Covariates from these models: ECLS-K 1998--1999 and 2010--2011

ECLS-K 1998–1999 ECLS-K 2010–2011
Socioeconomic status (SES). The SES is a composite variable reflecting the socioeconomic status of the household at the time of data collection. SES was created using components such as father/male guardian’s education and occupation; mother/female guardian’s education and occupation; and household income (see Tourangeau et al. 2009, 7-23–7-30). We use five SES quintiles dummies that are available. We use the following labels in the tables and figures: “Low SES” indicates the first or lowest socioeconomic quintile, “Middle-low SES” indicates the second-lowest quintile, “Middle SES” is the third quintile, “High-middle SES” indicates the fourth quintile, and “High SES” represents the highest or fifth quintile.  Socioeconomic status (SES). The construct is based on three different components (five total variables), including the educational attainment of parents or guardians, occupational prestige (determined by a score), and household income (see more details in Tourangeau et al. 2013, 7-56–7-60). We use the quintile indicators based on the continuous SES variable (we construct them).
Child living in poverty. Information about whether the child’s household lives in poverty is obtained from a household-level poverty variable. The household’s income is compared with census poverty thresholds for 2006 (which vary by household size) and the household is considered to be in poverty if total household income is below the poverty threshold determined by the U.S. Census Bureau poverty threshold (Tourangeau et al. 2009, 7-24 and 7-25). Child living in poverty. Information about whether the child’s household lives in poverty is obtained from a household-level poverty variable. This variable indicates whether the household income is below 200 percent of the U.S. Census Bureau poverty threshold. More details are provided in Tourangeau et al. 2013 (7-53 and 7-54).
Gender. A variable indicates whether the student is a girl or a boy. Gender. A dummy indicator represents whether the child is a boy or a girl.
Race/ethnicity. A variable indicates the race/ethnicity of the student—whether the child is white, black, Hispanic, Asian, or another ethnicity. Hispanic children are divided into two groups, those whose families speak English at home and those whose families do not. (This latter decomposition was first described and utilized by Nores and Barnett [2014] and Nores and García [2014]). Race/ethnicity. Our analysis includes dummy indicators of whether the race/ethnicity of the child is white, black, Hispanic, Asian, or “other.” Hispanic children are divided into two groups, those whose families speak English at home and those whose families do not.
Age of student. Age of the student calculated in months. Age of student. Age of the student is calculated in months.
Language at home is not English. A variable indicates whether the language the student speaks at home is a language other than English. Language spoken at home. Our analysis includes a dummy indicator that represents whether the language spoken in the child’s home is a language other than English (we call a child in this setting an English language learner, or ELL), versus whether the language spoken at home is English or English and other language(s).
Disability. A variable indicates whether the child has a disability that has been diagnosed by a professional (composite variable). Questions in the parents’ interview about disabilities ask about the child’s ability to pay attention and learn, overall activity level, overall behavior and relationships to adults, overall emotional behavior (such as behaviors indicating anxiety or depression), ability to communicate, difficulty in hearing and understanding speech, and eyesight (Tourangeau et al. 2009, 7-17). Disability. A dummy indicator represents whether the child has been diagnosed with a disability.
Type of family. A variable indicates whether the child is living with two parents, or with one parent or in another family structure. Type of family. A variable indicates whether the child lives with two parents versus living with one parent or in another family composition.
Prekindergarten care in a center-based setting. A dummy indicator represents whether the child was cared for in a center-based setting or attended Head Start during the year prior to the kindergarten year, compared with other options. These alternatives include no nonparental care arrangements and care provided through other means (by a relative or a nonrelative, at home or outside the home, or a combination of options). Prekindergarten care in a center-based setting. Our analysis includes a dummy indicator of whether the child was cared for in a center-based setting (including Head Start) during the year prior to the kindergarten year, compared with other options. These alternatives include no nonparental care arrangements and care provided through other means (by a relative or a nonrelative, at home or outside the home, or a combination of options). Any finding associated with this variable may be interpreted as the association between attending prekindergarten (pre-K) programs, compared with other options, but must be interpreted with caution. These coefficients should not be interpreted as the impact of pre-K schooling because the variable’s information is limited and the model uses it as a control-only variable. For a review of the extensive literature explaining the benefits of pre-K schooling, see Camilli et al. 2010.
