Early Childhood Research Quarterly 19 (2004) 413–430 Parent involvement in Head Start programs: the role of parent, teacher and classroom characteristics Dina C Castro∗ , Donna M Bryant, Ellen S Peisner-Feinberg, Martie L Skinner FPG Child Development Institute, The University of North Carolina at Chapel Hill, 105 Smith Level Road, CB# 8180, Chapel Hill, NC 27599-8180, USA Abstract The purposes of this study were to determine the extent and types of parent involvement in Head Start programs, and to examine the relations between parent participation and family, teacher and classroom characteristics Parents (n = 1131) and teachers (n = 59) from four Head Start programs participated Data were gathered through volunteer logs, parent interviews, teacher questionnaires, and classroom observations The most frequent type of parent involvement activity was helping out in the classroom, followed by attendance at parent meetings This pattern was consistent across the year (fall or spring), and across total amount of participation (i.e parents participating one, two, three or more times in the year) Parent employment was the strongest predictor of parent involvement compared to other parent characteristics Among teacher and classroom characteristics, classroom quality was the strongest predictor of parent involvement Also, teachers with more years of experience in Head Start had more total hours of volunteering in their classrooms and had volunteers returning more times Teachers’ reports of the involvement of parents in their classrooms were moderately correlated with volunteer logs, while parent self-reports of their involvement were only modestly correlated with volunteer logs, indicating that teachers may be more accurate than parents when reporting parent involvement activities © 2004 Elsevier Inc All rights reserved Keywords: Head Start programs; Preschool classroom; Parent involvement Parental involvement has been a significant cornerstone of Head Start since its beginnings (Zigler & Styfco, 2000) Its two-generational strategy distinguished it from many other early childhood intervention programs in existence in the early sixties (Consortium for Longitudinal Studies, 1983) Head Start’s pioneering perspective regarding parent involvement as key to both children’s and families’ growth and ∗ Corresponding author Tel.: +1 919 962 7363; fax: +1 919 966 7532 E-mail address: dina castro@unc.edu (D.C Castro) 0885-2006/$ – see front matter © 2004 Elsevier Inc All rights reserved doi:10.1016/j.ecresq.2004.07.005 414 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 development was consistent with Bronfenbrenner’s ecological approach to child development, a philosophy that was gaining increasing acceptance by early childhood educators and researchers at the time Bronfenbrenner (1979, 1988) argued that to be effective and have any lasting impact, early intervention programs would need to involve the children’s parents and communities, so that all environments affecting children would foster similar goals Though the majority of studies showing the positive effects of the involvement of parents in their children’s development and school performance have examined the elementary through high school years (e.g Eccles & Harold, 1996; Epstein & Dauber, 1995; Epstein, 1996; Henderson & Berla, 1994; Scott-Jones, 1995), a number of studies of the early childhood years have demonstrated that parental involvement has beneficial and lasting effects on both children and their parents (Marcon, 1999; Miedel & Reynolds, 1999) These findings have also been replicated in Head Start (Lamb-Parker et al., 1997) Roskos and Neuman (1993) suggested that active involvement in children’s educational activities is most strongly linked to improved parental attitudes, understanding, and behaviors related to how to interact with their children In a study with parents of disadvantaged children participating in the Chicago Longitudinal Study, Miedel and Reynolds (1999) found that even after controlling for family background, greater parent participation in preschool and kindergarten activities was later associated with higher reading achievement, lower rates of grade retention, and fewer years in special education when children were in eighth grade Marcon (1999) reported the positive impact of parent involvement on the development of low-income, urban children attending public pre-kindergarten or Head Start programs This study found that both greater parent involvement and more active types of involvement were associated with better language, social, motor, and adaptive development and greater mastery of early basic school skills of their children Another study conducted with Head Start children and their parents found that parental involvement was predictive of children’s gains in learning skills and classroom behavior (Taylor & Machida, 1994) Research has clearly demonstrated that greater parent involvement in their children’s early childhood programs is associated with greater gains in the cognitive, language, and socio-emotional domains Accordingly, the research field is moving from the question “Is parent involvement important or beneficial for children’s development and school success?” to the question, “How can we make parent involvement happen?” The present study focuses on the precursors of parent involvement in Head Start programs, examining a number of family, teacher and classroom characteristics Although the research evidence supports the positive effects of parent involvement and the Head Start performance standards emphasize parent involvement, Head Start programs face many challenges in trying to increase the number of parents who actively participate in Head Start activities Some research has been conducted on barriers to parent involvement and family factors associated with parent involvement in Head Start Employment, conflicting schedules, child care needs and mental health problems have been among the barriers most frequently reported by parents (Lamb-Parker et al., 2001) Pyle (1988) found maternal education to be the most important family factor associated with parental involvement While some of the ways Head Start programs encourage parent involvement are surely more effective than others, there is limited knowledge in this area In addition, the specific types of parent involvement activities offered by Head Start programs and the