The Multiple Cues or "Searchlights" Word Reading Theory: Implications for Reading Recovery®
Keith T. Greaney
Introduction
While there appears to be strong support for Reading Recovery® (RR) in several countries (see What Works Clearinghouse, 2008) there is also research evidence suggesting that such support is not universal (Chapman, Greaney, & Tunmer 2007; Iversen & Tunmer, 1993; Tunmer & Chapman, 2003; Tunmer & Chapman 2004b; Chapman, Tunmer, & Prochnow, 2001).
In fact, there is an interesting irony about RR in New Zealand, the "birthplace" of the program. While RR was designed to help the lowest achieving students in literacy, there is evidence to suggest that it has had little impact on this group of students. RR was first introduced in New Zealand schools on a national scale in the early 1980s at a time when the country ranked first in the 1970 International Association for the Evaluation of Educational Achievement (IEA) international survey of literacy achievement. Yet in the last 25 years following its introduction New Zealand's performance on the subsequent international literacy surveys such as PIRLS (see Mullis et al., 2003) has continued to drop from 1st in 1970, to 6th in 1990, to 13th in 2003, and to 26th in 2006 (Greaney, 2004; Tunmer & Chapman, 2004a:, Tunmer et al., 2008). Given that RR has been in operation for approximately 25 years and given that such a program was designed to assist the very children who comprise the lowest achievers in early literacy, it seems ironic that New Zealand's performances in international literacy surveys have continued to decline during this period.
While there have been several concerns about RR discussed in the research literature (Reynolds & Wheldall 2007; Shanahan & Barr 1995; Glynn, Crooks, Bethune, Ballard, & Smith, 1989; Chapman et al., 2007; Tunmer & Chapman 2004b; Iversen & Tunmer 1993; Tunmer & Chapman 2003), it is implied in this article that a major reason why RR has failed to have an impact on New Zealand's widening literacy achievement gap may be due to the program's heavy reliance on the multiple cues theory of word reading. A critical discussion of the multiple cues theory is important because this theory also underpins both the assessment and teaching practices of all RR teachers in whatever country the program operates. Furthermore, in New Zealand at least, (and possibly many other countries where RR operates), as a "flow-on" effect from RR, regular junior class teachers' reading assessment and teaching practices are also heavily influenced by this theory. (Children in New Zealand begin school on or soon after their fifth birthday and the first two-year groups are often referred to as junior classes.) The reason for the "flow-on" effect is that most RR teachers are also regular class teachers so it would be expected that their class literacy practices are influenced by many of the philosophical underpinnings of RR.
This article begins with an outline of the characteristics of the multiple cues theory of word reading. Second, some examples of how the running records assessment practices are influenced by this theory are presented and critically examined. Finally, a critical discussion of some examples of how the word identification teaching practices, particularly regarding the selection and use of teaching prompts, are also influenced by the multiple cues word reading theory are also presented.
. . . the multiple cues model of word reading has emerged and developed a particularly strong populatrity within whole language classroom evironments.
The Multiple Cues or "Searchlights" Theory of Word Reading
The relative importance that should be given to the explicit teaching of word identification strategies during instructional reading and tutor lessons is an area of debate among both teacher practitioners and researchers. Those who subscribe to the whole language (or meaning-emphasis) theory of reading claim that fluent readers use multiple cue sources of information including, activating prior knowledge, using sentence context and syntax cues, cues from any accompanying illustrations, and visual/grapho-phonic cues from the words (Clay, 2005a, 2005b; Smith & Elley, 1997; Smith, 1979; Hood, 2000; Ministry of Education, 2003). However, those who subscribe to the code-emphasis view of reading claim that, while these multiple cue sources may have some relevance, it is the visual/ grapho-phonic cues that are of prime importance and should therefore be explicitly taught as priority cues (Greaney & Arrow, 2010; Greaney, 2001, 2002; Pressley, 2006; Tunmer & Chapman, 2004a, 2004b; Tunmer & Greaney, 2010). For the past 40 years the multiple cues model of word reading has emerged and has developed a particularly strong popularity within whole language (i.e., meaning-oriented) classroom environments. One view of the multiple cues approach to word identification suggests that the multiple information sources available to readers enable them to predict or guess most unfamiliar words. Tunmer and Greaney (2010) state, for example, that "multiple cues theorists incorrectly assume that skilled reading is a process in which minimal word-level information is used to confirm predictions about upcoming text based on multiple sources of information" (p.230). Even though the model has been rejected by the scientific reading research community, the theory, as demonstrated later in this article, underpins how teachers both assess oral reading progress (i.e., running records) and how they instruct for word identification (i.e., teaching prompts).
