Wednesday, July 29, 2015

A moral dilemma and a solution (2005)

Stephen Krashen
Letter to the Editor, Reading Today, June/July 2005, p. 19

"Teachers dip into their own pockets for school supplies" (April/May issue, p. 41) should have been a lead story on page one of Reading Today, as well as in every newspaper in the United States. The National Education Association finding that the teachers polled spent average of nearly $1,200 out of their own pockets yearly in non-reimbursed school expenses confirms that schools are underfunded and also demonstrates the dedication of members of the teaching profession.
Of special interest to IRA members is how much teachers spend on books. The NEA reported that teachers spend an average of $250 of the total $1,200 on "books and videos" for their students. This is quite close to what others have found. In a recent article published in Reading Horizons, Christy Lao of San Francisco State University reported that the New York City teachers she interviewed spent an average of $378 per year of their own money on classroom library books for their students. If these data are typical, teachers are spending more on books than school libraries are. According to a survey published in the School Library Journal in 2003 by Marilyn Miller and Marilyn Shontz, school libraries now spend about $9 per year per student on books. There are about 50 million public school students in the United States. From this we can estimate that about $400 million is spent on books in school libraries, assuming 90% of all students in the U.S. have access to a school library.
If the average teacher spends $250 on books for students, this amounts to three quarters of a million dollars (3 million teachers nationwide). If Lao's figure of $378 is typical, they spend about a billion dollars, more than double the amount spent for books in school libraries. If teachers' practices reflect the need, this figure suggests that we should spend at least double the amount we are now spending on books.
Teachers face a serious moral dilemma. If they don't spend their own money on books, equipment, and even toilet paper, the students suffer, especially students from low- income families who often attend seriously underfunded schools and have little access to books outside of school. If teachers do spend their own money, there is no pressure on the system to supply these essentials. The only solution is to create pressure by doing studies such as the NEA and Lao did and by publicizing the results.

Monday, July 27, 2015

When one exposure is enough

Sent to Scientific American, July 27, 2015

The idea that we remember things better when we retrieve them more frequently from memory, as claimed in "Building The 21st-century Learner," (July 15) applies only to facts and concepts that are irrelevant to us. When a fact or concept solves a problem that is of genuine interest, one exposure is enough. That's why this poem is nonsense:

Do you love me?
Or do you not?
You told me once.
But I forgot.

Let's stop worrying about better ways of getting students to master material that is irrelevant to them. Let's make school more intellectually compelling.

Stephen Krashen

original article:

For full text of the article, plus Susan Ohanian commentary:

Sunday, July 26, 2015

Can a quick auditory test predict future reading ability?

Stephen Krashen July, 2015

The claim has been made that a short, 30 minute test, can predict future reading success. I argue here that this test "that can look in to a child's (reading) future" (Turner, 2015) only predicts the child's performance on measures of phonological awareness and other non-reading tasks, not reading comprehension.

Specifically: the test (The Auditory Processing in Noise test) measures the ability of the child to "tune out competing sounds to tune into speech", and the speech sounds were syllables such as "da" (Turner, 2015; White-Schwoch, Woodruff Carr, Thompson, Anderson, Nicol, Bradlow, Zecker and Kraus, 2015). Children who do better on this task also do better on a number of tests than their peers, controlling for demographic factors (not discussed in detail): tests of phonemic awareness (e.g. hearing a syllable such as /pa/ and being able to pronounce what is left after removing the initial consonant), automatized naming (how quickly the child can name objects, pictures, colors, etc), and memory for spoken sentences. None of these tests are tests of reading comprehension.

Performance on such tests does predict how well children eventually read for meaning, but there is good evidence that these abilities are not the cause of true reading ability: Rather, both reading ability and high scores on these tests are caused by something else: Actual experience in reading for understanding. This possibility is, in fact, mentioned by White-Schwoch et. al.: Although they accept the view that "phonological processing is a necessary foundational skill for reading development," they note that there is evidence that is consistent with the hypothesis that in school-aged children "reading subskills mature as a function of reading experience" (p. 9).

Evidence against the hypothesis that phonemic awareness is "a necessary foundational skill  for reading development" is straight-forward:

In true experiments in which one group of children gets phonemic awareness (PA) training and the other does not, the PA-trained group is not significantly better on tests of reading comprehension, administered months or years later. Truly significant effects were found in only one study with only 15 children in each group, in Hebrew as a first language (Krashen, S. 2001a; for second language studies, see Krashen and Hastings, 2011).

