Phonemic Awareness Studies 1970s

Page history last edited by PBworks 16 years, 6 months ago



Cynthia Boles

Deborah Louie

Suzanne Pfeiffer





Liberman Study

Jenkins, Bausell, and Jenkins Study

Fox and Routh Study

Haddock Studies

Templeton Study



Phonemic Awareness Studies 1970s




Phonemic Awareness (phoneme detection) is the ability to notice, think about, and work with the individual sounds in spoken words. Before children learn to read print, they need to become aware of how the sounds in words work. They must understand that words are made up of speech sounds, or phonemes. During the 1970s, reading research studies focused largely on the importance of teaching consonant blending (see Consonant Blends) specifically, as well as the manner in which the instruction is presented to the child. Another area of interest was the interplay between orthography and phonology, and the value of orthographic instruction.


The following studies represent an overview of phonemic awareness research conducted during the 1970s.


Liberman Study


Isabelle Liberman and her colleagues wanted to conduct an experiment to test the assumption that "young children do, in fact, find it difficult to make an explicity phonemic analysis of the spoken word and also that this ability comes later and is more difficult than syllabic analysis". Forty-six preschoolers, 49 kindergarteners and 40 first graders (ages 4, 5 and 6) were divided into two experimental groups, one assigned to phoneme segmentation and the other to syllable segmentation. The procedure was in the form of a tapping game whereby the child was required to repeat a word or sound spoken by the examiner and to indicate, by tapping a wooden dowel on the table, the number of the segments (phonemes in one group, syllables in the other) in the stimulus items. All the children were tested close to the end of the school year.


The results showed that the test items were more readily segmented into syllables than into phonemes. At age four, none of the children could segment by phonemes, while nearly half could meet the stringent criterior with the syllables. The ability to perform phoneme segmentation successfully did not appear at all until age five, and then it could only be demonstrated by 17 percent of the children. In contrast, almost half of the children at that age could segment syllabically. Even at age six, only 70 percent succeeded in phoneme segmentation, while 90 percent were successful in the syllable task. It was suggested that the sharp increase from 17 percent age five to 70 percent at age six in the number of children passing the phoneme task is probably due in large part to the intensive concentration on reading and readiness activities in the first grade.


Lieberman and her colleagues suggested that a "lack of awareness of phonemic segmentation may be one serious roadblock to reading acquisition". She concludes by stating that early teaching methods which emphasize the intensive teaching of the phonemic structure of the word before the introduction of letter forms should be considered (Liberman,I., 1973).



Jenkins, Bausell, and Jenkins Study


This study compares letter-name training and letter-sound training to see which method enables students to read faster. These two differing methods of training are referred to as transfer variables. To eliminate prior letter or sound knowledge as a possible threat to validity the authors replaced Standard English Orthography with graphemes (symbols) that represented the letters I, T, N, and A. Combinations of these graphemes resulted in transfer words, called such because the students’ knowledge either of letter-name or letter-sound was meant to transfer meaning and result in the students’ ability to read (See Figure 1 below.)



The researchers’ first experiment confirmed their hypothesis that letter-sounds are substantially harder to master than letter-names. Subsequent experiments revealed that phonic blending training coupled with letter-sound training (Group P) produced significantly higher word transfer than no training (Group C) or letter-name training alone (Group N), drawing the conclusion that the “acquisition of a reading vocabulary progresses more rapidly after phoneme training” (Jenkins,J., Bausell,R., Jenkins,L., 1972) (Jenkins, Bausell, Jenkins, 1973) .




Fox and Routh Study


The Fox and Routh study (Fox, B., Routh, D., 1976) (Fox & Routh, 1976) employed graphemes to examine the role phonemic analysis and synthesis play in the “word attack” or decoding of words. Two groups of 4-year-olds, one skilled in segmenting syllables into individual phonemes and one lacking this ability, were trained in phonic blending. Dividing the groups as such allowed for the simultaneous study of phonemic analysis and phonemic synthesis as a child decodes a new word.

Those children who were skilled at segmenting benefited from blending training, decoding the second list of words given significantly faster than those skilled at segmenting who received no training and the nonsegmenter group both with and without training. The study suggests a correlation between segmenting ability and initial decoding (reading) ability. Those students who are weak at segmenting must first develop orthographic knowledge before blending training can be expected significantly to improve their decoding skills.



Haddock Studies


Haddock found evidence that children must be able to analyze parts of words and recombine them into new units in order for transfer to the reading of unfamiliar words to occur. High correlations were also found between reading achievement and the ability to blend a given sound-letter correspondence with a common base (Haddock, M., 1976 & Haddock, M., 1978).


During the 1970s, the education community was still uncertain as to what type of segmentation–blending instruction would be most effective. In 1976, Haddock conducted an experiment with 80 prereading children from three private preschools (Haddock, M., 1976). Three methods of instruction were used:


  • An auditory method, where teachers pronounced training words in parts, for example, sh – eet and asked the children to recombine them into the whole word, sheet. Children did not look at the printed word while working on these tasks but did review daily the basic set of sound-letter correspondences so that memory did not become a significant factor.


  • An auditory-visual method, where teachers used manipulatives to point out how sounds and letters can be substituted for one another to make new words. An example would be showing the word feet and saying, “My word is feet – now I’m going to make the word…” and here the teacher would fold down a flap on which the letters sh were written so as to make the word sheet.


  • The third group of children practiced the basic set of sound-letter associations but were given no formal instruction in blending.


The training period lasted three week, 10 minutes per day, four days per week. Results indicated that training both with the auditory-visual method and the auditory method was significantly more effective than practice on sounds and letters in increasing children’s ability to pronounce the list of synthetic words. The auditory-visual method was also significantly more effective than the auditory method.


These conclusions were supported by another Haddock study that expanded on the first. The two studies along with previous evidence from laboratory studies strongly suggested that beginning readers would profit from specific instruction in blending sounds and letters, along with other basic decoding skills. Only two of the children in the control group of the original study were able to induce the principle on their own by practicing sound-letter associations. Based on the children’s performance, it appeared that an auditory-visual method was most effective. In addition, pronouncing sounds in isolation is not detrimental to children’s blending ability (Haddock, M., 1978).


Secondary questions concerned generalization which might be made about children’s ability to blend in specific situations. For example, if children are taught to blend single consonants (e.g., s,o,r,p) does this ability transfer to the ability to blend single consonants when they are presented in clusters (e.g., sp)? The second and third questions considered whether the position of the consonant in a word (initial or final) or the type of consonant (stop of continuant) made a difference in its blendability. In a 1970 study, Coleman (as cited in Haddock, M.,1976) reported that it was easier for his subjects to blend auditorily vowel-consonant (VC) combinations rather than consonant-vowel (CV) combinations and to blend continuant (those whose sounds can be prolonged, e.g., s, sh, f) rather than stop (those are formed by complete closure of the air passage e.g., t,k, p) consonants. A confirmation of the Coleman data in the present study would have given a definite direction to a sequence for blending skill instruction.


Explicit training in blending, whether by an auditory or an auditory-visual method, resulted in significantly superior blending ability than did training which consists of practice only on sound-letter correspondences. The results further indicated that an auditory-visual method of blending instruction was considerably more effective than an auditory method. At that time, the data strongly supported the importance of including some type of blending component in beginning reading programs (i.e., those dealing with decoding).


Templeton Study


The Templeton study (Templeton, S., 1979) examined the way in which orthographic and phonological knowledge interact in older students. The purpose of the study was to address two questions. First, does an awareness of the means of orthographically representing higher order phonological patterns precede a productive awareness of how these patterns are to be pronounced? Second, is Orthography a more efficient system for encoding and obtaining information about the higher order features of vowel alternation and vowel reduction than is speech or surface phonetic representations? Answering yes to both of these questions would support the assertion that individuals come to organize much of their information about words and word structure according to orthographic, as opposed to phonological, criteria.


The study consisted of students in three grades (six, eighth and tenth) and three separate schools. The researchers chose the subjects by using two criteria: (1) teacher’s recommendations concerning those students who were most likely to be good spellers, and (2) a brief screening spelling test to determine ability.


Twelve real English words and twelve pseudo words were selected. These words were randomly allocated to either a “context” or an “isolation” condition. Six real words were presented together with a sentence containing a blank in which the derived form of the real word could occur. The presentation of six pseudo words was also accompanied by a sentence in which the derived pseudo word could occur. The semantic features of the real words were theoretically delimited; for the pseudo words, an effort was made to construct sentences that would avoid providing possible semantic information for a real word that, by analogy, might be similar to the derived pseudo word.


As hypothesized, the orthographic, or visual, presentation of a base word significantly increased the probability of correct pronunciation of the derived words. At all three grades, presenting a sentence in which a derived word could occur significantly affected correct vowel alternation and vowel reduction. This finding suggested a relationship between higher order phonological knowledge and orthographic knowledge. Seeing base words (see root words) as opposed to hearing them, seems to provide a more direct link with the appropriate phonological rules that apply to derivatives of the base word. That this occurred, even among tenth graders, suggests the weak nature of vowel alternation knowledge. If the appropriate rules were fully internalized, then both conditions of presentation, visual and auditory, should have equally facilitated correct pronunciation. The researcher found that the constancy of orthographic knowledge over phonological knowledge was more pronounced than expected. This was evidenced by most students’ ability to spell most of the pseudo words correctly, while the pronunciation of these words, even at grade ten, was significantly lower than for real words.


Orthographic knowledge may be a necessary condition for higher order phonological knowledge. Orthography clarifies, where pronunciation may confuse, relationships among words. This study supports the claim that the more students know about the logic of word structure, the more sophisticated and adaptive will be their interaction with printed language.





A large body of the research conducted in the 1970s focused primarily on the importance of teaching consonant blending and the most effective way to deliver the instruction. This research showed that specific training in blending resulted in improved blending ability. Based on these findings, researchers recommended including a blending module along with decoding instruction in beginning reading programs.

Supplemental information can be found at the following links:



Phonemic Awareness Studies 1980's

Phonemic Awareness Studies 1990s

Phonemic Awareness Studies 2000s



Fox, B., Routh, D.K. (1976). Phonemic Analysis and synthesis as word-attack skills. Journal of Educational Psychology. 68(1), 70-74.


Haddock, M. (1976). Effects of an auditory an an auditory-visual method of blending instruction on the ability of prereaders to decode synthetic words. Journal of Educational Psychology. 68(6), 825-831.


Haddock, M. (March 1978). Teaching blending in beginning reading instruction is important. The Reading Teacher. 654-658.


Jenkins, J.R., Bausell, R.B., Jenkins, L.M. (1972). Comparisons of letter name and letter sound training as transfer variables. American Educational Research Journal. 9(1), 75-86.


Liberman, I. (1973). Segmentation of the Spoken Word and Reading Acquisition. ''Bulletin of the Orton Society. 23,'' 65-77.


Templeton, S. (1979). Spelling first, sound later: The relationship between orthography and higher order phonological knowledge in older students. Research in the Teaching of English, 13(3). 255-264.

External Links


From Phonemic Awareness to Fluency: Effective decoding instruction in a research-based reading program.


Literacy Connection: Website with phonemic awareness (phonological awareness) resources for teachers.


International Reading Association: Association home page with resources on improving the quality of reading instruction.


CORE: Website that discusses the scientifically based research behind CORE.


Phonemic Awareness: Website with information and resources on phonemic awareness.


Phonemic Awareness Activities: Website with phonemic awareness activities.



Comments by Clark E. Barrow


Hi all,

This is a comment specifically to Susan Pfeiffer (#12) and Cynthia Boles (#13). However, since it is a group project, it includes Deborah Louie also. I like the way you present your studies. In the first study by Jenkins, Bausell, and Jenkins, I noticed the figure is not included. I am sure it shows Group C, P, and N and it would be useful to see. I tried to include a table and figure in my article last week and it never would upload and look correct, so I omitted it. Your citation, “ … raining” (Jenkins,J., Bausell,R., Jenkins,L., 1972).” should read (Jenkins, Bausell, & Jenkins, 1972) according to APA style.


In addition, the citation for the Fox and Routh study should be (Fox & Routh, 1976). I notice the Fox and Routh study “suggests a correlation between segmenting ability and initial decoding.” Did they tell the correlation coefficient? I think this is interesting that the study shows combining different strategies. This is similar to what Rasinski and Hoffman present in an article that I read.


In your Templeton study, I see where orthographic and phonological knowledge are examined as they affect older students’ learning. I found this information on phonological knowledge and think it would be a good addition to your article.


There exist many tasks used to measure the development of phonological knowledge. Some of these tasks are “sound deletion tasks, same/different judgment tasks, and segment counting tasks” (Goswami, 1996, p. 252). The tapping task uses an item such as a wooden dowel for children to “tap out the number of sounds in words at different phonological levels” (p. 252). Another task of measuring phonological knowledge is the oddity task where “children have to listen to a group of spoken words and then select the word that has a different sound from the others” (p. 252). The same-different judgment task is where children listen to pairs of words and judge whether they have the same sound between them.


To deepen the thought of phonological knowledge, Stanovich (1992, as cited in Goswami, 1996) suggested thinking about phonological knowledge as a continuum. Stanovich said the oddity task would measure shallow phonological sensitivity whereas phoneme segmentation would measure deep phonological sensitivity. Snider (1997) agrees with Stanovich’s statement that oddity task is an easier measure of phonemic awareness than other tasks such as phonemic segmentation and deletion.




Goswami, U. (1996). Chapter 17 Phonological and Lexical Processes. In Handbook of Reading Research, Kamil, M. L., Mosenthal, P. B., Pearson, P. D., & Barr, R. (Eds.) (Vol. 3, pp. 251-264). Mahwah, NJ: Lawrence Erlbaum Associates.


Snider, V. E. (1997). The relationship between phonemic awareness and later reading achievement. The Journal of Educational Research, (90)4, 203.



End of Comments by Clark E. Barrow

Commentary by Paul Stewart

This commentary is specifically for Cynthia Boles (13) and Deborah Louie (14). However, since it is a group project, it also includes Susan Pfeiffer (12). I think your group did a great job of reviewing PA in the 1970’s. I have a couple of suggestions, the first one is to delete the Fox and Routh Study and expand on the Coleman work as cited in Haddock, M. 1976. Coleman noted that blending is a strategy that students can apply to many different words, but direct instruction in the blending strategy using many sounds is necessary before students will acquire the generalized skill. Coleman, E. (1970). "Collecting a data base for a reading technology", in Journal of Educational Psychology Monograph, 61(4), Part 2. Secondly, I think that you should expand on your conclusions once you add Coleman if you decide to. The reference for Haddock 1978 needs to have March deleted from the date and the volume number added to the reference also. This journal did not have a volume number, only a month. Not very clear on this, sorry.Appreciate the help!!

End of commentary by Paul Stewart

Commentary by Stacee Jennings


Wow! Excellent job guys. I had to read over your report 3 times to find something to comment on. Anyways, I'm sure you are aware, but your tables are not coming up. I figured out how to upload a file and link to it so if you need help with that, just send me an email. The only other thing I would suggest is providing a link for phonemic awareness activities for teachers to use in the classroom. Phonemic Awareness Studies 1990s has some really good ones to reference. This website also has some good activities Phonemic Awareness Activities.

Commentary by Kathi Crittenden

I really like the introduction. It was very clearly and concisely written. I also liked the way the paragraphs on the studies are written; they begin with a sentence that explains the premise of the study and then delve into further detail. This setup made it easy to follow the rest of the paragraph. The major improvement that could be made to the paper would be for the conclusion to be fleshed out and provide more of a synopsis of the paper as a whole.

Comments (0)

You don't have permission to comment on this page.