“Literacy/reading activities” index. This index captures the variance on a wide set of family early literacy practices. Using an index of activities instead of the underlying questions the index is composed of overcomes potential problems of multicolinearity and therefore improves the properties of our specifications. (This has an alpha of 0.6716). In particular, parents are asked the frequency (“not at all,” “once or twice a week,” “three to six times a week,” or “every day”) with which they engage with the child in the following activities: reading books; telling stories; singing songs; and talking about nature or doing science projects. Parents are also asked how often the child reads picture books outside of school, and reads to or pretends to read to himself or to others outside of school. “Literacy/reading activities” index. This index captures the variance on a wide set of family early literacy practices. Using an index of activities instead of the underlying questions the index is composed of overcomes potential problems of multicolinearity and therefore improves the properties of our specifications. (This has an alpha of 0.6948.) In particular, parents are asked the frequency (“not at all,” “once or twice a week,” “three to six times a week,” or “every day”) with which they engage with the child in the following activities: reading books; telling stories; singing songs; and talking about nature or doing science projects. Parents are also asked how often the child reads picture books outside of school, and reads to or pretends to read to himself or to others outside of school.
 “Other activities” index. Parents are asked the frequency (“not at all,” “once or twice a week,” “three to six times a week,” or “every day”) with which they engage with the child in the following activities: playing games or doing puzzles; playing sports; building something or playing with construction toys; doing arts and crafts; or doing science projects. (This has an alpha of 0.5972.)  “Other activities” index. Parents are asked the frequency (“not at all,” “once or twice a week,” “three to six times a week,” or “every day”) with which they engage with the child in the following activities: playing games or doing puzzles; playing sports; building something or playing with construction toys; doing arts and crafts; or doing science projects. (This has an alpha of 0.5527.)
Mother’s educational attainment. This is coded as “below high school (8th–12th grades); high school graduate or equivalent; vocational/technical program/some college; bachelor’s degree/graduate or professional school with no degree; and graduate (master’s, doctorate, or professional) degree.” Mother’s educational attainment. This is coded as “below high-school (8th–12th grades); high school graduate or equivalent; vocational/technical program/some college; bachelor’s degree/graduate or professional school with no degree; and graduate (master’s, doctorate, or professional) degree”.
Income. We adjust the income brackets in 2010 for inflation. We use the continuous variable to construct the 18 categories to make it comparable to the variable in 2010. We calculate a continuous income variable using the midpoint between the minimum and maximum for each category (equal to the values in 2010 adjusted by inflation). We calculate the income quintiles using this variable. Income. The original income variable comes in 18 categories. We calculate a continuous income variable using the midpoint between the minimum and maximum for each category. We calculate the income quintiles using this variable.
Parents’ education expectations. This is coded as “HS or less; 2 or more years of college; BA; MA; PHD or MD.” Parents are asked, “How far in school do you expect your child to go? Would you say you expect {him/her} to {attend or complete a certain level}?” Parents’ education expectations. This is coded as “HS or less; 2 or more years of college/attend a vocational or technical school; BA; MA; PHD or MD.”
Number of books the child has. This is represented by a continuous variable (0–200) and a categorical variable coded as “0 to 25; 26 to 50; 51 to 100; 101 to 199; more than 200.” For the regression analysis, the variable is divided by 10. Parents are asked, “About how many children’s books {does {CHILD} have/are} in your home now, including library books? Please only include books that are for children.” Number of books the child has. This is represented by a continuous variable (0–200) and a categorical variable coded as “0 to 25; 26 to 50; 51 to 100; 101 to 199; more than 200.” For the regression analysis, the variable is divided by 10.

Source: ECLS-K, kindergarten classes of 1998–1999 and 2010–2011 (National Center for Education Statistics)

View the underlying data on epi.org.