Publications>
Research Bulletin

Phi Delta Kappa Center for Evaluation, Development, and Research
December 2001, No. 31

Effects of Three Kindergarten Schedules on Achievement and Classroom Behavior
by Charlene Hildebrand

There is a good deal of interest throughout the country in modifying the prevailing daily half-day kindergarten schedule to a full- or alternate-day one in the hope that increased time in school will net academic gains for young children. Child development specialist David Elkind notes, however, that "full-day kindergarten is a day-care initiative, not an academic initiative."1 In fact, Elkind contends, when a kindergarten child comes from a home already rich in educational experiences, the kindergarten schedule is not going to make much difference. But for the child who does not, the extra time does help.2 Does the additional time result in benefits for young children that warrant the additional expense? Unfortunately, there exists little empirical evidence to help guide decision making in this area.

In order to understand how best to support five-year-olds in a learning environment, one school district in a Midwestern city with a population of approximately 25,000 collaborated with a nearby university to study the relative effects of three different kindergarten schedules (full-day, alternate-day, and half-day) on children's academic achievement and classroom social behaviors. The study established benchmarks in that a common kindergarten curriculum was used in three different kindergarten schedules within one school district. Further, individual standardized assessments were used to collect pre- and posttest data. Informal data from parents and teachers supported the general dimensions of the kindergarten schedule issue.

ABOUT THE STUDY

The purpose for conducting this kindergarten study was to collect comparative data within one school district applying a common kindergarten curriculum in order to better understand the general dimensions of an early childhood education issue. The study was conducted in a central Nebraska school district consisting of seven elementary schools, three of which participated in the study. School principals and the superintendent made kindergarten schedule assignments to each participating school based on their perceptions of children's needs. Five veteran kindergarten teachers were involved in the study, all with college credits beyond the bachelor's degree; none had earned a master's degree. The teachers were consulted in the summer regarding their assignments, and parents were informed of their children's kindergarten schedule. Forty-seven children (29 males and 18 females) attended the full-day program, which was taught by two teachers in adjacent classrooms. Forty-four children (24 males and 20 females) attended the half-day program, which was taught by the same teacher in the morning and afternoon. Fifty-six children (31 males and 25 females) attended the alternate-day schedules, either Monday, Wednesday, and Friday or Tuesday, Thursday, and Friday. These alternate-day groups were taught by two teachers teaming in one large instructional area. The mean ages of the all-day, half-day, and alternate-day groups were 5.7, 5.7, and 5.9 years, respectively.

PROCEDURES AND INSTRUMENTS

Data were collected from two standardized assessments in reading and math, a teacher rating scale of classroom social behaviors, an informal writing assessment, an observation inventory of developmentally appropriate classroom practices, a parent questionnaire, and a teacher's self-response of reading orientation. The reading, writing, and math assessments were administered individually in October and April. Classroom social behavior ratings were collected in April, and observations of developmentally appropriate classroom practices were conducted throughout the academic year. The parent survey was distributed in February at the second parent-teacher conference.

Non-random assignment effects were reduced by using an analysis of covariance statistical procedure, with children's chronological age and pretest scores as covariates in reading and math. Multivariate analysis of covariance, with chronological age as the covariate, was used to analyze the classroom behavior data.

The Test of Early Reading Ability-2 (TERA-2) was administered to individual kindergarten students in October and April.3 This norm-referenced test, containing two equivalent forms appropriate for children ages three through nine years, measures ability to attribute meaning to printed symbols, knowledge of the alphabet and its function, and understandings of the conventions of print.

The Test of Early Mathematics Ability (TEMA) was administered individually in October and April.4 The TEMA is an untimed test normed for ages three through nine years that measures the following skills: (a) understanding concepts of greater than and less than; (b) counting; and (c) calculation. The TEMA items use both pictures and manipulatives to provide insight into children's mathematical ability.

In April the teachers were asked to rate their perceptions of children's social competence using the Hahnemann Elementary School Behavior Rating Scale (HESB).5 The HESB is a 60-item instrument designed to provide a standard system for identifying and measuring children's classroom behaviors that interfere with, facilitate, or reflect their ability to cope with academic expectations. Each of 14 behavior categories is defined by three, four, or five items to be rated by the teacher: (1) originality; (2) independent learning; (3) involvement; (4) productive with peers; (5) intellectual dependency with peers; (6) failure anxiety; (7) unreflectiveness; (8) irrelevant talk; (9) disruptive social involvement; (10) negative feelings; (11) holding back/withdrawn; (12) critical/competitive; (13) blaming; and (14) approach to teacher. Teachers rate students on each item using a Likert scale from one to five or one to seven, depending on the item. Item raw scores are an indication of the amount of that factor the kindergarten teacher judged the child to exhibit in classroom behaviors.