The multiple cues theory of word reading model is often represented in diagrams similar to that presented in Figure 1 (see Hood, 2000; Smith & Elley, 1997; Clay, 2005b; Ministry of Education, 2003).
According to the Ministry of Education (2003) "Fluent readers and writers draw on their prior knowledge and use all available sources of information simultaneously and unconsciously" (p.30). While some researchers and commentators suggest that the multiple cues are viewed as being of equal importance, it is worth noting that the visual/grapho-phonic cues are always placed at the bottom of the diagram suggesting that perhaps these particular cues may be of less importance. In support of this concern Beard (2003) has noted that "Adams (1998) describes the way in which this diagram has been adopted in teacher education and is concerned that it may sometimes be used to underplay the role of phonics, as 'grapho-phonic cues' are tucked away at the bottom of the model, perhaps suggesting that such cues are a last resort in teaching" (p. 204).
Smith & Elley (1997) also maintain that "context cues are emphasized in junior classrooms [and that] reading is easier when cues that come from the meaning or the sentence structure help the child fill any gaps" (p. 26). Furthermore, Clay (1998) also suggests that there is a definite hierarchical order of the cues that young readers use when they read. She claims that beginning readers
need to use their knowledge of how the world works; the possible meanings of the text; the sentence structure; the importance of order of ideas, or words, or of letters; special features of sound, shape and layout; and special knowledge from past literacy experiences before (emphasis added) they resort to left to right sounding out of chunks or letter cluster, or, in the last resort, single letters (p. 9).
According to Clay's view, the word-level cues should be viewed as the least important. Frank Smith (1979) also views the word-level cues to be of least importance when he claims, for example, that "the first alternative and preference is to skip over the puzzling word. The second alternative is to guess what the unknown word might be. And the final and least preferred alternative is to sound the word out. Phonics in other words, comes last" (p.26).
The Multiple Cues or "Searchlights" Theory and Its Impact on Running Record Assessment Practices
Running records are the most common form of oral reading assessment used in elementary schools and in many tutor programs. (A running record is a recorded account of a student's oral reading behavior that is taken by the teacher as the student reads the text protocols.) Furthermore, running records are taken on a daily basis as an integral part of every RR lesson to monitor reading progress. A standard format for analyzing the error responses is used to calculate error and self correction rates and to analyze the extent to which the students are thought to be using the various (multiple) cues when attempting to identify unfamiliar words. Furthermore, the results from running records enable teachers to identify areas of weakness and to design instructional tasks to address these needs. Symbols are used to highlight the particular cues that students are deemed to have used when encountering unfamiliar words and these symbols represent the three main cues (M for meaning, S for sentence structure, and V for visual/grapho-phonics).
An example of a student's reading errors presented in an instructional resource text about running records for teachers (Ministry of Education, 2000, p.24) has the following story text and reading error responses and is presented to illustrate the level at which the multiple cues theory influences the analysis of the responses. The complete story is not inserted here, as only the sentences with the reading errors are the focus of the analysis. The reader's errors are printed above the words.
For the errors during the reading the student had identified only the d (for dropped), childrens (for children), walk (for wake) and wake for woke). The first error analysis showed that the student has used the visual (visual/grapho-phonic) cue for identifying the d (dropped). For the error childrens, (for children) the student has used both the meaning and visual/graphophonics cues, because the error indicates that the reader has retained the intended meaning and the error has retained grapho-phonemic similarity to the focus word. Similarly the error walk (for wake) is deemed to represent both sentence structure and close visual/grapho-phonic similarity. The final error wake (for woke) suggests that the reader has retained the meaning intention of the sentence and the two words are visually/grapho-phonemically similar. The emphasis on recording peripheral word-level information may be enough in and of itself to discourage teachers from further focusing on teaching for accurate word identification.
In a critique of running records procedures, Blaiklock (2004) also notes this concern when he states that "following these guidelines provides a misleading picture of a child's reading. A child who has difficulty decoding some words may still be assessed as making effective use of visual information if there is any visual connection between the child's errors and the correct words" (p. 248). Regardless of the quality of any reading errors, teachers should still recognize (when analyzing running records) that word reading errors are by definition, still incorrect responses and therefore, the problem should be viewed first and foremost, as one of inadequate decoding. In support of this concern (even when errors retain full meaning), McKenna and Picard (2006) also note that "teachers should view meaningful miscues (like substituting pony for horse) as evidence of inadequate decoding skills and not as an end result to be fostered" (p. 399).