level of involvement achieved may be related to selected parent and teacher characteristics Knowledge about such relations could help programs design approaches to increase parent involvement, thus strengthening the benefits of Head Start for both children and families The different ways in which parents are involved in their children’s education have been studied, mostly with parents of school-age children, and several classifications have been proposed For example, Epstein D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 415 (1992) proposed six types of involvement: parenting, communicating, volunteering, learning at home, decision-making and collaborating with community Hester (1989) suggested five ways of involving parents: communication with parents, parents as teachers, parents as supporters of activities, parents as learners, and parents as advocates Greenwood and Hickman (1991) discussed five types of parent involvement: parent as audience (e.g parent–teacher conferences, open houses), parent as volunteer (e.g classroom volunteer), parent as teacher of own child (e.g helping with homework at home), parent as learner (attending workshops on parenting, child development, etc.), and parent as decision maker (e.g parent council) Henderson, Marburger and Ooms (1986) summarized the different types of parent involvement in two broader categories, one that includes parent activities aimed at supporting the school program (e.g classroom volunteering, fundraising, policy council), and the other that includes parent activities for providing assistance to own child (e.g helping with homework, attending parent–teacher conferences and school events) Although a number of studies have been conducted on the topic of parent involvement, they leave many questions unanswered due to methodological weaknesses that affect many of the studies Baker and Soden (1998) reviewed 200 parent involvement research studies, identifying methodological limitations in the majority of them While recognizing that “the sound studies that exist have consistently found strong parent involvement effects” (p 1), they noted that confusion exists among practitioners and researchers regarding what constitutes effective parent involvement, mainly because many studies lack scientific rigor According to these authors, one of the major methodological limitations of existing parent involvement research has been the use of parent and teacher retrospective reports to assess parent involvement rather than using more objective measures such as direct observations or ongoing standard record keeping Another limitation identified was the inconsistency in defining parent involvement across studies, which makes it difficult to assess cumulative knowledge across different studies With these considerations in mind, we used both objective and self-report measures of parent involvement from both parents and teachers to (a) determine to what extent and in which ways parents are involved in Head Start programs, and (b) examine the relations between parent volunteering and family, teacher, and classroom characteristics The present study uses a variety of measures from multiple sources to address the following research questions: (1) What are the patterns of parent involvement in Head Start? (2) What are the relations between parents’ and teachers’ reports and objective measures of parent involvement? (3) What are the relations between parent characteristics and participation in Head Start? and (4) What are the relations between teacher and classroom characteristics and parent participation? Methods This research was conducted as part of the Head Start Quality Research Center, established to conduct research about the relations among program quality measures and between program quality and child and family outcomes in Head Start programs 1.1 Participants Parents and teachers from four Head Start Programs in a Southeastern state participated in this study These programs represented the range of demographic characteristics found in the state, and served about 1600 low-income children and their families from urban, suburban and rural areas The families served by 416 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 these programs tend to work in service-level jobs for business, in food and hotel services, or in agriculture The racial/ethnic distribution of the Head Start families in all four programs is slightly greater than 80% African–American, with small percentages of White, Latino, and Asian families All four programs use a center-based service delivery model with h/day classrooms serving 3- and 4-year-old children and their families Altogether volunteer data (volunteer logs) were available from 1491 Head Start volunteers (including parents and community volunteers) Of the total number of volunteers 1131 were parents whose children were enrolled in one of 62 classrooms Of the 780 parents who volunteered in the spring, 127 (16%) participated in parent interviews In addition, of the 62 teachers, 59 (95%) completed the teacher questionnaire and 35 (56%) completed the teacher questionnaire and participated in classroom observations Among parents who were interviewed (n = 127), 56.5% were employed, 66% were married, and the majority had a family size of 3–4 members (65%) The mothers’ education levels varied, with 45% having more than high school education The average monthly family income was US$ 1228 The teachers completing the teacher questionnaire (n = 59) had an average of 14.5 years of education, ranging between 12 and 20 years and an average of 8.6 years working in Head Start, ranging from 0.6 to 32 years 1.2 Procedures and instruments Table summarizes the outcome and predictor variables used in this study and their corresponding instrument source 1.2.1 Volunteer logs To obtain data on the frequency and types of parent involvement activities in Head Start, volunteer logs were collected monthly in 62 Head Start classrooms in the four participating Head Start programs Information on volunteer activities is recorded on a regular basis by all Head Start programs as one way of meeting their federally required in-kind match (volunteer hours are given a cash value that constitutes a match contribution from the agency/community) Each of the participating programs agreed to gather these data using a standardized “Volunteer Data Form” that was developed together by