum of total salaries for each age range

$210,000 + $240,000 + $280,000 + $310,000 + $330,000 + $325,000 + $325,000 + $315,000
= $2,335,000 in lifetime earnings

Of course, these estimates come with necessary caveats and limitations. Education alone does not account entirely for a worker’s lifetime earnings, and, in some cases, sex and race/ethnicity play an even larger role in influencing earnings than one’s degree. In a future blog post, we will address the effects of sex and race/ethnicity on earnings potential over the course of a lifetime.

More broadly speaking, these estimates examine the median lifetime earnings of a worker with a given level of education. In other words, this means that half of individuals in each age range earn less than this amount, while half earn more. This is not meant to represent the lived experience of any specific worker.

What we learned: There are gaps in lifetime earnings based on the highest level of education that someone has completed

Any level of postsecondary attainment translates to greater lifetime earnings compared to an individual with a high school degree alone, and additional education beyond an associate or bachelor’s degree is correlated with higher earnings. Lifetime earnings for all workers in North Carolina range from just over $1 million for individuals with less than a high school diploma to nearly $4 million for professional degree-holders, such as lawyers and doctors.

This gap in lifetime earnings has widened over the last decade. For full-time, year-round workers, bachelor’s degree-holders in 2007 could expect to earn 78% more than workers with just a high school diploma/GED and 33% more than workers with an associate degree. As of 2017, those numbers had grown by one percentage point and four percentage points, respectively: workers in North Carolina with a bachelor’s degree in 2017 could expect to earn 79% more than workers with just a high school diploma or GED certificate and 37% more than associate degree-holders.

For every education group, anticipated lifetime earnings in 2017 were lower than in 2007.

We also found that over the last decade, median lifetime earnings for full-time, year-round workers have decreased for every level of educational attainment.

Workers with an associate had the largest drop in lifetime earnings. Those with an associate degree in 2017 make 7% less than those holding the same degree in 2007.

But full-time, year-round employees in North Carolina made less at every degree level. Compared to 2007, expected lifetime earnings in 2017 were:

  • 5% less for those with less than a high school diploma;
  • 7% less for those with a high school diploma or GED certificate;
  • 5% less for bachelor’s degree-holders ;
  • 8% less for master’s degree-holders;
  • 3% less for doctoral degree-holders; and
  • 7% for professional degree-holders.

Reasons for this are likely complex, but without fully accounting for the composition of the workforce, it is hard to definitively know what this means. It may be related to wages being slow to recover since the Great Recession. Age hiring trends may also play a role, as much of the hiring at the national-level has skewed towards younger age groups (16-24 and 25-34). These newer workers tend to earn less than older age groups. While this does not speak to North Carolina specifically, it is likely that the state is following similar trends as the nation.

The American Community Survey from the Census Bureau is a nationwide representative survey that collects data on a wide variety of social and economic factors, such as ancestry, educational attainment, housing characteristics, and employment.

Posted in Carolina Demographics | Comments Off on Over the past decade, returns to postsecondary degrees have diminished nationwide and in NC