Effects of Selected Even Start Family Literacy Programs in Texas
on ParticipatingChildren and Parents

By
Don F. Seaman
Texas Center for the Advancement of Literacy and Learning
EAHRD, College of Education
Texas A&M University
October 2000

Presented at the National Even Start Association Conference
San Diego, California
dseaman@tamu.edu

Family Literacy programs, like other educational programs, are under increased pressure to document their positive effects upon participating families. However, both "Evaluation" and "Accountability" sometimes receive negative feelings from program staff when educational programs are reviewed for their effectiveness in delivering what is supposed to be delivered. At the same time, both are becoming increasingly important and are receiving more attention from policy-makers at federal and state levels.

Some evidence of the successes of family literacy programs have begun to emerge, but most of it is related to short-term effects, usually some kinds of measures of knowledge or skill increase in one or two years, for both children and adults. Early evidence of the use of qualitative measures of family literacy success was documented by Taylor (1983)  in her in-depth study of six white middle-class families living in suburban homes within a fifty-mile radius of New York City. Seaman, (1991) followed with an evaluation of the model used at the National Center for Family Literacy. However, more than a decade later, we are still looking for ways to determine to what extent desired change has occurred in achieving both short-term and long-term outcomes in family literacy (Benjamin and Lord, 1996). This need has also been documented by Hayes (1996) who has been evaluating family literacy programs for many years.

Staff at the Texas Center for Adult Literacy and Learning, Texas A&M University, provide external evaluation services to a number of Even Start programs in the state and the study sample was derived from  those programs. "Family" in this study refers only to the parents and the pre-school children participating in those programs included in this study.

The population sample for this research study consisted of two groups:

1. The first group consisted of 631 parents who were enrolled in 11 Even Start programs in Texas during the 1998-99 and 1999-2000 program years. Because of a variety of problems working with Even Start families, random selection was not possible. Instead, program staff, in cooperation with the researchers, selected two specific days in late spring (usually April) when all parents in attendance those days responded to a survey instrument, either in English or Spanish. Therefore, the sample was limited to those who were in attendance on those days. 

2. The second group consisted of 582 parents who had exited the same Even Start Programs during the 1996-1999 program years, for whatever reason, e.g., GED attainment, leaving the program to take employment, moving out of the community, etc. Because of the difficulty of locating this group, and since no external funds had been acquired to conduct this research, a telephone-interview survey was utilized to acquire the data. If more than 50 families had exited a program, a table of random numbers was used to select the group to be contacted. The target was 25 families per program. In some cases, whoever could be found was interviewed because many families could not be located or did not have telephones. The sample was limited to those who had telephones.

Therefore, the data cannot be generalized beyond those programs included in this research study. Readers should take from the data only what is meaningful to them and not try to project beyond what is contained in this report.

Both survey instruments were field-tested and revised, as needed, before being used to acquire the data.

Academic gains, measured by standardized tests, of those same parents and their children were determined by pre-and-post-testing while they attended the program. Data were compiled and analyzed in various ways. Frequencies and percentages were calculated, and various statistical procedures found in Martella, Nelson, and Marchand-Martella were used (1999). These included Chi-square, t-test, and  "standardized gain" calculations.

Finding 1: Parents have definite goals when enrolling in Even Start. However, those goals have changed somewhat during the past four years.

Table 1

Goals of parents who were surveyed while enrolled in Even Start (1998-2000) and those who had exited the program (1996-1999).

Goal	1996-1999 (n=582)* Follow-up Survey	Percent of Total	1998-2000 (n=631)* Parent Survey	Percent of Total
Learn English	234	28%	372	30%
Obtain GED certificate	364	43%	394	31%
Improve parenting skills	133	16%	234	18%
Obtain employment	94	11%	255	20%
No response	19	2%	6	1%

*Parents often had multiple goals

Finding 2: In general, parents feel they achieve their goals while in Even Start.

Table 2

Parents who had exited the program perceive that their goals were reached as follows:

a. Learn English	46%
b. Obtain GED	44%
c. Improve parenting	95%
d. Obtain employment	60%

Those who attain the GED usually pursue further educational skills.  Of the 189 who attained the GED certificate, 71, or 38% had enrolled in some kind of community college, technical school, or special training program.

Finding 3: Parents learn about Even Start from a variety of sources.

Table 3

Sources of information from which parents learn about Even Start (n=631)

Source of information*	Number	Percent
Family	90	13%
Friends	247	35%
Brochures from school	183	26%
Recruited by ES staff	111	16%
Invite by program parents	72	10%

*Some heard about the program from more than one source.

Even though friends and brochures sent from school were most effective, it appears that recruitment is effective when a variety of techniques is utilized.

