Feature Article - Poverty and Inner-City Education

Ben Levin, University of Manitoba1

At the time of writing, Ben Levin was a professor in the Faculty of Education at the University of Manitoba.

Numerous studies have consistently shown that the strongest single predictor of educational achievement and attainment is the socio-economic status (SES) of the student’s family. Thus, children in inner-city areas, characterized by low SES, face significant obstacles to educational success, with concomitant threats to other important life outcomes: employment, income, health, longevity, and civic participation.

Concern about inner-city education is not new, yet this long-standing concern hides a range of ideas about the nature, causes, and potential remedies for the problems of education in our cities.

Causes of and Remedies for Poverty

Thinking about the nature and causes of poverty tends to fall into one of two broad camps. One sees poverty as a shortcoming of individuals who will not or cannot do what is required to maintain a reasonable life. In this view, poverty is often a moral failing, and measures to provide extra supports to poor people likely encourage slothfulness and a lack of initiative, thus making the problem worse. A second view holds broadly that poverty arises mainly from systematic inequities in the economy and society, and is largely the result of forces beyond the control of individuals, such as the lack of work, low wages, or discrimination. Alleviating poverty is, therefore, seen as requiring either or both (there is disagreement on this point) extra supports to families and individuals, or structural changes in the economy and society. Public opinion and government policy tend to oscillate between these two positions.

Ideas about solutions to, or strategies for, addressing issues of poverty are largely shaped by people’s ideas about causes. The range of solutions or strategies falls into four broad categories. One set of policies provides some basic level of income and other supports for poor people through social assistance, minimum wages, tax credits, or a range of other vehicles. Many of these policies have little or no ameliorative element. They are simply designed to try and make life bearable, or at least possible, for people who are otherwise destitute.

Another set of strategies is intended to support greater individual effort by poor people. Examples include both positive incentives (e.g., support for training and education, savings plans, targeted programs) and negative incentives (e.g., denial of various benefits for non-employment). These strategies are based on a model of poverty as being largely the result of individual problems or incapacities.

A third, less frequently used set of strategies aims less at individuals and more at neighbourhoods or communities. Initiatives in this category derive from the belief that poverty is largely a result of structural factors rather than individual choices, and must therefore be addressed by trying to change structures, especially at the local level. Examples include economic development initiatives, neighbourhood action groups, school improvement, and employment creation.

Finally, anti-poverty strategies may be developed at the macro-social level. If the basic theory is that overall prosperity is the best way to reduce poverty – an idea that has been very powerful in mainstream thinking – then efforts to reduce unemployment or stimulate overall economic growth, even including tax cuts, could be seen as the primary policy levers to be used.

The Role of Education in Anti-Poverty Efforts

While more education is clearly linked to desirable life outcomes, these outcomes vary a great deal, even among those with similar levels of education. The last 20 years have shown that countries can simultaneously have increasing levels of education and increasing levels of economic and social inequality.

Still, efforts to encourage poor individuals to improve their situation often have the acquisition of more education as central to opening up other opportunities (though in practice many poverty support programs put substantial barriers in the way of people who want to return to education). Initiatives at the community level often involve efforts to improve education, whether in the schools or through extension into adult education or early childhood development. However, some community advocates see schools as part of the establishment that oppresses the poor, and are suspicious of school efforts given the evident failure of existing educational provisions to make large changes in patterns of inequality. Other research shows the ways in which schools sometimes support or even exacerbate social inequalities (e.g., Natriello et al., 1990).

At the macro-policy level, the importance of education to overall economic growth and prosperity has become an article of faith among governments and other agencies. Whether the rhetorical commitment is matched by active educational policy is quite another matter. Many jurisdictions, including Canadian provinces, have actually reduced the level and proportion of public resources flowing to public schools and post-secondary education, while early childhood development and adult education remain marginal to the overall educational enterprise in terms of policy, institutional structures, and resources.

Inequities in Schooling: The Achievement Gap

How big a problem do we have in urban education in Canada? Levels of child poverty in Canada remain very high by international standards (Bradbury and Jäntti, 2001). The Campaign 2000 Coalition reported that despite the commitment by Parliament to end child poverty, between 1989 and 1999 the number of poor children in Canada rose by 39%, including a large increase in families with at least one person in full-time employment.

