The results of the most recent literacy survey of Canadians 15 years and older showed that 16 percent functioned at Level 1 (out of five Levels), the most basic level of reading prose type material, with another 26 percent at Level 2 (Literacy, Economy and Society, 1995). Mishra (1987) expressed concern that the increase in low-literacy in adults in Canada each year approximates 30,000 due in large measure to a 30 percent dropout rate of Canadian high school students. In order to address the problem of literacy standards, it is necessary to understand why young Canadians can spend years in school and yet not acquire adequate reading and writing skills. It is likely that some factors leading to failure and dropout surface when a student reaches high school. It is more likely that such factors have their beginnings in the early stages of schooling and even in the home/community, prior to schooling (Speece and Cooper, 1990).

    The purpose of this study was to identify which factors influenced the literacy development of a sample of grade 9 students in two Canadian provinces: Newfoundland and Alberta. These two provinces were chosen since literacy surveys show that Alberta enjoys one of the highest adult literacy rates in Canada, while Newfoundland experiences one of the lowest. This is a cross-regional and not a cross-cultural study and is pertinent in light of current cross-Canada testing in various subject areas, such as the evaluation project of the Councils of Ministers of Education Canada (McEwen, 1993). Provincial results are often compared without regard to the many factors that contribute to success or failure.

    For the purpose of this study, literacy was defined as competency in reading and writing, including both the strategies or procedures which subjects employ in reading and writing, and the use of reading and writing in various contexts. The influences on literacy development and use, including affective reactions and interpersonal-social relations, were considered essential to understanding literacy achievement. Part of this definition is not unlike that used in the National Assessment of Educational Progress studies in the United States (NAEP Profiles of Literacy, 1985), in which literacy is defined as using printed, written or spoken language to function in society to achieve one's goals and to develop one's knowledge potential. Part of the definition is also consistent with Schiefflin's (1986) notion that literacy is a "social and cultural phenomenon, something that exists between people and something that connects individuals to a range of experiences" (p. viii).

    It is important to understand the model which provides the analytical framework for interpreting those factors which affect literacy achievement. The model chosen for this study was termed the Support Systems Model and was developed by the author on the basis of his earlier research (Fagan, 1987, 1988, 1989a, 1989b, 1990a, 1990b). The model consists of three categories of variables, referred to as Phases. One phase was termed Tertiary in the sense that forces at this level were generally no longer present in the student's life, such as one's early home environment and early community experiences. The Secondary Phase includes school related experiences. The Tertiary and Secondary Phases are distinguished by sequence of time. The Primary Phase variables may be considered as encompassing residual affects/effects from the other two phases. They were designated Primary because, in the studies which led to the development of the model, they were the factors first mentioned when low-literate adult subjects were questioned about the nature of their school and home experiences. Furthermore, the factors were still dominant in the orientation or attitude of these adults towards literacy development or academic upgrading programs. An underlying assumption of the model is that each individual experiences support (positive) and non-support (negative) factors or forces during the course of his/her academic development. The influence of such experiences on child development is aptly described by Cochrane, Cochrane, Scalena, and Buchanan (1985):

Each of us is made up of pieces that we get from one another. Everyone we meet has an impact on us, no matter how small, and the experience becomes a piece of what we are. Our experiences with other human beings are the means whereby we gain our own humanity (p.134).

    In order to overcome a weakness of models identifying factors correlative with literacy development, it was necessary to understand the interrelationships of these variables and how they operated across different grade levels (sequence). The importance of accounting for sequence and interrelationships is supported by Reynolds (1981) who studied the early schooling of children at risk and also by Willett and Singer (1991) in their research on student dropout and teacher attrition. The guiding question for the latter was "when?" rather than "whether?", that is, noting "when" a significant event occurs as opposed to noting "whether" or not it has occurred. Such a question, according to Willett and Singer, is best addressed through methods known as survival analysis, event history, or hazards modelling (p. 409). One adaptation of Willett and Singer's work for this study was the use of the concept of "at-risk probability." At-risk probability was calculated as a percentage of the non-support factors which the students experienced within each grade level. This allows the researcher to identify which year or years the students were most vulnerable and were most at risk in terms of school failure and whether these years occurred in isolation or in clusters.

