A new study was just released using Your Baby Can Read! (YBCR). Some of the stories written about the study contain strong quotes such as: “These children do not have the developmental capacity to learn how to read,” said Susan Neuman, the study’s lead author. Parents in her own study disagreed with her and stated that their babies were reading. However, she dismissed their results by stating that “parents may have interpreted imitation and mimicking as an indicator of word learning.” Here, you will see that a more likely result is that the parents were correct that their babies were reading and the researchers’ tests were not accurate for measuring reading.
Why did the researchers and parents disagree?
The researchers used a variety of tests to measure mostly “pre-reading” skills such as naming the letters of the alphabet, recognizing the baby’s name, identifying the sounds of individual letters, etc. that they knew were not covered by the program. On a test of reading, the researchers selected a word that was only in Volume 5 even though most of the babies had only watched the first 3 or 4 DVDs – meaning many babies had never seen the word or had only seen it a few times. Instead of simply asking the babies to point to the correct words, the researchers designed tests that required the YBCR group babies to look at a “target” word for most of the time while a novel stimulus was also present. Since infants generally look much longer at novel stimuli compared to familiar stimuli, the test was not a good measure of reading.
On the other hand, the highly-educated parents (e.g., 78% of the moms and 75% of the dads had at least 4-year degrees) had six months to analyze whether or not their babies were reading after being randomly assigned to use the program. Three decades of research show that parents are experts at analyzing their infants’ language and cognitive abilities. The parents may have noticed their babies reading in natural settings or they may have tested them. In natural settings, parents may have noticed that their babies saw words and either said them or acted them out prior to hearing the words being spoken. If parents designed tests to determine their babies’ reading abilities, they would be far more likely to use words that they knew their babies had seen instead of testing them on words that they had not yet seen. Parents would not have tested their babies on information that they hadn’t learned such as the names of the letters of the alphabet, three-word phrases, or the other tests the researchers made up.
These are the actual findings according to the data in the study:
- The babies in the Your Baby Can Read! group could read according to their parents who had numerous opportunities to test them on fair tests. Three decades of research show that parents are experts at determining their infants’ language and cognitive abilities.
- The babies in the Your Baby Can Read! group outperformed the babies in the control group on expressive vocabulary according to their parents.
- In all of the other tests, the researchers found a series of “null results” (no differences between the groups) – null results do not prove anything and they can occur for many reasons, including using insensitive tests that aren’t able to differentiate between groups. The researchers are ignoring the significant differences in their study that favored the Your Baby Can Read! group (that showed the babies learned to read and they learned vocabulary skills). Instead, they focus only on the null findings. Some of the probable causes for the null results are detailed below.
The researchers’ tests were clearly biased against the YBCR group and/or flawed:
- The researchers tested the babies primarily on “pre-reading skills” such as naming the letters of the alphabet. However, the YBCR program does not teach the names of the letters of the alphabet to babies because that is too abstract for babies who can’t yet read.
- The researchers gave the wrong instructions to the parents. They told the parents to watch the DVDs less than half of the recommended viewing time listed in the Parents’ Guide and on the Viewing Schedule for months 2 and 3, for example.
- Instead of asking the babies to read the words, touch or point to words, or act out the words, the researchers measured how long the babies looked at letters, words, toys with words on them, and/or very long strands of letters or symbols. The current study used mean looking time measures which were not appropriate and confounded their tests. In science, tests are confounded when an extraneous variable is added that interacts differently with the two groups. For example, ignoring the variable of novelty vs. familiarity in infant experiments is a major flaw since infants prefer novel stimuli. This variable – of being a novel or familiar stimulus – was introduced into the tests. Instead of only measuring if the babies would look at the “target” word or the “foil” word, the extra variable of novelty was added into the test only for the YBCR group, but not the control group (where both stimuli were novel since those babies had not used YBCR). This means that even if the babies from the YBCR group knew the “target” word, the natural inclination of infants is to look at novel stimuli which would make them wrong on the researcher’s tests.
- The unusual looking-time measure was used for most of the tests. Generally, the study had two words – a “target word” and a “foil word” – placed in the infants’ views while the parent wore headphones and blinders. Anyone who has an infant or has conducted research with infants probably realizes that infants would likely look at and touch the headphones and blinders since these were likely very novel. The babies were asked to look at the “target word” and the amount of time babies spent looking at the words was calculated in the first five seconds. There were no differences between the groups for most of the measures. The authors interpreted this to mean that the babies didn’t know how to read the target words. However, infants prefer novel stimuli so the babies who had used YBCR were likely interested in looking at the novel stimulus more than the more familiar “target word.” Even if the babies correctly answered the questions by looking at the familiar target word, they also looked at the more novel “foil word” for about the same amount of time. The control group likely looked at the two words at a chance level. Infant researchers would have used the accepted habituation paradigm1 in order to determine if differences existed between the groups, but the traditional reading specialists who conducted the study made up their own tests. The methodology that was used almost assured that there would be no statistical differences between the groups. This in no possible way means there were not actual differences between the groups as evidenced by the parental reports that found the YBCR group babies were reading and they learned vocabulary from the program. The tests were not properly designed to detect any differences and in many of the cases the tests were measuring material not in the program.
- Babies were not tested primarily on the high-frequency words in the program. Instead, babies were actually tested on a word the authors stated was in the program, but they failed to mention that the babies had rarely, or most cases never, viewed the word. The word “face” is not in the first four Volumes of the DVDs, word cards, books, or sliding cards. The word “face” is only in Volume 5 of the DVDs and most of the parents had never used Volume 5 of the program at the time they were tested – yet the researchers selected “face” to test the babies. In spite of this, the babies just missed statistical significance at p = 0.064 on the “reading with meaning” test when the p level was set at 0.05. This was considered to be a very difficult test and the YBCR did better than the control group on this test. While only the YBCR babies correctly answered any of the “familiar” questions, they barely missed statistical significance. Remember, the researchers selected a word most of the YBCR babies had never seen as a “familiar cue.” If the researchers selected words all of the babies had consistently viewed, then it is likely the YBCR group would have done even better on this test.
- The YBCR group looked longer at a string of symbols (which in this case was the “foil word”) than they looked at actual words. Most babies who used YBCR have not seen symbols such as #%^&* and they looked at them longer than the words. This could be interpreted as evidence that babies who had used YBCR actually figured out some patterns of written language very early in infancy. This is an important finding, but the authors are not infant researchers and they did not acknowledge its relevance. Infant researchers would have recognized that this was likely because the babies had learned a generalizable pattern of how words typically look from watching the DVDs and looking at words. I predicted that babies would be learning patterns of written language similar to this in the Parents’ Guide (see pages 22-23) that was included in the materials that Dr. Neuman and colleagues cited. Again, the researchers ignored their own data. The control group babies looked at novel strands of symbols and novel strands of letters about the same amount of time showing that they did not differentiate the symbols from words.
- Another “test of reading” was to see if the babies could read their names. Obviously, we do not have the babies’ names in our DVDs, books, or word cards. Instead of asking babies to read their names, they confounded the test. If the baby’s name was “Nathan” the baby was told to “Get Nathan’s Car” when looking at two cars – neither of which was the child’s car. One car was labeled “Nathan” and the other car was given a different name with the same number of letters.
- The new study reveals a bias toward the traditional approach to learning to read (e.g., learning the names of the letters first, followed by the sounds of the letters, etc.) by testing information that is thought to be important when older children are taught to read using a traditional approach. The lead author has edited several books on this traditional approach to teaching reading. Her definition of reading is “Reading = Decoding x Comprehension” which means that evidence of decoding was necessary in order for her to acknowledge that the babies were reading. This may help explain why she so easily dismissed the data in her own study and other studies on babies reading that also show babies are reading. This traditional approach is not working for 66% of 4th graders in the US who read below grade level according to a 2014 report2.
The parents in this study stated that their babies were reading after using Your Baby Can Read! for six months. Parents also stated that their infants learned vocabulary by using YBCR. Decades of research shows that parents are experts at knowing their own infants’ abilities. It is likely that the parents were selecting words that their babies had actually seen in the videos to determine if their babies were reading.
The parents’ scores were highly reliable (in other words, their scores agreed with the researchers) in areas not related to reading. In spite of all of this, the highly-educated parents’ findings were dismissed by saying, “There was a belief among parents that their babies were learning to read and that their children had benefited from the program with their expressive vocabulary development.” The researchers go on to state that the highly-educated parents “may have interpreted imitation and mimicking as an indicator of word learning.” However, the parents had six months and many more opportunities to determine if their infants were reading. The parents likely measured babies on words they had actually seen instead of finding words in the program that the babies had never seen. Parents likely came up with fair tests to determine if their babies could read words instead of asking the baby to look at one word while a novel stimulus is also present and expecting their baby to stay focused only on the target word. In general, parents are great at figuring out what their babies know.
A dozen other studies on Your Baby Can Read! conducted by many different researchers all show positive effects for babies, toddlers, and preschoolers after using YBCR. In fact, the new study admits they couldn’t find any negative results and they reported positive effects (including reading and vocabulary) for the babies who used YBCR.
Thanks for taking time to read our response. Please feel free to comment. Thanks again for your support.
Dr. Bob Titzer
1 The habituation paradigm uses an accepted research methodology to determine if there are differences between groups. The methodology goes like this: a baby watches an event and the eye tracking device measures how long the baby is focused on the event, the same event occurs repeatedly and the infant’s mean looking time decreases until a baseline occurs (where the infant’s mean looking time is fairly constant). Once each individual baby has a baseline measure, then the researcher can show a new event, if the baby’s looking time increases significantly, then it is determined that the infant could perceive the difference from event 1 to event 2. In these cases variability in the baby’s looking time is actually controlled in the experiment because the baby doesn’t get to see a new event until a baseline is fairly constant. In the present study, two new stimuli appeared at once. The researchers somehow came up with 5 or 10 seconds and applied it to all of the tests regardless of how long it might take for the baby to answer the questions. For example, the YBCR group of babies could have looked at both stimuli, then looked at the correct answer (the target word). Next, the babies could have looked at the novel stimulus (the foil word) for most of the remaining amount of time. If the infants explored by looking at the novel stimulus (which is what babies do) then they were counted wrong on the test.
2 The Annie E. Casey Foundation. (2014). Early Reading Proficiency in the United States. Baltimore, MD: Author. Retrieved from www.aecf.org.