Finding 4: Parents in Even Start increase their academic abilities.

Table 4

Increase in parent's English skills after participating in Even Start as measured by the Basic English Skills Test (BEST) 1998-1990 and 1999-2000

Subject	n	Pre/post-test	Mean	Std. Deviation	Gain	Std. Gain
Listening	389	Pretest	11.3	14.92	2.8*	.19
		Posttest	14.1	15.90
Communication	377	Pretest	21.6	12.96	8.4*	.65
		Posttest	30.0	12.82
Fluency	397	Pretest	7.5	7.01	5.4*	.76
		Posttest	12.9	12.21
Reading	350	Pretest	23.1	15.80	7.3*	.46
		Posttest	30.4	16.30
Writing	326	Pretest	13.7	7.68	5.8*	.76
		Posttest	19.5	11.06

*Statistically significant, p < .001

Gains in listening, communication, fluency, reading and writing were statistically significant at the .001 level of confidence.  This indicates that the gain could have been achieved by chance less than 1 time in 1000.

A standardized gain calculation reveals how much of an effect the intervention provided by the program had on gains made by the participants, i.e., the higher the standardized gain, the bigger the impact of the intervention on the scores gains.  The standardized gain is said to be small (.20), medium (.50), or large (.80).

Table 5

Increase in parent's academic skills after participating in Even Start as measured by the Test of Adult Basic Education (TABE) 1998-1990 and 1999-2000

Subject	n	Pre/post-test	Mean	Std. Deviation	Gain	Std. Gain
Reading	318	Pretest	518	91.31	18*	.20
		Posttest	536	91.12
Math	330	Pretest	502	95.81	24*	.24
		Posttest	526	91.49
Language	317	Pretest	508	98.53	20*	.20
		Posttest	528	94.22

*Statistically significant, p < .001

Gains in reading, math and language were statistically significant at the .001 level of confidence. This indicates that the gains could have been achieved by chance less than 1 time in 1000. Less confidence can be held about those gains because of lower standardized gain scores.

Finding 5: Participating in Even Start increases the amount of time parents read to their children.

Table 6

The amount of time parents read to their children increased after participating in Even Start (n=211)

	None	1time/wk.	2times/wk.	4times/wk.	Every day
Before ES	24%	37%	21%	10%	8%
After ES	4%	8%	26%	31%	31%

The percentage of parents who read to children four times per week or every day more than tripled after participating in Even Start.

Finding 6: Participating in Even Start increases the amount of time parents help their children with learning activities.

Table 7

Amount of parental help with their children's learning activities before and after participating in Even Start (n=284*)

Time	None	1 time/wk	2 times/wk	4 times/wk	Every day
Before ES	63 (22%)	40 (14%)	46 (16%)	48 (17%)	87 (31%)
After ES	8 (2%)	9 (3%)	35 (12%)	80 (28%)	152 (54%)

*There were 32 non-respondents. (These are 1998-99 survey data.)

In the 1999-2000 survey, the age groups of children of Even Start parents were divided into two age groups - ages 0-4, and kindergarten and higher grades - for the question about parental help with learning activities.

Table 8a

Amount of parental help with their children's learning activities before and after participating in Even Start (ages 0-4)

Time	None	1 time/wk	2 times/wk	4 times/wk	Every day
Before ES n=232	18 (7%)	36 (16%)	48 (21%)	60 (26%)	70 (30%)
After ES n=232	2 (1%)	10 (4%)	14 (6%)	58 (25%)	148 (64%)

Table 8b

Amount of parental help with their children's learning activities before and after participating in Even Start (Kindergarten and higher grades)

Time	None	1 time/wk	2 times/wk	4 times/wk	Every day
Before ES n=181	18 (10%)	31 (17%)	34 (19%)	30 (16%)	68 (38%)
After ES n=180	8 (5%)	4 (2%)	16 (9%)	40 (22%)	112 (62%)

One of the goals of Even Start is to promote more involvement of parents in their children's education. The data from this study document that that goal is definitely being attained. Statistical analysis indicated a significant difference at the .001 level which means that that goal could have been achieved by chance less than 1 time in 1000

Follow-up data from families who had exited the program during the two previous years showed very similar results. Therefore, the data document that parents continue to help their children even after exiting the program, a trend that is desired by all stakeholders in Even Start family literacy.

Finding 7: Participating in Even Start increases the amount of time parents spend with their children's teacher.

Table 9

Number of times per week parents visit with their children's teacher before and after participating in Even Start (n=258*)

Time	None	1 time/wk	2 times/wk	4 times/wk	5 times/wk
Before ES	97 (38%)	95 (37%)	40 (16%)	9 (3%)	17 (6%)
After ES	21 (8%)	88 (34%)	90 (35%)	28 (11%)	31 (12%)

*There were 58 non-respondents. (These are 1998-99 survey data).