Many children with low or even zero PA develop into competent readers (Krashen, 2001b; for second language studies, see Krashen and Hastings, 2011). A clear example of this is provided by Campbell and Butterworth (1985). Their subject, R.E., was a university student who “reads at least as well as her fellow undergraduates” (p. 436). This university student graduated from London University with second-class honors in psychology, and performed above average on standardized tests of reading. She had great difficulty in reading nonsense words, and while she knew the names of all the letters, she had difficulty with the sounds corresponding to the letters. She also performed poorly on tests of phonemic awareness and phonemic segmentation.

Campbell and Butterworth concluded:
“Since R.E.’s word reading and spelling are good, strong claims based on the necessity of a relationship between phonemic segmentation and manipulation skills, on the one hand, and the development of skilled reading and writing, on the other, must be weakened” (p. 460).
This good reader would probably have done poorly on the Auditory Processing test.

Falling behind

White-Schwoch et. al. consistently point out that their test will help predict which children will "fall behind" in school. This reflects the obsession with grade level and the assumption that all children should progress at a similar rate. All that really counts is whether children eventually learn to read and become lifetime readers. Both are dependent on finding interesting and comprehensible reading material.

Fink (1995/6) studied 12 people who were considered dyslexic when they were young, who all became “skilled readers”. Out of the 12 people, nine published creative scholarly works and one was a Nobel laureate. Eleven out of these people reported that they finally learned to read between the ages of 10 and 12 (p. 273), and one did not learn to read until the 12th grade.

According to Fink, these readers had a lot in common:
“As children, each had a passionate personal interest, a burning desire to know more about a discipline that required reading. Spurred by this passionate interest, all read voraciously, seeking and reading everything they could get their hands on about a single intriguing topic" (pp. 274-275).


The Auditory Processing in Noise test is associated with performance on tests of phonological processing and other measures. There is no solid evidence showing that performance on these tests leads to better reading for understanding. There is, however, no evidence showing that PA training leads to better reading, several studies show that children can learn to read quite well with little or no PA.  There is also considerable evidence showing that the path to learning to read and improving reading is doing plenty of reading for understanding of texts that the reader finds interesting (see also Krashen, 2004).

One of the researchers on the Auditory Processing in Noise projects suggests that all children be tested at birth (Turner, 2015). I suggest we invest instead in libraries, librarians and excellent literature instruction.

Campbell, R. & Butterworth, B. (1985). Phonological dyslexia and dysgraphia in a highly literate subject: A developmental case with associated deficits and phonemic processing and awareness. The Quarterly Journal of Experimental Psychology, 37A, 435-475.
Fink, R. 1995/96. Successful dyslexics: A constructivist study of passionate interest
reading. Journal of Adolescent & Adult Literacy 39 (4): 268-80.
Krashen, S. 2001a. Does “pure” phonemic awareness training affect reading comprehension? Perceptual and Motor Skills, 93, 356-358. Available at:
Krashen, S. 2001b. Low PA can read OK. Practically Primary, 6(3), 17-20. Available at:
Krashen, S. and Hastings, A. 2011. Is Phonemic Awareness Training Necessary in Second Language Literacy Development? Is it Even Useful? International Journal of Foreign Language Teaching 7(1). 
White-Schwoch T, Woodruff Carr K, Thompson EC, Anderson S, Nicol T, Bradlow AR, Zecker SG, Kraus N (2015) Auditory processing in noise: A preschool biomarker for literacy. PLOS Biology 13(7): e1002196. Available at:
Turner, C. 2015. The test that can look into a child's (reading) future.

Thursday, July 23, 2015

Wrong cure for a non-problem

Stephen Krashen

Duval (Florida) superintendent  wants K,1,2 children to attend academic summer school because half of county's third graders don't read on grade level.
1.     Grade level, by definition = median score, the 50th percentile.  One-half must read below grade level if grade level based on district's scores. If based on national scores, Duval third graders are exactly at the national average. We have a math crisis among administrators.
2.     Academics is the wrong cure even if there was a real problem: The path to the highest levels of literacy starts with (1) stories, read-alouds, and then (2) massive self-selected free voluntary reading, which prepares students for (3) heavy, academic and specialized reading.  It is not just the most pleasant way, it is the only way (Krashen, S. 2012. Developing academic proficiency: Some hypotheses. International Journal of Foreign Langauge Teaching, (2): 8-15. (available at

Available with Ohanian commentary:

Monday, July 20, 2015

American educators just can't win

Sent to the Christian Science Monitor, July 20
The US won the Math Olympiad, but American educators just can't win: Instead of telling us how these young people did so well and celebrating their accomplishments, the Monitor’s report of the Olympiad reminded readers that US math scores are not at the top of the world. The article then presented suggestions for improvement in math education ("US wins Math Olympiad for first time in 21 years. Is math education improving?"July 19).
We are always interested in improving our teaching methods, but pedagogy is not the main problem. The main problem is poverty. Study after study shows that when we control for the effects of poverty, American academic performance, including performance on math tests, is among the best in the world. American scores are unspectacular because of our rate of child poverty, now an astonishing 25%, is far higher than the poverty rates in top performing countries.
Poverty means food deprivation, lack of health care, and underfunded libraries, often without librarians.
The best teaching in the world will not help when students are hungry, ill, and have little or no access to books.
Stephen Krashen

Sunday, July 19, 2015

Fulfilling the goals of NCLB

Sent to the Los Angeles Times, July 19, 2015

There is a consensus that No Child Left Behind didn't work ("Can two federal bills finally fulfill goals of No Child Left Behind?" July 19), but the congressional proposals for the new education law either focus on the wrong issues (federal vs. state authority) or keep bad features of NCLB (required annual high stakes testing). 
Study after study has shown that the main cause of low school achievement not a lack of local control or a lack of tests. The main cause is poverty.
The level of poverty in some districts exceeds 95%. The best teaching in the world has little effect when students are hungry, ill, and have little or no access to books.
Let's replace No Child Left Behind by insisting on a living wage that leaves No Child Left Unfed, No Child Without Adequate Health Care and No Child Without Easy Access to a Good Library.

Stephen Krashen
Susan Ohanian

Original article:

Saturday, July 18, 2015

The way to replace No Child Left Behind

Sent to US News, July 18

The way to replace No Child Left Behind is with a law that addresses the main cause of low school achievement: Poverty.  The rate of child poverty in the US is an astonishing 25%, the second highest of all economically advanced countries, and exceeds 95% in many of our urban and rural schools.
Let's protect children from the impact of poverty by insisting on a living wage that leaves No Child Left Unfed, No Child Without Adequate Health Care and No Child Without Easy Access to a Good Library.

Stephen Krashen
Susan Ohanian


Friday, July 17, 2015

Congress ignores major cause of low school achievement

Sent to the New York Times, July 17

  In the versions of the education law passed by both houses of congress, annual testing is still required for reading and math, and for science every three years ("Senate Approves a Bill to Revamp ‘No Child Left Behind’," July 16).
 This means that whether the common core standards are in place or not,  school will remain in a test-preparation mode for reading, math and science. 
  Nothing will be done about the major cause of low school achievement: poverty.
Instead,  we will continue to spend taxpayer money on online testing that uses untested technology and that will continue bleed funds that are badly needed elsewhere.
  There will be no effort to determine whether spending money on online tests is more beneficial than investing the money in ways that have been demonstrated to protect students from the impact of poverty; there will be no serious effort to make sure that no child is left unfed, all children have adequate health care, and all have access to libraries.

Stephen Krashen

A lousy way to rate teachers

Sent to the Los Angeles Times, July 17

"Group sues 13 school districts for not using test scores in teacher evaluations,"(July 16) should stimulate discussion of whether using student test-score gains to evaluate teachers should remain state law. I suggest that the discussion include these two points:
A number of studies have shown that rating teachers using test score gains does not give consistent results. Different tests produce different ratings, and the same teacher’s ratings can vary from year to year, sometimes quite a bit.
In addition, using test score gains for evaluation encourages gaming the system, trying to produce increases in scores by teaching test-taking strategies, not by encouraging real learning.
This is like putting a match under the thermometer and claiming you have raised the temperature of the room.
We are all interested in finding the best ways of evaluating teachers, but using student test-score gains is a lousy way to do it.

