Defining and Measuring Parenting for Educational Success: A Critical Discourse Analysis of the Parent Education ProfilePennsylvania State University
The Parent Education Profile (PEP) is an instrument used by family literacy programs to rate parents’ support for children’s literacy development. This article uses Critical Discourse Analysis to examine how the PEP constructs the ideal parent, the text’s underlying assumptions about parenting and education, and its ideological effects. The analysis shows how many features of the PEP evaluate parents according to a middle-class, predominantly White model of parenting and family-school interaction. Furthermore, the PEP tends to assume a universal, normative model of parental support for literacy, parental (mothers’) responsibility for educational outcomes, equal access to resources required to meet the PEP standards, and a limited parental role in assessment. In so doing, the PEP lends support to several dominant discourses regarding poor and minority families, such as the discourse of parent involvement and the "mothering discourse," which encourages mothers’ supplementary educational work. Implications for policy, research, and practice are discussed.
Key Words: family/home education • women’s issues • assessment • literacy • social class • minorities With the advent of No Child Left Behind and the National Reporting System for Adult Education, federally funded family literacy and adult education programs must now provide evidence that participants have made gains on standardized educational tests. Geared primarily toward low-income mothers with young children, federal Even Start and state-funded family literacy programs typically integrate adult education, early childhood education, interactive parent-child literacy activities, and parent education.1 Until recently, however, these programs had few measures to assess growth in parenting skills. To address this need, many states have adopted the Even Start Family Literacy Parent Education Profile (commonly referred to as the Parent Education Profile or PEP), an instrument that classifies parenting practices on a scale of 1 to 5 "in approximate order of development," from "least supportive of literacy outcomes" to "most supportive" (RMC Research Corporation & New York State Department of Education [RMC], 2003, p. 11). RMC and the New York State Department of Education developed the PEP to organize research findings about the effects of parents on children’s literacy development in a way that Even Start programs could consistently track and report on parents’ progress. We were motivated by the then new requirement that state agencies develop performance indicators to measure the performance of Even Start programs. (2003, p. ii) This statement reveals that the PEP is embedded in the discourse of assessment and accountability in education (Linn, 2000). Its origins as a response to federal accountability policies suggest this instrument and others like it are mechanisms for governing nonformal education programs and participants. Drawing on Foucault’s analysis of "governmentality," Graham and Neu (2004) argue that "standardized testing programmes, by rendering the participants visible and subjecting them to public scrutiny, contribute to the construction of governable persons" (p. 295). Parenting assessment tools similarly construct and discipline parents and practitioners through observation, classification, and rating, thereby encouraging them to internalize the assessment results and social norms embedded in the instrument and to adjust their behavior to meet its standards (Graham & Neu, 2004; see also Gipps, 1999, for a sociocultural analysis of assessment). Although this study is grounded in these broader concerns about the unintended consequences of rating parents for accountability purposes, we focus here on how the PEP conceptualizes ideal parenting practices. With its focus on parents’ support for literacy, the PEP is also related to the dominant discourse of parent involvement in education, which assumes that parents are primarily responsible for children’s literacy and educational achievement, that they enact this responsibility in specific ways, and that education in the home and school should be closely aligned (de Carvalho, 2000; Nakagawa, 2000; Schlossman, 1978, 1983). This discourse is evident, for example, in school policies that require various forms of parent involvement and hold families accountable for children’s school success (Nakagawa, 2000) as well as in research, popular culture, and everyday conversations among parents and educators. Paradoxically, this discourse tends to portray parents as both the primary cause of children’s educational problems and failing schools and the key to academic achievement and high-quality schools (Fine, 1993; Nakagawa, 2000). Although they may appear natural, conceptions of parenting, literacy, and family-school relationships are historically and culturally situated (S. Auerbach, 2007; Lareau, 1987; Panofsky, 2000). As such, educators and scholars concerned with cultural responsiveness, respect for participants, and the democratic inclusion of learners in education should examine how assessment instruments like the PEP define "good parenting" as it relates to education and the types of roles and identities it constructs for both parents and professionals. In this article we use Critical Discourse Analysis (CDA) to examine the following questions: How does the text of the PEP construct the ideal parent? What assumptions about parenting and education are evident in the PEP? What are the ideological effects of these assumptions? Our study is the first analysis of this widely used instrument. Critical analysis of the PEP is needed because texts promote particular discourses and ideologies (Fairclough, 2003; van Dijk, 2001; Wodak & Meyer, 2001), for example, by privileging certain views of parental roles and more or less desirable literacy and parenting practices. Circulated through assessment instruments, educational policies, research, popular culture, and informal interactions, these perspectives become taken for granted, shaping how educators view parents and how parents view themselves and each other (see Griffith & Smith, 2005). Texts also structure power relations among policy makers, researchers, practitioners, and program participants (Fairclough, 1989), granting parents more or less power to assess themselves and decide how to raise and teach their children. In this sense, assessment instruments are regulating practices (Foucault, 1977/1995; Sparks, 2001) that promote a tacit set of norms and values and guide the actions of practitioners and program participants. We support family literacy practitioners’ efforts to help families succeed while also recognizing that the materials and instruments used in educational settings may contradict or undermine even the best intentions. Analysis of the PEP is important precisely because it may unconsciously shape professional-participant relationships and the practice of family literacy and parent education. Examining how people internalize, contest, adapt, or promulgate the PEP and its conceptualization of parenting and literacy is an important research topic. Our task here, however, is to illuminate the PEP’s underlying assumptions, to link these to current discourses about parenting and education, and to trace the ideological consequences of rating parents’ compliance with this instrument. The social identities of family literacy participants make it especially important to examine how the PEP constructs the ideal parent. The majority of family literacy participants are mothers who do not have a high school diploma or who speak English as a second language. In 2000–2001, 84% of the families in federal Even Start programs had incomes at or below the federal poverty level, and 70% of participants were racial/ethnic minorities (46% Latino/Latina, 19% African American, 3% Native American, 2% Asian/Pacific Islander; U.S. Department of Education, 2006). Although most participants are voluntary, some are mandated to attend by Child Protective Services, welfare, or the court system. Family literacy participants, then, are likely to experience multiple forms of oppression, placing them in a vulnerable position vis-à-vis educators and policy makers. Historical analysis of parent education in the United States (Schlossman, 1978, 1983) underscores the need to scrutinize how assessment tools conceptualize ideal models of parenting and assign responsibility for educational success. This study reveals that many aspects of the PEP tend to promote and evaluate parents according to the mainstream (middle-class, predominantly White) model of parenting. The article shows how the text uses scientific discourse to validate this construction of the ideal parent and then analyzes the PEP’s assumptions about the kinds of practices that foster literacy development and academic success, the assignment of responsibility for children’s educational outcomes, the kinds of resources required to engage in the prescribed practices, and the parent’s role in assessment. We find that the ideological effects of these assumptions may inadvertently lend support to the dominant discourse of parent involvement, the "mothering discourse" (Griffith & Smith, 2005) that encourages dependence on mothers’ supplementary educational work, deficit perspectives of nonmainstream parents, and individualistic explanations of educational disparities. We conclude with implications for research and practice, suggesting that educators and scholars view parenting practices as adaptations to collective histories and contemporary socioeconomic conditions (García Coll et al., 1996).
The PEP includes the following four scales and 15 subscales:
The PEP primarily measures parenting practices (Darling & Steinberg, 1993), or the behaviors used to socialize children toward specific goals. Some items pertain to parenting style (emotional climate) and knowledge, awareness, and material resources such as the availability of books or a safe home environment. Although the PEP focuses on "parent education for literacy purposes" (RMC, 2003, p. 8), it addresses multiple dimensions of child rearing including emotional tone, linguistic style, and play—dimensions that vary cross-culturally (e.g., see Gonzales, Cauce, & Mason, 1996; Johnston & Wong, 2002). This multidimensionality makes the instrument holistic but also potentially increases the risk of subjective interpretation based on observers’ own cultural values and social location. The PEP draws on the Even Start parent education framework, which in turn is based on Powell and D’Angelo’s (2000) analysis of research on parent education and children’s literacy outcomes, "the Equipped for the Future frameworks related to parenting," and "the stages of parent development synthesized from New York’s longitudinal evaluations of family progress in Even Start" (RMC, 2003, p. 6). According to the manual, professionals must be trained to use the PEP, and the "ratings should be made by a team [of professionals] that knows the parent well and based on evidence of behaviors from logs, portfolios, interactions, and interviews or discussions with the parent over a several month period" (RMC, 2003, p. 10). Professionals should observe parents in "everyday activities" and "routine program opportunities" (RMC, 2003, p. 10) and provide written documentation justifying each rating. Furthermore, ratings should be based on the highest level of "typical" or "consistently observable" behaviors. Examples of documentation for each scale and subscale from the pilot study are provided in the manual. The RMC Web site states the PEP is being used in "Even Start and other family literacy programs, Title I and other school parent education programs, and adult education programs" (RMC Research Corporation, 2008, para. 2). Currently, the No Child Left Behind reporting guidelines for Even Start (U.S. Department of Education, n.d.) list only the PEP as an instrument that states can use to measure the "percentage of parents who show improvement on measures of parental support for children’s learning in the home, school environment, and through interactive learning activities" (p. 58). While federal legislation does not appear to mandate the PEP, to the best of our knowledge at least 11 states require federally or state-funded family literacy programs to use this instrument (e.g., see Colorado Department of Education, 2005).2 At the time of press at least 12 other states used the PEP as a program performance indicator or were pilot testing it for other purposes.3 When states use PEP scores as a program performance indicator, program effectiveness depends on parents meeting a certain PEP level, with higher scores and/or gains indicating greater effectiveness. In turn, programs that meet more performance standards are more likely to be funded. Some programs also use the PEP as a pre/post measure to demonstrate gains in parenting skills. For instance, one Even Start program reported, "families are consistently at level 2 for 14 of the 15 sub-scales, with parenting behavior performing at 40% to 47% of the highest possible score" (University of Vermont, 2003, p. 4). Yet tying program performance to PEP scores and the use of pre/post comparisons without a control group seems questionable, as the PEP edition (RMC, 2003) we analyzed had not been tested for validity and reliability.