Visiting their children's teacher(s) should become important to parents after participating in Even Start. The data in table 9 indicate that the program fosters that kind of activity. Again, the score differences between the data sets were statistically significant at the .001 level.

In the 1999-2000 survey, the age groups of children of Even Start parents were again divided into two age groups - ages 0-4, and kindergarten and higher grades - for the question about parental visit of their children's teacher.

Table 10a

Number of times per week parents visit with their children's teacher before and after participating in Even Start (ages 0-4)

Time	None	1 time/wk	2 times/wk	4 times/wk	5 times/wk
Before ES n=197	75 (38%)	62 (31%)	34 (17%)	11 (6%)	15 (8%)
After ES n=197	18 (9%)	49 (25%)	53 (27%)	36 (18%)	41 (21%)

Table 10b

Number of times per week parents visit with their children's teacher before and after participating in Even Start (kindergarten and higher grades)

Time	None	1 time/wk	2 times/wk	4 times/wk	5 times/wk
Before ES n=164	38 (23%)	59 (36%)	40 (24%)	10 (6%)	17 (11%)
After ES n=162	13 (8%)	47 (29%)	35 (22%)	32 (19%)	35 (22%)

Follow-up data from those parents who had exited the program during previous years showed similar results, except that higher percentages were found at the upper levels, i.e., 4 and 5 per week.  This indicates that parents continue to visit their children's teacher even after exiting the program, a positive effect.

Finding 8: Children in Even Start increase their academic abilities.

Table 11

Increase in children's academic abilities after participating in Even Start as measured by the Preschool Language Survey (PLS-3) (n=455)

Subject	Before/After	Mean	Std. Dev.	Gain	Std. Gain
Auditory Compr.	Before After	87.0 92.5	14.71 14.75	4.5*	.39
Expressive Communic.	Before After	87.3 93.9	16.35 15.97	6.6*	.40
Standard Score	Before After	175 186	28.85 28.63	11*	.39

*Statistically significant, p < .001

Gains in auditory comprehension, expressive communication and standard score were statistically significant at the .001 level of confidence.  This indicates that the gains could have been achieved by chance less than 1 time in 1000.

Finding 9: Even Start children improve in school as perceived by their parents.

Of 344 parents who indicated they had children in school (many of the parents had only children 0-4 years of age), 275 responded to a question about whether their children had improved in school since the family had entered the Even Start program. Of the 275, 256, or 93% indicated that improvement had occurred, whereas 19, or 7% believed their children had not improved. Ways in which improvement had occurred included:

a. Better grades
b. Improved ability to speak English
c. Lower absenteeism

These improvements can affect economics as well as academics.  For example, in one school district, the program coordinator, through data, determined that after participating in Even Start, the absenteeism rate of the children in school of families in the program was reduced by 2.5 days per year. This didn't seem like much until the number of children per family, the number of families enrolled, and the Average Daily Attendance rate allowed by the state were calculated. This amounted to about $26,000 per school site that the school gained by the higher attendance rate, and there were multiple sites in the program. You can be assured that administrators in that school really support Even Start.

Finding 10: Parents are more likely to become employed after participating in Even Start programs.

Follow up data in regard to employment on those families who had exited from the program (for whatever reason) are discussed in the tables below.

Table 12

Employment status of Even Start programs before and after participation in the program.

	Total number	Employed	Unemployed	No Response
Before ES	574	110 (19%)	295 (52%)	169 (29%)
After ES	562	256 (46%)	306 (54%)	0 (0%)

Although employment of parents is not a stated goal in the Even Start legislation nor of many families when they enroll, as seen in Table 1, it becomes the goal of some parents. Economic status is one of the criteria in the "most in need" formula used by many program staff to determine who can attend the program. Simply stated, many parents desperately need the money and they seek employment before acquiring their GED. This issue is also reflected in the recent welfare reform legislation.

The types of work held by most parents when the telephone survey was administered included the following:

Housekeeping, custodian Waitress, cook Teacher's aide Sales Correction employee

Cashier, clerk Childcare/eldercare worker Assembly line Driving trucks, other vehicles (there were others)

When asked the types of work they desired 5 years in the future, their responses included the following:

Any better paying job	Sales
Teacher	Hospital or other medical work
(Some wanted to remain at home until their children enter school.

Table 13

Weekly wages earned by Even Start parents.

	Hourly average	Hours per week	Total number employed	Total wages per week
Before ES	$ 6.22	33.4 hrs	110	$ 21,148
After ES	$ 6.60	30.8 hrs	256	$ 53,856

As seen in Table 13, even though the hourly wage isn't greatly different and the total hours worked per week by the parents is almost the same, when factoring into the equation the total number of parents employed, a significant difference is found in the total weekly wages earned by all of the parents after they had participated in Even Start. These data show a positive association between participation in Even Start and the income level of the many parents. 

