Since you are looking at this site with an older browser, you will not be able to see any graphics or formatting. For better results, please upgrade your browser.

A National Portrait of Chronic Absenteeism in the Early Grades

Authors: Mariajosé Romero and Young-Sun Lee
Publication Date: October 2007
This analysis is part of NCCP’s Pathways to Early School Success Project which identifies, synthesizes and promotes “on the ground” use of emerging knowledge that can reduce the achievement gap.

It has long been recognized that chronic absenteeism and school truancy in middle and high school are significant problems with highly visible negative consequences for youth, and ultimately, for their employability as adults.1 Little is known, however, about chronic school absenteeism among early elementary school students, as well as among children in preschool programs. This brief reveals a significant level of absenteeism in the early school years, especially among low-income children, and confirms its detrimental effects on school success by examining children from across various incomes and race/ethnicity groups in a nationally representative sample of children entering kindergarten—The Early Childhood Longitudinal Study (Kindergarten Cohort)2—in 1998.

How Widespread is the Problem of Early Absenteeism?

On average, children missed 5 days in kindergarten, 4.5 days in first grade, and 3.7 days in both third and fifth grades. However, almost 14% of kindergartners, 12% of first graders, 11% of third graders, and 10% of fifth graders were at-risk absentees: they missed an average of 12 to 18 days during the school year. Over 11% of kindergartners, almost 9% of first graders, 6% of third graders, and 5% of fifth graders were chronic absentees: they missed at least 18 days or more of the school year. In total, one-quarter of all kindergarten children were either at-risk or chronic absentees.

How Does Family Income3 Impact Early Absenteeism?

Absenteeism rates were higher among low-income children

Figure 1: Absenteeism rates were higher among low-income children

In all grades, the lower family income, the higher the absenteeism rates. Living in a poor family (one whose income is below the federal poverty level—FPL), or a low-income family (one whose income is up to 200% FPL)4 greatly increased the chances of being an at-risk or chronic absentee.

Poor children were more likely to be chronic absentees than their highest income peers

Figure 2: Poor children were more likely to be chronic absentees than their highest income peers

Further, while attendance improved for all income groups through the elementary grades, in all grades a greater proportion of poor children were chronic absentees than other income groups. This proportion diminished between kindergarten and third grade, but increased again in fifth grade.

In kindergarten, children in poor families were four times more likely to be chronic absentees than their highest income counterparts; this proportion decreased to 3.6:1 in first grade, and 2.6:1 in third grade, but rose again to 5.3:1 in fifth grade.

How Do Race and Ethnicity Impact Early Absenteeism?

Absenteeism rates were higher among American Indian children

Figure 3: Absenteeism rates were higher among American Indian children

In all grades, American Indian children had the highest absenteeism rates. For all racial groups, attendance improved over the elementary grades. By the fifth grade, however, American Indian children continued to miss, on average, twice as many school days as their peers.

Does Early Absenteeism Affect Later Absenteeism?

The majority of chronic absentees in first grade had been chronic absentees in kindergarten

Figure 4: The majority of chronic absentees in first grade had been chronic absentees in kindergarten

Data suggest the emergence of trajectories of absenteeism early in children’s careers in formal schooling: the greater the number of absences in kindergarten, the greater the number of absences in first grade.5 Over one-half of chronic absentees in kindergarten also were chronic absentees in first grade.

Socioemotional development and functioning as reported by teachers, and incipient school refusal behaviors as reported by parents, surfaced as important factors related to absenteeism. Children who exhibited less mature socioemotional functioning in the classroom, as indicated by teachers’ scores on approaches to learning, interpersonal relations, self-control, as well as on externalizing and internalizing problem behaviors, had higher absenteeism rates than those with greater socioemotional maturity. Likewise, children who, according to their parents, complained about school, were upset to go to school, or claimed to be sick to stay home had higher absenteeism than those who did not engage in these behaviors.6

What is the Impact of Early Absenteeism on Academic Achievement?7

Chronic absentees in kindergarten had the lowest academic performance in first grade

Figure 5: Chronic absentees in kindergarten had the lowest academic performance in first grade

Early absenteeism negatively impacted on academic achievement in reading, math, and general knowledge in the early school years. Greater absenteeism in kindergarten was associated with lower achievement in reading, math, and general knowledge at the end of first grade.8 On average, children missing 10% or more of the school year scored five points less than did those who were absent up to 3% of the school year in kindergarten.9

Chronic absenteeism in kindergarten was detrimental particularly for the reading performance of Hispanic children in first grade

Figure 6: Chronic absenteeism in kindergarten was detrimental particularly for the reading performance of Hispanic children in first grade

Low-income and Latino children were especially at a disadvantage in first grade, particularly when they were chronic absentees in kindergarten: they scored, on average, 10 points less in first grade reading achievement than did white children with the best kindergarten attendance.10

Concluding Remarks

This brief paints an aggregate national portrait. There are real limitations to using the ECLS-K data: for example, it did not oversample low-income or black and Latino children. Also, this analysis included only children for whom there were complete data on absenteeism. These children were more likely to be white and slightly higher in income, and had better health status than the entire ECLS-K sample. Yet, even with these limitations, the data are robust enough to sound a clear challenge to educators, policymakers, and researchers concerned with early school success. Efforts to improve access to and the quality of early care and learning also need to encompass strategies to ensure that children are actually in settings that can start them on successful early learning trajectories. If children are not in school, the odds that they will succeed are lessened.