Conceptual Approaches to Family Literacy and Parent Education The PEP sits squarely in the debate regarding the role of family literacy, parent involvement, and parent education in alleviating educational disparities rooted in class, race, language, and culture. Most parents join family literacy and parent education programs because they want to help their children succeed in school. While these programs may engage parents in similar activities such as book reading, they are rooted in distinct philosophical approaches regarding the causes of and solutions to educational inequities, the program’s ultimate aims, the best ways to support poor and culturally diverse families, the maintenance of mainstream parenting and literacy practices as a normative ideal, and the inclusion of parents in planning, assessment, and evaluation. We use Elsa Auerbach’s (1995) typology to identify three approaches to working with families—multiple literacies, social change, and intervention-prevention—while recognizing that in practice these categories often overlap. Drawing on sociocultural theories of literacy, the "multiple literacies" model defines the problem as the cultural mismatch or discontinuities between home and school literacies and, concomitantly, educators’ limited recognition of families’ funds of knowledge (Moll, Armanti, Neff, & Gonzalez, 1992), including oral and written literacy practices that differ from the mainstream. Educators in these programs therefore seek to validate and build on families’ cultural identities and culturally specific child-rearing and literacy practices (particularly those used in out-of-school settings), while also enabling them to learn new skills they deem important. Prominent examples of this approach include the Intergenerational Literacy Project (Paratore, 2001) and Project FLAME (Rodriguez-Brown, 2003). The "social change" model (e.g., Delgado-Gaitán, 1990; Orellana, 1996; Reyes & Torres, 2007) locates the root cause of educational disparities not in families or cultural discontinuities, but rather in political, social, and economic systems that maintain social inequalities (E. R. Auerbach, 1995). For example, Olivos (2006) contends nonmainstream parents are less involved in schools because they have been excluded by school systems and society. Although this model also emphasizes cultural pluralism and views literacy as a social practice, its chief aim is to increase parents’ power and control in their educational programs, children’s schools, and communities. Parents in these programs shape programmatic goals, curricular content, and evaluation, and the curriculum enables them to analyze social problems and advocate for their interests, for example, by changing school policies. Supported by many family literacy and parent education proponents (e.g., Epstein, 2001; Haskins & Adams, 1983; Henderson & Berla, 1994; Ponzetti & Bodine, 1993), the dominant "intervention-prevention" model locates the problem in parents’ ability to promote positive literacy attitudes and interactions in the home. Thus, educators can best prevent literacy problems and interrelated social problems through interventions that help parents adopt the behaviors and attitudes characterizing high-achieving mainstream families. Specifically, parents learn child-rearing practices and oral and written literacy practices that are associated with language and literacy development and academic achievement such as dialogic book reading, school involvement, helping with homework, and specific parenting strategies (e.g., Burchinal, Peisner-Feinberg, Pianta, & Howes, 2002; Bus, van Ijzendoorn, & Pellegrini, 1995; Hoover-Dempsey et al., 2001; Lonigan & Whitehurst, 1998; Morrow & Young, 1997; Sénéchal, LeFevre, Thomas, & Daley, 1998; Snow, Barnes, Chandler, Goodman, & Hemphill, 1991). Parents often appreciate the new strategies and knowledge they learn in such programs (L. M. Phillips, Hayden, & Norris, 2006). Upon closer examination, the research base underlying the PEP and the intervention-prevention approach reveals a complex portrait. There is evidence showing that many practices prescribed by the PEP, such as authoritative parenting style, joint book reading, and abundant literacy materials, are related to literacy development and school success, especially for White, middle-class children (e.g., Bennett, Weigel, & Martin, 2002; Bus et al., 1995; Hoover-Dempsey et al., 2001; Jordan, Snow, & Porche, 2000) and in some cases for low-income or minority children (e.g., Dearing, McCartney, Weiss, Kreider, & Simpkins, 2004; Jeynes, 2005; Lonigan & Whitehurst, 1998; Miedel & Reynolds, 1999; Morrow & Young, 1997; Zellman & Waterman, 1998). Nonmainstream children appear to benefit most from these practices when other conditions are met, such as high-quality classrooms (Snow et al., 1991) and close relationships with teachers (Burchinal et al., 2002). Some scholars have expressed concern that many of the studies supporting the dominant family literacy model (a) use correlational research as a basis for intervention, even though such studies do not establish causation (de Carvalho, 2000; Hayes, 1996; Mattingly, Prislin, McKenzie, Rodriguez, & Kayzar, 2002; Panofsky, 2000); (b) reveal an "exaggerated belief in the power of individuals to overcome material limitations" (de Carvalho, 2000, p. 14); or (c) are used comparatively and normatively to show how nonmainstream families fall short (Panofsky, 2000). Other studies conducted with U.S.-born and immigrant children and youth reveal that the effects of parenting practices and styles on academic outcomes differ considerably by race/ethnicity, socioeconomic status (SES), generational status, and sociocultural setting (Berlin, Brooks-Gunn, Spiker, & Zaslow, 1995; Desimone, 1999; Kao, 2004; Spera, 2005). For instance, the parenting subscales in the HOME (Home Observation for Measurement of the Environment) Inventory better predicted cognitive and social outcomes for White than for Latino or African American children, and different parenting practices explained child outcomes for distinct racial/ethnic groups (Sugland et al., 1995; see also Berlin et al., 1995). Similarly, Goldenberg, Gallimore, Reese, and Garnier (2001) found that Latino immigrant parents had high educational expectations, yet these did not influence children’s school performance. In his review of multivariate studies on parent involvement, Domina (2005) concluded no single parent activity consistently predicted children’s outcomes. His analysis of the National Longitudinal Survey of Youth showed that "parental involvement does not independently improve children’s learning, but some involvement activities do prevent behavioral problems" (Domina, 2005, p. 233). Mattingly and colleagues’ (2002) analysis of 41 evaluations of K–12 parent involvement programs found little evidence that they improved academic achievement (see Jeynes, 2005, for an alternative view). In terms of parenting styles, Dornbusch, Ritter, Leiderman, Roberts, and Fraleigh (1987) found authoritative parenting was more strongly associated with higher grades for White adolescents than for African Americans and Latinos, and not at all for Asians; the results also varied by gender among Latinos/Latinas (see also Pong, Hao, & Gardner, 2005; Steinberg, Dornbusch, & Brown, 1992). Similarly, Kao (2004) cautioned against assuming that the relationship between parental behavior and educational outcomes for White families is "completely transferable to minority and immigrant households" (p. 447). Another study showed authoritarian parenting in fact contributed to academic achievement for U.S. and Australian children whose parents did not have a college degree (Leung, Lau, & Lam, 1998). Taken together, these studies underscore the need to investigate the assumptions about parenting and education embedded in assessment instruments like the PEP, in particular, whether they uphold a specific cultural model of parenting as the ideal pathway to literacy, academic success, and other desired outcomes (García Coll & Patcher, 2002; Kao, 2004).
Critiques of the Dominant Model Second, some scholars contend that research showing how poor and minority families fall short of the mainstream standard has helped perpetuate the deficit model (García Coll & Patcher, 2002), referring to the idea that "students, particularly of low-SES background and of color, fail in school because they and their families have internal defects, or deficits, that thwart the learning process" (Valencia & Black, 2002, p. 83). In particular, the deficit model presumes that poor parenting practices and failure to value education cause both educational underachievement and other social problems (E. R. Auerbach, 1995; Ogbu, 1982; Powell et al., 2004; Sigel, 1983; Taylor, 1997; Valdés, 1996; Valencia & Black, 2002; Vernon-Feagans, Head-Reeves, & Kainz, 2004). This line of inquiry challenges us to explore how assessment instruments support and/or challenge deficit perspectives of poor, minority, and immigrant families.4 Researchers have also examined the implicit messages about mothering in family literacy and parent education (Smythe & Isserlis, 2004; Sparks, 2001). Two guiding concepts are especially relevant to examining assumptions about gender, mothering, and parental responsibility in the PEP. Created by psychologists, educators, and other professionals and then popularized by women-oriented media, the "mothering discourse" coordinates mothering work in the home and teachers’ work in public schools by mobilizing mothers’ "work, care, and worries" toward their children’s education (Griffith & Smith, 2005). Similarly, the "ideology of intensive mothering" (Hays, 1996), which holds that a good mother is the central caregiver and spends "copious amounts of time, energy, and material resources on the child" (p. 8), powerfully shapes how many Americans think about mothering. Both concepts position the ideal mother as one who is immersed in supporting her children’s education. Finally, some scholars have questioned the ethics of intervention programs that seek to change the parenting and literacy practices of nonmain-stream families (E. R. Auerbach, 1989; Reyes & Torres, 2007; Schlossman, 1978; Sigel, 1983; Sparks, 2001; St. Clair & Sandlin, 2000; Taylor, 1997; Valdés, 1996). Their primary concern is parents’ ability to adopt the child-rearing and literacy practices of their choice, to maintain their cultural identities if they so choose, and to shape parent education and family literacy programs. Accordingly, we examine whether the PEP enables parents to influence their own ratings and pursue alternative ways of parenting and supporting literacy.
Cultural Models of Child Rearing and Involvement in Education
In the "concerted cultivation" logic of child rearing, prevalent among middle-class families, parents tend to follow expert advice by "deliberately try[ing] to stimulate their children’s development and foster their cognitive and social skills" (Lareau, 2003, p. 5). Similarly, the middle-class ideal of "intensive mothering" is "child-centered, expert-guided, emotionally absorbing, labor-intensive, and financially expensive" (Hays, 1996, p. 8). In the "accomplishment of natural growth" model, more common among poor and working-class families, the parent cares for children, provides them with boundaries, and allows them to grow, while also meeting their basic needs and providing a safe, nurturing environment (Lareau, 2003).7 Following García Coll and colleagues (1996), we contend both models are embedded in an "adaptive culture" that is the "product of the group’s collective history (cultural, political, and economic) and current contextual demands" (p. 1904) such as racial segregation or poverty. That is, parents raise children to respond to the demands of their respective environments. Language use and parent involvement in literacy activities. Lareau’s research (2003) shows that middle-class parents tend to reason and negotiate with their children about routine and important matters alike (see also Hays, 1996). Consequently, middle-class children not only help make decisions but are also apt to challenge parents’ statements and decisions. In the concerted cultivation model, parents often turn everyday events into learning experiences or object lessons, eliciting extended verbal responses from children (Lareau, 2002; Lareau & Weininger, 2007). The use of indirect statements, often in the form of a question, is a key feature of the mainstream linguistic style (e.g., "Isn’t it time for your bath?" Delpit, 1995, p. 34). Linguistic style is important because the Expressive and Receptive Language PEP sub-scale rates the use of commands, discouragements, open-ended questions, verbal encouragement, and the like. A hallmark of the concerted cultivation style is the re-creation of school-like experiences in the home (Lareau, 2003), for example, labeling objects in books and everyday life, eliciting what-explanations ("What is this?"), and engaging in interactive storybook reading (Heath, 1983, 2001). Bedtime storybook reading is the "paradigmatic reading routine" for mainstream families (Griffith & Smith, 2005, p. 43). Although this practice is culturally specific (Hammer, Nimmo, Cohen, Draheim, & Johnson, 2005; Janes & Kermani, 2001; Panofsky, 1994, 2000), many middle-class parents consider it "a natural way for parents to interact with their child at bedtime" (Heath, 2001, p. 319). In short, the mainstream child-rearing model emphasizes the development of school literacies (Delpit, 2003; Gadsden, 1994; Paratore, 2001)—practices that match how many schools and educators define acceptable literacy practices, linguistic style, and display of knowledge (Cairney & Ruge, 1998; Heath, 2001; Lareau, 1987; Panofsky, 2000; Paratore, 2001; Valdés, 1996; Vernon-Feagans et al., 2004). The nonmainstream child-rearing approach tends to emphasize clearly defined parent and child roles and deference to parental authority rather than negotiated decision making. Accordingly, verbal interaction is often more directive than democratic (Delpit, 1995; Heath, 2001; Lareau, 2002; Lareau & Weininger, 2007; Valdés, 1996). Additionally, Lareau (2002) found that most poor and working-class parents did not "focus on developing their children’s opinions, judgments, and observations" (p. 763), reflecting the view that teachers were responsible for fostering children’s cognitive development. These patterns may reflect cultural perspectives about the value of parental control (e.g., Chao, 1994) or a response to socioeconomic conditions (García Coll et al., 1996), such as the need to prepare children for jobs that reward obedience (Hays, 1996, pp. 93–95) or to protect them from dangerous neighborhoods (Cook & Fine, 1995). Home-based (out-of-school) literacies such as storytelling, consejos ("spontaneous homilies designed to influence behaviors and attitudes"; Valdés, 1996, p. 125), and use of print for nonschool and communal purposes are common among nonmainstream families (Cairney & Ruge, 1998; Paratore, 2001), although many also engage in schooled literacies. Challenging the view that poor and minority children are linguistically deficient, some scholars (e.g., Heath, 2001; Panofsky, 2000; Purcell-Gates, 2004) highlight their contextualized knowledge and oral language abilities, skills that standardized instruments may overlook (Campbell, Dollaghan, Needleman, & Janosky, 1997). Books are often scarce in poor and working-class homes (Heath, 2001; see Rogers, 2003, for an exception), and some nonmainstream parents—especially those with limited schooling—consider storybook reading uncomfortable or unfamiliar (DeBruin-Parecki, 1999; Janes & Kermani, 2001; on reading practices, see Hammer et al., 2005; Panofsky, 2000; Powell et al., 2004; Reese & Gallimore, 2000). Parent involvement in school. Lareau (1987, 2000) posits that middle-class families have an interdependent relationship with schools, as their literacy practices, daily routines, and organized activities center on schooling and they often socialize with school families (see also Griffith & Smith, 2005). Mainstream parents typically have comfortable relationships with school staff, whom they regard as social equals (S. Auerbach, 2007; Lareau, 1987). Additionally, they tend to intervene in educational institutions to "customize" children’s education and seek favorable treatment (Brantlinger, 2003; Lareau, 1987, 2000; McGrath & Kuriloff, 1999). Importantly, middle-class parents’ forms of engagement—for instance, classroom volunteering and attending school functions—align with how schools define legitimate involvement (S. Auerbach, 2007; Lareau, 1987; Nakagawa, 2000; Valencia & Black, 2002). However, school professionals may also view overly involved parents as too demanding (McGrath & Kuriloff, 1999). By contrast, poor and working-class parents tend to have an independent relationship with schools, viewing family and school as separate spheres (Lareau, 1987, 2000). According to this child-rearing logic, parents are responsible for nurturing children’s moral development and physical, social, and emotional well-being, while professionals are best equipped to develop their academic abilities (Delgado-Gaitán, 1990; Delpit, 1995; Heath, 2001; Lareau, 2002; Valdés, 1996; Valencia & Black, 2002). Some studies show that poor and minority parents have often had negative experiences with representatives of schools, courts, and other institutions, engendering mistrust, reluctance to initiate contact with schools, or sporadic, formal interactions with teachers (S. Auerbach, 2007; Diamond & Gomez, 2004; Lareau, 1987, 2002; Lareau & Horvat, 1999; Olivos, 2006). While many nonmainstream parents do advocate for school change (Diamond & Gomez, 2004), these efforts do not necessarily bear fruit (Lareau & Horvat, 1999). In sum, both cultural models present viable ways of raising children and engaging in literacy, but the concerted cultivation model is typically accepted in professional and popular discourse as the ideal. The next section explains how we used these ideal types and CDA to investigate the PEP.
CDA is well suited to our focus on the PEP’s conceptualization of the ideal parent, its assumptions about parenting and education, and its ideological effects. We used CDA because it allowed us to examine the tacit meanings, values, and assumptions embedded in a text and to link these to wider ideologies and discourses (Fairclough, 1989, 2001, 2003; Gee, 1999; L. J. Phillips & Jorgenson, 2002; van Dijk, 2001; Wodak & Meyer, 2001). In this article we are chiefly interested in implicit meanings, or those that "are not openly, directly, completely or precisely asserted" (van Dijk, 2001, p. 104). CDA is based on the premise that language, knowledge, and power are inextricably linked (Foucault, 1980). Thus, it "focuses on how language as a cultural tool mediates relationships of power and privilege in social interactions, institutions, and bodies of knowledge" (Rogers, Malancharuvil-Berkes, Mosley, Hui, & Joseph, 2005, p. 367). CDA posits that oral and written language promotes particular ideologies, or "the system of ideas and values that reflects and supports the established order and that manifests itself in our everyday actions, decisions, and practices, usually without our being aware of its presence" (Brookfield, 2005, pp. 67–68). Because language privileges certain values, ideas, and practices over others (L. J. Phillips & Jorgenson, 2002; van Dijk, 2001), those associated with a dominant discourse become the norm and appear to be the inevitable, natural order of things—the best or only way parents should be involved in schools, for example. The ways people interpret and use texts like the PEP promulgate specific views of parenting, gender roles, educational disparities, literacy, and so on. As a social practice, spoken and written language reflects and shapes relations of power in everyday interactions, institutions such as educational programs, and society (Fairclough, 1989, 2001, 2003).
