Even Start Family Literacy: Providing Success For
Low-Income, Minority Families

by

Elizabeth B. Thompson, CFLE
Texas State Even Start and Family Literacy Coordinator

and

Don F. Seaman, Ph.D.
Research Consultant

Texas LEARNS
The Texas Adult Education and Family Literacy Partnership
6311 Irvington Blvd., Houston, Texas. 77022-5618 | 713-694-6300 | 1-866-696-4233

Presented at the Annual Conference
National Center for Family Literacy
Orlando, Florida

March, 2007

Even Start Family Literacy: Providing Success for
Low-Income, Minority Families

Brief Introduction

It is no secret that a definite gap exists between the educational achievements of low-income children and their middle-class peers. Klein and Knitzer (2006) indicate that often, low-income children enter public school early childhood programs, quickly fall behind their middle-class peers, and remain there throughout school. This is particularly true among minority children and a strong relationship seems to exist between falling behind and eventually dropping out of school before attaining a high school diploma.

This is a big problem for border states, especially one as big as Texas, and the national media have discussed the problem and its accompanying issues. For example, of those students entering high school, Texas had the lowest percentage of high school graduates in 2003 and 2004 (Spencer, 2004). According to Spencer, school administrators in Houston estimated that as many as 40% of students, mostly Hispanic, never graduate from high school. This situation had only worsened two years later on a national basis as indicated by data released by the Associated Press (2007). “The dropout rate among Blacks, Hispanics, and other low-income students is about 60% . . . Experts warn that the high dropout rate will lead to economic and social problems” (The Eagle, 1).

So what is needed to help alleviate this problem? One solution is in the development and implementation of a strong, effective program to provide educational opportunities for families, both parents and children to attain their needed learning skills. Parents can then transition into productive family and community members while their children can enter early childhood programs more able and ready to learn. In Texas, the Even Start family literacy program has provided those opportunities and families, both parents and children, have been successful in making the most of those opportunities.

Follow-up Research on Parents

During 2005, an investigation was initiated to determine if participation in Even Start family literacy programs in Texas made any difference in the lives of the parents in those families (Seaman and Chen, 2005). This investigation was initiated because of allegations that previous national evaluations of the Even Start program showed that families who had participated in the program were no better off than families who had not participated as measured by test scores and similar assessments. However, no research had been conducted to determine how parents in Even Start had used their newly acquired educational skills to make their families’ lives better.

The investigation of Even Start in Texas began in 2005 with the steps found below.

1. A random sample of 20 programs was drawn from all non-first-year programs;
2. Two random samples of 30 parents were drawn within each program – the second sample being a “back-up” to assure that at least 30 parents would be available to be interviewed.
3. A telephone survey was developed and field-tested to acquire information from the parents about how their lives had changed since participating in Even Start.
4. Staff in each program interviewed the parents using the survey instrument and interviews continued until either 30 were attained or until both lists of names had been completed.
5. Findings from the investigation included:

• The two reasons indicated most by the parents for enrolling in Even Start were (1) to learn English, and (2) to earn their GED certificate. In general, almost all parents had either reached their goals or had made progress toward them. Over 90% of those parents interviewed were minority.
• Parents who had not read to their children nor helped with homework before participating in Even Start did so regularly even after they had been gone from the program for several years.
• After participating in the program, the percent of parents engaged in the workforce increased 184%. Their combined weekly wages had increased 287%.
• Twenty-five percent of those who had left the program had enrolled in further education or training. For many, they were the first in their family to ever do so.
• There was a 30% decrease in the number of families on public assistance across the 20 programs in the research sample after participating in Even Start.
• The number of parents who were spending time volunteering, mostly in schools, had virtually doubled since they had participated in Even Start. This provided the schools with some valuable volunteer assistance.

Need for Children’s Research Data

When data from the follow-up study of parents was shared with various National Centers and related organizations, the response was always positive, but was usually followed by the question, “That’s great, but what data do you have that shows that Even Start makes any difference with children”? From annual reports by independent evaluators, the State Coordinator could document that children in Even Start programs make academic gains each year and enter Pre-Kindergarten (Pre-K) ready to learn. However, how does this compare with children who have not participated in the program, hence the need for a study with an experimental (treatment) and a control (non-treatment) group.

Feasibility Study

During July and August 2006, to determine if a state-wide investigation comparing Even Start children to children who have not participated in the program was warranted, a “feasibility” study was implemented as outlined below.

1. To attain a “critical mass” of data for analysis purposes, nine Even Start programs, a ten percent non-random sample of all programs in the state, were selected to participate in the feasibility study.
2. To be selected in the feasibility study, the staff in the nine programs must have tested all Even Start children who entered Pre-Kindergarten (Pre-K) in fall, 2005, with the Peabody Picture Vocabulary Test (PPVT). The PPVT measures receptive vocabulary development for Standard English and is a screening test of verbal ability. Vocabulary development is a pre-cursor to reading comprehension which is crucial to success in school.
3. When a list of all of the standard scores of all children was attained from a local program, Coordinators had designated those children who had participated in Even Start the year before (2004-2005) for at least four months. The remaining scores were from children who had not participated in Even Start the previous year because the family was not enrolled in the program.
4. The mean scores of each group were compared through statistical analysis to determine if a significant difference existed.

