1. Introduction

For the past decade, United States federal, state, and district education policy change has been driven by a reform movement bent on improving schools and educational outcomes through standards-based accountability systems (No Child Left Behind, Common Core Standards) and market-like competitive pressures (charter, voucher, and open-enrollment programs, Race to the Top). Poll numbers suggest that the public, ‘increasingly supports efforts to create new schooling options, overhaul teacher pay and evaluation systems, and provide strong incentives for improvement’. Today, ‘ideas such as charter schools, performance pay, and consequential accountability are much more widely accepted than they were a decade ago’ (Hess, West and Petrilli, Citation2011, 58).

Though the education reform movement appears to have shaped public opinion on policy options, both policymakers and the public can attest that true renovation and results rest on initiatives making their way through a complicated system with multiple interests and players. If a disconnection exists between innovative solutions and citizen support for and knowledge of specific actions needed to successfully embody such reforms, progress will halt. In this study, utilizing data from an original survey of public school parents, I ask whether it makes sense for policymakers to put faith in education reforms that rely heavily on parent knowledge and initiative. Using No Child Left Behind (NCLB) (the federal education policy responsible for structuring the educational experience of families served by the nation’s 100,000 public schools) as a test case, I evaluate the depth and distribution of parents’ policy understanding and find that a lack of policy knowledge across diverse schooling environments and variance in parent awareness of and opinions on school experiences and policy may contribute to NCLB’s limited ability to meet its ultimate goals.

NCLB is not the most recent large-scale plan to address obstacles in preparing a twenty-first century workforce, (NCLB waivers, Race to the Top, and Common Core Standards are now layered atop NCLB). But understanding how the policy operates at full capacity – by design NCLB relies on administrators, teachers, and parents to contribute to policy change (parents of children in chronically failing schools may transfer their children to higher performing institutions, access educational support services or take part in improving and restructuring their child’s school) – allows us to weigh the consequences of policies structured to rely heavily on an informed citizenry as well as those that may not go far enough to incorporate citizens and school leaders – the Common Core of State Standards Initiative faced a great deal of backlash from these constituencies when attempts to avoid many of NCLB’s top-down design pitfalls failed to engage target populations. A deeper understanding of NCLB’s operations might also help us better anticipate the challenges associated with other potential policy mechanisms.

This study proceeds as follows. First I review the existing literature on citizen opinion and understanding of policy and highlight competing theories on how we might expect parents to understand and respond to NCLB. I then describe NCLB’s logic and mechanisms in greater detail. I follow with an introduction to the original survey utilized for an exploration of parents’ policy response, then report on parent policy knowledge, attitudes, and evaluations. I conclude by discussing the limitations of my approach and reflecting on what parents and policymakers can learn from a closer look at this policy in practice.

2. Existing research as motivation for exploration

For decades, political scientists and policy scholars have engaged in attempts to measure and describe the depth and significance of Americans’ understanding of politics and public policy. Scholars of civic understanding focus on the process by which citizens develop a solid policy comprehension, linking information and ‘self-interested’ or ‘enlightened’ policy evaluations. In their seminal work on Americans’ civic knowledge, Delli-Carpini and Keeter (Citation1996) argue that citizens’ ability to articulate their interests depends upon both the supply and quality of information available and the ability to understand the relationship between this information and the policy driving personal experience in this policy realm. In a related and growing body of work on policy feedback and political learning, scholars exploring public experiences with, perceptions of, and reactions to policies ranging from the G.I. Bill to welfare reform and Social Security suggest that policy designs and administration shape public wants, citizens’ self-perceptions, how citizens view and relate to one another, and how individuals understand and act toward political systems and institutions (Soss, Citation2000; Campbell Citation2002; Mettler, Citation2005). Taken together, these perspectives suggest that for most citizens, basic policy knowledge or awareness is a necessary first step on the way to enlightened opinions but this basic grasp does not suffice in isolation. A crucial next step in forming self-interested policy evaluations is an understanding of the channels through and ways in which a policy shapes one’s lived experience.