The Concepts of Writing, an informal assessment tool, was administered to all kindergarten children in October and April. This 10-item questionnaire is designed to survey children's ability to construct meaning using alphabetic symbols. Writing samples were rated using a nine-point holistic scoring guide that reflects the following developmental stages of writing: (1) produced mainly pictures; (2) produced letter-like forms (scribbling); (3) produced random letters; (4) wrote own name correctly; (5) wrote names of family members and friends and/or environmental print items; (6) wrote one word (other than own name) using invented spelling; (7) wrote single, nonenvironmental print, correctly spelled word; (8) wrote multiple random unfamiliar words using conventional and/or invented spelling; and (9) wrote sentence-like response.

Observations of kindergarten classrooms were made by a rater trained to use the Developmentally Appropriate Practice (DAP), a 16-item rating scale derived from the Nebraska Kindergarten Position Paper and consistent with the views of the National Association for the Education of Young Children.6 This informal instrument is designed to gauge the emphasis and emotional climate of kindergarten curricula. Each of 16 factors is defined and rated on a nine-point continuum ranging from "appropriate practices" to "inappropriate practices." The scores on the 16 individual factors are summed for a total raw score. The highest possible score on the DAP is 64.

During the spring parent-teacher conferences, the kindergartens distributed a Family Reading Inventory to survey parents' participation in their children's literacy development. The 14-item questionnaire provided data concerning reading habits and materials in the home, educational and occupational background of the parents, and children's previous childcare experiences.

Each kindergarten teacher's theoretical orientation to reading instruction was identified using the DeFord Theoretical Orientation to Reading Profile (TORP).7 The TORP uses a Likert scale for 28 statements about reading to classify teacher beliefs into phonics, skills, or whole language orientations.

RESULTS OF STUDY

Nonrandom, or purposive, sampling strategies were appropriate for this type of research, which is intended to answer questions for a specific situation. In order to increase the precision of the research design, an analysis of covariance (ANCOVA) statistical procedure for reading, math, and writing was used to determine the differences between three kindergarten groups, employing pretest scores and chronological age as covariates.

No significant differences for group means were found for reading raw score (p <. 014) and reading percentile (p <. 050). No significant differences were found for math and writing posttests. Post hoc analyses using t tests indicated significant differences in reading between all-day and half-day groups. Nonsignificant differences were found between all-day and alternate-day and between alternate-day and half-day schedules. The significant and nonsignificant differences between reading groups on percentile scores are shown in Table 1.

Results from the HESB revealed significant differences between groups using a multivariate analysis of covariance for the 14 subscales of behavior. Age was used as a covariate in this statistical procedure. A significant main effect was found for originality, independent learning, involvement, productive with peers, intellectual dependency, failure anxiety, approach to teacher, inattention, and academic expectation.

Post hoc analyses of the HESB results using the Newman-Keuls procedure indicated children attending the half-day program scored significantly higher on all four factors considered to facilitate learning (originality, independent learning, involvement, and productive with peers), while the alternate-day group scored significantly lower on all four of these factors. On the four factors considered to show a child's ability to cope with academic expectations (blaming, approach to teacher, inattention, and academic expectation), there were significant differences between groups on three factors. Regarding children's approaches to their teacher, the students attending the half-day program had a higher positive score, whereas students attending the all-day program had the median score, and the students attending the alternate-day program had the lowest score. Regarding inattention (teachers' measures of limitations of attentiveness), students attending the half-day kindergarten exhibited the least amount of inattentive classroom behaviors (M = 4.0), while students attending the alternate-day (M = 4.8) and all-day (M = 5.2) programs were perceived to demonstrate more tendency toward inattentive behaviors. Regarding academic expectation (teacher rating of children's achievement), there were significant differences between the half-day group and the alternate-day and all-day group. Students attending the half-day program scored significantly higher.

Results of the TORP self-report instrument measuring teachers' philosophic perspective of reading instruction indicated that the two alternate-day teachers oriented their beliefs about reading at the "whole language" level. The teachers of the full-day and the teacher of the half-day program oriented their philosophic perspective at the "skills" level.

An informal observational rating scale, DAP, was created for this study to rate teachers' use of developmentally appropriate practices. Teachers in the alternate-day schedule provided the highest level of developmentally appropriate practices. Teachers in the all-day schedule used the median level of developmentally appropriate practices. Teachers in the half-day schedule appeared to make least use of developmentally appropriate practices.

Answers to the parent survey, completed by 113 parents (77% response rate), identified the following themes:

• Families that reported having more children's books available at home tended to spend more time reading and less time watching television.
• If the father indicated he did not like to read, there tended to be fewer children's books and less time spent reading to children. Fathers who indicated that they preferred reading books or had to read on a regular basis for work, school, or other reasons tended to spend more time reading to their children on a regular basis.
• Parents who read magazines and newspapers on a regular basis were more apt to be seen reading to their children. There did not appear to be a similar relationship if either parent preferred to read books.
• The educational level of the mother was related to the number of children's books that were available in the home and the amount of time spent reading to children.

Statistical comparison of survey data with TERA (reading) scores revealed that students whose fathers and/or mothers were high school graduates performed significantly higher on the TERA posttest than did students whose fathers and/or mothers were not high school graduates.

Correlations of selected parent survey responses and kindergarten teachers' ratings of achievement for TERA (reading), TEMA (math), and writing posttest raw scores indicated that only kindergarten teachers' ratings of achievement were significantly correlated with children's reading, math, and writing achievement levels (p <. 0001). Mothers who were at home and Head Start participation were positively correlated with academic achievement. Some negative correlations were found between early child care and kindergarten academic progress.

DISCUSSION

This study was conducted by a Midwestern school district in collaboration with a nearby university to collect empirical evidence documenting the effectiveness of three different kindergarten schedules implemented in three elementary schools during one academic year. Those responsible for the care of young children -- teachers, parents, and administrators -- had expressed concern to the local board of education that other school districts were reporting success with modified kindergarten schedules while local children remained in the traditional schedule. In a review of the literature, the board discovered a lack of empirical evidence to help guide decision making in this area and conflicting results in existing kindergarten studies. This study was initiated to add to the kindergarten research base, as well as to explore local community dynamics.

The study's findings suggest that there are no clear differential effects of kindergarten schedules on both academic achievement and classroom social behaviors. Therefore, school districts can consider socioeconomic, financial, philosophical, and other factors in their decisions regarding kindergarten schedule.

The first major question addressed by this study concerned the academic progress of young children in three different kindergarten schedules. No significant differences between the standardized test scores of children in each group were found in math and writing. The full-day kindergarten group scored significantly higher in reading than did the alternate-day and half-day groups. This finding supports Elkind's observation that some children, especially those from homes and daycare centers with limited access to educational toys and books, may benefit from extra time to develop literacy skills.8 It also substantiates the National Reading Panel's emphasis on skills development in the early stages of reading acquisition.9

The study explored teachers' orientation to reading instruction. Teachers of students attending the full-day schedule, the group that demonstrated the highest reading raw scores and reading percentiles on the TERA-2, reported a "skills" preference toward reading instruction. However, teachers on this schedule used developmentally appropriate practices with young children inconsistently.

The second major question in the study concerned the effects of the three schedules on children's classroom behavior at the end of the school year. Unlike other kindergarten studies reporting few differences between factors of the HESB, multivariate analysis of covariance for the 14 subscales of behavior showed significant differences between groups favoring the half-day schedule on nine factors.10 Children in the half-day group scored significantly higher on factors considered to facilitate learning, with no differences between scores of the three groups on most factors considered to interfere with learning. Students in the half-day program were found to demonstrate the least "intellectual dependency" and "failure anxiety" in the classroom. They were most likely to "voluntarily approach the teacher." Children's "academic expectations," as perceived by the teacher, were reported to be significantly higher for the half-day kindergarten schedule. In comparison, the half-day youngsters were stronger than were the full-day and alternate-day children in prosocial behaviors that support learning in a classroom.

Research has demonstrated that parents have a strong influence on the literacy development of their children. Self-report data from the parent survey suggests that parents are aware of the importance of making children's books available in the home, reading to children, modeling reading by creating opportunities for children to see them read a variety of texts, limiting television viewing, and seeking at least a high school diploma themselves. Parents documented differences between genders regarding levels of education, with mothers reporting a higher level of some college/trade school experience (40%) than did fathers (27%). Analysis of covariance on children's reading achievement scores revealed a significant difference if the father or mother was a high school graduate. Children's academic achievement was positively correlated with mothers being full-time homemakers.