    In summary, the type of learner an individual becomes depends on the cognitive and affective meaning which he/she abstracts from the impact of school and community experiences (positive or negative) over a particular course of time. Vygotsky (1981) describes the formation of the learner as follows:

We could say that it is through others that we develop into ourselves. The individual develops into what he/she is, through what he/she produces for others. This is the process of formation of the individual (pp. 161-162).

    A survey research design guided the choice of procedures for the collection of the data. The specific measures within the survey design included interview, strategy tasks in reading, and a Concepts of Literacy Task.

   Interview.   The interview was the main data gathering instrument and consisted of a number of questions on the factors at each phase of the model. Mishler (1986) states that the goal of an interview is to provide an opportunity whereby interviewer and respondent can jointly construct meaning based on an understanding of shared contexts. By having a common core of questions "you are confident of getting comparable data across subjects. . . " (Bogdan and Bicklen, 1982, p. 136). A structured interview, according to Mishler (1986), need not be limiting or prevent "storying". Mishler states that "Telling stories is far from unusual in everyday conversations and it is apparently no more unusual for interviewees to respond to questions with narratives if they are given some room to speak" (p. 69).

    The interview methodology may be considered a modification of the critical incident technique (Borg and Gall, 1989) which allows the interviewer to tap time related events of the respondents. In this study, subjects in a particular group were asked to report incidents which they felt were significant (critical) in their schooling. The use of open ended questions ("Think back over your time in school - what was important in helping you do as well as you have done") in the interview facilitated the collection of this type of data.

    One difficulty with interviews is the possible discrepancy that may arise between perception and reality. One attempt to deal with this was to pose the same questions in different format at different times throughout the interview so that responses could be cross-checked. Also, whenever possible, the respondents were asked to describe specific happenings and the nature of feelings engendered were probed in relation to the experiences encountered. The interview was developed by generating a number of questions for each of the category variables in the three phases of the model. The questions were then arranged randomly and an independent rater was asked to assign them to the different categories of variables. The percent of agreement was 94.6.

    Strategy Tasks in Reading and Writing. There were five reading tasks that were written so that they measured the use of seven particular cognitive processes: word analysis and identification, meaning derivation, synthesis, prediction, inference, generalization, and monitoring.

    Concepts of Literacy Task. This task consists of eight photographs of people engaged in reading or writing activities. In an attempt to lessen any bias in the subjects' responses, the photographs had been selected to represent a variety of reading and writing tasks, and to portray people of different ages and gender.

    For both the Strategy Tasks in Reading, and the Concepts of Literacy Task, reliability had been obtained in an earlier study by readministering the tasks to 24 subjects after a six month interval. The average percent of agreement for respondents across both administrations was 89.

Data Collection Procedure

    Each subject was seen individually. The Strategies Tasks and the Concepts of Literacy Task were interspersed within the interview. In addition to yielding specific data, they also served the purpose of leading into another topic of the interview.

Subjects

    Two groups of low-achieving grade 9 students of 20 subjects each were chosen from Alberta and Newfoundland. Grade 9 was chosen because it is the last year of junior high and a time when many students may drop out of school.

    The students from Alberta were chosen in cooperation with two large urban school boards. Low-achieving students were defined as those who scored at least one-half standard deviation below the mean on a standardized reading test administered by the school boards (usually, the Canadian Test of Basic Skills). Students were chosen randomly from those who received parental/guardian permission to participate. The rankings of the students were confirmed by the students' homeroom or classroom teachers. Academic achievement over the course of a year was also taken into account. End of year tests were used to verify the achievement status of the students and only those students who were consistent in their low-achievement placement prior to the research project and at the end of the year were included in the sample.

    All subjects were at least of average IQ as measured by a standardized test, usually the Canadian Cognitive Abilities Test, although a large number of low-achieving students had been administered the WISC-R. Information was collected on socioeconomic status, chronological age, and gender. Students who were recent immigrants and who had little experience with the Canadian school system were excluded from the sample.