A look at the scholarship on citizen understanding and evaluation of tax policy further demonstrates how these bodies of work relate to one another. In 2001 and 2003, President George W. Bush signed two of the largest tax cuts in history. Many of the specific provisions in these tax reductions disproportionately benefited wealthy taxpayers, and yet, a majority of the American public supported these reforms. Two independent examinations of citizens’ response to these tax initiatives (Bartels Citation2005; Hacker and Pierson Citation2005) suggest that Americans’ ‘unenlightened’ policy evaluations were likely the result of citizens’ inability to connect their own economic position and existing economic inequalities with macro-level tax initiatives. That is, citizens did not see how these policies might shape their personal circumstances. In a more recent piece, Suzanne Mettler investigates levels of citizen support for the home mortgage interest tax deduction, the retirement savings contribution tax credit, and the earned income tax credit while purposely manipulating the type and amount of policy information to which survey respondents are exposed. She finds that a basic description of policy goals and mechanisms and information on the distribution of policy benefits by income group enables respondents to form more rational policy opinions than similar respondents not privy to such information (Mettler Citation2011). John Sides’s (Citation2010) experimental work on estate tax policy yields similar conclusions. Taken in concert, these studies suggest that policy design and administration influence how people understand and relate to policy (Campbell Citation2011). Tax policy and the tax code can be obscure and many other factors may influence evaluations of one’s economic position, but attempts to make tax information more accessible or relatable may influence citizens’ perceptions of tax reforms and future tax policy.

This tax policy scholarship does not necessarily lead to clear expectations for parents’ perceptions of education policy. On one hand, we might expect more clear and in-depth understandings of policy in the education realm simply because through their children’s schools, parents constantly confront policy in action. As I describe in greater detail in the following section, NCLB carries non-trivial consequences for all public school families. Under the policy every student in every school throughout the country sits for tests that align with state content standards and this alone likely shapes each family’s school experience. In Title I schools, those that by design and definition serve a significant population of low-income families, sanctions meant to address academic failure in a specific context may completely steer school culture. Here, teachers may begin focusing their instruction exclusively on tested subjects (reading, math and science), test preparation sessions may be held throughout the school year, and school staff may structure professional development around testing and accountability. If these schools fail or continue to fail to meet proficiency goals, these activities may go into overdrive. School assemblies and parent teacher conferences may focus exclusively on academic performance. Communication with parents may focus solely on reporting failure and alternative education options, students may transfer, and schools may be closed or restructured. Thus, in schools identified for improvement (SIFI) under NCLB, we might expect even greater awareness of the policy.

An alternative hypothesis, one that might explain the puzzle of low parent participation in policy sanctions and opportunities which I describe in greater detail in Section 3, comes from the literature exploring public opinion and understandings of various school choice programs. This body of work suggests that clear-cut policy understandings may be more of a hope than a reality. Schneider, Teske and Marschall’s (Citation2000) scholarship on school choice illuminates substantial variance in the depth of parent knowledge of education opportunities as well as the resources and incentives driving parents to make sense of their options. Moe’s (Citation2001) work highlights the correlation between resources, knowledge, expectations, and understanding of and views toward voucher programs. In a recent Education Next poll on charters, researchers note growing support for this form of choice but also emphasize the public’s very limited understanding of the services these public institutions provide (Howell, West, and Peterson Citation2010; Hess, West, and Petrilli Citation2011). Though all public school parents experience NCLB and failing school parents should, given the policy’s design, have more opportunities to confront and recognize the policy than their more resource-advantaged peers, a limited understanding of policy consequences and opportunities or inability to link experience with policy may preclude parents’ participation in school improvement. It would not be unreasonable for parents to take NCLB at face value and trust that the policy would ensure their children did well in school. Next, I present the layout and logic of NCLB to add nuance and context to these competing expectations.

3. The layout and logic of NCLB

In 2001, the Elementary and Secondary Education Act (ESEA) was amended and reauthorized as NCLB.Footnote¹ The amendment marked a notable shift toward high stakes accountability in federal education policy. Through the act, the federal government mandates the development of academic proficiency standards and standardized testing of all students in all schools to monitor whether schools and districts are making adequate yearly progress (AYP) toward these goals. Title I schools – designated such due to their high concentration of ‘disadvantaged’ students – continually failing to make expected academic progress may be ‘identified for improvement’ and subjected to a series of increasingly harsh policy sanctions: After two consecutive years of failure, schools must offer families an opportunity to transfer their children to a more successful public institution (school choice).Footnote² After a third consecutive year of failure, families must be offered the opportunity to pursue free supplemental education services (tutoring). Continued failure in subsequent years leads to the imposition of a series of school-level reforms designed to completely restructure a school (restructuring).