Finding 11: Parents in Even Start programs believe they have become good role models for their children.

Many parents indicated that they had never believed they could be a "role model" for the kind of person they wanted their children to become. However, they gave several examples of how they believe they are now performing that function.

1. Showing their children the value of education. By attending classes, reading to their children, and emphasizing the importance and pleasure of learning, parents model the kinds of behavior they want their children to follow. In addition, their self-confidence that grows as they attain new learning skills is, in turn, acquired by their children.
   
2. Speaking better English. The majority of Even Start parents in Texas speak another primary language and learning English is a major goal (see table 1). They make significant gains each year as indicated by pre-and-post-test scores on the Basic English Skills Test, the ESL assessment instrument required in adult literacy education in Texas. They also indicated that they try to speak more English at home, but that isn't always an option.
   
3. Becoming better parents. In previous years, some program's coordinators had indicated that parents wanted less parenting and more adult education. However, as shown in Table 1, that is changing. When asked, parents can discuss new parenting techniques they now use at home that were learned in Even Start. They also express surprise and pleasure about how well those new skills have improved communication among family members.

Finding 12: Parents who participate in Even Start programs become better consumers.

When asked whether they had acquired better shopping skills from participating in Even Start, of the 531 parents who had exited the program, 394 or 74% stated that they had acquired better skills, whereas 119 or 23% stated that they had not acquired better shopping skills. Eighteen or 3% did not respond to the question. Among those skills indicated most often were:

a. Can read labels better
b. Now know how to use coupons
c. Know how to look for "specials"
d. Can now compare prices among products
e. Can now budget money more wisely

As indicated above, not only are former Even Start parents now earning more money, most feel they can also use it more wisely. In addition, many feel they now have better eating habits at home.

Finding 13: Even Start fosters some community-related activities by parents.

Through follow-up interviews of those who had exited a program, researchers learned that of the 582 individuals responding, 252 (43%) said they were registered to vote. Of those 252 parents, 100 (40%) indicated that they had registered after enrolling in the Even Start program. Therefore, of the total respondents, 17% had registered to vote after participating in Even Start. (It is quite possible that some of the parents were not eligible to register).

In addition, of 402 respondents, 77 (19%) indicated that they had acquired a driver's license while in Even Start. That is critical because many of those who responded that they didn't have a license also indicated that they drive on a regular basis. In Texas, Department of Public Safety employees will come to programs and offer instruction for acquiring a driver's license. They also want safe drivers on the highway.

Finding 14: Even Start produces healthier families.

Almost 75% of the parents felt that their families were healthier because they had participated in Even Start. When asked "why", they responded:

a. "I know how to contact a doctor;"
b. "I can answer the doctor's questions and explain where I hurt;"
c. "I can read prescriptions and know how to follow directions;" and
d. "We know more about how to stay healthy and avoid getting sick so much."

Finding 15: Families who participate in Even Start feel they have more control over their lives.

When asked about this issue, of the 574 parents who responded, 525 (91%) felt that they now have much more control over their lives since entering the Even Start program. Reasons for that feeling included:

1. They have better control over their children, using positive approaches learned in the program;
2. They know when and how to borrow money, avoiding "credit traps";
3. They know how to "stretch" their money through wise shopping techniques;
4. They can speak English and can converse with sales clerks, "check-out" personnel, and other strangers in public; and
5. They know now to access community social services, when needed.

The data in this limited study document that Even Start family literacy is significantly associated with positive changes demonstrated by the participating families. Not only does the program meet the basic tenets of the legislation, it goes beyond it by enabling families to acquire the kinds of skills needed to become more functional in their daily lives. Not only are the families benefiting, but so are their local communities.

Benjamin, L. A., and Lord, J. (1996). Family Literacy: Directions in research and implications for practice. Washington, DC: U. S. Office of Education.

Hayes, A. (1996). "Longitudinal study of family literacy program outcomes". In L. A. Benjamin and J. Lord (Eds.), Family Literacy: Directions in research and implications for practice. Washington, DC: U. S. Office of Education.

Martella, R. C., Nelson, R., and Marchand-Martella, N. E. (1999). Research methods: Learning to become a critical research consumer. Boston: Allyn and Bacon.

Seaman, D. F. (1991). Follow-up study of the impact of the Keenan model of family literacy. Adult Basic Education, 2, 2, 71-83.

Taylor, D. (1983). Family literacy: Young children learning to read and write. Portsmouth, NH: Heinemann.

Back to Documents Index