The Multiple Cues or "Searchlights" Theory and Its Impact on Word Identification Teaching Practices
The extent to which explicit instruction in phonemically based decoding strategies should occur in the teaching of reading in the early years of instruction should be of interest to all teachers. This is because all general class teachers teach reading and they all use teaching prompts when instructing for word identification. When students encounter unfamiliar words during regular class instruction, the teacher can do several things including telling the word or offering prompts to assist the reader to either arrive at or decode the word themselves. However, some teaching prompts resemble nothing more than simple global clues similar to those used in a game of treasure hunt, while other prompts encourage the reader to activate specific word-learning strategies. The amount of metacognitive strategy involvement that the reader is encouraged to use is therefore dependent upon the quality of the prompts that the teacher selects. In other words, an effective word identification teaching prompt is more likely to strategically empower the student to be able to successfully identify the same word on subsequent encounters and in different texts. On the other hand, an ineffective prompt is unlikely to have any strategy-enhancing value for the student when the unfamiliar word is encountered on subsequent occasions.
Smith and Elley (1997) acknowledge that the predominant philosophy underpinning reading instruction in New Zealand has often been described as whole language. These authors define the whole language reading philosophy as being based on the theory that learning to read is as natural as learning to speak. They claim that, "educators who support this model of the reading process usually advocate a natural or whole language approach to teaching, arguing that reading and writing are best acquired naturally in the same way as we learn to speak and listen" (p. 77). Furthermore, teachers who subscribe to this naturalistic development viewpoint are less inclined to include explicit teaching in phonological-based word identification strategies in their classes.
In her most recent text for RR teachers Clay (2005b) states that when instructing for word identification "in your first attempts [italics in original] to call features of print to the child's attention, prompt for sentence structure, and then prompt for the message" (p. 111). In her companion text Clay, (2005a) further warns that "undue attention to the details of letters can block the child's ability to use his language knowledge and meaning of the text, as part of his information base for decision making" (p.25).
Discouraging the use of grapho-phonic cues is particularly problematic given that most of the students who are admitted on to RR programs have the greatest phonological-based learning needs. Chapman et al. (2001) found in a longitudinal study of RR that the students who failed to make satisfactory progress showed the most severe deficits on all phonological processing measures at the outset. Furthermore, Center, Freeman, and Robertson (2001) investigated whether the efficacy of RR outcomes varied as a function of classroom literacy program type. Whole language and code-emphasis classes were compared and the results showed that those RR students who had come from code-emphasis classes outperformed the RR students who had come from the whole language classes. It appeared that the code-emphasis instruction offered in the classes had compensated for the lack of attention given to this aspect within the RR lessons. However, for those students from the meaning-oriented classes where no code-emphasis instruction was present, the RR program had failed to address the phonological-based problems at all.
Despite the large amount of international research over the last 25 years that demonstrates the importance of (and the necessity to explicitly teach to some students), specific word-level identification skills, it is surprising that there is still so much attention given to the less efficient multiple cues approach. Of even more concern, is the continued and deliberate relegation of the visual/grapho-phonic cues to that of least importance. Several researchers share this concern. Pressley (2006) states for example that "the scientific evidence is simply overwhelming that letter sound cues are more important in recognizing words than either semantic or syntactic cues, [and that] a heavy reliance on the latter is a disastrous strategy for beginning readers" (pp. 16 and 36).
Furthermore, Moats (2007) notes that "Contextual guessing strategies are supported by the cueing systems model of word recognition which has no basis in reading science. According to this theory, students are said to use grapho-phonic cues, semantic or meaning cues, and syntax or contextual cues to recognize words. In practice, the emphasis is on anything but the links between speech sounds and spelling. Unfortunately, balanced literacy students are learning strategies that poor readers rely on, not what good readers know" (p. 20).
When investigating teacher preferences for word identification prompts Greaney (2001) demonstrated that teachers did prefer to use the context-based cues ahead of the phonologicalbased cue sources when instructing for word identification strategies. However, Morris et al. (2000) argue that "we should not dismiss the possibility (as Clay seems to do) that some children might benefit from studying a single information source (e.g., spelling patterns) in isolation while simultaneously being offered the chance to integrate this knowledge in contextual reading and writing" (p. 251).