Several assumptions underlie NCLB’s policy prescriptions. The law first assumes that high stakes testing can accurately identify schools in need of improvement. The law also supposes that pre-established sanctions applied within schools serving the most ‘disadvantaged’ populations will lead to school-level improvements: School choice will offer families a way out of chronically failing institutions and provide schools with market-based incentives to improve. Supplemental education services such as tutoring, remediation, and enrichment activities will provide an opportunity for individual students to improve academic capacities. Changes brought about through corrective action and restructuring – new school leadership, curriculum, or staff, conversion to a charter school, or state control – will greatly alter school culture and increase achievement.Footnote³ Importantly, each of these options relies on parents’ policy participation. Finally, the law suggests that, taken together, testing and sequenced sanctions for disadvantaged populations will lead to improved academic outcomes for all (Forte Citation2010).

A decade after NCLB’s enactment, well over half of the nation’s schools are failing to make AYP toward state standards and a third of schools have been identified for improvement due to repeated failure to hit policy targets (Dillon Citation2010; Usher Citation2012).Footnote⁴ In addition, nearly half of Americans view the policy unfavorably and less than a quarter express positive policy opinions (Bushaw and McNee Citation2009). Further, few families have taken advantage of the policy’s purported opportunities through formally designated channels such as school choice and tutoring services.Footnote⁵

Researchers have shown where flaws in policy logic may contribute to unfavorable policy views and low take-up rates. Personal experiences in and connections to existing schools preclude families’ school transfer (Vernez et al. Citation2009) as do school leaders’ efforts to minimize NCLB’s impact on district finances and operations (after all, keeping children in their residentially determined schools keeps funding in place and reduces the potential administrative burden of high transfer rates) (Hess and Finn Citation2004; Krueger and Ziebarth Citation2004; Howell Citation2006). Variance in service delivery method, intensity, and mechanism may also correlate with confusion and low enrollment in supplemental education services (Forte Citation2010).

But in the previous section, I outlined an alternative and underexplored explanation for diminished public support for the policy and low participation rates in sanction-decreed opportunities (all this prior to Race to the Top, NCLB Waivers and the Common Cores State Standards Initiative). I suggested that the magnitude of response to the policy also hinges on parent policy awareness. Rather than NCLB directly structuring school experience and arming parents with the knowledge necessary to take advantage of opportunities for improvement, perhaps the policy is designed and delivered in such a way that parents are simply under-informed about this policy and its potential consequences and opportunities.Footnote⁶

An examination of federal, state, and district department of education websites suggests great variation in the language and content of such notifications (Lavery Citation2014). In addition to testing requirements and improvement sanctions, NCLB requires districts and schools to notify families about school progress (failing schools must present parents with information on opportunities for transfer, tutoring, and participation in restructuring).Footnote⁷ Some question whether or not districts follow through on this directive. Even if accessible information is widely disseminated, parents’ limited grasp of policy operations (demonstrated in existing work on charter and voucher programs) may be partially to blame for the policy’s failure to reach its touted potential (Schneider, Teske, and Marshall Citation2000; Moe Citation2001).Footnote⁸

In what follows I introduce the survey designed to assess parents’ policy knowledge, attitudes, and evaluations of NCLB and the following hypotheses:

4. Survey design

During the 2010–11 school year, all 56 public elementary schools in Seattle, Washington, were invited to participate in a study of ‘parents’ attitudes towards schools, education policy, and government’.Footnote⁹ The principals of 13 elementary schools across the district agreed to participate. Seven schools identifieid for improvement (SIFIs) under the policy and thus subject to direct, concentrated policy exposure and 6 institutions not marked for sanctions under NCLB (Non-SIFI) comprise my final sample. Four hundred eighty-four parents from these elementary schools were surveyed in-person at existing, evening, all-school events such as math nights, literacy nights, art fairs, and holiday concerts between October 2010 and April 2011.Footnote^10,Footnote¹¹

Much of the work on parent involvement relies on very large samples composed of families from a variety of school types and locations. Parents were recruited to this study in-person at existing all-school events in order to increase response rate and ensure an engaged and informed sample. That is, rather than aim for a representative sample in which the participating parent population mirrors the underlying student body, I recruited parents already drawn to school events. This approach guards against finding a lack of knowledge where a knowledge base truly exists (I have selected the sample most likely to have knowledge of the policy and therefore bias should be in favor of finding high rather than low knowledge) and helps establish a baseline from which we might speculate about the policy’s influence in other contexts. A small group of well-informed parents may be enough to move school quality or make this policy work. If engaged parents, willing to partake in a research study on school grounds have low knowledge, it is quite likely that the non-participating parent population is equally, or perhaps more, uninformed.