Findings

Seventy PPVT Standard scores were acquired from 8 Even Start programs. A basic data analysis is shown in Table 1.

As seen in Table 1, the mean scores of the Even Start children were substantially higher than those who had not been enrolled in the program. In addition, for the non-Even Start children, a median score lower than the mean score indicates that there are more individual scores lower than the mean than above it. There were 36 children who had been enrolled in Even Start and 34 who had not been enrolled.

The difference in mean scores for the Even Start children and the Non Even Start children was statistically significant at the .001 level. Data are presented in Table 2 below. This indicates that the probability of this difference occurring could be expected to happen 999 times out of 1000 due to the treatment provided (Even Start participation) with similar populations. A difference this large would not be considered probable due to chance. It would be due to the learning which had occurred as the children participated in the Even Start program.

The findings of this feasibility study indicated that Even Start makes a significant difference in preparing children for success in school. However, because the findings could not be generalized to all Even Start programs in Texas, it was recommended that this study be replicated with all Even Start programs in the state using data from the 2005-2006 program year.

Statewide Study

In fall, 2006, a statewide study, paralleling the feasibility study was implemented. All 60 Even Start programs in Texas were asked to participate using the same procedures followed in the feasibility study. However, the scores were from all Even Start children, four years of age who entered Pre-Kindergarten by September 1, 2006 instead of 2005. The September 1 cutoff data was selected because it mirrors the cutoff date for Texas compulsory attendance age. Again, program coordinators indicated which children participated in the program during 2005-2006, and which had not participated.

When all data had been received, researchers were able to use those from 51 of the 60 programs (85%). Some programs sent incomplete data, two sent percentile scores instead of standard scores, and some did not have any children who were four years of age on September 1, 2006. The data analysis and findings are found below.

As seen in Table 3 the results were quite similar to those in the feasibility study. Although the mean and median scores for both groups were slightly lower, the differences between the two groups remained relatively the same.

Table 4 provides information in regard to the statistical analysis performed upon the differences between the mean scores of the two groups. Again, the results are virtually the same as those found in the feasibility study.

Based upon the findings of the statewide research study, it can be concluded that (a) given the same population of children, and (b) through providing the intervention found in Even Start programs, if an individual predicted that the differences that occurred would occur 999 times in one thousand investigations, that individual is likely to be incorrect only one time. In addition, a relatively high amount of confidence can be put into these findings as indicated by the “effect size”. According to Cohen, cited in Spatz (2005), an “effect size index” provides a mathematical way to determine how much difference really exists between two figures. In this analysis, the effect size fell between .5, a median effect size, and a .8, a large effect size, indicating that confidence can given to the findings.

These data clearly show that through participating in an Even Start family literacy program, children are significantly better prepared for success in school than children who have not had this kind of intervention. Referring to the data cited in the introduction to this report, this intervention could have a positive effect in reducing the incidence of school dropouts in later years in school because children are entering school more able to learn through increased vocabulary as measured by the Peabody Picture Vocabulary Test (PPVT).

To further document how Even Start benefits parents and children, a correlation between the PPVT scores of those Pre-K children who had been in the program for at least four months in 2005-2006 and those of their parents on their respective language test was conducted. As shown in Table 5, no significant relationship was found between the PPVT scores and the parents’ scores on the Basic English Skills Test (BEST) oral subtest, or the Language subtest of the Tests of Adult Basic Education (TABE). However, as shown in Table 5, a statistically significant relationship was found between the PPVT scores of the children and the scores of their parents who took the BEST PLUS. This indicates that children’s performance on the PPVT is directly related to their parents’ achievements in the classroom.

Adding these data to those acquired through the parent follow-up study cited below (Seaman and Chen, 2005) and briefly described above, it is quite apparent that the Even Start family literacy program, at least in a large state like Texas, can help low-income, minority families achieve in their educational endeavors and prepare for a better future. Based upon these findings, it is possible to project that the same thing is happening in other states where Even Start is functioning.

REFERENCES

Associated Press (2007). Experts: Dropout rate a big problem. Bryan-College Station, Texas: The EAGLE, January 15, 1.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. As cited in Chris Spatz: Basic statistics: Tales of distributions (2005). Belmont, CA: Wadsworth Thomson Learning, 77-78.

Klein, L., and Knitzer, J. (2006). Effective preschool curricula and teaching strategies. Columbia University: National Center for Children in Poverty, Mailman School of Public Health.

Seaman, D.F., and Chen, C. (2005). Follow-up data on parents who have exited Even Start programs in Texas, 2004, and 2005. Louisville: National Center for Family Literacy; and Washington, D.C.: National Even Start Association.

Spencer, J. (2004). Texas graduation rate worst in nation again. Houston Chronicle: January 10, 2007, 1.