In an analysis of students at-risk in Canada, Levin (2004) concluded that several different measures – poverty, high school dropouts, and the vulnerability index created from the National Longitudinal Study of Children and Youth (NLSCY) – all led to a similar conclusion that about 25% of students in Canada has some notable level of risk with about 10% to 15% likely to face serious challenges. However, these risk factors are not evenly distributed geographically. In high-poverty communities, the levels of challenge could be much higher. It is well known that Aboriginal school completion rates remain much lower than those for other Canadians.

Evidence on student outcomes shows that gaps in achievement remain substantial in Canada, and socioeconomic status is an important factor in those gaps. Willms’ (2003) analysis of data from the NLSCY showed that socio-economic status has substantial effects on children’s educational skills, and these effects are stronger in communities with high overall levels of poverty. The Program for International Student Assessment (PISA)2 study also found large gaps between the top and bottom performing 15 year olds in Canada. Moreover, these gaps grow over time. Data from the Manitoba Centre for Health Policy (MCHP) showed huge disparities in results on the Grade 12 provincial examinations for students coming from the poorest parts of Winnipeg (Brownell et al., 2004).

An important issue is whether schools in poor neighbourhoods suffer systematically, as they do in many places in the United States, from poorer facilities, less qualified teachers, and generally weaker levels of resourcing. Impressionistic evidence would suggest this is not the case in Canada, because financing tends to be more equal across school districts.

How Much Can Schools Do?

Schools are often seen as vital contributors to efforts to reduce poverty and achievement gaps. Some argue that it is unreasonable to expect schools to overcome, to any significant extent, the powerful social and economic forces that create and sustain inequality. Another line of argument is that the target and demand must be for schools to do much more than they have in the past to equalize opportunities for poor children. Any other position is defeatist and allows schools to make excuses for poor performance. Advocates of higher expectations for schools point out that school results vary enormously even within similar SES communities, suggesting some schools are managing to produce much better outcomes.

How much improvement in student outcomes might reasonably be expected given sustained efforts by schools in high poverty areas? Teddlie and Reynolds (2001) suggested that 10% to 15% of the variation in pupil outcomes is attributable to all the things schools do (or do not do). However, this does not mean we can necessarily improve outcomes by this amount through changing school practices. Their estimate might be close to a ceiling on what is possible by way of school improvements.

Various studies examined schools that seem to be doing better than expected. These studies typically noted features of these successful schools, such as a common vision, high expectations for students, strong leadership focused on student success, use of data to guide planning, and strong ties with the community. However, schools making the greatest gains are, by definition, untypical. A reading of the literature on various forms of deliberate school improvement work indicates gains in student outcomes are quite modest in most schools, even after extensive efforts over several years. Moreover, the improvement process will often be more difficult in schools in challenging circumstances, because such schools face higher staff and pupil turnover, lower levels of overt parent support, and a history of lack of success. Recent US evidence (Bracey, 2004) showed that sustained improvement over time in high-poverty schools is rare, despite claims by studies of exceptional schools.

Conceptualizing Strategies to Address High Levels of Poverty in Schools

Over the years, schools and school systems adopted a variety of measures to address equity concerns. Indeed, there is very little in the current lexicon of school supports that was not advocated in the 1960s and early 1970s, if not well before that. The measures taken can be thought of as falling into three categories. The first includes in-school changes in programming or supports, such as feeding or clothing students, mounting special programs, greater outreach to parents, and whole-school reforms in teaching and learning. The second category includes system measures designed to provide positive or negative incentives for better performance, such as testing of students and teachers, choice of schools, and extra funding or other financial incentives tied to performance. The third category includes measures to extend the scope of schooling into other areas, such as early childhood, adult education, and community economic development.

This kind of general categorization understates the importance of local context. The nature of poverty and the kinds of challenges it presents to schools vary from one community to another across Canada. These differences suggest that strategies and approaches by the schools also need to be different.

What do we Know about Efforts by Schools?

In all categories, most efforts at any level to address needs in high-poverty schools have been short term and on the margins of the basic enterprise of schooling. The same proposals reappear regularly, but do not seem to affect mainstream schooling processes. Ideas from 40 years ago, such as integrated services, community development, or parental involvement, are once again being advocated.

We do not have a good base of evidence as to the impact of these various measures. For example, there is very little evidence on the impact of initiatives, such as feeding children or providing additional counselling. It seems reasonable to think there would be effects from ensuring children eat properly, are dressed warmly, or have access to programs that give some recognition to their language or culture. The limitation of all these programs is that they do not directly address academic achievement so, while probably important, they are unlikely to make the necessary difference in outcomes.

Another common response of schools to poverty has been the development of a whole range of special or targeted programs. Many such programs create separate classes or teaching arrangements for children and youth seen to be at risk. These approaches are adopted largely because they are the least disruptive to existing arrangements and, in many cases, are supported by specific funding programs.

Once again the empirical evidence on outcomes is weak but, on the whole, research evidence does not support segregated or withdrawal programs as effective interventions (Knapp et al., 1995).

Current thinking on effective school change tends to focus on basic approaches to teaching and learning, and to student-teacher relationships. Some researchers think changes in regular classroom practices are key.

[A]chievement in school is made more likely when: teachers teach for mastery; curricula are relevant to students’ present and future needs; authentic assessment practices are used; democratic classrooms are created where students contribute to the rule-making and governance; rational, humane and consistent behaviour management techniques are adopted; teachers are warm, approachable, fair and supportive and a range of ways of being successful are made available to students (Howard et al., 1999, p. 316).

In the PISA 2000 study, more motivated students did better regardless of their background or the school’s teaching.

Changing mainstream programs is a relatively new approach. The more popular changes include the creation of teacher advisory systems in secondary schools to try to ensure stronger adult-student connections and the use of new approaches to early literacy in elementary schools including Reading Recovery and balanced literacy. Schools may also need to distribute resources differently in high-need communities. However, these changes in the everyday nature of school activities and, especially, in the work of teachers are not easy to do and may challenge current practices. Many studies evaluating large-scale efforts to improve schooling for at-risk students show how hard such changes are to make and sustain.

Some of what are known as whole school reform models offer evidence of positive effects, such as Success for All,3 the School Development Program, or Accelerated Schools, but are still controversial, partly because of their reliance on test score results as the main criterion, and partly because the number of truly independent evaluations of most programs remains small. A massive effort in England through the National Literacy Strategy and National Numeracy Strategy included additional funding, supports, incentives, and accountability pressures. Test scores improved rapidly in the first few years, but since then have stayed flat at levels below the targets, raising concerns about whether these strategies could raise achievement to desired levels (Earl et al., 2003a).

Extending the Scope of Schooling

Almost all the literature on education and poverty assumes that anti-poverty work should take place primarily in schools. Yet the discussion of the limits of school improvement raises the possibility that schools may not be the most efficacious sites for efforts to improve and equalize educational outcomes. The most significant impact on student outcomes may come from other measures, such as reducing lead poisoning from substandard housing, reducing the incidence of fetal alcohol syndrome, or ensuring that children in isolated communities get an adequate diet.

There is at least some awareness in many inner-city settings of the need to take a broader approach as evidenced by efforts over many years to include early childhood, adult education, parental involvement, and community development in school programs.

Many intervention programs in inner cities have focused on the preschool years. Although there is a strong belief that early intervention will create significant improvements in outcomes, the evidence is actually not so straightforward (Bradley and Whiteside- Mansell, 1997). Preschool appears to be an important area for action, but not sufficient in itself. Canada lags behind many other countries in its effort and infrastructure to support very young children. On a per child per year basis, schools in Canada get about six times more public funding than do various programs for very young children. Moreover, the infrastructure that does exist for early childhood in Canada and many other countries is largely at ages three to five, whereas both need and impact appear greater when children are even younger.

Adult education is another promising area in high-poverty communities, especially for adults with low levels of formal education, because improving the skills of adults usually also improves the lives of the children in their care. The development of a network of adult learning centres in Manitoba illustrates the extent to which adult participation is connected to desires to assist their children in being more successful in school. Although many school systems in Canada operate adult learning programs, support for such programs by provinces and school systems tends to be very modest, and programs often lack a solid base in policy that would promote wider accessibility and high quality.

Parental involvement has been recognized increasingly as an important element in building school success. Educators may tend to blame parents for children’s problems and seek to counteract the influence of parents. The inability of such an approach to work should be evident based on its abject failure in Aboriginal education in Canada. The stresses in children’s lives are intimately connected to the conditions of their parents, which means schools have no choice but to try to build strong connections with parents. We have learned a great deal more in recent years about how to build these connections, including how to work effectively with parents who may themselves be struggling. However, with very few exceptions, the development of parent involvement in schools in Canada is a poorly resourced, add-on activity that is well down on the priority list of schools. Combining early childhood and adult education leads to what are called two generation programs (Dunst and Trivette, 1997). The US National Governors’ Association,4 for example, advocates family literacy programs that involve parental support for children’s literacy and school support for parent learning.

Schools may need to look at ways to participate in larger efforts to build strong communities by, for example, hiring local people, purchasing goods and services locally, and supporting community efforts to improve housing or create jobs. Such strategies have considerable promise, but are not often attempted for a range of reasons.

Another often-advocated strategy is the integration of various educational and social services through the school to make services more available to those requiring them. Efforts to implement integrated service models go back to the beginnings of public schools. While the idea has an intuitive appeal, like many institutional changes, it is very hard to do effectively and sustain over time (Volpe, 2000).

One should avoid developing an excessively pessimistic conclusion from all this information. In particular, the evidence reviewed does not suggest that spending money on efforts to improve high-poverty schools is a poor strategy. As one group of researchers put it:

The fact is that there is virtually no evidence of the consequences of colossal increases in the educational resources to which disadvantaged children are exposed, because this strategy has never been systematically adopted (Natriello et al., 1990, p.192).


Everything we know about poverty and education tells us that it has been very difficult to make any lasting change in the link between children in poverty and poor educational outcomes. Despite huge efforts by many people in schools, children who grow up in our inner cities are still at much higher risk for a whole series of adverse outcomes. We do not have enough evidence to know how schools might be able to compensate for these disadvantages. This paper has suggested that current expectations for schools are probably unrealistic, and our strategies may need re-examination. At the very least, we would benefit from a broader discussion about what steps might be most useful and most feasible.


Anderson, L., and L. Pelliger. 1990. “Synthesis of Research on Compensatory and Remedial Education.” Educational Leadership 481 (September): 10-16.

Barnett, W. 1996. Lives in the Balance: Age-27 Benefits-Cost Analysis of the High/Scope Perry Preschool Program. Ypsilanti, MI: High/Scope Press.

Bracey, G. 2004. “The Trouble with Research, Part 2.” Phi Delta Kappan 85, no. 8: 635-636.

Bradbury, B., and M. Jäntti. 2001. “Child Poverty Across Twenty-Five Countries.” In The Dynamics of Child Poverty in Industralised Countries, ed. B Bradbury, S. Jenkins, and J. Micklewright. Cambridge: Cambridge University Press/UNICEF, pp. 62-91.

Bradley, R., and L. Whiteside-Mansell. 1997. “Children in Poverty.” In Handbook of Prevention and Treatment with Children and Adolescents: Intervention in the Real World Context, ed. T. Ammerman and M. Hersen. New York: John Wiley and Sons, pp. 13-58.

Brownell, M., N. Roos, R. Fransoo, A. Guevremont, L. MacWilliams, S. Derksen, N. Dik, B. Bogdanovic, and M. Sirski. 2004. “How Do Educational Outcomes Vary with Socioeconomic Status?” Winnipeg: University of Manitoba Centre for Health Policy. <www.umanitoba.ca/centres/mchp>.

Coleman P., and J. Collinge. 1998. Parent, Student and Teacher Collaboration. Thousand Oaks, CA: Corwin.

Dodge, D. 2003. “Human Capital, Early Childhood Development, and Economic Growth: An Economist’s Perspective.” Speech at Sparrow Lake, May 2003.

Dunst, C., and C. Trivette. 1997. “Early Intervention with Young At-Risk Children and Their Families.” In Handbook of Prevention and Treatment with Children and Adolescents: Intervention in the Real World Context, ed. T. Ammerman, and M. Hersen. New York: John Wiley, pp. 157-180.

Earl, L., N. Watson, B. Levin, K. Leithwood, M. Fullan, and N. Torrance. 2003a. Watching and Learning 3: Final Report of the OISE/UT Evaluation of the Implementation of the National Literacy and Numeracy Strategies. Prepared for the Department for Education and Skills, England. Toronto:OISE/University of Toronto. <www.standards.dfes.gov.uk/literacy/publications>.

Howard, S., J. Dryden, and B. Johnson. 1999. “Childhood Resilience: Review and Critique of Literature.” Oxford Review of Education 25, no. 3.

Hunter, H. 2000. “In the Face of Poverty: What a Community School Can Do.” In Solutions that Work: Fighting Poverty in Winnipeg, ed. J. Silver. Winnipeg: Canadian Centre for Policy Alternatives, pp. 111-125.

Knapp, M., P. Shields, and B. Turnbull. 1995. “Academic Challenge in High-Poverty Classrooms.” Phi Delta Kappan 76, no. 10: 770-776.

Levin, B. 1995. “Education and Poverty.” Canadian Journal of Education 20, no. 2: 211- 224.

———. 2003. “Approaches to Policy for Equity in Lifelong Learning.” Paper prepared for the OECD, Paris.

———. 2004. “Students at Risk: A Review of Research.” Paper prepared for The Learning Partnership, Toronto.

McCain, M., and J. Mustard. 1999. Early Years Study. Toronto: Publications Ontario.

Natriello, G., E. McDill, and A. Pallas. 1990. Schooling Disadvantaged Children: Racing Against Catastrophe. New York: Teachers College Press.

Teddlie, C., and D. Reynolds. 2001. “Countering the Critics: Responses to Recent Criticisms of School Effectiveness Research.” School Effectiveness and School Improvement 12, no. 1: 41-82.

Volpe, R. 2000. “What Have We Learned Documenting and Evaluating School-Linked Services for Children and Youth at Risk?” Paper presented to the Pan-Canadian Education Research Agenda, Ottawa. April. <www.cmec.ca>.

Willms D. 2003. Ten Hypotheses about Socioeconomic Gradients and Community Differences in Children’s Developmental Outcomes. Report SP-560-01-03E. Applied Research Branch, Human Resources Development, Canada, February.


  1. This paper is part of a larger research project on poverty and inner-city education, conducted by Ben Levin and Jane Gaskell, and funded by the SSHRC. Further details and references can be obtained by contacting either author.

  2. <www.pisa.gc.ca>.

  3. <www.successforall.net>.

  4. <www.nga.org>.

Bookmark - Social Networks, Social Capital and Social Exclusion

“Policy-makers and practitioners have become aware over the years that the unit of their concern is neither the isolated individual nor a theoretically bounded group such as a household or a community. Themes of interdependence and interconnectedness have come to the fore in recent exchanges between researchers and policy-makers, especially in the investigation of social exclusion. The three terms ‘social network’, ‘social capital’ and ‘social exclusion’ (and ‘inclusion’) are linked in a variety of complex and interesting ways and all of the work reported here focuses, with different degrees of emphasis, on these interchanges.”

“The chapters included bring together:

  • an overview of the social network literature, summarizing the main sociological arguments and traditions;

  • a review of the range of social phenomena which social networks seek to explain;

  • examples of quantitative and qualitative studies using a broad network approach; [and]

  • a discussion of the implications for social and public policy of a network perspective.”

Chris Phillipson, Graham Allan, and David Morgan, eds. 2004. Social Networks and Social Exclusion: Sociological and Policy Perspectives. Aldershot, UK: Ashgate Publishing. Quotes from the Introduction.