Seattle was chosen for its urban context, diversity, and the availability of schools across the income and improvement spectrum. Seattle is unremarkable in terms of both school performance (student test scores do not greatly surpass or undershoot those of children in other large urban school districts to the extent scores are comparable across state policy contexts) and district efforts to communicate policy to parents (the district choice and supplemental services template prepared by the Seattle Public Schools is quite similar to those prepared by other districts across the country) (The Eli and Edythe Broad Foundation Citation2008; U.S. Department of Education Citation2009). However, the city’s ethnic composition and residents’ relatively high income levels and liberal political leanings suggest that results may not generalize well to other locations, a potential limitation I discuss further in my concluding remarks.

Elementary schools were targeted for several reasons. First, NCLB requires that schools assess student learning in reading and math every year from grades 3 through 8 and once in grades 10 through 12. These testing requirements interact with grade span to encourage greater policy exposure at the elementary than middle or high school level. In addition, Seattle, like most districts, has more elementary schools than middle or high schools (as multiple elementary schools feed into schools at higher levels). This allows parents at the elementary school level more schooling options to choose from under policy sanctions. Finally, research suggests that parents of elementary school children are more likely than parents of older children to be intimately involved in the day-to-day schooling decisions of young children and participate in on-campus evening events (Jeynes Citation2005, Citation2007; Stewart Citation2008).

Surveys assess parents’ NCLB knowledge, attitudes toward their child’s education and the general state of schools, and evaluations of NCLB.Footnote¹² In the next section, I compare characteristics of the parent sample to the underlying student population (district and participating school averages).

5. Sample

Parents participating in this research study were asked about their familiarity with and attitudes toward NCLB. Surveys also provide information about civic behaviors and demographic characteristics such as parent’s partisan affiliation, gender, age, educational background, employment status, race/ethnicity, native language, the number of children present in each household, and total household income. As noted in the previous section, given my purposive sampling method, I expect the parent sample to differ in predictable ways (greater knowledge, higher SES, less diversity) from the underlying student population. In Table 1, I consider sample and student population comparisons for key demographic characteristics also drawing out distinctions between populations with different levels of policy exposure – parents whose children’s schools have been identified for improvement and are thus likely to have concentrated policy experience (SIFI) and those for whom NCLB is likely to represent little more than an annual, concentrated exam period (non-SIFI).

The top portion of Table 1 displays demographic and socioeconomic summary statistics for several relevant student and parent populations. Column 1 presents demographic background information for all district students from the 2010–11 school year. Column 2 displays summary statistics for all district SIFI under NCLB.Footnote¹³ Column 3 shows sample school characteristics. A comparison of columns 1 and 3 suggests that the sample school student population is fairly representative of the district’s student population. English language learners, Hispanics, those who qualify for free or reduced price lunch and White students are slightly overrepresented in the sample, while Black and Asian populations are slightly underrepresented. Column 4 displays summary statistics for sample parents.Footnote¹⁴ Note that because parents are not eligible for free or reduced price lunch, I use college education, employment status, and income as indicators of adult socioeconomic status.Footnote¹⁵ Comparing columns 3 and 4 makes clear that racially and socioeconomically advantaged groups are overrepresented in the sample. For a better understanding of how these characteristics are distributed across school type, I draw attention to columns 5 and 7. This comparison demonstrates that the parents sampled at SIFI henceforth SIFI parents (Column 5), are more racially and economically diverse than parents sampled at non-SIFI schools (Column 7) (as we might expect given the Title I status of these institutions). Comparing respondents with underlying student populations at sample schools of each type (Column 6 isolates the student population from SIFI, while Column 8 presents student averages for non-sanctioned schools) further demonstrates the sample population’s relative advantage. Respondent parents across institution type are more likely to be White, less likely to speak English as a second language, more likely to hold a college degree or have full- or part-time employment, and less likely to earn less than $20,000 per year than we might expect if parents share background characteristics with the students at these schools.

The lower portion of Table 1 displays information on parents’ gender, age, partisan affiliation (7-point scale with 1 indicating a respondent identifies as a strong Democrat and 7 as a strong Republican) as well as three measures of civic activity known to correlate with school-related outcomes – PTA membership status, religious membership status, and an indicator of political activity (4-point scale – one point each for contacting a political official, working on a campaign, contributing money to a campaign, and registering to vote).Footnote¹⁶ The information displayed across columns 4, 5, and 7 demonstrates that fewer fathers completed the survey at SIFI than at more successful schools. Parents at these schools tended to be younger and more politically conservative than their peers whose children attend higher performing institutions.Footnote¹⁷ Finally, parents from failing school communities are less likely to be PTA members or participate in political activities and more likely to identify as members of religious communities than their peers representing more successful schools.