It is unrealistic to expect teachers to have to teach students to read every unfamiliar word they encounter. This is because of what Juel and Minden-Cupp (2000) refer to as the "orthographic avalanche of print" they face. This refers to the amount and complexity of words that early readers encounter as they progress through school. However, this is the very reason why it is important that students should be taught to use effective word identification strategies that allow them to transfer such skills from the taught instructional situation to help them identify other unfamiliar words on subsequent encounters without needing to rely on assistance from teacher prompts. Because different texts present (even previously taught) words in different contexts, the reliance on the context-based cues that may have been successful in one situation (e.g., in conjunction with context-based prompting) will almost always not be helpful when the student encounters the same or similar words in subsequent texts. This is because regardless of the context in which hitherto unfamiliar words may reappear within subsequent texts, the visual/grapho-phonic representations of those words (i.e., spelling), remain constant, and it is the constancy of those visual/grapho-phonic representations that teachers should be encouraging students to focus on. This is what makes some prompts more strategically valuable for students.
Conclusions
While decoding is not sufficient for reading comprehension, it is nevertheless, necessary, (Gough & Tunmer, 1986). Because efficient readers are first and foremost, efficient decoders of print, it makes sense to focus on instructional strategies that are most likely to enhance the student's ability to independently decode the unfamiliar words while reading connected text. To do this it is important for teachers to understand that the greatest clue to a word's identity is the word itself. Unfortunately, this realization is unlikely to be fully accepted by many teachers while the predominant theory of reading both within regular class programs and in tutor programs like RR, continues to be dominated and influenced by the ineffective multiple cues theory of word reading.
As mentioned earlier, RR has been in operation in New Zealand since the early 1980s and since its inception over 25 years ago, our literacy achievement gap has actually continued to widen. Furthermore, in other countries where RR operates (with the same heavy reliance on the multiple cues theory), it would be logical to expect the same problems identified in this article, to also be present in relation to the teaching of word identification strategies to students with reading difficulties. Finally, it is incomprehensible to accept the claims that promote the RR program as being an effective and acceptable intervention for students with early reading difficulties when the program has such a heavy multiple cues bias within the teaching and assessment components. Of more concern is that while the international scientific community has rejected the multiple cues theory of word reading, the evidence for this rejection continues to be overlooked by RR proponents and many school systems.
References
Beard, R. (2003). Uncovering the key skills of reading, In N. Hall, J. Marsh & J. Larson (Eds.), Handbook of early childhood literacy. London, England: Sage Publications.
Blaiklock, K. (2004). A critique of running records of children's oral reading. New Zealand Journal of Educational Studies, 39(2), 241–253.
Center, Y., Freeman, L., & Robertson, G. (2001). The relative effect of a code-oriented and a meaning-oriented early literacy program on regular and low progress Australian students in Year 1 classrooms which implement Reading Recovery. International Journal of Disability, Development and Education, 48, 207–232.
Chapman, J. W., Greaney, K. T., & Tunmer, W. E. (2007). How well is Reading Recovery really working in New Zealand? New Zealand Journal of Educational Studies, 42(1&2), 17–29.
Chapman, J. W., Tunmer, W. E., & Prochnow, J. E. (2001). Does success in the Reading Recovery program depend on developing proficiency in phonological processing skills? A longitudinal study in a whole language instructional context. Scientific Studies of Reading, 5, 141–176.
Clay, M. M. (1985). The early detection of reading difficulties. Auckland, New Zealand: Heinemann.
Clay, M. M. (1998). An observation survey of early literacy achievement. Auckland, New Zealand: Heinemann.
Clay, M. M. (2005a). Literacy lessons designed for individuals: Part one. Auckland, New Zealand: Heinemann.
Clay, M. M. (2005b). Literacy lessons designed for individuals: Part two. Auckland, New Zealand: Heinemann.
Cowley, J. (1991). Uncle Timi's sleep. Wellington, New Zealand: Learning Media.
Glynn, T., Crooks, T., Bethune, N., Ballard, K., & Smith, J. (1989). Reading Recovery in context. Wellington, New Zealand: Department of Education.
Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading and reading disability. Remedial & Special Education, 7(1), 6–10.
Greaney, K. (2001). An investigation of teacher preferences for word identification strategies. Australian Journal of Language and Literacy, 24(1), 21–30.