the researchers and the Head Start staff This form was a monthly log that included information on category of volunteer (e.g parent, other relative, community member), number of times volunteered, type of volunteer activity performed each time (e.g helping in classroom, going on field trip, attending parent meeting, preparing materials at home), and time spent at each activity for every volunteer activity that occurred Teachers required all volunteers to sign in each time they volunteered, and helped them so, if needed If parents helped out from home preparing materials or doing laundry, teachers recorded this activity for them Researchers gathered these data from each classroom once a month For this study, we used only the parent volunteer data (n = 1131) not data from community volunteers At the classroom level, both the total number of parents who volunteered at least once, and the total number of volunteer hours were calculated 1.2.2 Parent interviews Parent interviews were conducted in the second half of the spring semester as part of the Head Start Quality Research Center’s activities, to gather information about various parent characteristics A letter inviting parents to participate in this interview, along with a consent form, were sent by teachers to all parents who had volunteered at least once in some capacity in the spring semester (n = 780) Parents D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 417 Table Study variables description and instrument source Study variables Outcome variables Number of hours parents volunteered Number of times parents volunteered Number of parent volunteers per classroom 4.Parent report of involvement in Head Start Description Instrument source Included volunteering in any type of activity at the Head Start center, and helping from home Total number during a Head Start year Parents who volunteered at least once in each classroom during a Head Start year Score on a 14-item scalea Volunteer logs Volunteer logs Volunteer logs Parent interview Predictor variables Parent characteristics Maternal education Parent employment status Parent satisfaction with Head Start Parent activities with child at home Barriers to participation in Head Start Teacher characteristics Teacher education Teacher experience in Head Start Teachers’ perceptions of involvement of parents in their classrooms Teachers’ reports of parent involvement Classroom characteristics Classroom quality a (a) less than high school (b) high school graduate, or (c) more than high school (a) unemployed (b) working part-time, or (c) working full-time Score on an 8-item scalea Score on a 21-item scalea Score on a 14-item scalea Parent interview Parent interview Parent interview Parent interview Parent interview Years of education Years working in Head Start Score on a 17-item scalea Teacher questionnaire Teacher questionnaire Teacher questionnaire Includes different types of activities offered by the Head Start program (e.g open house, home visits, children performances, field trips) Teacher questionnaire Score on the 43-item scalea Classroom observation with ECERS-R These scales are described in Section 1.2 who agreed to participate (n = 127) sent their signed consent form to the teacher Teachers notified researchers and a trained research assistant contacted parents to arrange the interview date and time, and conducted the interview with parents A number of questions from The Head Start Family and Child Experiences Survey (US Department of Health and Human Services, 1998) were included in the parent interview protocol Data used in the present study include family demographic information (i.e parents’ education and employment, family income, and household size), parents’ satisfaction with their child’s Head Start program (mean score on questions concerning Head Start’s helpfulness in areas such as child growth and development, preparation for school, and family support where = least helpful and = very helpful), barriers to parent involvement (total number of Yes responses to 14 potential barriers that have kept parents from participating in Head Start such as need for child care, interference with work or 418 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 school schedules, need for transportation, etc.), parents’ activities with their child at home (score on a 21-item scale, based on the National Household Educational Survey (1999), that contains questions about activities parents did with their child during the past week and month such as reading at home, telling stories, playing games, going to the library, visiting a zoo or aquarium, going to a park or to the movies, attending a community event), and parent reports of their volunteer activities in Head Start (score on a 14-item rating scale about how often parents have participated in Head Start activities such as helping out in the classroom, going on field trips, attending meetings and events, preparing materials, among others) The parent interview protocol is available from the first author 1.2.3 Teacher questionnaire During the spring semester, all teachers from the four participating programs (n = 62) were invited to complete a staff questionnaire that included questions on parent involvement Fifty-nine teachers (95%) completed the questionnaire These data were used to explore associations between teacher characteristics and parent volunteering Demographic information from teachers includes education and experience in Head Start Teacher perceptions about involvement of parents in their classrooms were measured using the Teachers’ Beliefs About Parents Scale (Feldman & Gerstein, 1988) This is a 17-item scale that includes questions about the percentage of parents in the classroom who are interested in their child doing well, are easy to involve in Head Start, are able to help their child learn, work with their child on learning activities at home and other similar items, using a 10-point scale from to 100% The scale has good internal consistency as reported by the authors (Cronbach’s alpha = 0.95) The internal consistency reliability of the scale based on data from this study was also good (Cronbach’s alpha = 0.96) The teacher questionnaire also included reports of parents’ participation in different types of activities offered by the Head Start program (e.g open house, helping in the classroom, parent–teacher conferences, home visits, field