Taken together, the summary statistics provided in Table 1 demonstrate that, as expected, the sample parent population is more socioeconomically advantaged and less racially and ethnically diverse than the underlying school population. Importantly, given the hypothesis that policy exposure will breed familiarity, the sample preserves expected differences between SIFI and non-SIFI respondents. That is, respondent parents representing failing schools are less likely to be White, more likely to speak English as a second language, less likely to hold a college degree or have a full-time job, and more likely to earn less than $20,000 per year than parents from higher performing schools.

6. Knowledge

The first step in assessing whether information deficiencies exist in parents’ understanding of NCLB and shape their policy-related attitudes and behaviors is gauging their policy familiarity. Howell (Citation2006) conducted a similar survey of parents’ NCLB knowledge shortly after sanction enforcement began. Following Howell, I assess the depths of parents’ policy knowledge drawing contrast between parents served by schools of different policy status to highlight potential consequences of policy design. When asked via survey at a school event, most Seattle public school parent participants – over 70% as shown in the ‘% All Parents’ column of the first panel of Table 2 – expressed a basic level of familiarity with NCLB. This is not surprising considering this has been the big banner federal education policy for nearly a decade and parents with young children may not have experience with any other education policy regime. The second and third columns of Table 2 offer a more nuanced view of policy familiarity, comparing responses from parents of children in SIFI for continued failure to make AYP (2 or more consecutive years) with those from parents of children at higher performing public schools (non-SIFI). Here, we see that SIFI parents are slightly more likely to have heard of NCLB.Footnote^18,Footnote¹⁹

A high proportion of parents from all school types sampled profess NCLB awareness. But for various reasons (default reactions, social desirability), from a policy perspective the extent of this knowledge matters.Footnote²⁰ If parents have heard of NCLB but cannot speak to the relationship between NCLB and their children’s education, then professed knowledge carries little meaning. Parents must know about the act’s requirements for schools in general, but more importantly, they must be familiar with policy tenets that may influence their child’s school (in the case of parents of children in SIFI, the availability of school choice, supplemental education services, and parental involvement in school improvement) if they are to take advantage of policy opportunities or otherwise express policy concerns (Howell Citation2006).

The structure of the survey designed to assess parent knowledge and attitudes is conditional. That is, if a parent responded that he or she was not familiar with NCLB, he or she was asked to skip directly to the survey section on opinions on his or her child’s elementary school.Footnote²¹ Those expressing some policy familiarity were first asked to attempt to identify their child’s school’s policy status.

The second panel of results in Table 2 demonstrates that fewer than half of parents across school type claim school-specific policy knowledge. Though the percentages differ slightly by school type with parents of children attending higher performing schools now professing greater policy knowledge than their peers whose children attend SIFI, these differences are again quite small. If we take advantage of the information provided in parents’ responses to the questions in both of the first two panels adding the number of parents who expressed no familiarity with NCLB to the number of respondents who admit they do not know their child’s school’s policy status (118 + 156), it becomes clear that in fact, nearly 70% of all parents are NOT familiar enough with the policy to confidently assess an indicator of school quality essential for further action.

The most interesting and consequential knowledge-related results here are reported in the third panel of Table 2. To gauge the extent of parents’ policy specific recall, I directed parents expressing some confidence in their policy knowledge to identify that school’s status given the following options: (1) This school made AYP based on last year’s test scores; (2) this school failed to make AYP based on last year’s test scores but has not yet entered school improvement status; (3) this school failed to make AYP based on last year’s test scores and has been identified for improvement; or (4) I do not know this elementary school’s status under the NCLB Act.

Though at first glance, an examination of the last panel of Table 3 suggests that across school type, a majority of parents get school status right, by the numbers, less than 20% of school improvement parents correctly identified their child’s school as a chronically failing institution formally identified by the state for repeated failure to meet academic targets.Footnote²² That is, if we take the number of parents correctly identifying their child’s school’s status and divide it by the total number of parent respondents for that school type (as shown in the final row of Table 2), the picture provided is drastically different. We can assume that this level of familiarity would be required for parents to make decisions on switching schools or obtaining supplemental education services. This inability to identify status linked to opportunity occurs even though the federal government has mandated that districts and schools send school improvement status letters home to those parents whose children attend underperforming schools. Still more discouraging is the reminder that these results were collected based on surveys that parents chose to fill out at large-scale school events they chose to attend. These parents are very likely more attentive to school-related issues than many of their same-school peers (Howell Citation2006).