Greaney, K. (2002). Commentary: That reading debate again. New Zealand Journal of Educational Studies, 37(1), 101–108.
Greaney, K. (2004). First, to fourth to thirteenth and (in all probability), still dropping? New Zealand's international literacy results: Some personal thoughts about the reasons for the gap. DELTA, 56(2), 53–64.
Greaney, K., & Arrow, A. (2010). Why the new national literacy standards won't close our literacy achievement gap. New Zealand Journal of Teachers' Work, 7(1), 29–37.
Hood, H. (2000). Right to read: An open letter to teachers. Palmerston North, New Zealand: Dunmore Press.
Iversen, S., & Tunmer, W. E. (1993). Phonological processing skill and the Reading Recovery programme. Journal of Educational Psychology, 85, 112–125.
Juel, C., & Minden-Cupp, C. (2000). Learning to read words: Linguistic units and instructional strategies. Reading Research Quarterly, 35, 458–492.
McKenna, M. C., & Picard, M. (2006/2007). Does miscue analysis have a role in effective practice? The Reading Teacher, 60, 378–380.
Ministry of Education, (2000). Using running records: A resource for New Zealand classroom teachers. Wellington, New Zealand: Author.
Ministry of Education, (2003). Effective literacy practices in years 1 to 4. Wellington, New Zealand: Learning Media.
Ministry of Education, (2010, October). The New Zealand Curriculum Update, Issue 2. Wellington, New Zealand: Author.
Moats, L. (2007). Whole language highjinks: How to tell when "scientifically-based reading instruction" isn't. Retrieved from Thomas B. Fordham Institute website: http://
www.edexcellencemedia.net/publications/ ... gehijinks/ Moats2007.pdf
Morris, D., Tyner, B., & Perney, J. (2000). Early steps: Replicating the effects of a first grade reading intervention programme. Journal of Educational Psychology, 92, 681–693.
Mullis, I. V. S., Martin, M. O., Gonzalez, E. J., & Kennedy, A. M. (2003). PIRLS 2006 international report. Boston: International Study Centre, Lynch School of Education, Boston College.
Pressley, M. (2006). Reading instruction that works: The case for balanced teaching. New York, NY: Guilford Press.
Reynolds, M., & Wheldall, K. (2007). Reading Recovery 20 years down the track: Looking forward, looking back. International Journal of Disability, 54(2), 199– 223.
Shanahan, T., & Barr, R. (1995). Reading Recovery: An independent evaluation of the effects of an early instructional intervention for at-risk learners. Reading Research Quarterly, 30(4), 958–996.
Smith, F. (1979). Reading without nonsense. New York, NY: Teachers College Press.
Smith, J. W. A., & Elley, W. B. (1997). How children learn to read: Insights from the New Zealand experience. Auckland, New Zealand: Longman.
Stuart, M., Stainthorp, R., & Snowling, M. (2008). Literacy as a complex activity: Deconstructing the simple view of reading. Literacy, 42(2), 59–66.
Tunmer, W. E., & Greaney, K. T. (2010). Defining dyslexia. Journal of Learning Disabilities, 43(3), 229–243.
Tunmer, W. E., & Chapman, J. W. (2003). The Reading Recovery approach to preventive early intervention: As good as it gets? Reading Psychology, 24, 337–360.
Tunmer, W. E., & Chapman, J. W. (2004a). Why the reading achievement gap in New Zealand won't go away: Evidence from the PIRLS 2001 international study of reading achievement. DELTA, 56(2), 69–82.
Tunmer, W. E., & Chapman, J. W. (2004b). Reading Recovery: Distinguishing myth from reality. In R. M. Joshi (Ed.), Dyslexia, myths, misconceptions and some practical application (pp. 99–114). Baltimore, MD: International Dyslexia Association.
Tunmer, W. E., Nicholson, T., Greaney, K. T., Prochnow, J. E., Chapman, J. W., & Arrow, A. W. (2008). PIRLS before swine: A critique of New Zealand's national literacy strategy. New Zealand Journal of Educational Studies, 43(2), 105–119.
What Works Clearinghouse. (2008, December). Reading Recovery (WWC intervention report). Washington, DC: Institute of Education Sciences.
Keith Greaney, Ph.D., is a senior lecturer at Massey University College of Education. Keith was a primary school teacher for 28 years and since coming to Massey in 1998 has been teaching and researching in the areas of early literacy development and the assessment and teaching in literacy to children who have literacy-related learning difficulties.