Data Selection
Data Analysis and Trustworthiness
Gee’s (1999) and Fairclough’s (1989, 2003) methodological frameworks were useful ways to examine both smaller units of language and the messages conveyed in the text as a whole. We followed Gee’s recommendation to begin with a small piece of data, formulate hypotheses, and apply the hypotheses to larger pieces of data. Beginning with the Preface, we carefully read each section and developed preliminary hypotheses about what was (or was not) being communicated. For instance, we considered how the authors wanted the PEP to be used and interpreted, which audience(s) it addressed, which concepts were (de)emphasized, and how parents were described. We then examined other parts of the text to see whether our initial hypotheses were supported or challenged. Based on Gee’s (1999) "building tasks" of CDA, we studied how pieces of language (words, phrases, sentences, and longer stretches of text) were used to make certain things significant or not (e.g., categorizing parental practices by levels); to enact specific activities, identities, and relationships (e.g., the roles parents and staff are to play in assigning ratings, the parenting and gender identities the PEP creates); to distribute social goods (e.g., what the PEP’s creators consider good or normal, or what Gee terms politics); to make connections among or to separate concepts (e.g., linking parents’ ability to reach higher levels on the PEP to children’s academic success); and to privilege or discredit specific sign systems and forms of knowledge (e.g., the use of scientific language, stance regarding parents’ knowledge). This process helped us discern how the PEP conveys what is normal, right, good, proper, appropriate, valuable, or high status—in short, what it means to be a good parent. We also used a macro-line technique, which entailed examining sentences in their entirety (Gee, 1999). This analysis upheld the microanalysis of specific words (for example, see our analysis of the "Message to Parents" and "Including the Parent Perspective"). Finally, we used the 10 questions posed by Fairclough (1989) to analyze the PEP’s vocabulary and grammatical features, including word choice and repetition, use of pronouns and active or passive verbs, and relationships between subject, verb, and object, among others. These questions help scholars identify the experiential, relational, and expressive values of words (i.e., vocabulary) and grammatical features in a given text. Experiential value refers to the ways language encodes the producer’s knowledge or "experience of the natural or social world" (Fairclough, 1989, p. 112), including ideological representations. The relational value of a text reflects and creates social relationships (e.g., between professionals and parents), while the expressive value conveys the producer’s appraisal or evaluation of reality or truth. By posing Fairclough’s questions we saw more clearly, for instance, how images of women and the use of female pronouns established mothers as the instrument’s main audience. To enhance the rigor and quality of data analysis, we used several strategies common in the CDA literature, namely, explicating our methodological choices, theoretical position, and analytical procedures (Cheek, 2004; Gee, 1999; Meyer, 2001; Rogers et al., 2005); providing text-based examples to support claims (Meyer, 2001; van Dijk, 1993, 2001); and connecting textual analysis to scholarly literature (Gee, 1999; van Dijk, 1993). Finally, reflexivity (Fairclough, 2001; Rogers et al., 2005) involves examining our own assumptions (for example, through the review process) and locating ourselves as White scholars from lower-middle-class and middle-class backgrounds, respectively, who are committed to equity in family literacy and other educational settings.
According to Foucault (1977/1995), disciplines exercise power by classifying, characterizing, arranging on a scale, disqualifying, and invalidating particular practices, thereby establishing what is normal or abnormal. By examining which practices the PEP places at the highest and lowest levels of the scale, we can identify what the authors consider desirable and undesirable or, in Gee’s (1999) terms, how social goods are distributed. We found that in many respects the PEP’s construction of the ideal (Levels 4 and 5) parent aligns with the mainstream parenting model, while the nonideal (Levels 1 and 2) parent shares some similarities with the nonmainstream model (see Table 2). Thus, the PEP upholds specific features of the middle-class, predominantly White parenting style as a normative ideal, yet does not present this as a culturally specific style that may produce different outcomes across class, race, and culture (García Coll & Patcher, 2002). Because the instrument is designed as a one-size-fits-all model, it does not appear to allow for conceptions of effective parenting, appropriate parental and child roles, and pathways to literacy development that deviate from the PEP, at least as they relate to the scale items. Much of the language describing ideal parenting practices, particularly Scale III (Supporting Children’s Learning in Formal Education Settings), reflects the dominant discourse of parental involvement, which holds that good parents support their children’s education in specific ways and are actively involved in schools, according to the school’s definition of involvement. Although the PEP emphasizes formal school involvement, it is also based on the premise that parents can support education behind the scenes. For instance, one of the subscales refers to communicating high educational expectations, something parents of many cultural and socioeconomic backgrounds already do (Goldenberg et al., 2001). The inclusion of this item helps convey a more expansive definition of educational support and involvement. Analysis of vocabulary in the scales showed that the descriptors of Level 4 and 5 parents (Table 2) evoked an image of stability (e.g., parents engage in the activities regularly, ongoing, everyday, daily), involvement, (e.g., the parent is interested or participates in school activities), and initiative (e.g., the parent initiates learning activities or contact with the school). That these attributes are located at the highest levels of parental development indicates they are more valued and desirable than others (Gee, 1999). Like the concerted cultivation of parenting logic (Lareau, 2003) and the child-centered, labor-intensive ideology of intensive mothering (Hays, 1996), the PEP portrays the ideal parent (mother) as one who continuously seeks opportunities to develop children’s literacy and cognitive skills, vigilantly monitors children’s learning, regularly intervenes in schools and attends school functions, supplies abundant literacy materials, gives children choices, and reasons with children in a give-and-take, democratic manner. Because the PEP’s ideal parent is deeply involved in school-based literacies such as homework and storybooks, this parenting style assumes parents’ time and energy are structured more by the "institutional regime" of schooling (Griffith & Smith, 2005) than by household survival. Finally, the PEP’s ideal parent, especially the Choices, Rules, and Limits subscale, parallels several dimensions of Baumrind’s (1971) authoritative parenting style, in which parents "establish and firmly enforce rules and standards for their children’s behavior," "consistently monitor conduct and use nonpunitive methods of discipline," are "warm and supportive," "encourage bidirectional communication, validate the child’s individual point of view, and recognize the rights of both parents and children" (Glasgow, Dornbusch, Troyer, Steinberg, & Ritter, 1997, p. 508). This model prioritizes children’s independence and self-expression, key facets of the middle-class parenting style (Chao, 1994). For example, the PEP describes a Level 5 parent as one who "consistently provides opportunities for child to make choices within limits, e.g., age, safety" and "actively engages the child in discussion." The PEP provides some evidence that Baumrind’s model is considered "the prototype for appropriate parenting," one widely promoted by intervention programs (Chao, 2001, p. 1832). In sum, the descriptors for Level 4 and 5 parenting practices underscore Gee’s (1999) notion that language makes certain activities and identities significant. Many of the Level 1 and 2 parenting practices, on the other hand, overlap with the nonmainstream style more common among poor and working-class families. For instance, the PEP constructs the nonideal parent as one who mainly uses commands, has few literacy materials in the home, does not engage in regular book reading or focus explicitly on developing school-based literacies, rarely contacts teachers, and does not have extensive knowledge of the school’s expectations or the child’s educational progress. We do not believe these attributes inherently indicate poor parenting skills or lack of support for education (Kao, 2004). However, the PEP descriptors implicitly construct Level 1 or 2 parents as lacking awareness, knowledge, understanding, or ability (see Table 2), a message that undercuts family literacy practitioners’ claims and efforts to "oppose deficit perspectives and embrace family strengths" (Nakagawa, McKinnon, & Hackett, 2001). Moreover, the descriptors for Level 1 and 2 parents are primarily negative, including nothing, no, not, doesn’t, not aware, does not know, no role, not formed, and negative. Like the early literacy research Panofsky (2000) analyzes, the PEP frames some differences (e.g., using commands) as deficits. These deficits take on a moral dimension because, as Panofsky explains, parents who do not read to their children (or adhere to other aspects of ideal parenting) are perceived as "neglectful" and "are by implication ‘bad’ parents" (2000, p. 191). The Expressive and Receptive Language subscale in Scale II exemplifies how the PEP positions the middle-class parenting and linguistic style as an ideal. A Level 5 parent "actively engages the child in discussion, using strategies such as paying attention to the interests of the child, using open-ended questions, providing verbal encouragement, or giving the child an opportunity to process information" (RMC, 2003, p. 16). On the other hand, a Level 1 parent’s "verbal interactions with child are predominately commands or discouragements [and the parent] responds inconsistently to child’s verbal or behavioral clues." Neuman and colleagues (1995) observe that the "ideal" middle-class mother . . . has often been described as one who is engaged in inquiry-like verbalizations, rarely making negative, corrective, or punitive statements, compared to the poor lower status mother who is seen as controlling, directive, and intrusive. . . . Implicit in these studies [focusing on verbal interaction and linguistic style] is the notion that lower status parents are failing their children linguistically and cognitively by not providing an enriching environment for learning. (pp. 802, 804) The PEP seems to promote this perspective: Higher level parents ask open-ended questions and let children make choices, while lower level parents tell children what to do, have "limited verbal interaction," and use "simple sentences" and "yes/no" questions (see Dudley-Marling, 2007, for a critique of Hart & Risley’s research on language development). By focusing on the linguistic form rather than the content of educational messages (Neuman et al., 1995), this conceptualization may overlook legitimate cultural differences in parent-child communication style and beliefs about appropriate parental and child roles (Johnston & Wong, 2002; Powell et al., 2004; Valdés, 1996). In sum, many (but not all) parts of the PEP evaluate parents according to middle-class standards, presuming these are equally appropriate and beneficial for all families.
Use of Scientific Discourse Like the parent involvement policy texts Nakagawa (2000) analyzed, the PEP uses scholarly research (the endnotes cite 31 studies dated 1982 to 1997) to support its construction of the ideal parent. Specifically, the text infers direct, causal relationships among complex factors that may only be correlated (see Hayes, 1996)—a flaw few practitioners are likely to notice. For example, the PEP includes the following excerpt from Powell and D’Angelo’s (2000) report on parenting education in Even Start: Research indicates that children’s success in school is related [italics added] to their active involvement in joint book reading at home (a parenting practice with child), parents’ expectations of their child’s educational attainment (a parent belief), and the predictability and routines of the home setting (parent as manager of environment). (p. 64) The PEP then prescribes these practices and tells parents, "the higher levels you reach on PEP, the more likely your child will do well in school" (RMC, 2003, p. 92)—a clear causal inference. The risk of such inferences is that educators and parents may incorrectly assume that the symptoms (specific behaviors) cause school success. Indeed, Hayes (1996) contends that findings from correlational family literacy research—for example, studies examining presence of reading materials, amount of time spent reading, and TV viewing—"have led to overly simple conclusions about how to intervene." Similarly, the PEP assumes specific parental beliefs and practices directly cause academic success or low achievement. The discursive strategy of using scientific language accomplishes several purposes: It assures the intended audience (policy makers, administrators, practitioners, and, indirectly, parents) that the PEP is an impartial, value-neutral tool; it suggests the effectiveness of the recommended practices is beyond question; and it encourages the audience to accept these practices as natural and commonsensical (Fairclough, 2003). Such reification of scientific research, however, fails to recognize that knowledge and scholarly inquiry are socially constructed, embedded in culturally specific, value-laden understandings of child and literacy development and optimal parenting. Indeed, research has been instrumental in establishing the deficit discourse (García Coll et al., 1996; Tate, 1997; Valencia & Black, 2002) and the taken-for-granted views of child development underlying the mothering discourse (Griffith & Smith, 2005).
A Universal Set of Parenting Practices Supports All Children’s Literacy Development The PEP recognizes that people have distinct cultural views of parenting and education yet simultaneously suggests all parents should adopt the recommended practices. The "Cultural Context" section, addressed to staff, illustrates this point: The expectations of the parent role embedded in the PEP are related to children’s success in American schools and so are important for parents from all backgrounds to understand. Of course, a parent’s cultural or family background may make it difficult for her to readily embrace some behaviors. For example, some parents do not feel comfortable engaging young children in conversation or helping children with work from school. In those cases, staff may need to work extra hard to help parents understand the desired behaviors and reasons for their importance. (RMC, 2003, p. 59, all italics added) The text accurately notes that some parents may feel discomfort. However, the excerpt portrays parents’ cultural identities only as a potential obstacle and implies that in matters of supporting children’s education, professional knowledge carries more weight than that of parents (Gee, 1999). The text does not suggest that parents may have reasons for resisting certain practices or that educators could help them identify alternative ways to support school success, as Powell and colleagues recommend (2004). The excerpt implies the intended goal—embracing the prescribed behaviors—is not open to debate. The PEP, then, presumes there is a universal model of how U.S. parents should interact with their children to promote literacy development. The suggestion that parents should strive to overcome discomfort resulting from cultural beliefs and adopt the recommended behaviors tacitly encourages cultural assimilation. For example, the passage does not advise professionals to help parents weigh the benefits and disadvantages of replacing their beliefs, values, and practices with those of the dominant culture and decide which practices they want to change. As previously stated, many of the prescribed practices are related to positive academic outcomes, especially among White, middle-class children. However, the PEP does not appear to reflect research showing that there are multiple pathways to school achievement (Carreón, Drake, & Barton, 2005; Chao, 1994; Kao, 2004; Okagaki & Frensch, 1998; Powell et al., 2004; Szalacha, Marks, LaMarre, & García Coll, 2005; Wright & Smith, 1998) and that the same parenting behaviors may affect children differently depending on the family’s race/ethnicity (Berlin et al., 1995; Chao, 2001; Deater-Deckard, Dodge, Bates, & Pettit, 1996; García Coll & Pachter, 2002; Panofsky, 1994; Sugland et al., 1995). Furthermore, by discursively linking the prescribed practices to school success (what Gee, 1999, calls connections), the text implies that failure to follow the PEP will contribute to academic underachievement. Finally, the concepts of relational value (Fairclough, 1989) and relationships (Gee, 1999) reveal how the language in this excerpt constructs a hierarchical relationship whereby professionals convince parents to change ("work extra hard to help parents understand"), while parents "understand" and "embrace" the desired behaviors. The "Cultural Context" section, coupled with terms such as "intervention" elsewhere in the text, sends the message that parents need professional assistance to become fully competent (E. R. Auerbach, 1989). Many educators would no doubt disagree with this message, but it is nevertheless implied in the text and thus is likely to penetrate practice in unconscious ways. The manual includes sample anecdotal notes justifying ratings, written by staff members who were part of the PEP pilot study. Several of these examples reveal how professionals in some cases evaluated parent behaviors through their own cultural lenses. For instance, one person provided the following rationale for assigning a Level 1 rating on Scale III, Parent’s Role in Supporting Child’s Learning in Education Settings: "Parent expects school to teach alphabet to child but does not see that it is ‘her job’ to work with the child" (RMC, 2003, p. 46). These notes reflect the dominant perspective that positions parents as adjunct teachers and encourages school-like activities in the home (Lareau, 2003). However, Lareau (1987) suggests that for parents who struggle with literacy or are burdened with life demands—as many poor, working-class, and immigrant parents are—it is reasonable to expect teachers to impart academic knowledge (see also Valdés, 1996). Conceivably, then, a parent could receive a lower rating because some of her parenting, literacy practices, and beliefs about appropriate roles for parents, children, and teachers diverge from the PEP.
Parents (Mothers) Are Primarily Responsible for Educational Outcomes Other sections of the text reinforce parental responsibility for literacy and school achievement. "A Message to Parents About PEP" illustrates this point: You have a big job to do in helping your child be successful in school. You make a difference by: how you talk with your child; how you set and use rules; how you play with your child; how you use reading and writing . . . and in many other ways! The Even Start staff want to help you with the education part of being a parent. The staff know about research that links the actions of parents to the success of children in school. They will share that information with you. Changing how you talk and work with your child takes time and effort. To keep track of your progress and know what to do next, the Even Start staff will use the Parent Education Profile—PEP, for short. The PEP show [sic] steps to success in four areas of parent activity. The staff will figure out what step you are on now. They will discuss goals with you, see you work with your child, and watch you try out new ways of talking and reading with your child. The higher levels you reach on PEP, the more likely your child will do well in school. (RMC, 2003, p. 62, all italics added) The text tells parents that they have a "big job to do" and should change how they "talk and work" with their child so she or he will do well in school. These statements imply that parents are potentially part of the problem and the solution, a common refrain in the parent involvement discourse (Nakagawa, 2000, p. 456). Furthermore, the statement linking higher PEP levels to school performance makes an unsubstantiated causal inference, as we are not aware of any research investigating causal or correlational relationships between the PEP instrument and academic achievement. This inference may convey an overly simplistic message to parents and teachers: Parents need only change their behavior to ensure children’s literacy development and academic success. The assumption of parental responsibility in the PEP serves to promote an individualistic understanding of educational disparities, while obscuring the need to alter institutions, policies, and social structures (Luttrell, 1996, p. 354; Panofsky, 2000). A notable exception to the individualistic focus is a Level 5 descriptor: "The parent works with others to promote system improvements for quality education for all children" (RMC, 2003, p. 18). The purpose of the PEP is to assess parenting practices, not structural changes or school reform. However, the supplementary materials addressed to practitioners and parents do not mention that factors outside the family—for instance, teacher perceptions and practices such as reading group placement (e.g., Condron, 2007; Panofsky, 2003) or school-level factors such as segregation (e.g., Betts, Rueben, & Danenberg, 2000; Parcel & Dufur, 2001; Portes & Hao, 2004; Roscigno, 1998)—also influence educational success. The PEP does not suggest, for example, that parents could support educational achievement by advocating for highly qualified teachers or equitable funding—an alternative vision of parent involvement (Olivos, 2006). Despite frequent references to "parent(s)," the images, sample documentation of parent ratings, and underlying assumptions reveal that mothers (or other female caretakers) are the instrument’s intended audience. The use of gender-neutral language (parent, he/she, and his/her) in the PEP scales is commendable, as it suggests the instrument and prescribed practices are applicable to both men and women. However, the vast majority of family literacy participants are mothers, and in North America mothers are chiefly responsible for caretaking and supervising children’s education (Fineman, 1995; Kline, 1995). This suggests that in reality the PEP is mainly directed toward and used with women; it is women who will be observed and rated, whose time and energy will be directed toward children’s schooling. Furthermore, the 11 photos in the PEP depict racially diverse women (presumably participants, staff, or volunteers) interacting with children, but none depict men (on fathers in family literacy, see Gadsden, 2003). Images convey ideas about how the world is and ought to be. The problem, then, is not that the pictures portray reality but that they also normalize it, supporting the assumption that literacy and caretaking should be women’s work (Luttrell, 1996). Finally, of the 128 examples of documentation for rating parent behaviors, 3 (2%) mention fathers, 4 (3%) are gender neutral, 24 (18%) refer to family or parent(s), and 96 (75%) mention mother(s) (e.g., "Mom makes up games to play with the kids . . ."). This pattern likely reflects women’s overrepresentation in family literacy programs, including those who participated in the PEP pilot study. In these ways the images and text tend to reinforce the mothering discourse as a normative ideal (Dudley-Marling, 2001; Griffith & Smith, 2005; Smythe & Isserlis, 2004; Sparks, 2001). They help circulate the ideology of intensive mothering (Hays, 1996) by emphasizing reliance on expert advice and prescribing labor- and emotionally intensive, child-centered activities as an ideal child-rearing model. Prior research suggests many North American mothers internalize a sense of responsibility for children’s academic achievement, often engendering guilt, anxiety, and a propensity for comparing how they measure up to other mothers and to school standards (Dudley-Marling, 2001; Griffith & Smith, 2005).
All Parents Are Equally Able to Engage in the Prescribed Practices The descriptions of Level 4 and 5 practices reveal that parents must have access to specific resources to attain higher ratings on several subscales (RMC, 2003): Home has a variety of materials for reading, writing, and drawing that are accessible to child. Materials are used daily. (Scale I, Level 5, p. 14) To attain a high rating on the first four criteria, a parent (mother) needs free time, disposable income, a flexible work schedule, transportation, and, if she has other children, affordable and accessible child care, conditions that scarcely describe the lives of most poor and working-class families (see Dodson, Manuel, & Bravo, 2002; Heymann, 2000). Because the ratings are based on observed behaviors and living environment, the instrument cannot account for extenuating circumstances such as single-parent families or limited financial resources. The fifth criterion assumes parents belong to a social network of people who are knowledgeable about the educational system and their children’s school. Lareau (1987) found that the cultural capital and social networks of middle-class families facilitated intervention in schools and the sharing of information about schooling. Poor and working-class families, however, had limited information about children’s school experiences (partly because they socialized with relatives, not school families) and often felt powerless in their interactions with school personnel. Lareau’s findings suggest parents’ social class and social network composition are likely to shape their ability to comply with PEP expectations. Documentation notes for the PEP pilot study ratings (RMC, 2003) provide additional evidence that adequate resources help parents attain higher levels on some subscales: Parents provide much exposure to books, computer, outdoor play, and exploration. The family takes frequent trips to the library and park. (Scale I, Level 4, p. 33) A low-income family that lives in an unsafe neighborhood with no museums, parks, or libraries, for instance, would be hard pressed to provide the learning opportunities noted in the first example. Poverty also subjects families to unsafe housing, food insecurity, and health problems (second example). To take children to the doctor whenever they are ill, a parent needs extra income, transportation, accessible medical facilities, health insurance, flexible schedules, and other resources (see Dodson et al., 2002).8 The PEP, however, does not enable raters to distinguish between parents who do not "improve the health and safety of the [home] environment" (RMC, 2003, Level 5, p. 20) due to limited resources, and those who do not care or understand why this is important. Instead, the PEP attributes Level 1 practices on this and other subscales mainly to lack of awareness or understanding: "Parent is not yet aware that issues in the home settings/environment have a negative effect on child’s learning and development, e.g., domestic violence, substance abuse, nutrition, smoking" (RMC, 2003, p. 20). These examples illustrate Griffith and Smith’s (2005) point that the practical "conditions of mothering" such as low wages, precarious housing, or insufficient time do not seem to lessen mothers’ responsibility for children’s schooling (or their health and safety). The assumption that families have the resources required to meet the PEP standards may unintentionally equate higher SES with better parenting on some subscales (see Berlin et al., 1995, for a similar discussion of the HOME). By focusing on parents’ knowledge, awareness, and actions apart from material circumstances, the PEP tends to shift our attention from the inequitable distribution of resources among poorer and wealthier families to individuals’ compliance with the ideal parent model (Kline, 1995). Our concern is that the instrument may unwittingly hold parents accountable for social and economic circumstances over which they have little control.
The Parent’s Role in Assessment Is Limited to Behavior Change The PEP is designed to capture patterns of important behaviors—most of which take a good deal of practice to change. Therefore, it makes sense to formally assess changes in a parent’s behavior only after he or she has had enough time to learn and practice new skills. . . . (RMC, 2003, p. 57) In states that use the PEP as a program performance indicator, program quality is contingent on parental behavior change, that is, reaching higher PEP levels. Behavior modification, which has a long history in adult (Fingeret, 1989) and parent education (Sigel, 1983), is problematic because it can facilitate social control and undermine self-determination. A central concept in CDA is that language and texts both reflect and alter social relations, establishing particular identities and ways of relating (Fairclough, 2003). Several excerpts illustrate how the language and intended use of the PEP tend to construct imbalanced power relations between staff and participants (Wodak, 2001). Paradoxically, the "Message to Parents" simultaneously credits parents with responsibility for shaping educational outcomes yet positions them as passive primary agents. An active agent is able to make things happen and exercise control over others, whereas a passive agent is subjected to processes and the actions of others (Fairclough, 2003). The 22 uses of you and your in the text establish parents as primary agents, meaning they are supposed to change their behavior in accordance with the PEP. Yet other parts of the PEP suggest parents do not have a say in formulating or negotiating the standards by which they are evaluated. The absence of active verbs also indicates parents have little agency in the administration of the PEP. In assigning parents and mothers considerable responsibility but little or no control, the PEP mirrors the prevailing discourses of parent involvement (Nakagawa, 2000) and mothering (Griffith & Smith, 2005). The "Message to Parents," quoted earlier, positions staff members as keepers of specialized knowledge ("research that links the actions of the parents to the success of children in school") but does not mention that parents also have valid knowledge to contribute. This passage illustrates how texts privilege certain forms of knowledge (Gee, 1999): In this case, expert, scientific knowledge of child development and parenting appears to supersede "mothers’ experiential understanding of their children" (Griffith & Smith, 2005, p. 36). There is much we can learn from research on parenting. However, the elevation of expert advice reflects the belief, emerging in the 1800s, that mothers need to be "scientifically trained" to ensure proper child rearing, a belief grounded in a growing national confidence in scientific, "expert-guided" solutions to social problems (Hays, 1996; see also Schlossman, 1978, 1983). Analysis of verbs in the "Message to Parents" provides additional evidence of hierarchical relations. As observers and evaluators, staff members know, share, keep track, figure out, discuss, see, and watch (see Foucault, 1980). Parents, on the other hand, are framed mainly as recipients of knowledge who are subject to the actions of professionals and responsible for adopting new behaviors. By repeating such messages throughout the text, the PEP scripts distinct activities, identities, and roles for participants and professionals (Fairclough, 2003; Gee, 1999). Because people are active agents, they could use the PEP in unconventional ways and break out of these scripted roles to create more equitable relationships, for example, by including parents’ self-assessment. We elaborate on this point in the conclusions. Macro-line (whole sentence) analysis (Gee, 1999) revealed that the PEP explicitly encourages the inclusion of the parent’s viewpoint (RMC, 2003, pp. 6, 57), yet the instructions for administration appear to undermine this suggestion. A section titled "Including the Parent Perspective" demonstrates this contradiction: A parent’s perspective on his or her progress is a valuable addition to the discussion of achievements. An individual staff member might gather information from discussion with a parent in advance of an assessment meeting and bring that information into the assessment discussion. If the staff disagree with the parent’s assessment of progress, it is important to explain to the parent where the differences are—use this opportunity to discuss what a parent might do differently with their children to improve their skills. (RMC, 2003, p. 57) According to the text, staff members should include parents in the discussion, but if there is disagreement, the PEP implies professional assessment should prevail ("use this opportunity to discuss what a parent might do differently"). An inclusive approach would mean that (a) both parents and professionals might reconsider their perspectives and the final rating and (b) that parents’ self-assessment would be formalized and included in the final assessment. Another excerpt eliminates the parent’s perspective altogether: Because the PEP is based on staff perspectives formed during ongoing interactions with parents, the assessment can be completed even after a parent has exited the program. . . . All staff who have substantive contact with the parent [e.g., home visitor, adult educator, teachers of parent’s children, program coordinator] should work together to determine the level that a parent has attained on each sub-scale. (RMC, 2003, p. 57, all italics added) In this selection, the assessment process is not described as a collaborative effort in which parents can assess themselves and influence their rating; rather, the staff administers the instrument based on "what they have seen the parent do and say" (RMC, 2003, p. 58). In the former approach to assessment, the teacher-learner relationship is based on "power with," whereas the latter illustrates "power over" (Gipps, 1999). That staff members can assign a rating after a participant leaves the program constrains the parent’s agency. The choice of prepositions in another excerpt underscores the parent’s limited agency: The instructions suggest staff "might use one or two scales on [italics added] all families to begin to familiarize staff with the process of PEP administration" (RMC, 2003, p. 59). The preposition on—as opposed to with—again positions parents as passive agents (Fairclough, 2003), the objects of others’ actions. This is not only a matter of language, for we have learned that some educators have observed and rated parents without informing them, in one case as part of a nonresearch pilot study in several programs and in another as a routine program practice, reportedly to help parents feel more at ease. Such uses of the PEP—or any observational measure—diminish participants’ agency and undermine their right to know they are being assessed. The assumption that parents should change their behavior to attain higher PEP levels, coupled with their limited ability to influence the rating, demonstrates how measures like the PEP can facilitate regulation and government oversight of the parenting and literacy practices of marginalized women and men (Sparks, 2001)—even if this is not what educators intend. The PEP text constructs a specific set of social relations for parents and staff: the organization and evaluation of parents’ (mothers’) activities according to the "institutional regime" of schooling (Griffith & Smith, 2005), activities that may replace other, equally valid forms of teaching, learning, and communicating. The instructions suggest parents may offer their perspective, but if neither the final rating nor the conception of the ideal parent is negotiable, it is unclear what purpose parental input serves. In short, the instructions for administration construct power relations that appear to circumscribe parents’ role in the assessment.
This study illuminates how the PEP discursively constructs the ideal parent and its underlying assumptions about parenting and education, demonstrating how many of its features perpetuate discourses and ideologies that may inadvertently uphold power inequities between poorer and wealthier families, and between participants and professionals. In addition to explicating the content of the ideal parenting model, our analysis shows how the instrument itself—the rating system, standardized format, and instructions for administration—serves to rank and normalize parental practices and guide staff and parents toward particular identities, roles, and behaviors. As such, this study raises timely questions about the appropriateness of using the PEP to assess how parents—primarily poor women and women of color—support children’s literacy development and, in turn, the effectiveness of family literacy programs. Our analysis complements research by Nakagawa et al. (2001) showing how programs’ use of assessment tools can unwittingly promote deficit views of parents, contrary to practitioners’ stated beliefs. In this case, the image of the ideal parent, coupled with the rating system, implicitly labels as deficient parents who do not or cannot engage, for whatever reason, in PEP practices. We do not believe educators intend to send these messages, making it all the more important to examine the underlying assumptions of assessment instruments. We have argued that many aspects of the PEP promote and evaluate parents according to features of the mainstream (concerted cultivation) model of parenting, while some of the practices more common among poor and working-class parents, such as the use of directives, are labeled inadequate and unsupportive of children’s literacy development. The supplementary materials use scientific discourse—without information about reliability, validity, or bias—to establish the PEP’s legitimacy and to present the recommended practices as a natural, normative ideal. By evaluating family literacy participants according to this standard, the instrument reinforces such practices as the use of school-based literacies in the home, negotiation and reasoning with children, child-centered activities, formal involvement in schools, and reliance on mothers’ unpaid, supplementary work to ensure educational success (Griffith & Smith, 2005). These findings resonate with the normative assumptions underlying the prevailing discourses of parenting (Chao, 1994; Griffith & Smith, 2005; Hays, 1996; Schlossman, 1978, 1983), family literacy (Smythe & Isserlis, 2004), early literacy (Panofsky, 2000), child development (García Coll et al., 1996), and parent involvement in education (Nakagawa, 2000). The assumption of a single, universal set of parental practices that best supports children’s literacy development tacitly tends to encourage the adoption of mainstream beliefs, values, and practices, or cultural assimilation. Furthermore, it implies that following the PEP will, ipso facto, produce equally positive results regardless of family background, school and community setting, or socioeconomic conditions—factors that powerfully mediate the influence of parental beliefs and behaviors on educational success (Berlin et al., 1995; Chao, 1994; Kao, 2004; Okagaki & Frensch, 1998; Sugland et al., 1995). To reiterate, the PEP does include practices associated with literacy development and academic achievement in middle- and upper-class families. However, this relationship is less well understood among nonmain-stream families. This study reinforces Okagaki and Frensch’s (1998) assertion that educators cannot assume that what works for one group of families will necessarily work for another group. The social and economic context along with the global constellation of beliefs parents hold regarding multiple aspects of life (e.g., the importance of family, principles of child development, education, perspective of work, and their general world view) may make intervention strategies that work in some family contexts ineffective in other family contexts. (p. 142; see also García Coll & Patcher, 2002) By assuming that parents are chiefly responsible for children’s educational outcomes, the PEP subtly shifts our attention to parental practices and away from structural and institutional factors that contribute to educational disparities. Such a shift lays the groundwork for blaming the victim (Ryan, 1971). The PEP scales justifiably focus on ways parents can support children’s education, and there is much we can all learn in this regard. However, by implying that educational achievement is mainly the result of parental practices, the supplementary materials do not help educators understand that literacy development and family-school-community relationships are mediated by social stratification (García Coll & Pachter, 2002; Wright & Smith, 1998). Alternatively, the text could have framed literacy development as a complex process influenced by numerous factors, including child attributes, school conditions, community setting, and parental beliefs and practices—the focus of the PEP. Such a stance would help professionals and participants understand that parents are a part, but not the only part, of a multifaceted web that shapes children’s learning. Because it is implicitly directed toward mothers and encourages a particular style of child rearing, the PEP also tends to support the view that mothers are accountable for children’s achievements and struggles in school and in life (Griffith & Smith, 2005). In practice, of course, people use and interpret texts in unpredictable, surprising ways. Future research might examine whether the PEP enhances or diminishes mothers’ sense of adequacy and competence. The third assumption, that parents are equally able to engage in the prescribed practices, is problematic because poor and working-class families have limited access to the resources many of these activities require and are less able than wealthier families to mobilize material, social, and cultural resources to their children’s advantage (Lareau, 1987, 2002; Lareau & Weininger, 2007; McGrath & Kuriloff, 1999). Most family literacy programs do provide books, educational games, free field trips, and other valuable resources, but PEP observers are not instructed to account for chronic conditions such as scarce economic resources. It is important, then, to consider whether the PEP’s expectations (e.g., to participate in school-connected activities 4 to 6 times per year) are reasonable for poor families and what kinds of resources are needed to comply with these standards (de Carvalho, 2000; Lareau, 1987). We believe that assessment tools should convey high expectations and that such activities as providing a literacy-rich environment, contacting teachers, and caring for children’s health and safety are important. What we question is the fairness of rating parents’ behaviors without also considering their life circumstances or reasons for their actions. Finally, by establishing behavior change as its goal and limiting parents’ ability to influence their rating, the PEP instructions appear to construct a power-over relationship for professionals and participants (Gipps, 1999). We suggest that minimally, programs should obtain participants’ consent to observe, document, and rate their parenting practices, for, as Foucault (1977/1995/1980) argued, observation is a mechanism of power. Parents who internalize the "inspecting gaze"—the knowledge that they are being watched—may come to monitor and regulate their own behavior and see themselves through the eyes of the instrument. This disciplinary power also characterizes other forms of standardized testing and assessment (Graham & Neu, 2004). As educators, we support the ability of families to help their children learn and affirm educators’ efforts to help families do so. As Panofsky (2000) puts it, we are not opposed to reading to children or school involvement, but rather "to the misinterpretation and misappropriation of research in ways that naturalize and normalize the cultural practices of some while stigmatizing and marginalizing the cultural practices of less-powerful others" (p. 195). Our concern is that some of the PEP assumptions may undermine educators’ ability to recognize families’ strengths, respect their cultural identities, and include them in important decisions. Adherence to the PEP model could also blind teachers to alternative ways of fostering literacy development and school success (Carreón et al., 2005; Chao, 1994; Kao, 2004; López, 2001; Moll et al., 1992; Okagaki & Frensch, 1998; Powell et al., 2004; Wright & Smith, 1998), for example, using culturally specific book reading styles (Hammer et al., 2005), providing verbal and nonverbal support for academics (Ceballo, 2004), or encouraging regular school attendance and participation in cultural activities (Szalacha et al., 2005). Simply put, caution and cultural sensitivity are needed when using instruments that prescribe, monitor, and rate parental support for education and literacy. Since programs (and possibly researchers) are likely to continue using the PEP, we conclude with suggestions for using it in a more collaborative, culturally responsive manner. First, policy makers should reconsider mandating the PEP and using it as a program performance indicator, as these conditions expand state power to regulate the private sphere, including how marginalized parents (mothers) talk to their children, use literacy materials, participate in schools, and so on. In programs that use the PEP or are considering its adoption, staff and parents would benefit from deliberating together about the issues raised by this study: (a) To what extent does the PEP allow for culturally diverse models of child rearing and literacy support? How could it be adapted to accommodate differing perspectives? (b) Does the PEP offer a complex understanding of the myriad factors influencing literacy development and academic success, including but not limited to the family? If not, how could such information be provided for educators and parents? (c) Can programs provide parents with, or help them access, the resources needed to achieve higher scores on specific subscales? If not, how could the instrument be adjusted to reflect mitigating circumstances? (d) Does the PEP give parents an active, substantive role in the assessment process and enable professionals and participants to relate collaboratively ("power with")? If not, how could parents play a more active role? Practitioners and researchers using the PEP would benefit from examining their beliefs about child rearing and literacy development, as these are likely to affect their ratings—especially if they differ from participants in social class or race/ethnicity (e.g., see Gonzales et al., 1996). Furthermore, as Gadsden (2004) notes, focusing on the parental practices that "appear to be barriers [to literacy] or that are inconsistent with our own perceptions of the world" (p. 422) may produce cultural misunderstanding about the underlying factors that cause literacy problems. As García Coll and Patcher (2002) remind us, "We have to be particularly cognizant of possible stereotypes we bring due to our unfamiliarity of [ethnic and minority] families and our own class and cultural biases" (p. 13). We are more likely to see the strengths of poor and racially/ethnically diverse families if we first recognize these assumptions. Similarly, to learn more about families’ life conditions practitioners could ask parents about extenuating circumstances that may prevent them from engaging in specific PEP activities. In this way, educators could identify—and help parents acquire—the material and social resources needed to provide an enriching educational environment. Insofar as possible, educators should involve parents in defining what good parenting means and in assessing themselves (see Gipps, 1999). Learning new forms of parent-child interaction and literacy practices profoundly shapes the parents’ and children’s identities, values, family routines, relationships, language, and membership in a community (Valdés, 1996). In one sense, all education involves change; the question is who decides which practices should (or should not) be changed and how such decisions are made. To that end, practitioners and scholars should consider whether assessment procedures and instruments foster inclusion (the ability to influence important decisions) and self-determination (the ability to decide whether to maintain or change specific literacy and parenting practices). We suggest family literacy participants have the right to identify "what aspects of ethnic [and, we would add, class-based] parenting they wish to retain and those they wish to relinquish in favor of the dominant culture’s parental values, attitudes, and practices" (García Coll et al., 1996, p. 1904). Educators can help parents navigate these decisions by using the PEP as a springboard for mutual learning. The Freirean-inspired approach recommended by Reyes and Torres (2007) demonstrates how collective reflection on the strengths and limitations of different child-rearing practices can assist both parents and educators in transforming their perspectives and roles. Powell and colleagues’ (2004) suggestions for respecting cultural differences in family literacy evaluations are relevant to this study. They recommend that staff and evaluators consider whether "desired parent outcomes are consistent with the cultural norms of the participants" (p. 560). For instance, if parents do not engage children in discussion during reading because they believe children should be "quiet and attentive" when adults read, then staff could decide to teach active participation anyway, while also acknowledging their cultural values, or teach alternative strategies to foster literacy development. These insights suggest educators should be willing to modify goals and instructional strategies, especially if a specific outcome measure (e.g., parental self-efficacy in book reading) is not culturally relevant and does not appear to support the end goal (e.g., increased frequency of book reading; Powell et al., 2004). Since the PEP allows staff to omit a subscale that does not apply to a specific family, practitioners could use this strategy to accommodate cultural differences. A limitation of this study, and CDA more broadly (Rogers et al., 2005), is that we did not examine how the PEP is used in programs. Future studies should explore how practitioners use, adapt, or resist this tool (e.g., using it in more collaborative ways). The divergent responses (anger, appreciation) to our analysis of the PEP at a national conference suggest it is a hotly contested topic in the field, especially in states where its use is mandatory. Further investigation is needed to elucidate how participants respond to the PEP’s content and the observation and assessment process; how its use influences staff-participant relationships and the ways they see each other and themselves; and how the adoption of PEP practices shapes educational success, interpersonal relations, and beliefs about parenting across socioeconomic and racial/ethnic groups. Several steps can be taken to enhance the PEP’s rigor and cultural appropriateness. It should be tested with economically and culturally diverse parents for (a) reliability and validity and (b) class, cultural, and gender bias, and then revised to mitigate any bias.9 For instance, if the PEP included pictures of men, educators and parents could more easily envision alternative gender roles. Evidence of reliability and validity should be included. Second, developers could elicit feedback from family literacy participants and experts—especially scholars of color—who conduct research on parenting and literacy development with nonmainstream families. Third, introductory materials could include information about competing perspectives on mainstream and nonmainstream child-rearing models vis-à-vis literacy and school success and help professionals and parents consider which of the PEP practices may have more or less influence on these domains. Finally, future studies should examine the methodological rigor and demographic composition of the PEP research base. Scale items that are based on correlational studies or that have limited evidence of effectiveness across SES, race/ethnicity, and immigrant status should be reconsidered. In conclusion, the PEP evokes broader ethical concerns about using measures that prescribe, assess, and seek to alter the parenting practices of poor and culturally diverse families (St. Clair & Sandlin, 2000; Valdés, 1996). We concur with García Coll and colleagues (1996) that "cultures and lifestyles different from the white middle-class mainstream are not pathological, deviant, or deficient relative to the mainstream but rather legitimate and valuable in their own right" (p. 1895). Instead, educators and researchers could view parenting and literacy practices in their socioeconomic, cultural, and historical contexts, interpreting them as reflections of deeply rooted cultural orientations and as adaptations to specific conditions such as scarce or abundant material resources. Based on this contextualized, historical understanding, assessment, like education itself, can become a "process of cross-cultural communication, negotiation, and mutual learning" (Fingeret, 1989, p. 14).
A previous version of this article was presented at the 2006 National Conference on Family Literacy. We wish to thank the editors and anonymous reviewers for their insightful comments and suggestions. All the views expressed in this article are those of the authors and do not necessarily represent those of the Goodling Institute.
ESTHER PRINS is an assistant professor in the Adult Education Program, Department of Learning and Performance Systems, Pennsylvania State University, 305B Keller Building, University Park, PA 16802; e-mail: esp150{at}psu.edu. She is also co-director of the Goodling Institute for Research in Family Literacy. Her research examines the social and cultural dimensions of adult and family literacy and participatory approaches to education and research, especially concerning how adult education reproduces or mitigates gender, racial, class, and cultural inequalities. BLAIRE WILLSON TOSO is a PhD candidate in the Adult Education Program, Department of Learning and Performance Systems, Pennsylvania State University, 305B Keller Building, University Park, PA 16802; e-mail: bwt121{at}psu.edu. Her research interests focus on issues of power and agency, particularly as they pertain to literacy, class, gender, and culture.
1 This four-component model is common in Even Start. However, family literacy initiatives include many other models and occur in a host of settings, including schools, libraries, and community-based organizations. Parents enrolled in Even Start must have a participating child aged 8 years or younger. This article focuses on programs that receive state or federal funding and that include parent education, as these are most likely to use the Parent Education Profile (PEP).
2 These states include Colorado, Georgia, Illinois, Iowa, Michigan, Montana, New Jersey, New York, North Dakota, South Carolina, and Washington.
3 These states include Alaska, Arizona, Florida, North Carolina, Pennsylvania, Tennessee, and Vermont, among others.
4 Virtually all family literacy programs affirm the importance of building on parents’ strengths and respecting their culture. Despite these stated intentions, an implicit "neo-deficit" discourse has emerged in family literacy (E. R. Auerbach, 1995; Dudley-Marling, 2007).
5 Lareau’s typology includes five categories. To create ideal types for this article, we omitted two of Lareau’s categories that were unrelated to the PEP ("Organization of Daily Life" and "Consequences"), changed "Intervention in Institutions" to "Parent Involvement in School," and added the category "Parent Involvement in Literacy Activities" based on the literature review.
6 We do not discuss Scale IV, Taking on the Parent Role, because it is implicit in the other literature we have reviewed and less relevant to literacy development. 7 As used here, care entails providing a safe, nurturing environment and meeting children’s basic needs. Lareau (2003, p. 66–67) explains: The limited economic resources available to working-class and poor families make getting children fed, clothed, sheltered and transported time-consuming, arduous labor. Parents tend to direct their efforts toward keeping children safe, enforcing discipline, and, when they deem it necessary, regulating their behavior in specific areas. 
8 In a separate study, a family literacy coordinator told the first author that many low-income participants had to travel 45 minutes or more to another county to see a doctor. Many did not have cars, and the county had limited public transportation. In such cases, children’s health problems are the result of economic circumstances rather than the parent’s lack of awareness.
9 We are grateful to a reviewer for suggesting this and several other recommendations here. Received for publication June 6, 2006. Revision received October 9, 2007. Accepted for publication January 23, 2008.
This version was published on September
1, 2008 American Educational Research Journal, Vol. 45, No. 3,
555-596 (2008)
 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
TOP
Abstract
The PEP