    An attempt was made to choose a Newfoundland sample as similar as possible to the Alberta sample. Students were also chosen in cooperation with two large urban school boards and the same number of students per group (20) was chosen randomly from those for whom parental/guardian permission was given to participate in the study. While the Alberta students lived in a large city (population greater than 500,000), the Newfoundland students lived either in a city of approximately 100,000 which also encompassed schools in a rural area, or a in city of approximately 20,000. Unlike one of the criteria for the Alberta sample, standardized reading and IQ test scores were not always available for all students and the criteria for choosing low achieving students were homeroom teacher and counsellor judgments and the general achievement record. This was considered acceptable since the goal of the study was to focus on less-successful students. As in the case of the Alberta sample, end of year results were also examined to note consistency of achievement placement. Teachers of students in both provinces considered the "mix" of students in their classrooms to be "typical" of in that province.

    Other characteristics of both samples of low-achieving students are given in Table 1.

*1-Professional/Technical; 2=Skilled Trade; 3=Clerical/Sales; 4= Manual Labour; 5=Unemployed

---

Data Analysis

    An assumption underlying the results was that low-achievement is due to a lack of support within one or more phases of the Support Systems Model. The responses from the interview questions were converted to numerical scores. For example, if in response to the question, "Do you remember having books in your home when you were small, before you went to school?", the student answered "No", this was coded as 1; if the answer was "Yes", it was coded as 0. A rating of 0 was interpreted as the absence of a non-support or interfering factor. Only questions that were common across all subjects were converted to numerical scores. All scores were summed within Phases of the model.

Phases of the Model

    The analysis of variance data for the low achieving students on the three Phases of the model are given in Table 2.

    The means and standard deviations are reported in Table 3. The students in Alberta and Newfoundland differed on the Primary Phase variables of the model.

    During their school career, the Alberta students experienced almost twice the mean number of non-supports as the Newfoundland students in terms of Primary Phase variables: attitude, attribution, and self-helplessness. It seems that the lack of success of students in both provinces were likely to be equally affected by factors within the Tertiary and Secondary phases of the model. However, the Alberta students were more likely to be influenced by factors of a more personal and affective nature.

Sequence and Clustering of Non-Support Factors

The distribution of non-support factors across grade levels is given in Table 4.

    A very definite pattern emerges for both low-achieving groups and differs by group. The Alberta students experienced a greater percentage of non-support factors at the elementary grade levels; the corresponding peak for the Newfoundland students occurred at the junior high school levels.

    An analysis of the sequence of non-support factors in terms of their occurrence across consecutive grades showed that they were likely to occur in clusters of grades rather than being experienced in a single grade. Alberta students experienced considerable non-support in grades 4, 5, and 6, while Newfoundland students experienced such non-support in grades 7, 8, and 9. If a student has a "bad year" that is immediately followed by a year of support, it is possible that the effects of the "bad year" may be offset. However, when students encounter "bad years" three grades in a row, this is likely to have a more significant effect on their subsequent achievement.

    In order to obtain additional information on the significance across different grade levels, the students were asked which grades they remembered as "worst" in their school career. The results are shown in Table 5.

    Not only did the Alberta low achieving students experience a greater percentage of non-support factors at the upper elementary grades, but these factors apparently had a residual or continuing effect on the students as most remembered these grades as the worst grades experienced. The Newfoundland students' data on grades least liked were also consistent with those grades when a relatively greater percentage of non-support factors were experienced at the junior high school levels, particularly grades 8 and 9. However, Grades 1 and 2 were also noted by a greater number of Newfoundland students as the worst grade experienced. It seems that from the very beginning, the Newfoundland low-achieving students perceived school as a less than satisfying experience.

Perceived Significance of Support and Non-Support Factors

    Responses to questions probing the students' perceptions of which factors were most significant in supporting (fostering) or non-supporting (hindering) their literacy development are categorized and percentages are given in Tables 6 and 7.

    Both groups were fairly similar in their identification of perceived significant support factors; effort, personal/family, and teachers were the top choices of both groups.

    The Alberta students also identified peers/friends, while the Newfoundland students believed that the nature of the school/program was a significant factor contributing to success.

    While both groups saw lack of effort as the factor most likely to interfere with success, this factor was cited almost three times as often by the Alberta students. Both groups were fairly comparable in noting the effect of poor attitude and little motivation. The Alberta students also mentioned teachers and low self-concept/affect as being significant non-support factors.

    Data on concerns of the low achieving students are given in Table 8.