Stephen Krashen
Professor Emeritus, University of Southern California

original article:

Some sources: Different tests produce different ratings: Papay, J. 2010. Different tests, different answers: The stability of teacher value-added estimates across outcome measures. American Educational Research Journal 47,2.
Vary from year to year: Sass, T. 2008. The stability of value-added measures of teacher quality and implications for teacher compensation policy. Washington DC: CALDER. (National Center for Analysis of Longitudinal Data in Educational Research.) Kane, T. and Staiger, D. 2009. Estimating Teacher Impacts on Student Achievement: An Experimental Evaluation. NBER Working Paper No. 14607

Monday, July 13, 2015

Libraries: Not just access to digital technology

Posted at: 
S. Krashen
Ben Lee argues convincingly that "Public libraries play a central role in providing access to data and ensuring the freedom of digital knowledge."
But we should not devalue "reading for the sake of reading."  There is substantial research showing that voluntary "pleasure reading" makes a profound contribution both to literacy development and to our knowledge of the world. 
Study after study shows that those who engage in free voluntary reading read better, write better, spell better, have larger vocabularies, and better control of complex grammatical structures.  I reviewed this research in Krashen (2004), and a recent study by Sullivan and Brown (2014), who reported that the amount of reading done at age 42 is a clear predictor of vocabulary test scores. Sullivan and Brown also reported that reading high quality fiction was a very strong predictor of vocabulary knowledge, and reading "middle brow" fiction was also a good predictor, confirming the value of fiction.
Studies by Stanovich and colleagues (reviewed in Krashen, 2004) show that those who read more know more about literature and history and have more "cultural literacy." This is no surprise, but readers also know more about science and even have more “practical knowledge."
Lee points out that those living in poverty have little access to digital resources. The same, of course, is true of access to pleasure reading material.

Krashen, S. (2004). The power of reading. Portsmouth, NH: Heinemann and Santa Barbara: Libraries Unlimited (second edition).
Sullivan, A. & Brown, M. (2014). Vocabulary from adolescence to middle age. London: Centre for Longitudinal Studies, University of London.

United Opt Out and the ESEA: It is time for a Real Testing Moratorium

 I agree with United Opt Out's position on ESEA [] and I want to add a point.

It is possible that the proposal that passes congress will have some kind of reduction in the amount of high-stakes testing required.  If this happens, the testing industrial complex continue to make massive profits. MOREOVER, THEIR PROFITS WILL BE NEARLY THE SAME AS EVER, AND WILL CONTINUE TO INCREASE OVER TIME.

The main cost of testing is the infrastructure, which will remain more or less the same regardless of the amount of testing done. As long as tests are administered online, the boondoggle will continue and grow and tests will continue to bleed funds that are badly needed elsewhere: Every time a new operating system or new hardware is required, we, the taxpayers, will pay the costs. This is why the "reformers" will cheerfully agree to a reduction.

We must insist that no test be given to students unless it has been demonstrated that the test is helpful, and it has been demonstrated that the investment in the test is more beneficial that investing the money elsewhere (e.g. health care, food programs, libraries).  It is time for a Real Testing Moratorium.

Stephen Krashen
July 13, 2015

Sunday, July 12, 2015

The ELL "Problem" and the Common Core "Solution"*

Stephen Krashen

   There is a problem with ELLs/Emergent Bilinguals (henceforth ELLs), but it is not the problem that is typically stated in the media. And there is a solution, but it is very different from the one that is being currently offered by the Common Core. In fact, the Common Core promises to make things much worse for ELLs, as well as for all students in public schools today.
   According to Uro and Barrio (2013), there are two big problems with ELLs:
   The first is that ELLs aren't as proficient as fluent English speakers:  Test results show "wide gaps in reading and mathematics between ELLs and non-ELLs"  (p. 100) and that only five to six percent of ELLs score at or above proficient on fourth grade reading tests in several cities.
This kind of comment shows as astonishing lack of understanding of what an ELL is. If the results did not show gaps between ELLs and non-ELLs, the ELLs would not be ELLs.  Also, if an ELL scores at or above proficient, that ELL should not be classified as ELL.  If five percent of a group of ELLs score at or above proficient, that five percent have been misclassified.
   A second "problem" is equally irrelevant: The Great Cities reports that " ….  trend lines suggest that ELLs have not made meaningful progress academically between 2005 and 2011 …" (p. 100).  We would not expect ELLs as a group to "improve"; when ELLs make sufficient progress, they are reclassified as non-ELL. The group average test score should stay about the same.
   There are, however, real problems with ELLS: First, we are not using the best pedagogy – study after study has informed us that comprehension-based methods are far superior to skill-based methods for second language and literacy development, but much instruction remains skill-based. In addition, despite overwhelming evidence, we have not taken advantage of education in the first language, a powerful means of accelerating literacy development and making second language input more comprehensible (Crawford and Krashen, 2015).
   A second problem is the fact that a large percentage of ELLs live in poverty (Crawford, 1997; Batalova, 2006). Poverty means inadequate diets, inadequate health care, and little access to books; all of these have a devastating effect on school performance.  The best teaching in the world will not help if students are hungry, ill, and have no access to reading material.
   The Common Core will do nothing to solve these problems, and will do a lot to make things worse.  The Common Core language standards are in general hostile to a comprehension-approach to language development (Krashen, 2013), and the Common Core approach for ELLs is to force students to deal with demanding and difficult nonfiction texts in order to promote earlier mastery of "academic language" (Maxwell, 2012). There is no evidence that making reading harder produces better results and plenty of evidence that the route to academic language includes a great deal of self-selected, recreational reading, which now is nearly impossible to include in the current version of the Common Core (Krashen, 2013).
   For standards to be enforced, we must have tests, and the testing demands of the common core are incredible: In fact, the Common Core will insist on more testing than we have ever seen on this planet: The US Department of Education asserts that we will have testing at all grade levels, all subjects, interim tests, and maybe even pre-tests in the fall to be able to measure improvement through the academic year (Krashen, 2011). The tests will be administered online, an untested plan that will cost billions, and that will demand more and more taxpayer money as today's computers become obsolete, and as new "advances" in technology are developed (Krashen and Ohanian, 2011), draining money from projects and approaches that would actually help students.
   The Common Core, a product of the business world, not professional educators, is such an extreme and misguided proposal that we cannot even discuss implementation. We can only discuss resistance.