trips) The teacher questionnaire is also available from the first author 1.2.4 Classroom observations These data were used to examine the relationship between classroom quality and parent volunteering All teachers in the four participating Head Start programs were invited to participate in the spring classroom observations Each teacher received a letter of invitation and a consent form Thirty-five of the 62 teachers (56.4%) agreed to participate The Early Childhood Environment Rating Scale (ECERS-R, Harms, Clifford, & Cryer, 1998) was used to measure overall classroom quality The two research assistants who collected ECERS-R observation data were trained by one of the gold standard trainer/researchers on this measure, who was a member of the research team They both had demonstrated greater than 85% inter-rater reliability in a previous study The ECERS-R includes 43 items rated on a 1–7 scale from inadequate to excellent, with scores of or above generally considered indicative of practices in the developmentally appropriate range This scale measures seven major areas: space and furnishings, personal care routines, language-reasoning, activities, interaction, program structure, and parents and staff (includes opportunities for professional growth and parent involvement practices) The total mean item score was used in these analyses 1.3 Data analysis Descriptive analyses were conducted on data from volunteer logs and parent interviews to determine to what extent and in which ways parents were involved in Head Start programs, as well as to identify the D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 419 characteristics of parent volunteers Correlations were calculated among parent reports (from the interviews), teacher reports (from the questionnaires) and number of volunteers and number of volunteering hours (from volunteer logs) to determine the extent to which these different sources of data were related Next, the relations between parent characteristics and their participation in Head Start were examined First, bivariate correlations were performed to examine the association between parent characteristics and parent volunteering Second, a multivariate analysis of variance (MANOVA) was performed to examine whether parent characteristics, including demographic characteristics (maternal education and employment) and parent beliefs and behaviors (satisfaction with Head Start program, barriers to involvement in Head Start, and activities with child at home) predict their volunteering in Head Start (total number of hours from volunteer logs and self-reports of volunteering from interviews) Subsequently, the associations of teacher and classroom characteristics with parent volunteering were examined First, bivariate correlations were estimated between teacher and classroom characteristics and the parent volunteering variables Second, a MANOVA was performed to test whether teacher characteristics (time working in Head Start, years of education, and perceptions about involvement of parents in their classrooms) and classroom characteristics (classroom quality) predict parent volunteering (total hours per classroom and number of individuals volunteering) Results 2.1 What are the patterns of parent involvement in Head Start? Parents were, by and large, the most important source of volunteers for the Head Start programs in this study They constituted 76% of the total number of volunteers (1131 of 1491) As seen in Fig 1, most parents (59%) volunteered only one or two times in a year, another 9% volunteered three times, and 32% volunteered four or more times (up to more than 100) Also, the total number of parent volunteers decreased from fall to the spring, with the largest number of parents (47%) volunteering in both fall and spring, while fewer parents volunteered only in the fall (31%) or only in the spring (22%) (see Fig 2) Fig Frequency of parent volunteering (n = 1131) 420 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Fig Parent volunteers by time of year volunteering (n = 1131) As shown in Fig 3, helping out in the classroom was the most frequent type of volunteer activity for parents in this study (35%), while attendance at parent meetings, including program and center-wide meetings (e.g policy council, parent orientation and parent education meetings) and classroom meetings (e.g classroom committee and parent–teacher conferences), was the next most frequent type of activity Fig Parent involvement by type of volunteer activity (n = 1131) D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 421 Fig Parent involvement by type of volunteer activity and number of times volunteered (n = 1131) (24%) Helping on field trips accounted for 14% of volunteering, while 12% was accounted for by volunteering from home (e.g preparing class materials, doing laundry) The remaining 15% of parent volunteering included a variety of other activities such as bus monitoring, fundraising, and helping out at festivals, and other special events This pattern of frequency in the type of volunteer activities for parents was consistent when data were analyzed according to the number of times parents volunteered (see Fig 4) and the time of year parents participated (see Figs and 6) In both cases, helping out in the classroom was the most frequent activity in which parents participated, followed by parent meetings Attendance at parent meetings was about the same for parents who volunteered one time and those who volunteered four or more times, while the latter group helped out in the classroom much more often From the fall to the spring semesters, helping out in the classroom decreased but attendance at parent meetings stayed about the same (see Fig 5) Fig Parent involvement by type of volunteer activity and time of year volunteering (n = 1131) 422 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Fig Parent involvement by type of volunteer activity and pattern of participation (n = 1131) 2.2 What are the relations between parents’ and teachers’ reports and objective measures of parent involvement? Comparisons were conducted between data from the volunteer logs (daily records of each parent volunteer activity) and parent interview data (parents’ recollections when asked about how often they had participated in a number of activities listed) There were modest but significant correlations between these two sources of data for the total number of hours volunteered, r(127) = 36, P < 001, and the number of times volunteered, r(127) = 23, P < 01 These results suggest that parents were somewhat accurate reporters of their involvement in Head Start activities Teachers’ reports of parent involvement (estimations of the number of parents in their classrooms who volunteered in activities offered by the Head Start program) were moderately and significantly associated with data from the volunteer logs, for the number of volunteers per classroom, r(59) = 55, P < 01, and the total number of volunteering hours per classroom, r(59) = 34, P < 05 These results suggest that teachers seemed to be fairly accurate reporters of parent volunteering activities in their classrooms Even though questions about volunteering that were asked of parents and teachers had the same content, they were not asked in the same way, and those variations may be related to the differences found 2.3 What is the relation between parent characteristics and their participation in Head Start? As seen in Table 2, results of the bivariate correlations between data from volunteer logs and parent interviews indicated that employed parents were less likely to volunteer in any of the activities in the Head Start programs They volunteered fewer times, r(127) = −.35, P < 001, and for fewer hours, r(127) = −.39, P < 001, than parents who were unemployed Also, mothers with more years of schooling were less likely to volunteer in Head Start, r(127) = −.19, P < 01, consistent with the fact that mothers with more years of schooling were more likely to be employed, r(127) = 39, P < 01 Parents’ reports of their involvement in Head Start were negatively associated with their reports about barriers to involvement, so that those who reported fewer barriers were more involved, r(127) = −.28, P < 01 In addition, parents’ reports of their involvement in Head Start were positively associated with home activities (from the parent Maternal education Parent employment Satisfaction with Head Start Barriers to participation Activities with child at Home Involvement self-report a ∗ ∗∗ ∗∗∗ Parent volunteering variables Parent characteristics variables Times volunteered Hours volunteered Parent employment Satisfaction with Head Start Barriers to participation Activities with child at home Involvement self-report −0.19∗∗ −0.35∗∗∗ −0.14 −0.10 −0.10 0.23∗∗ −0.18∗ −0.39∗∗∗ −0.16 −0.09 −0.10 0.36∗∗∗ 0.39a ∗∗ – – – – – −0.003 0.11 – – – – 0.007 0.05 −0.21∗ – – – 0.05 −0.002 0.04 −0.00 – – −0.03 −0.31∗∗ −0.10 −0.28∗∗ 0.31∗∗ – Maternal education and parent employment, both categorical variables, were estimated using a Phi coefficient P < 05 P < 01 P < 001 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Table Correlations between parent characteristics and parent volunteering variables (n = 127) 423 424 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Table Means and standard deviations for parents’ reports of involvement and total time volunteered by employment status Parent employment n Parents’ reports of involvement Total time volunteered Full-time Part-time Unemployed 41 17 45 1.77 (0.27) 1.93 (0.34) 2.07 (0.44) 22.83 (41.79) 35.94 (64.72) 112.41 (130.29) interview), so that parents who reported doing more activities with their children at home were also more involved in their children’s Head Start program, r(127) = 31, P < 01 There were no significant relations between the measures of involvement from the volunteer logs and either parents’ reported barriers to involvement or activities with their children at home A multivariate analysis of variance was conducted to examine whether parent characteristics were associated with parent volunteering, including both the number of hours from volunteer logs and the parents’ reports of involvement These results indicated that there was a significant multivariate effect of parent employment, F(4,188) = 5.33; P < 001, for both variables In addition, parents’ reports of their satisfaction with Head Start, barriers to participation, and activities with their children at home were significantly related to parents’ reports of their involvement in Head Start, F(7,102) = 5.12; P < 001, but were not significantly associated with the number of hours volunteered Also, there were no significant relations between maternal education and either measure of parent involvement1 Two t-tests of least squared means (LSM) were performed as post-hoc tests to identify the significant differences in parent involvement for three subcategories of parent employment (full-time, part-time, and unemployed) According to parents’ reports of their involvement in Head Start, parents who were unemployed were significantly more involved than parents who worked full-time (LSM difference = 3.03, P < 01) There were no differences between the involvement of parents who worked part-time and either those who worked full-time or those who were unemployed Similarly, in the case of the number of hours from volunteer logs, parents who were unemployed volunteered significantly more hours than parents who worked full-time (LSM difference = 4.25, P < 001), or parents who worked part-time (LSM difference = 2.57, P < 01) The difference in the number of hours volunteered between parents who worked full-time and part-time was not significant The effect size for the difference between unemployed and full-time employed parents’ reports of involvement was 0.83 The effect sizes for the differences in total number of hours volunteered between unemployed and full-time employed parents and between unemployed and part-time employed parents were 1.01 and 0.68, respectively Table shows the means and standard deviations for parents’ reports of involvement and total number of hours volunteered by employment status 2.4 What are the relations between teacher and classroom characteristics and parent volunteering? Results of the bivariate correlations, shown in Table 4, indicate that teachers with more years of experience working in Head Start had more total volunteer hours in their classrooms, r(59) = 42, P < 001, and had volunteers returning more times, r(59) = 40, P < 01 Also, classrooms with a larger number A preliminary analysis was conducted to determine if family income and family size were important factors to control for Neither one significantly predicted either measure of parent involvement Because about a third of parents did not report income, we chose to exclude these two demographic variables from further analysis in order to maximize the sample size Parent volunteering variables Teacher education Head Start experience Teachers’ parent involvement reports Teachers’ perceptions of involvement of parents Classroom quality ∗ ∗∗ ∗∗∗ P < 05 P < 01 P < 001 Teacher and classroom characteristics variables Number of unique volunteers Number of volunteer hours Percentage of returning volunteers Head Start experience Teachers’ parent involvement reports Teachers’ perceptions of involvement of parents Classroom quality 0.11 0.19 0.55∗∗ −0.20 0.42∗∗ 0.34∗ 0.12 0.40∗∗ −0.16 −0.42∗∗ – – −0.20 0.27 – 0.14 0.10 0.65∗∗∗ −0.16 0.14 0.03 0.09 0.01 0.03 – – – 0.04 0.41∗∗ 0.22 0.26 – – – – D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Table Correlations between teacher and classroom characteristics and parent volunteering variables (n = 35) 425 426 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 of different individuals volunteering had a larger proportion of volunteers who returned, r(59) = 26, P < 05 There were no associations between teachers’ years of education and the number of volunteers or the amount of volunteering time, perhaps because teachers with more years of education were also more likely to have less experience in Head Start, r(59) = −.42, P < 01) Regarding classroom quality, higher quality classrooms had a larger number of different volunteers, r(35) = 41, P < 01, but there were no differences in total hours of volunteer time A multivariate analysis of variance was performed to examine whether teacher and classroom characteristics, as dependent variables (Head Start experience, years of education, perceptions of involvement of parents in their classroom, and classroom quality) predict parent involvement outcome variables (number of individual volunteers and total hours of volunteer time) This analysis was conducted on data from 35 classrooms for which both classroom observations and teacher questionnaire data were available There was a significant multivariate effect of classroom quality, F(2,30) = 3.41, P < 05 Based on post-hoc analysis, this result is primarily attributable to the relationship between classroom quality and the number of different volunteers per classroom, t(1,31) = 2.58, P < 05, B = 3.44 The magnitude of this association is such that, a one point increase in quality is associated with a 3.44 increase in the number of parents who volunteered No significant associations were found between any of the predictor variables and total hours of volunteer time per classroom Discussion The present study documents the different ways in which parents participate in Head Start programs and identifies family, teacher, and classroom characteristics related to parent involvement A strength of the study is the relatively large sample of parents on which a number of these findings are based The use of multiple sources of information was another strength in this study and addresses some limitations in research on parent involvement noted by other researchers On the other hand, the present study did not include information about parents who did not participate at all in Head Start activities This information would have been important to include because many employed parents in our study managed to be involved, albeit at a lower rate than those who were unemployed Future research could examine in which ways parents who are involved differ from those who are not Our findings seem to show some progress in the participation of parents in Head Start We found that 47% of the parents who volunteered accounted for 53% of the total number of volunteer activities Previous studies have reported the uneven nature of parent participation in Head Start, with smaller percentages of the parents accounting for most of the total time volunteered Powel (1989) cited a 1975 report to Congress by the US Comptroller General that presented findings from a study with six Head Start programs That study found that 35% of the parents accounted for 71% of the total time volunteered Ten years later, the Head Start Synthesis Project also reported an uneven participation of parents with “a core of parents contributing a disproportionate share of time” (McKey et al., 1985, p 17) In our study, almost half of the parents who volunteered were involved in Head Start activities throughout the school year Regarding the types of activities in which parents participate, a study with African–American Head Start parents found that they most frequently participated in the classroom or attending field trips; however, they were rarely involved in decision-making activities or parent meetings (Howard, 1992) Our results are consistent with the earlier finding that parents most frequently volunteered in the classroom; however, D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 427 we found that attending meetings was the next most frequent form of involvement, suggesting that there may have been some increase in this activity in the past decade In our study the parent meeting category included different types of meetings (i.e policy council, center-wide and classroom committees, parent–teacher conferences, parent orientation, parent education), which does not allow us to provide precise information about parents’ participation in each particular type of meeting Even so, the finding that parents seem to be participating more in meetings indicates progress in the involvement of families in the Head Start program in general, and more specifically serves as an indicator of increased opportunities for family and staff communication It is important to note that while parents’ participation in other volunteering activities decreased from fall to spring in our study, their participation in meetings remained at the same level The finding that working parents volunteer less is not surprising as they have less discretionary time It is interesting that parent employment emerged as the strongest predictor across measures of parent involvement, above and beyond maternal education and parents’ beliefs about the program and childrearing behaviors Our finding of the effect of parents’ employment on their participation in Head Start programs is consistent with other research studies A study of the involvement of parents in three rural Head Start programs found that higher hours