    Again there was similarity in the pattern of responses for both groups. However, while both groups were concerned about future success/ achievement, this constituted over one-half of the concerns of the Alberta students but less than one-third for the Newfoundland students. Alberta students, while low-achievers, were more ambitious that the Newfoundland students in terms of future goals. Both groups were similar in identifying obtaining a job as a concern. "No concerns" made up 21 percent of the responses of the Newfoundland students.

    Low-achieving students may be understood in terms of the lack of support they receive from their beginning school years, and even before coming to school. Support factors are defined in terms of positive experiences, and non-support factors in terms of negative experiences which they have encountered. These may range from physiological difficulties (vision) to the attitude they develop toward school and school tasks.

    The results suggest that different patterns of non-support experiences may contribute to lack of school success. The Newfoundland students tended to be more at risk in the very early grades and again at junior high. The Alberta students, on the other hand, felt that the elementary grades represented their most negative school experiences. The influencing factors on both groups of students tended to cluster rather than occurring in single years. A single "bad year" may not be too crucial in determining a child's lack of success, but a sequence of such years would likely be a compounding factor, so that without special help or intervention, a child would likely continue a pattern of failure.

    The Alberta and Newfoundland students differed in the number of Primary non-support factors: the attitude, attribution, affect, etc. that they encountered. The fact that a large number of Alberta students came from homes which had experienced divorce or separation could have resulted in their being more inclined to examine their feelings and security. The findings raise a number of questions. Do Newfoundland students who are low-achievers become more complacent by junior high, while the Alberta students are more likely to express their negative feelings? The Alberta students were more conscious of non-support factors at this stage of their lives than were the Newfoundland students. In spite of their low-achievement, the Alberta students set higher level goals for their future than did the Newfoundland students. The Alberta students had been more at risk in the elementary grades but, whether through intervention or some other reason, were feeling more positive about their current grade experiences. Is a combination of setting high level goals and overcoming the greatest at-risk period of their school lives more likely to lead to better academic performance in the high school years and a lower likelihood of dropping out of school? On the other hand, are Newfoundland students who are feeling at-risk at junior high, who are more complacent in their attitude and feelings about their achievement status, and who have set lower goals for their future, more likely to "give up" and - unless a major change occurs in terms of intervention and academic support - may not complete high school?

    The Support Systems Model provides a suitable framework for understanding literacy development (or educational development, in general). This model can be used as an evaluative model and allows teachers not only to identify factors which the students perceive as significant in influencing their success (or lack of it), but they can also identify the years when students were more at risk and whether several "bad years" compounded their chances of success. If this were the case, such as students having a bad experience in grades 4, 5, and 6, the teacher would then be more likely to understand the skills and knowledge which the students had not mastered. This also points out the importance of teachers understanding the school continuum in terms of expected learning outcomes.

    Teachers, through preservice and inservice training, should develop a greater awareness of the impact of various socio-cultural factors on student achievement and learn to capitalize on the strengths (supports) and compensate for weaknesses (non-supports). Unless a pattern of failure is interrupted, it will only become exacerbated. Once students begin to experience difficulty, the total context of that child's environment (as specified by the model) must be examined for the purpose of detecting where the child might be most at risk and how intervention might be most productive. Counsellors might note which non-support events continue beyond a year.

    The importance of success for continued success cannot be underestimated and may underlie such other factors as self-concept, attitude, and effort. Teachers should plan and arrange tasks so that at some time each student experiences success and this success is acknowledged. There is still much truth in the old adage, that nothing breeds success like success.

Almy, M. C. (1949). Children's experiences prior to first grade and success in beginning reading. New York: Teachers College, Columbia University.

Baker, L., & A.L. Brown (1984). Metacognitive skills and reading. In P.D.Pearson, R. Barr, M. L. Kamil, & P. Mosenthal (Eds.), Handbook of reading research (pp. 353-394). New York: Longman.

Berkowitz,S. J. (1986). Effects of instruction in text organization on sixth grade students' memories for expository reading. Reading Research Quarterly, 21, 161-178.

Bloome. D. (1983). Reading as a social process Awareness in Reading/Language Research, 2, 165-195.

Bogden, R. C., & Bicklen, S. K. (1982). Qualitative research for education: An introduction to theory and methods. Boston: Allyn and Bacon.

Borg, W. R., & Gall, M. D.(1989). Educational Research. New York: Longman.