The aggregate and comparative analyses above highlight differences in policy awareness and understanding for the parent populations examined, but it is likely that children enrolled in schools of varying performance levels come from different kinds of families that have different educational needs, capacities, and experiences (Howell Citation2006). An exploration of the relationship between these factors and policy knowledge helps to assess policy’s independent influence on parent awareness as well as the influence of other factors that may contribute to lower levels of policy understanding in failing school communities.

To determine how parent demographic characteristics and school status independently influence policy familiarity and more specific context knowledge, I regress each input on these knowledge-related outputs. In the multiple regression model displayed in Table 3, school status does not appear to assert an independent influence on parent policy awareness (though findings are in the expected direction and this may not be surprising considering the high proportion of parents across school type who claim familiarity with NCLB and the controls included in the model). However, the first column of Table 3 indicates that several background characteristics – parent income, PTA membership, religious membership, and the level of parents’ political participation – may influence parent policy familiarity.Footnote²³ PTA members and more politically active parents are more likely to be aware of NCLB than their peers. Results show a negative relationship between both income and religious membership and policy awareness.Footnote²⁴ The income result may appear surprising if one only considers the relationship between knowledge and education more generally. But in context, the result is less of an aberration as under NCLB, high-income parents are less likely to have children attending SIFI (only Title 1 schools, by definition schools with a significant number of high poverty students, are identified for improvement) and are therefore less likely to receive targeted policy exposure and information.

The second column of Table 3 presents the results of a logistic regression predicting a correct response to the school status question while controlling for school status and this same host of demographic variables.Footnote²⁵ When all of these factors are accounted for, school status again fails to exert a statistically significant independent influence on more context-specific knowledge (though the sign of the coefficient again indicates that parents of children in SIFI may be less, not more, likely to know school status than their peers at higher performing institutions). In this second model, two background controls correlate significantly with parents’ ability to correctly identify their child’s school’s status. Parents employed full- or part-time are less likely to correctly identify school status than their counterparts (unemployed or retired parents), perhaps because work leaves these parents with less time at schools or devoted to school-related activities.Footnote²⁶ Curiously, income, PTA membership, and political activity no longer reach statistical significance.

Taken together, these multiple regression models suggest that across school type, non-working, politically active, PTA parents are more likely to know their children’s school’s status than their peers. Given what we know about the work status and political involvement of the typical population clustered in failing schools by the cruel irony of policy design (sanctions only apply to low-income populations with characteristics most predisposed to low-testing outcomes), the lack of policy understanding in context should not be surprising (see Howell Citation2006). But, for those who hope to increase parent involvement in closing achievement gaps, such findings may be concerning.

7. Attitudes

Though when pressed, parents appear to know little about this policy or its greatest influence on schools (identifying and sanctioning failing institutions which may lead to stark changes in school culture), by virtue of parenting children in public schools, they are likely to hold opinions on experiences and outcomes at their child’s school, school sanctions, and the general state of schools. Even if policy does not appear to have a direct and independent influence on knowledge, does NCLB shape parents’ opinions of their child’s school or the public schools writ large? Do attitudes differ by school status and exposure to sanctions?

The parent participation survey asked about parent satisfaction with the overall educational experience and resources provided at their elementary school, whether or not parents believed the school or teachers focused too much on testing and standards, and whether or not parents believed the teachers in the school were qualified to perform their jobs. Parents were also asked to evaluate the relationships between themselves and teachers, administrators and other parents, and how involved they felt with the school community.Footnote²⁷ An examination of the first row of Table 4 demonstrates the relationships between school status and each outcome.Footnote²⁸ The coefficient on the variable capturing the relationship between school status and satisfaction with a child’s school experience fails to reach statistical significance, suggesting that parents’ school satisfaction is unlikely to be independently influenced by school status. Parents with more education and income and those who belong to a religious community report greater satisfaction with the educational experience provided at their child’s school than their counterparts. This may be because those with more education and higher incomes tend to live in more affluent areas which also tend to have better resourced schools.

Next, I focus on parent attitudes toward testing and teacher quality. Again, school status fails to reach statistical significance.Footnote²⁹ Education and income once more correlate significantly with parent attitudes, however, here, due to the phrasing of the survey questions, (Does this school focus too much on testing? Do the teachers in this school teach to standardized tests?) a positive relationship indicates less satisfaction with these outcomes. More educated parents are more likely to disapprove of the focus on testing and standards than their less educated peers and higher income parents are more likely than their lower income peers to condone the focus on assessments. This is an interesting finding as the coefficients on these variables often run in the same direction. English language learners, like more educated parents, resent the focus on testing in their schools.

The third column of Table 4 considers the relationship between school status and parent relationships with each other and the school staff and community at their child’s school. Here, school status is a significant predictor of parent attitudes. Parents in SIFI are less likely to get along with the teachers and administrators in the school, feel that parents and community members are actively involved, feel that school personnel reach out to them, feel that parents communicate regularly with one another, or feel involved with the educational or social events taking place in the school. A large literature on the influence of parent involvement and social capital on school outcomes suggests this may spell trouble for true school improvement (Coleman and Hoffer Citation1987; Braatz and Putnam, Citation1996). If parents do not trust in and feel a part of the school community or communicate with teachers, administrators, and each other and feel welcomed into school improvement conversations, they will not engage in building a school improvement program that best serves their needs and interests. Several control variables in this model also reach statistical significance. PTA members are significantly more likely to positively evaluate school engagement and relationships.

The fourth column of Table 4 reports parent views on school reform and sanctions. School improvement parents’ opinions on these outcomes are especially important as these are the formal channels through which failing school families can act when schools reach school improvement status. I find that school improvement parents are less likely than their peers in higher performing institutions to approve of the reforms and sanctions the policy prescribes, though differences do not reach traditional levels of statistical significance.Footnote³⁰ This is concerning because these sanctions are the primary channel through which parents can register policy discontent. If the parents who must actually pursue opportunities available through sanctions find these channels unpalatable, policymakers hoping to truly raise achievement in chronically failing schools might explore alternatives this population would find more satisfactory.Footnote³¹

The final column of Table 5 considers parents’ satisfaction with the general state of the country’s schools. Here, there are no statistically significant differences in evaluation between parents whose children attend schools of different status. However, White and Republican parents are significantly more likely to report less positive evaluations of the nation’s schools than their comparison groups, while English language learners and those with more education provide more positive evaluations than their counterparts.

Taken together, the models presented in Table 4 demonstrate that policy status influences parents’ opinions on some issues of consequence to families and school reformers. On average, families in SIFI are less satisfied with school relationships than their peers, and though sanction results do not reach conventional levels of statistical significance, these findings suggest that perhaps the types of opportunities made available through sanctions are not well matched to the parent population granted access through this policy. Because NCLB relies on parents to participate in school improvement, both school relationships and opportunities well matched to the needs and preferences of failing school families are essential precursors to reform.

8. Evaluation

The literature on citizen knowledge and political learning suggests that citizens form ‘enlightened’ or ‘self-interested’ policy preferences in response to an in-depth understanding of policy in relation to their individual context. Do parents with different levels of policy exposure (those in SIFI vs. those in non-sanctioned institutions) offer different policy evaluations? Do evaluations appear to align with reported attitudes suggesting a connection between experience and evaluation?

Table 5 shows the results of a model predicting parent attitudes toward NCLB with improvement status and the reported demographic variables used in each of the previous models.Footnote³² Results indicate that parents of children in SIFI have significantly more positive attitudes toward the policy and policy progress than their peers in more successful institutions. English language learners also rated the policy more positively than their peers. Additional significant findings further validate the model. Though the measures of partisanship (Republican), school involvement (PTA member), and political activity fail to reach statistical significance, all run in the direction one might predict. Republicans are more likely to approve of a Bush-era education policy than their Democratic counterparts and those who are involved at school or in politics are less likely to approve of the policy, which may actually explain their involvement.

Policy designers of the past decade have based their policy prescriptions on the assumption that testing and sanctions would change school climates and improve student outcomes. We know that NCLB influences school experiences: Testing occurs and sanctions must be at least superficially advertised to families in failing schools. Here, I demonstrate that families with different policy sanctioned school experiences report quite different evaluations of this experience. Given that earlier results suggest that parents of children in failing schools lack context-specific policy knowledge, it is possible that this lack of knowledge inspires a disconnection between evaluations of policy-structured experiences and evaluations of responsible policy. That is, without drawing attention to the relationship between displeasing policy tenets and policy prescriptions, it is likely that many parents buy into policy rhetoric, believing the concept within the name “NCLB”, and trusting in public institutions to ensure their children’s success. I close with a review and discussion of these findings.

9. Discussion & conclusions

In this article, I use original survey data to explore the depth and distribution of parents’ understanding of NCLB, the federal policy responsible for structuring the everyday experience of 50 million public school children and their families and putting education reform movement initiatives into practice. I report several important findings: (1) Most parents claim familiarity with NCLB regardless of the type of school their child attends. Yet few parents accurately understand how key provisions interact with their children’s school context to structure outcomes and provide opportunity. Most parents do not know whether or not their child’s school falls on the state’s list of SIFI and parents whose children attend failing institutions are even less likely than their peers to correctly identify school status; (2) Parents’ opinions on schools differ somewhat according to their child’s school’s policy status. Parents of children in SIFI are more likely to provide negative evaluations of certain aspects of their child’s educational experience (school-based relationships) and policy reform and sanctions than their peers whose children attend higher performing public schools, and; (3) Policy evaluations differ by school status in a manner that suggests that parents whose children attend a school identified for improvement may lack an ‘enlightened’ understanding of how NCLB shapes their school experience. SIFI parents provide more negative evaluations of their experiences with schools and offer greater disapproval of policy sanctions than their peers with much less concentrated policy experience, and yet, this same group offers more favorable evaluations of NCLB.

Before I turn to a discussion of the implications these findings hold for future research and policy, two potential limitations of my approach warrant additional mention. As noted in Section 4, Seattle is similar to other urban locales in terms of school performance and district efforts to communicate with parents about school status. This is an important strength of the sample. But several demographic issues specific to Seattle may influence this study’s generalizability to other urban school districts. Seattle is consistently ranked as one of the most literate of America’s largest cities, and Census Bureau data indicate that Seattle has a higher percentage of college graduates than any other major American metropolis (United States Census Bureau Citation2008). Because this survey focuses on views toward education, results may differ if questions were fielded in a place with lower aggregate educational attainment. Seattle also has higher per capita income, larger White and Asian populations, a larger recent immigrant population, and more liberal/progressive and less religious residents than the average American city (United States Census Bureau Citation2008). These factors may also bias results in predictable ways.

A second concern relates to selection bias. All elementary school principals in Seattle were invited to participate in this study. The principals who signed on may be more eager to please the district or appear transparent, or more confident in their approach to school leadership than those who chose not to participate. A principal’s leadership style and decisions may independently influence the school experience of families in participating schools in ways that make the attribution of results to policy status uncertain. Given this, all results should be interpreted in context.

This study may hold several important lessons for scholars, policymakers, and practitioners. NCLB relies on citizen action. Parents must request supplemental education services, transfer their children away from chronically failing schools, or show up to restructuring meetings to voice their concerns. I find low levels of concrete policy understanding across parent populations when relying on a potentially advantaged parent sample. As William Howell pointed out shortly after NCLB sanctions were first enforced, this may be partially the result of overreliance on district officials to inform parents of policy opportunities which may negatively impact district relationships, finances, and capacity. But even if districts make concentrated efforts to reach out to needy families, additional and alternative efforts may be necessary to inform citizens of policy-related responsibilities (Lavery Citation2014). NCLB Waivers and Race to the Top aim to address some of NCLB’s more visible failures, yet neither policy appears to address the potential mismatch between families’ needs, views, and concerns and the sanctions and opportunities thrust on and made available to failing school communities. Until policy designers take note of this important obstacle to implementation, it seems unwise to expect transformational change in failing institutions.

As the paragraph above suggests, one might read these results as motivation to further future policy initiatives tailored to encourage parents’ advocacy on their own behalf (this is the approach inherent in California’s 2010 Parent Empowerment Act). But existing scholarship offers limited assurance that direct and straightforward information delivery can in fact influence policy knowledge or behavior (Hastings and Weinstein Citation2008; Lavery Citation2014). Additional research on the type, tenor, and delivery of information policymakers provide to partners in improvement is necessary if a policy’s success depends on transparency and accessibility. In the meanwhile, researchers and policymakers might do well to work together on an alternative approach to policy evaluation and revision. Data on school-based attitudes and policy evaluations suggest that citizens have trouble linking experience to policy. Therefore, those most interested in a school reform as an ultimate end may want to alter policy goals, designs, and mechanisms to align with parents’ evaluation of school experiences, rather than require that parents uncover the link between these experiences and policy. Policymakers could tweak policy to rely less on parental action. Such an approach would directly address the no-longer ironic finding that families concentrated in disadvantaged schools either lack access to necessary and detailed policy information or the resources necessary to understand the relationship between information on policy sanctions and opportunities for their children and may lead to more significant and appropriate policy adaptations (Howell Citation2006). Perhaps the ends will justify the means.

Notes on contributor

Lesley E. Lavery, PhD, is an assistant professor of Political Science at Macalester College. She is also an affiliate of the Center for Education Data and Research at the University of Washington, Bothell. Her research focuses on education policy, public and social policy, and political behavior.