of employment and the associated greater income might compete with spending time in parent involvement activities (Driebe & Cochran, 1996) Another study found that Head Start parents who were employed full-time volunteered less often than parents who were unemployed or employed part-time (Hooker, 1993) These results from the 1990s are different from earlier research that found maternal education to be the most important family characteristic associated with parent involvement (Pyle, 1988) These more recent studies may seem to reflect an increase in the number of low-income parents being incorporated into the workforce, which may have had an impact on the parent involvement strategies used in Head Start programs The models of parent involvement are being revised and some authors have proposed models that incorporate both school-based and home-based involvement (e.g Epstein, 1996; Fantuzzo, Tighe, & Childs, 2000) It is evident that Head Start programs will need to create more opportunities to accommodate working parents’ schedules For example, some programs have started scheduling meetings or other events on Saturdays or weekday evenings, and are providing more ways for parents to participate from home The Head Start programs that participated in this study counted activities parents performed at home as part of their volunteering time However, they mainly included activities to help the program (e.g doing laundry, preparing classroom materials) It would be necessary to assume a broader definition of parent involvement to include not only parents’ activities to support the program, but also activities parents conduct at home to support their children’s development and education Expanding parents’ participation from home may include providing parents with ideas and resources for a variety of activities they can with their children at home, in connection with the school and the community This would have implications for developing effective communication strategies with parents When interpreting these findings, it should be taken into account that the parent interview data reported reflect the characteristics of parents who were involved in Head Start at higher rates compared with those who did not participate in the parent interview The average number of times volunteering in the year for interviewed parents (n = 127) was 21 while non-interviewed parents (n = 653) volunteered an average of six times Also, the percentage of parents agreeing to participate in the interview was low (16%) which should be considered a limitation of this study The finding that classroom quality was the strongest predictor of parent involvement (number of individual volunteers per classroom) among the set of teacher and classroom characteristics examined, 428 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 may indicate that high quality classrooms have the conditions or characteristics that encourage or promote parent volunteering On the other hand, the presence of many volunteers in the classroom may contribute to the quality of the classroom Classroom quality may be a proxy for several important aspects of teachers’ behaviors and classroom practices that increase the involvement of parents (e.g ways in which parents and teachers interact, what parents when helping out in the classroom) 3.1 Implications for practice Most early childhood programs – not just Head Start – struggle with parental involvement, especially with how to involve working parents Better dissemination of effective strategies for working with parents would be valuable Perhaps conducting events that involve the whole family, targeting particularly children and fathers, might generate more enthusiasm for participating in an after-work activity Involving parents in the planning and conduct of the activities should also enhance interest Although it cannot be interpreted causally, the finding that volunteering in Head Start was positively associated with parental involvement in home learning activities can be used to motivate staff efforts to more effectively involve parents in volunteering In this study, more experienced teachers had a greater number of volunteer hours in their classrooms and had volunteers returning more often Perhaps they have learned some techniques to promote a stronger culture of volunteerism in their classrooms We acknowledge that this is a rather modest finding (i.e teacher experience in Head Start was significant at the bivariate level but not in the multivariate analysis); however, it seems reasonable to suggest that encouraging experienced teachers to mentor other teachers about their parent involvement practices may be a strategy to help increase parent involvement in early childhood programs Higher quality classrooms were found to have larger numbers of parents volunteering Because of the implications of this finding for improving practice, the authors consider it important to replicate this component of the study in a larger number of classrooms 3.2 Implications for future research The volunteer logs included many types of parent involvement, including voluntary home activities, but the most frequent activity was the “traditional” helping out in the classroom Given this finding, how could this time best be used? Some questions for future research include studying what parents when they are in the classroom, whether teachers use this time to help parents learn new skills, and how parents perceive their classroom volunteering experience Greater knowledge of the interactions that occur within the classroom – between volunteers and teaching staff and volunteers and children – might help us understand how programs can best utilize this most frequent parent volunteering activity Also, higher quality classrooms had more parent volunteers Future research should examine in more detail the strategies that are used by teachers who are successful in involving parents in their classrooms and the ways in which teachers interact or relate with parents who are volunteering Some of these findings have implications for the methodology of studies using parent involvement as a predictor or outcome variable Predictor variables obtained from the parent interview predicted self-reported parent involvement, but did not predict parent involvement when data from the volunteer logs were used Further, the association between parent reports and volunteer logs