Cochrane, O., Cochrane, D., Scalena, S., & Buchanan, E. (1985). Reading,writing, and caring. Winnipeg, MN: Whole Language Consultants.

Clark, M. (1976). Young fluent readers: What can they teach us? London: Heinemann Educational Books.

Doake, D. B. (1981). Book experience and emergent reading behaviour in preschool children. Unpublished doctoral dissertation. The University of Alberta, Edmonton.

Fagan, W. T. (1987a). Adult versus high and low reader processing characteristics. Wisconsin State Reading Association Journal, 31, 57-63.

Fagan, W. T. (1988). Concepts of reading and writing among low-literate adults. Reading Research and Instruction, 27, 47-60.

Fagan, W. T. (1989a). Prisoners and non-institutional adults' perceptions of conditions affecting their learning. Journal of Correctional Education, 40, 152-158.

Fagan, W. T. (1989b). Literacy in the lives of two groups of low-literate adults. Journal of Educational Administration and Foundations, 4, 50-54.

Fagan, W. T. (1990a). Socioaffective factors and literacy development. In S. P. Norris & L. M. Phillips (Eds.), Foundations of literacy policy in Canada (pp. 227-244). Calgary, AB: Detselig Enterprises, Ltd.

Fagan, W. T. (1990b). A comparison of the writing processes of low-literate adults and grade 9 and grade 6 students. Australian Journal of Adult Education, 30, 76-83.

Garner, R. (1990). When children and adults do not use learning strategies: Toward a theory of settings. Review of Educational Research, 60, 517-529.

Goodman, K. (1969). Analysis of oral reading miscues: Applied psycholinguistics. Reading Research Quarterly, 5, 9-29.

Heath, S. B. (1983). Ways with words: Language, life and work in communities and classrooms. Cambridge, England: Cambridge University Press.

Johnson, P. H., & Winograd, P. N. (December, 1983). Passive failure in reading. Paper presented at the annual meeting of the National Reading Conference, Austin.

Literacy, Economy, and Society. (1995). Results of the first international adult literacy survey.Ottawa, ON: OECD and Statistics Canada.

McDermott, R. P. (1977). Social relations as contexts for learning in school. Harvard Educational Review, 47, 198-213.

McEwen, N. (1993). Educational quality indicators. Alberta Journal of Educational Research,39,
167-78.

Miller, S. D., & Yochum, N. (1991). Asking students about the nature of their reading difficulties. Journal of Reading Behavior, 23, 465-486.

Mishler, E. G. (1986). Research interviewing. Cambridge,MA: Harvard University Press.

Mishra, M. (1987). Southam report shocks Canadians: Special report-literacy in Canada. Wordlit, 78, 1-3.

NAEP profiles of literacy: An assessment. (1985). Princeton,NJ: National Assessment of Educational Progress.

Paris, S. G., Lipson, M., & Wixon, K. K. (1983). Becoming a strategic reader. Contemporary Educational Psychology, 8, 293-316.

Reynolds, A. J. (1991). Early schooling of children at risk. American Educational Research Journal, 28, 392-422.

Schickedanz, J.A., & Sullivan, M. (1984). Mom, what does U-F-F spell? Language Arts, 61,7-17.

Schiefflin, B. B. (1986). Introduction. In B. B. Schiefflin & P.Gilmore (Eds.) The acquisition of literacy: Ethnographic perspective. Norwood, NJ: Ablex Publishing Co.

Speece, D. L., & Cooper, D. H. (1990). Ontogeny of school failure: Classification of first grade children. American Educational Research Journal, 27, 119-140.

Torrey, J. W. (1969).Learning to read without a teacher: A case study. Elementary English, 46, 550-556, 658.

Vygotsky, L. S. (1981). The genesis of higher mental functions. In J. V. Wertsch (Ed.), The concept of activity in Soviet psychology. Armonk, NY:M.E. Sharpe.

Wilcox, K.(1982). Ethnography as a methodology and its application to the study of schooling: A review. In G. Spindler (Ed.), Doing the ethnography of schooling. New York: Holt, Rinehart, & Winston.

Willett, J. B., & Singer, J.D. (1991). From whether to when: New methods for studying student dropout and teacher attrition. Review of Educational Research, 61, 407-450.