Endnotes

1. See, for instance, Broadhurst, K.; Paton, K.; & May-Chahal, C. (2005). Children missing from school systems: Exploring divergent patterns of disengagement in the narrative accounts of parents, carers, children and young people. British Journal of Sociology of Education, 26(1), pp. 105-119; Hilton, Z. (2006). Disaffection and school exclusion: Why are inclusion policies still not working in Scotland? Research Papers in Education, 21(3), pp. 295-314; and Kane, J. (2006). School exclusions and masculine, working-class identities. Gender and Education, 18(6), pp. 673-685.

2. The Early Childhood Longitudinal Study, Kindergarten Cohort (ECLS-K), from the National Center for Education Statistics, is a national study following a sample of children who entered kindergarten in 1998, until they reached the fifth grade. The ECLS-K collects data on children’s development, family characteristics and functioning, and a host of characteristics in the school environment.

3. Family income was approached as an income-to-needs ratio, which is determined by the U.S. Department of Health and Human Services (HHS) on the basis of the total annual family income, the number of adults and children in the family, and how this income compares to the income level below which HHS considers a family of a specific size and child-adult composition to live in poverty. This ratio is adjusted for inflation every year. ECLS-K data contain one variable identifying children who live below the federal poverty level (FPL), but this variable lumps together all income groups above this cut-off. For the purposes of this analysis, income-to-needs ratio groups were calculated based on respondents’ annual family income for 1998, when baseline data were collected, as well as family size and the number of children and adults in the family, compared to U.S. Census Bureau, Current Population Survey. (1998). Poverty thresholds in 1998, by size of family and number of related children under 18 years. Retrieved September 7, 2006, from <www.census.gov/hhes/poverty/thresh98.html>.

4. The current federal poverty level for a family of 4 is $20,650 in 2007. Low-income families make up to twice that amount. For more information on measuring poverty, see NCCP’s state profiles at <www.nccp.org> and the U.S. Department of Health and Human Services <www.aspe.hhs.gov/poverty/07poverty.shtml>.

5. ß=.490, t=42.19, significant at p<.00069, two-tailed.

6. Negative attitudes toward school, ß=.097, t=7.56; Teacher perceptions of socioemotional functioning, ß=-.058, t=-4.47, both significant at p<.00076, two tailed.

7. The ECLS-K provides different measures to assess children’s achievement in reading, math, and general knowledge: (1) Number right scores, which simply count the number of items answered correctly; (2) Item Response Theory (IRT) scores, which assess children’s responses across items varying in difficulty; (3) Standardized scores T Scores), which assess children’s performance as compared to their peers; (4) Criterion-referenced proficiency level; and (5) Proficiency probability scores, which assess children’s performance in subtests evaluating specific skills. For the purposes of this analysis, T scores were used since they represent estimates of children’s achievement compared to the population, and can be used to examine changes in children’s achievement over time. T scores have a mean of 50 and a standard deviation of 10. See National Center for Education Statistics. (2001). User’s manual for the ECLS-K base year public-use data files and electronic codebook (NCES 2001-029 [revised]). Washington, DC: National Center for Education Statistics.

8. Separate regression models were run with T scores as dependent variable for each academic area and absenteeism rates as one of the predictors. Beta weights and t ratios for kindergarten absenteeism rates are as follows: with DV=reading, ß=-.072, t=-5.82, significant at p<.00063, two-tailed; with DV=math, ß=-.095, t=-8.089, significant at p<.00063, two-tailed; with DV=general knowledge, ß=-.045, t=-3.921, significant at p<.00063, two-tailed.

9. This relationship was independent of children’s income, race, disability status, attitudes toward school, socioemotional development while in kindergarten, age at kindergarten entry, the type of care and education experiences prior to kindergarten, or the type of program in kindergarten (morning, afternoon, or full day).

10. Regression models were run with absenteeism groups (i.e., 0%-3.3%; 3.33%-6.6%; 6.6%-10%; and 10% or more) instead of absenteeism rates. For the interaction term for Latino*absent 10% or more, ß=-.057, t=-3.83, significant at p<.000555, two-tailed.