*Official title: Advocacy in Common Core State Standards Implementation: Why is there a need for advocacy for English language learners/emergent bilinguals? In: Common Core, Bilingual and English Language Learners/Emergent Bilinguals: A Resource for Teachers. In G. Valdes, K. Menken, and M. Casto (Eds). Philadelphia: Caslon Publishing (pp.147-148).

Betalova, J. 2006. Spotlight on Limited English Proficient Students in the United States.  Migration Information Source.
Crawford, J. 1997. Best evidence: Research foundations of the Bilingual Education Act. Washington, DC: National Clearinghouse for Bilingual Education. Retrieved November 11, 2003, from nts
Crawford, J. and Krashen, S. 2015. English Learners in American Classrooms. Portland: DiversityLearningK12. Updated edition.
Krashen, S. 2012. How much testing? krashen-how-much-testing/
Krashen, S. 2013. Access to books and time to read versus the common core standards and tests.  English Journal 103(2): 21-39
Krashen, S, and Ohanian, S. 2011. High Tech Testing on the Way: A
21st Century Boondoggle? Living in Dialogue (Apr 8).
Maxwell, L.   2012. Language demands to grow for ELLs under new standards. Education Week, April 23, 2012.
Uro, G. and Barrio, A. 2013. English Language Learners in America’s Great City Schools: Demographics, Achievement, and Staffing. Washington DC: Council of the Great City Schools.

Thursday, July 9, 2015

TPRS: Contributions, controversies, problems, new frontiers (handout)

Stephen Krashen   twitter: skrashen     NTPRS, July 20, 2015

1.      Compelling comprehensible input: a history of compelling CI
2.      Pop-up grammar: functions of learned grammar = monitor, make input more comprehensible, appreciation.
a.      using conscious learning to solve real-world problems
3.      Books for FVR

1.      The urge for transparency
2.      The cure: theory and compelling CI
3.      Targeted grammar and vocabulary
a.      problems w. targeting: natural order, contrains interest, review, misses lots of grammar, denial of i+1
b.      Nontargeted input: contains i+1, easier to make input compelling, natural review, all rules eventually included.
c.      Institutional demands. Cure: share existing research, determine rules acquired from pure CI.
d.      Writing and Timed Writing: valid, but "washback" effect –  can increase confidence in acq
        Should we practice writing at all? Doesn't writing make you smarter?> don't require it.
4.      Output: forbidden? No. Not forced
a.      speaking helps indirectly: invites CI
b.      comprehensible output? Rare in the real world, acquisition without production
c.      when output doesn't emerge: language shyness
5.      Circling: Are we just doing ALM (audio-lingual method)?
a.      yes, at its worst: targeted structure, as "practice," forced output
b.      at best: confirms comprehension, moves the storu

NEW FRONTIERS: language classes – fewer constraints on subject matter
1. Expanding TPR: exercises, especially yoga instruction, self-defense, dance, magic tricks, juggling, cooking
2. Sheltered subject matter teaching: courses or modules
a. music: performance, theory, appreciation
b. Popular literature.
c. Second language acquisition research and theory, linguistics

1.      What do we do when students have different first languages?  As TPRS spreads to second language ...
a.      Use of L1 very helpful when L2 and L1 do not share cognates
b.      Other ways: visual contex -  pictures, films, real objects, movements of the body (TPR) & linguistic contex.
c.      Is context "misleading"?
(1)    acquisition is gradual. (5-10% each encounter: Nagy, W. et al, 1985.
(2)    Most contexts are helpful: 61% helpful, 8% misdirective – Beck et al, 1983.