was modest These findings indicate that the use of self-reported parent involvement may need to be re-examined Although D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 429 sign-in logs are likely not 100% accurate, they are more objective than retrospective self-reports Teachers’ reports, on the other hand, were found to be somewhat more accurate when compared with volunteer logs However, it is important to take into account that the questions to estimate parent involvement asked to parents and teachers were not exactly the same, which could explain these differences Further research focusing on the validity, reliability and feasibility of methods to study parent involvement is recommended References Baker A J L., & Soden, L M (1998) The challenges of parent involvement research ERIC/CUE Digest Number 134 (ED419030) Bronfenbrenner, U (1979) The ecology of human development: experiments by nature and design Cambridge, MA: Harvard University Press Bronfenbrenner, U (1988) Interacting systems in human development: research paradigms, present and future In N Bolger, A Caspi, G Downey, & M Moorehouse (Eds.), Persons in context: developmental processes New York: Cambridge University Press Consortium for Longitudinal Studies (1983) As the twig is bent .Lasting Effects of Preschool Programs Hillsdale, NJ: Lawrence Erlbaum Associates Driebe, N M., & Cochran, M (1996) Barriers to parent involvement in Head Start programs Paper presented at Head Start’s Fourth National Research Conference, Washington, DC Eccles, J., & Harold, R (1996) Family involvement in children’s and adolescents’ schooling In A Booth & J F Dunn (Eds.), Family–school links: how they affect educational outcomes? (pp 3–34) Mahwah, NJ: Lawrence Erlbaum and Associates Epstein, J L (1992) School and family partnerships In M Alkin (Ed.), Encyclopedia of educational research (6th ed., pp 1139–1151) New York: McMillan Epstein, J L (1996) Perspectives and previews on research and policy for school, family, and community partnerships In A Booth & J F Dunn (Eds.), Family–school links: how they affect educational outcomes? (pp 209–246) Mahwah, NJ: Erlbaum Epstein, J L., & Dauber, S L (1995) Effects on students of an interdisciplinary program linking social studies, art, and family volunteers in the middle grades Journal of Early Adolescence, 15, 237–266 Fantuzzo, J., Tighe, E., & Childs, S (2000) Family involvement questionnaire: a multivariate assessment of family participation in early childhood education Journal of Educational Psychology, 92(2), 330–337 Feldman, D., & Gerstein, L H (1988) A factor analytic study of three teacher belief scales Measurement and Evaluation in Counseling and Development, 21, 72–80 Greenwood, G E., & Hickman, C W (1991) Research and practice in parent involvement: implications for teacher education The Elementary School Journal, 91(3), 279–288 Harms, T., Clifford, R M., & Cryer, D (1998) Early childhood environment rating scale (Rev ed.) New York, NY: Teacher College Press Henderson, A T., & Berla, N (1994) A new generation of evidence: the family is critical to student achievement Washington, DC: National Committee for Citizens in Education Henderson, A T., Marburger, C L., & Ooms, T (1986) Beyond the bake sale: an educator’s guide to working with parents Columbia, MD: National Committee for Citizens in Education Hester, H (1989) Start at home to improve home–school relations NASSP Bulletin, 73, 23–27 Hooker, B W (1993) Characteristics of Head Start volunteers as related to parent involvement Masters Abstracts International, 32(02), 0393 Howard, R J (1992) The involvement of African–American caregivers in Head Start Dissertation Abstracts International, 53(05-A), 1389 Lamb-Parker, F., Piotrkowski, C S., Baker, A J L., Kessler-Sklar, S., Clark, B., & Peay, L (2001) Understanding barriers to parent involvement in Head Start: a research-community partnership Early Childhood Research Quarterly, 16, 35– 51 430 D.C Castro et al / Early Childhood Research Quarterly 19 (2004) 413–430 Lamb-Parker, F., Piotrkowski, C S., Kessler-Sklar, S., Baker, A J L., Peay, L., & Clark, B (1997) Executive summary: parent involvement in Head Start New York: NCJW Center for the Child Marcon, R A (1999) Positive relationships between parent school involvement and public school inner-city preschooler’s development and academic performance The School Psychology Review, 28(3), 395–412 McKey, R H., Condelli, L., Ganson, H., Barrett, B J., McConkey, C., & Plantz, M C (1985) The impact of Head Start on children, families and communities DHHS Publication No 85-31193 Washington, DC: CSR Inc Miedel, W T., & Reynolds, A J (1999) Parent involvement in early intervention for disadvantaged children: does it matter? Chicago longitudinal study Journal of School Psychology, 37(4), 379–402 National Household Educational Survey (1999) Early childhood program participation – parent survey Washington, DC: National Center for Education Statistics US Department of Education http://nces.ed.gov/nhes/pdf/quex/early/99parent.pdf Powel, D R (1989) Families and early childhood programs Washington, DC: National Association for the Education of Young Children Pyle, N S (1988) The relationship of parent involvement in Head Start to family characteristics, parent behaviors and attitudes, and preschool inventory scores Dissertation Abstracts International, 49(07-A), 1695 Roskos, K., & Neuman, S (1993) Enhancing Head Start parents’ conceptions of literacy development and their confidence as literacy teachers: a study of parental involvement Early Child Development and Care, 89, 57–73 Scott-Jones, D (1995) Activities in the home that support school learning in the middle grades In B Rutherford (Ed.), Creating family/school partnerships (pp 161–181) Columbus, OH: National Middle School Association Taylor, A R., & Machida, S (1994) The contribution of parent and peer support to Head Start children’s early school adjustment Early Childhood Research Quarterly, 9(3/4), 205–387 US Department of Health and Human Services (1998) Head Start Family and Child Experiences Survey In: Head Start: program performance measures Second progress report Washington, DC: Head Start Bureau Administration on Children, Youth and Families Zigler, E., & Styfco, S J (2000) Pioneering steps (and fumbles) in developing a federal preschool intervention Topics in Early Childhood Special Education, 20(2), 67–70, 78