Although mothers have traditionally been viewed as the major providers of literacy experiences in the home, this study found that fathers play key roles in a variety of ways. If the father did not like to read, there tended to be fewer books in the home and less time spent reading to children. Whereas, if the father preferred to read either for pleasure or job-related reasons, there tended to be more time spent reading to children. The number of children's books in the home and the amount of time spent reading to children were positively correlated with the mother's level of education. These findings confirm those found in seminal studies regarding parental involvement in early literacy acquisition.

The professional preparation of caregivers has been one of the important issues regarding the quality of early childhood education today. This study documents the need for stronger integration and articulation of preparation across professional settings, including community-based training, two-year programs, and four-year programs for caregivers.11 It also supports the professional preparation of individuals in programs, such as Head Start, that have made a demonstrated difference in children's childcare experiences. This study found that early childcare experiences and preschool experiences were negatively correlated with kindergarten children's academic achievement. Head Start was positively correlated with children's early academic achievement. This finding helps to establish that, at an early age, education and care are inseparable and that both components must be consistent in a young child's daily environment for either to be relevant to the needs of children and families.

Finally, a Midwestern school district found that more time spent "in school" did not ensure academic and social/emotional success for young children. It could be said that the quantity of time in a kindergarten day is no more important than the quality of time. This leads to careful scrutiny of professional preparation for teachers at the baccalaureate level and effective ongoing professional development for those in the classroom. Most early childhood educators would agree that two or three hours of active involvement in intellectual activities integrated with indoor and outdoor play and cooperative group activities is sufficient academic stimulation for most five-year-olds for one day. Community dynamics determine how such a program is to be delivered. If child care is a major issue with the community, full-day kindergarten programs probably afford better quality child care than might otherwise be affordable. If parents believe it is their responsibility to care for the kindergarten child during at least part of the day as opposed to the school's, the traditional half-day schedule is appropriate. For the most part, parents and teachers in this and other studies tend to respond negatively to the alternate-day schedule due to lack of continuity both at home and at school.12 In sum, this study contributed to the knowledge base and general dimensions of the kindergarten issue at a community level. Kindergarten scheduling is not only an educational issue, but also one of quality child care and parenting responsibility.

ENDNOTES

1. Annie Gowen, "All-Day Kindergarten Boosts Reading," Washington Post, 10 September 2001, sec. B, p. 1.

2. David Elkind, The Unhurried Child: Growing Up Too Fast Too Soon (Reading, Mass.: Addison-Wesley, 1981).

3. Kim Reid, Wayne Hresko, and Donald Hammil, Test of Early Reading Ability-2 (TERA-2) (Austin, Tex.: Pro-Ed, 1989).

4. Herbert Ginsburg and Arthur Baroody, Test of Early Mathematics Ability (Second Edition) (Austin, Tex.: Pro-Ed, 1990).

5. George Spivak and M. Swift, Hahnemann Elementary School Behavior Rating Scale (Chester, Pa.: Widener University, Institute for Graduate Clinical Psychology, 1975).

6. Nebraska State Board of Education, Position Statement on Kindergarten (Lincoln, Nebr.: Nebraska Department of Education, 1984); Sue Bredekamp, Developmentally Appropriate Practice in Early Childhood Programs Serving Children from Birth Through Age Eight (Washington, D.C.: National Association for the Education of Young Children, 1987).

7. Diane DeFord, "Validating the Construct of Theoretical Orientation in Reading," Reading Research Quarterly 8 (1985): 167-92.

8. David Elkind, "Full-Day Kindergarten," Young Children 42, no. 5 (1987): 2.

9. National Reading Panel, Teaching Children to Read: An Evidence-Based Assessment of the Scientific Research Literature on Reading and Its Implications for Reading Instruction (Washington, D.C.: National Institute of Child Health and Human Development, 2000).

10. Dominic Gullo and Douglas Clements, "The Effects of Kindergarten Schedule on Achievement, Classroom Behavior, and Attendance," Journal of Educational Research 78, no. 1 (1984): 51-56.

11. National Institute on Early Childhood Development and Education, New Teachers for a New Century: The Future of Early Childhood Professional Preparation (Washington, D.C.: U.S. Department of Education, 2000).

12. Ibid., 11.