2.      The potential of technology – use most obvious and inexpensive, not costly and unsupported by evidence:  reading, eg, our own sharing; movietalk ; bogus applications

References: (all Krashen articles at, "language acquisition"

Beck, I., McKeown, M. and McCaslin, E. 1983. Vocabulary development: Not all contexts are created equal. Elementary School
Journal 83: 177-181.
Krashen, S. 1998. Comprehensible output? System 26: 175-182.
Krashen, S. 1998. Language shyness and heritage language development. In S. Krashen, L. Tse, and J. McQuillan (Eds.) Heritage Language Development. Culver City: Language Education Associates. pp. 41-49.
Krashen, S. 2013. Rosetta Stone: Does not provide compelling input, research reports at best suggestive, conflicting reports on users’ attitudes.
International Journal of Foreign language Teaching, 8:1
Krashen, S. 2013. The effect of direct instruction on pronunciation: Only evident when conditions for Monitor use are met?   GiST: Education and Learning Research Journal   7: 271-275.
Krashen, S. 2014. Does Duolingo "trump" university-level language learning? International Journal of Foreign Language Teaching 9(1):13-15.
Nagy, W., Herman, P., and Anderson, R. 1985. Learning words from context. Reading Research Quarterly 17: 233-255.

Tuesday, July 7, 2015

A longer school day or invest in libraries?

Sent to Education Week, July 7, 2015

Bolgen Vargas & Sandra A. Parker ("Lessons From a Longer School Day (and Year)," July 7) argue that a longer school day will better prepare students for high tech jobs and prevent summer loss.
Summer loss is mostly concentrated in students living in poverty. Studies going back to 1975 consistently show that the major cause of summer loss in literacy among students living in poverty is a lack of access to reading material. 

The most obvious solution is to invest in public libraries filled with books and other kinds of material that students will read, as well as librarians who will help young readers find what is right for them.
We are living in a golden age of literature for young people. Let’s take advantage of it.

Stephen Krashen
Professor Emeritus
University of Southern California

Poverty and access to books: Neuman, S. and Celino, D. 2001. Access to print in low-income and middle-income communities. Reading Research Quarterly 36(1): 8-26.
Summer loss and poverty, more reading and gains:
Allington, R. and McGill-Franzen, A. 2012. Summer Reading: Closing the Rich/Poor Reading Achievement Gap. New York: Teachers College Press.
Heyns, Barbara. 1975.  Summer Learning and the Effect of School. New York: Academic Press.
Kim, Jimmy. 2003. Summer reading and the ethnic achievement gap, Journal of Education for Students Placed at Risk 9, no. 2:169-188.
Shin, Fay. and Krashen, Stephen. 2007. Summer Reading: Program and Evidence. New York: Allyn and Bacon. (Available for free download at

Friday, July 3, 2015

Uninformed about learning to read

Comment posted in response to: How Phonics is Taught Can Affect How Well a Child Learns To Read (the THE Journal).
July 3, 2015
Stephen Krashen

An article in the THE Journal reports on a study involving adults learning “made up scripts”: For one script, learners were asked to link each embedded letter to a sound within the word (known as a "grapheme-phoneme mapping" in learning parlance). For another script entire words had to be memorized.” Sounding out letters activated the left hemisphere, memorizing whole words the right hemisphere. The sounding out group did better reading new words.
Of course, in this study “reading” = pronouncing. And of course the THE Journal article and the researchers conclude that this is evidence for a phonics approach.

My brief comment, posted on their website:

The study described here compared two highly inefficient methods of teaching reading: explicit phonics and memorizing whole words. 

Not mentioned is the leading hypothesis about how children learn to read, developed by Frank Smith and Kenneth Goodman: by understanding what is on the page. A small amount of basic phonics instruction can sometimes help in making texts comprehensible, but there are severe limits on how much phonics can be 
learned and applied because of the complexity of many of the rules.

posted at: