There’s a fine art to biting your tongue and walking away and diplomacy is a worthy skill to acquire and hone. Recognizing that we all have differences, is another skill that is fabulous to have. However, sometimes, it’s necessary to find a way to correct a wrong, particularly when it’s a common wrong. (Are you confused yet? Bear with me…please?)
This post has been a while in coming…and I’m fairly sure it will exasperate some and anger others. I sincerely hope that most of you will find it enlightening rather than annoying. I have bit my tongue and walked away from the computer screen a lot lately…now I’m attempting to correct a wrong.I remember being in grad school and learning how to read articulation norm charts. You know, those Iowa-Nebraska norms, or the Sander’s norms (as adapted here). I remember being told that the majority of SLPs read them incorrectly. At that point in time, I remember thinking “yeah, right. It’s not that hard.” However, I didn’t take into consideration the discrepancies among the charts themselves…and, at that time, I didn’t know enough to really apply critical reasoning skills and see if the charts are even measuring the same thing.
Guess what…They don’t. The reason there is so much discrepancy between the charts is because while they all measure sound acquisition, they all have different criteria for what they call “mastery.” Some consider mastery to be at 75% some consider it to be 90%. Some only look at initial sounds, some look at all three positions.
Guess what else…My professor was right. After talking with people on the Facebook SLP pages and on twitter, I’ve come to realize many (maybe not the majority, but definitely a lot) SLPs read these norms incorrectly.
Gregory Lof has a phenomenal handout available called Confusion about Speech Sound Norms and Their Use. I’m going to attempt to explain parts of it here, but I highly recommend you go and read the presentation for yourself. It is fascinating!
Let’s take a look at the Sander’s chart.
As you can see here (hopefully) there is a range for when sounds are considered acquired. Most clinicians that I’ve spoken with believe that the bars indicate the development of the sound, so for instance, with R the child would start to have it at age 3 and it should be fully mastered by age 8. Makes sense, right?
Wrong! What it actually is showing is that 50% of the children age three (3) correctly produce the sound and by age eight (8) 90% of the children have the sound mastered.
Did you catch that? R begins earlier than 8…50% of kids have R at age 3… 3. Hmm. If we wait until 8 to even START to teach how to make the sound, we have put our clients at least 5 years behind the “norm.” Ouch!
Let’s look at what else I see. I often hear that /s/ is a “later developing sound” (I hate that phrase)…It actually has the same bar graph as /r/…so 50% of three-year olds can correctly say /s/. Another common sound that is “later” is “TH.” Voiced TH is developed for 50% of kids at age five, unvoiced by age four.
In fact, if you look at the chart, the majority of sounds are developed by age four. There are two sounds that are a bit later (“th” and “zh”) but that’s it. Two. At least 50% of typically developing 4 year olds have ALL but two sounds correctly produced. hmmm.
Also, another interesting note is that Sanders looked at data from both boys and girls, and included phonemes across words (so initial, medial, and final position). So for those 3 year olds that have /r/? They have them in all positions of the word. (This information can be found here on about page 15.)
Let’s look at the Iowa-Nebraska Norms. This is the norms chart that ND has implemented as their eligibility criteria. It has /r/ as being mastered by 90% of children by age eight (8), while /s/ is mastered by age seven (7). So…let’s see. I often hear that SLPs won’t work on /s/ until after the adult front teeth have emerged, which according to this is after age 7 or 8 because that’s when the central incisors fall out…yet, 90% of typically developing kids have it mastered by age seven (7). Hmm. So if we wait until a child is 8 to start to teach it…well, you can fill in the blank.
Interestingly enough, Smit et al. also found that “reversals” occurred for /b, g, s, z, and l/. Reversals were found when “a predetermined criterion for acquisition is reached by one age group, but not by an older age group.” In other words, even though /s/ was found to be mastered by 90% of children age 7, in reality it was mastered earlier – then enough kids “lost” the skill that they could no longer consider it mastered, and then it re-emerged as mastered at 7. There’s a couple of reasons this could happen; the authors stated perhaps it’s due to “phonologic regression,” or variation in sampling, because the listeners were willing to accept a slightly less precise /s/ for 3-4 year olds than they were for 6 year olds, or perhaps because it is mastered earlier but a slight shift in production occurred and they moved from correct production to error before reverting to an acceptable production again. (Personally, not that I think I’m better than all these famous researchers, I suspect the last is true…they have a good /s/ and then it shifts slightly to an error. This could be due to over emphasis, co-articulation…heck, lost front teeth???) Either way, this research shows that /s/ develops earlier than 7 for the majority of kids.
One fascinating tidbit from Greg Lof’s presentation was the difference between norms. Let’s take a fairly non-controversial sound. We all would expect the sound /p/ to be mastered fairly early. It’s an early developing sound, it’s a bilabial so it’s easily seen and mimicked, it’s unvoiced…easy! We expect toddlers to say “puppy.” Right? Yet, Prather, et al (1975) has /p/ developed at age 2;6. Templin (1957) has /p/ at age 3. Sander (1972) has /p/ at 3. Smit et al. (1990) has /p/ at 3. Goldman-Fristoe (2000) has /p/ at 6;6. (Yes…6;6. That was NOT a typo). The same range can be said for /t/ – anywhere from age 2;6 to age 6;0 depending on which chart you read. For /r/ the range is from 4;6 to 8. Hmm. Some charts have mastery for /r/ at age 4;6? Why do we wait until 4th grade to start to work on it then?
Another fascinating chart to look at is this compilation of the various developmental norms, the year they were done, the number of children who participated, and sample type. If you look, Templin (whose chart is a favorite of many) only had 480 participants and only looked at single-words. Where as Peter Flipsen had 320 participants and used connected speech.
So…where is this all heading? I’d love for you to let me know which developmental norms chart you use. I’d also like to know if your district/state requires you to “wait” for a year after the developmental chart shows the sound should be acquired. More importantly, I’d love for you to consider what you use as guidelines for articulation development. I realize that many of you (myself included) are somewhat limited by our districts/states as to when we can work on certain sounds…Heck, I realize some of you are limited and not allowed to work on articulation at all which is a whole different rant…but, will knowing the differences in norms and how they are often misread make a difference in your therapy? I’d love to hear (even if you disagree with me). Drop me a line in the comments!
Until then….Adventure on!
Mary
Speech Adventures 
Thank you for writing this and thank you for the links to further resources and reading! You have so brilliantly covered the thoughts that have gone through my head on many occasions and the frustration I frequently feel in regards to speech norms and the huge variability in the research and resources!! I sat on my couch frantically nodding my head in agreement while reading it. Much appreciated : )
I’m glad you found value in my writing. The variability is frustrating and confusing. Looking at what the norms really looked at helped me a lot.
Hi Mary, Great blog. I just read Gregory Lof’s presentation a few days ago when one of our FB friends posted the link. I agree, it is very fascinating to see all of the different “norms”. This makes it difficult to have consistent standards of practice as SLPs, because we are not coming from the same knowledge base.
I’ve used the Sanders norm chart, to explain speech sound development to parents and teachers. When my 2nd grade screenings reveal a student has not mastered the /r/, /s/, /th/ sound(s), we begin an intensive RTI program with parent involvement for these “later-developing sounds” (sorry, I use that term because these are the sounds that students do tend to master last).
I keep my own data, not scientific research by any means, I compare results from the student’s kindergarten screenings to their 2nd grade screening results. Last year, of the students who did not have the /s/, /r/, or /th/ sound mastered in kindergarten, 83% of them had independently mastered these sounds by their 2nd grade screening. Had I started therapy with these students in kindergarten, but what an outrageous caseload I would have, with students whose errors would probably resolve without intervention. I guess that’s where the norms are just our guideline, along with our experience, to determine when to intervene.
That being said, I had to be off work most of last year to battle breast cancer, and the SLP who covered for me did not share my philosophy. She put K and 1st grade students on IEPs with goals for artic. of /r/, or /s/, or /th/- (and even /b/ in the final position when it could be produced in all other positions). So, this conversation is interesting to me…to possibly rethink my standard of practice before I change all of these IEPs or exit students-and have to explain that to parents.
Mary, what norms do you use? And do you qualify students on IEPs at a younger age, with justification based on those norms?
Great reply Maryann, thank you.
ND requires the use of the Iowa-Nebraska norms, so I currently go by that.
I’m very blessed in that I have a fair amount of leeway in what I can and can’t do in therapy. In addition to an IEP, I have the option of enrolling a child in a Speech Intervention plan which was created specifically for those artic cases that need my specialized help, but perhaps don’t need an IEP. They have a 2 year maximum, so it’s necessary to make sure that the sounds will most likely be remediated in 2 years. Depending on the age of a student, for instance, I may consider a Speech intervention plan (SPI) for a child to work on “TH” or an interdental lisp, or /r/. If a child has a phonological delay, I would most likely start them on an IEP as there’s so often an impact on reading.
For a simple /s/, I’d most likely wait until the end of 1st grade – beginning of second, depending on the severity. I had a student with a significant interdental lisp; she had full tongue protrusion during all /s, z, sh/ productions. I watched and waited through most of kindergarten and the last quarter of K I enrolled her in SPED services. We just dismissed after 3 years of therapy. However, the only reason I put her on an IEP in K was because of the severity of her lisp.
I do believe that earlier intervention is easier. I’ve worked with students on S and R that are in 3rd or 4th (or later) grades and earlier…and earlier is always easier to eradicate bad habits and teach proper placement. If they have more than one error, I’m more likely to enroll them earlier rather than later.
I love this! Our district has its own standard developed based heavily on Smit et al (which is fabulous since we are in Kansas and I was fortunate enough to study with Dr Smit for a short time). Our artic chart is quite simple and indicates the age they expect the sounds to be mastered–without a stated mastery percentage. Fortunately our chart has all the sounds–blends included–mastered by age 7. That opens up a lot of options regarding treatment. We also seem to have a fair bit of leeway in treatment if we can justify our position. For instance, I have seen a number of preschoolers we are addressing /s/blends with to resolve both cluster reduction and tongue thrusting/lisping tendencies. Thus, we are able to target these “later developing” sounds a bit earlier. Bonus, the focus is off of /s/ and it can be addressed along the way.
That’s awesome Melissa. I’d love to see the chart if you are able to share it. It sounds like a great plan.
Love this! I am a firm believer and advocate of early intervention. Why wait when most kids can produce the sounds at a much younger age? And why wait for kids to continue to use sounds wrong, making it that much harder to remediate? I have always questioned the norms charts and felt that they were being misinterpreted! Can’t wait to share this information with some of my co-workers!
I’m glad you found it helpful. It was very interesting to really look at the charts and see what they were truly measuring.
Many years ago, Dr. Robert Lowe was a young professor at Bloomsburg University of Pennsylvania. I remember him joking about norms… If you wanted to pick up a kid, you used one set of norms. If you didn’t you used another set. That being said, I think you still have to consider that some kids will acquire the sounds later than others. If 50% of kids have acquired the sound! that means 50% haven’t. I like the idea of the Late Eight. It categorizes the sounds well. And teeth do make a difference in /s/ therapy, providing a larger border in most instances.
I at one time had a set of norms that talked about, normal at age ( ), delayed at age ( ). It got misplaced, and I haven’t been able to find it again.
I agree the “late 8” categories are nice. I just don’t like the term as I think it’s misleading. Teeth can make a difference in /s/, but it is possible to teach /s/ without the teeth as a marker.
As far as the 50% who have and haven’t…when I test someone for language and they have a standard score of 100, it’s where 50% of students have met the criteria and 50% haven’t. Why do we have different standards for articulation?
The purpose of the post isn’t to point fingers or say I’m right you’re wrong…It’s simply to bring attention to the misinterpretation of the norms and suggest critical appraisal of practices. We all have to do what we view as best practices, I just think we ought to have all the information before we decide what those best practices are. I’d love to see your set of norms that have “normal” “delayed” etc…that would be great. I’ve seen something similar for phono processes, but not for articulation.
Percentiles do not work like that. It means you have performed better than fifty percent of the population. Not that fifty percent have achieved a particular level.
I’m fairly sure I understand how percentiles work…however, if I take a 100 preschoolers age 5, and 50 have /r/, is that not the 50th percentile and 50% in that sample?
No, percentiles really have nothing to do with the criteria. They are measuring scores, and distribute scores out on a bell curve.
I stand corrected, thank you.
However, the point of the original post is still relevant.
This is a great post, Mary. However I think we also have to bear in mind that when 50% of children are correctly producing sound, 50% aren’t. So with your /r/ example, can we really call age 3 the “norm” when half the population can’t produce it?
I’m not saying we should work on /r/ at age 3 when it’s the only error. I’m saying the practice of waiting for a year or longer past the age when 90% of students have it cemented is wrong (in my opinion). When people say “R doesn’t develop until age 8 or later” the data shows that they are incorrect because for at least half of the child population R develops at 3. As far as “norm” absolutely…Look at a bell curve. We consider other performance right at a “standard level” when only 50% of students have it.
The purpose of the post is to point out the discrepancies in the articulation charts, the misconceptions that many people have regarding the charts, and to encourage SLPs (and others) to critically evaluate what they do in practice.
Hi Mary. Thanks for an insightful read, and for sharing the link to Gregory Lof’s presentation. His method of combining norms really highlights the variety in the research!
Whilst on clinical placement as a student I referred to norms for Australian children that were established by Kilminster and Laird (1978). This chart can be found on Caroline Bowen’s website: http://www.speech-language-therapy.com/index.php?option=com_content&view=article&id=32:table4&catid=11:admin&Itemid=119
Another chart I used to help guide therapy planning was one developed by the York Region Preschool Speech and Language (2011). http://childdevelopmentprograms.ca/backend/wp-content/uploads/Articulation-Developmental-Norms.pdf
Perhaps the articulation norm charts that clinicians use are geographically influenced. For instance, I am from Australia, thus the Kilminster and Laird (1978) chart based on a study of Australian children seemed more valid than those such as Smit et al., (1990) and Templin (1957) whose norms were based on studies of American children. As I have recently moved to the United States, perhaps I should concentrate my focus on the latter. Thoughts?
Thanks for the reply Maree. If you’re working in the schools, you will find that most districts have a guideline for which norms they want you to use. For instance, my state (North Dakota) suggests (strongly) that we use the Iowa-Nebraska norms for whatever reason. Others I know really like the Sanders norms. I really like the links you put forth, and I think they’re more accurate for when speech sounds should be developed…Unfortunately, I’m not sure if you’ll be able to continue using them here.
Thanks for your quick response, Mary. It will be interesting to find out once I start working over here.
In my experience, I find that waiting until 7 to work with the /r/ is the most appropriate. There seems to be a “window of opportunity” for the /r/…if you wait too long after 7, it’s next to impossible to correct. I have also found that approximately 90-95% (and I’m taking an educated guess here…I haven’t actually figured out the percentage) of students who have trouble with /r/ when they’re 5 or 6 will self-correct their error. I realize that other SLPs may have different opinions/experiences, but this works for me. Also, with an interdental lisp, waiting until the permanent teeth erupt works for me. I definitely don’t want to place a child in speech therapy when it’s not needed. My success rate would be higher, but so would my caseload!
Thanks for such a well thought out post!