Part of the NCA Commission on Accreditation and School Improvement Journal of School Improvement, Volume 2, Issue 2, Fall 2001
We Thought We Did Everything Right: What Went Wrong?

Robert L. Armstrong


About the Author: Dr. Robert L. Armstrong is professor emeritus of Arizona State University and research consultant for NCA Commission on Accreditation and School Improvement.

Editor's Note: The first three reviews on the present pages are on the subject of school improvement for all, which is the theme of our Spring 2002 issue.

 
Previous Article | Next Article | Contents, This Issue | Feedback | JSI Home | NCA Home
 

Pyryt, M., Sandals, L. & Begoray, J. (1998). Learning style preferences of gifted, average-ability, and special needs students: A multivariate perspective. Journal of Research in Childhood Education, 13(1), 71-76.

NCA outcomes accreditation and school improvement efforts, while notably successful in a number of ways, have also created some confusion. For example, one of the early and ongoing watchwords was "equity." What does "equity" mean? Does it mean teaching all students in the same way? Does it mean that all ethnic subgroups will produce the same achievement patterns? Does it mean using the same curriculum for all students? Actually, it means none of the above. The concept of "equity" means, very simply, that all ability levels of students will learn, will improve their achievement. Simple as it is, this is an important notion because the failure of low ability students to improve is often obscured and missed when average and above-average students are improving, and the means for the total group are rising.

This, then, raises a question; if the goals are the same for all, and if all are to improve, then should all be taught in the same manner? In a study of 172 elementary students in a western Canadian city, investigators Pyryt, Sandals and Begoray looked at this question from the standpoint of learning style. Their research question was; "Do students of different gender or differing learning ability differ in learning style?" Theirs was a highly controlled multivariate study in which gender (two levels) and learning ability (gifted, average-ability, special needs) were the independent variables, while the 22 scales of the Learning Style Inventory constituted the dependent variables.

The results might surprise the reader. While there were statistically significant differences for both ability and gender, there were only a few, and they were small. It seems that boys have a slightly greater tendency to prefer studying with peers and that gifted students tend to prefer learning alone, use more tactile approaches, and are better self-motivated. The large and substantive differences that might have been expected were not there. It appears that variations in learning style are primarily an individual thing. It seems that we cannot stereotype either gender or ability levels of our students when it comes to planning instructional methodology.


Elbaum, B., Hughes, M., Moody, S. & Vaughn, S. (2000). How effective are one-to-one tutoring programs in reading for elementary students at risk for reading failure? A meta-analysis of the intervention research. Journal of Educational Psychology, 92(4), 605-619.

One of the logistical questions in the concept of equity (all levels of students will improve in achievement) is that of logistical expenditure. How much can a school afford to expend to ensure that lower ability students do, indeed, improve in their learning? It must be apparent that this is not an easy question to answer. Different skills and different kinds of knowledge usually require different kinds of learning. What might be achieved with minimal effort in one field of study may require an extraordinary outlay of effort in another?

The investigators of this study turned their attention to just one area, reading improvement in at-risk students. In a well-planned meta-analysis they analyzed the findings from 30 studies. The results were informative, to say the least. For example, it was found that students taught on a one-to-one basis performed at a level .4 of a standard deviation higher than those in the comparison groups. If readers who are dismayed by the thought of the costs of one-to-one instruction, two other findings might be helpful. One is that college students and trained, reliable community volunteers were able to provide significant help to struggling readers. The use of such persons under the supervision of a trained reading teacher would offer real cost-cutting benefits.

The other finding was that when one-to-one tutoring was compared with tutoring in small groups (six or less), no advantage was found for the one-to-one treatment. This finding also offers substantial cost cutting without loss of achievement. Also in this study, Reading Recovery programs fared well with the students who completed them, but students who did not complete the program showed no more improvement than students in no program at all.

This study dealt only with reading for at-risk students. However, by implication it seems possible that in virtually any study area (numeration, writing, problem-solving, etc.) there may be creative ways in which students with special needs can be served without extravagant costs that may inhibit other facets of the curricular program.


McClanaghan, M. (2000). A strategy for helping students learn how to learn. Education, 120(3), 479-486.

Still another aspect of the "all student ability levels will improve in achievement" perspective is the question, "Do all students know how to learn?" The notion of learning how to learn was addressed by the investigator at the college level using a different set of characteristic learning styles than those in the Pyryt et al. (1998) study. The theoretical base of this study was the psychological type indicator, the MBTI (Myers/Briggs Type Indicator). It was adapted by J. Robert Hanson and Harvey F. Silver (1995, Learning Styles and Strategies, Princeton, NJ, Hanson Silver Strong) to create a spectrum of four distinct learning styles: (ST) sensing-thinking, (NT) intuitive-thinking, (SF) sensing-feeling, and (NF) intuitive-feeling.

Since these four personality types differ in their classroom needs, I have included a condensed version of instructional attributes associated with each type.

ST (sensing-thinking): Students want specific information and to know what is right and wrong. They need immediate feedback and lose interest if the pace slackens or the content doesn't seem useful. They learn well from repetition and drill, as well as authentic experience.

NT (intuitive-thinking): These folk are logical, analytical, and skeptical. They are independent workers who trust reason and hard evidence. They like challenge and the chance to be creative. Relevance is important, and they can be very persistent if they are absorbed in the work.

SF (sensing-feeling): Need to relate learning to personal experience. They like to learn cooperatively and are trusting and collegial. Harmonious relationships with their classroom peers may be more important than content mastery. They are sensitive to approval/disapproval and like to discuss topics.

NF (intuitive-feeling): These people are always looking for patterns to guide them and for connections with previous learning. They tend to value aesthetics, thus originality and uniqueness. They respond well to a flexible learning atmosphere that allows them to innovate. They often have trouble organizing their time, and they need to be shown the big picture. They are easily bored by routine.

Of course, no single student's learning style can be classified solely into one of these categories, but my own experience with the MBTI leads me to believe that this learning type categorization may have some real potential for use in our classrooms. If a teacher of physics, or literature, or advanced mathematics, or world history could learn the prevailing type in the classroom, and then adapt to it, the result might be invaluable in helping students learn to take advantage of their own characteristics in the study process.


Schweiker-Marra, K. (2000). Changing teacher attitudes and actions to promote better parent-teacher communications. ERS Spectrum, 18(3), 12-18.

In this short but pithy article the investigator reports on a single school case-study in which a large urban middle school addressed the problem of almost non-existent parent-teacher communication. The parents were dissatisfied, and a survey of seven items (e.g., the school encourages feedback from me about the quality of the program, this school does a good job of informing me about what is being taught in my child's classroom) was scored low by responding parents on all seven points.

A school committee attacked the problem by reviewing the results of the survey. The committee chair was the new technology teacher. She suggested capitalizing on the capabilities of technology to improve communication. The following seven suggestions were initiated: monthly school newspaper, parent hotline, web page, "good news" postcards, encouraging parent volunteers, open houses, special events.

The teachers tended to view their role in a negative framework (it is my job to report to parents when a student's grades are slipping or when they are in disciplinary trouble). As a result of this attitude, a mistrust of parents' ability to correctly interpret the teachers' classroom methods and the ever-present time constraints for teachers, not all of these efforts were equally popular with the teachers. The parent hotline and the web page did draw positive teacher reactions but the time expended by teachers was, at times, prohibitive. Teachers also were rarely comfortable with parent volunteerism. Overall, however, the result was strongly positive. The original survey of seven items was conducted again midway, and at the end. Parent expressions of satisfaction rose from an average of 39% (and a low of 26%) in the first survey to 83% (and 73%). By any measure, the effort was a success, and although the teachers still had some reservations in certain areas, their final response was much more positive than their initial one.

I might add that one glaringly absent means of communication that most of our schools are using in our improvement process is the involvement of parents in the planning and decision-making activities of the process. Score one for us.


Smith, B. (2000). Quantity matters: Annual instructional time in an urban school system. Educational Administration Quarterly, 36(5), 652-682.

I doubt anyone would any longer quarrel with the notion that it takes time to learn and that time is one of the more precious commodities in a school. But how many of you have any idea of the amount of instructional time "leakage" in a typical school day? I call it "leakage" because for the most part time doesn't get away in huge blocks, like days and weeks, but in the drips and dribbles of minutes and hours.

Investigator Smith utilized time and activity data from eight Chicago elementary schools gathered over a period of three years. In-class activities were recorded and classified as "instructional time segments" or "non-instructional time segments." For example, in a traditional mathematics class, lecturing, guiding students through problems, and having students work problems at their desks would be classified as "instructional time segments." During this three-year period Smith found that only 4% of the teachers averaged 10% or less of non-instructional time, while a full 20% averaged more than 30% non-instructional time.

In an alternative form of analysis Smith compared the time allocation of the official school day to that of the means of "well-managed" classrooms in an actual school day, and to the means of "poorly-managed" classrooms in an actual school day. The official school day allocated 30 minutes to daily logistics and 300 minutes to instructional activities. In fact, in "well-managed" classrooms daily logistics took 50 minutes and other non-instructional activities took 40 minutes, leaving 240 minutes for instruction. In the "poorly-managed" classrooms 60 minutes went to daily logistics, 80 minutes to other non-instructional activities and 190 minutes to instruction.

It doesn't take a graduate mathematician to figure out that the difference between the well managed and the poorly managed classrooms amounts to 150 hours of lost instruction time per year. Is this something we should be checking on in our schools? Plugging the time hole alone might do a lot to help improve student achievement.


Harris, A. (2000). What works in school improvement? Lessons from the field and future directions. Educational Research, 42(1), 1-11.

I don't know about you, but it always comes as a great comfort to me to realize that the problems in education seem to cut across all cultures and national boundaries. For that reason, I peruse the professional journals from other countries as assiduously as I do those in our own country. The present article is a case in point. Educational Research is published by the National Foundation for Educational Research in England and Wales, and author Harris is on the faculty of the School of Education, University of Nottingham.

From a careful analysis of current school improvement efforts Harris found some common themes that underpin the principles and practices of each of the several improvement approaches. Among the positive ones are:

  • Establish a vision.
  • An extended view of leadership that involves both teachers and senior managers in leadership and decision-making activities.
  • Matching programs to context, that is, the plan for any school has to be unique to the needs and contextual demographics of that school.
  • Focusing on specific student outcomes.
  • A multi-level approach wherein change at the school, teacher and classroom levels is encouraged.

Do these look familiar, or am I having a case of Yogi Berra's "déjà vu all over again?" It sounds to me as if the folk in the mother country are learning exactly the same lessons about school improvement policies and practices that we have learned right here in NCA.


Smith, F. (2001). Just a matter of time. Phi Delta Kappan, 82(8), 573-576.

I'm sure that all of you with classroom teaching experience have noticed that there are often students in class who never seem to quite be able to complete assignments in the allotted time or learn the lessons as quickly as the others. In fact, I feel like I'm talking down to all of you right now because the problem of learning time in the classroom is so pervasive and ongoing it should be classified as a chronic disabler or something of the sort.

According to Smith, a former professor at the Universities of Toronto and Victoria, the real problem isn't just that some students are slower than others. The real problem is the way we teach and schedule. Our school system, which began to assume its present structural form about 150 years ago, was patterned after the one-hen-in-a-cage assembly line approach that was becoming popular in industry at that time. Smith opines that this type of instructional structure encourages rote, factual learning, rather than the conceptual experiential learning that young people (in fact, all of us) need. He even quotes American physicist laureate Richard Feynman as saying that it would not matter if many aspects of science were neglected completely, provided children at some time had the opportunity to think like scientists.

Is this idea new? Hardly, but Smith's article does point up the fact that our conventional teaching-learning mode, as well as our high-stakes testing pattern, are antithetical to conceptual learning. Is there a way for schools to begin to increase experiential learning right now? Group and individual problem-projects, conducted in cooperation with community organizations or persons, are a way to start. Once we begin to get used to the notion of students having control of at least part of their time, the same attitude might begin to extend to on-campus time. This is no quick fix, but for those of you who want to begin to look at the long haul in educational improvement, this seems a grand place to begin.


Casavant, M. and Cherkowski, S. (2001). Effective leadership: Bringing mentoring and creativity to the principalship. NSSP Bulletin, 85(624), 71-81.

In our school improvement efforts we have moved far in the direction of faculty/staff leadership in the planning and decision-making processes. This has been a healthy, even a necessary, shift. No matter what anyone else (governing board, superintendent, principal) decides, it is the teacher who is in direct contact with the students on a daily basis. It is the teacher who controls the delivery of the curriculum, guides the discussions, raises the questions, evaluates the achievements, anticipates the needs, and comforts the discouraged. To deny those teachers primary decision-making responsibility would be folly, indeed.

But, what are we to make of our principals? Are they to become simple administrative errand boys, shufflers of paper, and accountants of bodies? The investigators of the present study, both doctoral candidates at the University of Saskatchewan in Canada, have other thoughts. They take the need of principal leadership as a given and then address the question of what forms a good principal leader and how such leaders might be developed. For example, in an earlier study Renihan and Leonard (2000, Is the Principalship for You?, Saskatoon, Saskatchewan, Canada, Saskatchewan Educational Leadership Unit) produced two sets of behaviors, those effective and those ineffective in good leadership. I won't list all of the effective behaviors, but here are a few, in particular those important to our style of school improvement:

  • Builds rapport with staff and students
  • Is supportive of teachers, other professionals
  • Empowers; gives credit to others
  • Has a vision of goals for the school
  • Is involved in professional development

A point that is easy to forget is that if we expect principals to become a different kind of leader, we must begin training our novices of today to be that kind. Casavant and Cherkowski have here made a strong case for mentoring. Calling on the studies of others, they cite the following attributes of protégé's that result from good mentoring:

  • Confidence, higher levels of credibility
  • More aware of their strengths and able to make tough decisions
  • Stronger management skills
  • Human resource skills and conscious of the dynamics of working with teachers
  • Better communication skills, as well as overall competence and confidence

Of course, there can be pitfalls in the mentoring process. The mentor and protégé must be compatible. The district (or districts) must be willing to fund the time needed for mentor and protégé to be together on a regular basis. Also, mentors need to be carefully chosen. One who is ultra-conventional may fail to stimulate a protégé to find new ways within self. Nevertheless, a well-planned mentoring period may be our best hope for developing new principals in the mode we need for our school improvement efforts into the future


Riehl, C. (2000). The principal's role in creating inclusive schools for diverse students: A review of normative, empirical, and critical literature on the practice of educational administration. Review of Educational Research, 70(1), 55-81.

As we look further down the long road of school improvement, I think it behooves us to also look beyond the many relatively short term problems and solutions that are so painfully at hand. The old adage, "those who forget the mistakes of the past are doomed to repeat them," has an analogy that I think applies to us now, "those who ignore the warning signs of the future are doomed to make serious mistakes when the future becomes now" (my apologies for my lack of alliterative skill). So, what am I talking about? I'm referring to the truly ethnically mixed population that is becoming the American norm, even in such far-flung Anglo outposts as Bismarck and St. Paul. We thought we were creating a melting pot in this country when our immigrants were largely northern European. The early comers were shaken when the ethnicity of the incomers shifted more to the Mediterranean. Today we are adding substantial numbers from Asia and the Middle East. We are becoming a truly universal nation for the first time in our history, and the schools are just beginning to bear the enormous burden that this imposes. I was recently told of an elementary school in Minneapolis where one particular classroom had 19 languages represented.

In a massive review of research, the investigator of this study looked at 194 previous studies with precisely this problem in mind, "What can a principal do to meet the needs of the diversity of students that are typically entering our classrooms?" Out of this mass of earlier investigations Riehl extracted three major administrative tasks: fostering new meaning about diversity; promoting inclusive school cultures and instructional programs; building relationships between schools and communities. The investigator views work of this magnitude and depth as a form of practice, where inclusive administrative practice is rooted in values of equity and social justice. Riehl also adds that such practice requires administrators to bring their full subjectivities to bear on their practice, and it implicates language as a key mechanism for both oppression and transformation.

In Riehl's opinion, such administrative practice goes well beyond any designation of skill such as "science of administration." It has both moral and ethical underpinnings and must address fundamental human needs. As you can see, this article offers no "how to," no "recipe," no "scratch'n'sniff" plan. Rather, this writing is a sobering and deeply insightful look into the future. Those of us in education who ignore these warning signs are doomed to err. Deep, long-range thought and discussion is needed now. If such thought and discussion is even able to help us avoid the stopgap but possibly damaging solutions that suggest themselves in the next few years, such ponderings will be well worth the time. Remember that school improvement is not a fad that will go away. It is here for eternity, if you will. Plan that way.


Brodney, B., Reeves, C., and Kazelskis, R. (1999). Selected prewriting treatments: Effects on expository compositions written by fifth-grade students. The Journal of Experimental Education, 68(1), 5-20.

In the jargon of the last decade, in student achievement improvement, it is usually "where the rubber hits the road." In teaching and learning, the rubber hits the road in the classroom, and it consists of just those things the teacher and students do together that enhance understanding and learning. With all other administrative, curricular or managerial decisions taken into account, in the end it's the daily interaction of teacher and students that ultimately determines the learning of students. In essence, the whole spectrum of school improvement boils down to one question, "How can we educators do things differently so that our students will learn more, remember more, be more competent when they leave us?"

The present article struck my eye because "writing" seems to be an area of difficulty for most schools, a goal that over the past several years has tended to show smaller gains than the other goals. In the present study the investigators performed a very simple experiment on 96 fifth-grade students. Four treatments were tried on a writing exercise; reading-prewriting (R/PT), reading-only (RT), pre-writing only (PT), and no treatment (COMP), the comparison group. All groups wrote an essay on a selected topic. "Reading" consisted of 10 minutes of reading on the topic, "pre-writing" activities consisted of 10 minutes of listing, organizing, or whatever activity seemed appropriate to the student.

It probably goes without saying that the R/PT group exceeded the other three groups on ideas, but they also exceeded the other three groups on organization, style, mechanics, and total words, as well. It appears that when you know better what you intend to write, you will write better in every respect. Should we have known that all along? I suppose, but it is the simple little ideas like this out of which superior teaching and student achievement is built.


Baumeister, R. (2001). Violent pride: Do people turn violent because of self-hate or self-love? Scientific American, 284(4), 96-101.

In the Spring 2001 issue of Journal of School Improvement I reported on an article about self-esteem authored by Ross and Broh. Based on a large clustered probability sample they noted that while a sense of personal control was significantly related to high grades and test scores, self-esteem was not. Thus, they identified one of the critical sources of frustration for our schools that have selected student self-esteem as a goal in past improvement cycles. Approaching the same construct from another point-of-view, Burmeister began his report by noting that an exhaustive survey of the literature offered absolutely no empirical evidence to support the age-old "common knowledge" belief among educators that violence in youth is related to low self-esteem. Further, he and his colleagues found no evidence to support another widely held belief that self-esteem fluctuates widely in individuals.

In a series of controlled sequential-phase studies, Baumeister and his colleagues found that (1) violence in young people is related to high self-esteem, (2) individual self-esteem is relatively stable, and (3) there is a second factor associated with high self-esteem that is more likely to produce violent behavior, narcissism. The latter finding is particularly important because it has always been fairly obvious that a lot of people with apparently high self-esteem are not violent, thus falsely bolstering the belief that violence is related to low self-esteem. In fact, the Baumeister studies have made a strong claim to the effect that violence in youth is strongly related to a combination of high self-esteem and what is sometimes referred to as "self-love." The problem seems to begin within an individual who thinks well of him/herself but is inordinately sensitive to criticism or any other apparent threat to this belief in self-worth. The two traits combined, appear to offer a "powder keg" situation. Well, forewarned is forearmed. Perhaps this understanding of the nature of violent students may help educators plan for means of meeting the problem.

Previous Article | Next Article | Contents, This Issue | Feedback | JSI Home | NCA Home


All material on this site © 2000-08 NCA Commission on Accreditation and School Improvement unless otherwise noted.
Questions may be directed to the Webmaster (webmaster@ncacasi.org).