Improvements for Individual Students (All)

Here is an overview chart comparing Visual Reading speeds at the beginning and end of the course for all participants. Each blue bar represent their reading speed before they start Visual Reading practice. The red bar next to it, is their final reading speed at the end of the course.

Average Weekly Improvement

The average weekly improvement seen in Dyslexic readers is analysed in this paper

"Analysis of Weekly Reading Progress for Neurodivergent Readers (Mainly Dyslexic) using Visual Reading Strategies with Structured Saccade Overlays©. (February 2023)."

The Visual Reading course is based on 3 main concepts. The first is that silent reading speeds can be dramatically improved by gaining control over the accuracy of saccades through using the Structured Saccade Overlays©. The second is that comprehension can be improved by good use of metacognitive reading strategies, even when speed increases. The third is that reading skills develop best when the reader is enabled to read texts that interest them.

The process involves 3 coaching elements.

1. Explaining how and why to use the Structured Saccade Overlays with their chosen reading material, and how to use and develop metacognitive reading strategies

a) This is primarily handled by the initial coaching session (around 40 minutes), and

b) reinforced through subsequent coaching.

2. Ongoing coaching to ensure that the reader gains maximum benefit from using the overlays and that their comprehension keeps improving

a) This is handled by a series of weekly coaching sessions (usually about 10 minutes long). How many the reader needs, depends on their ‘practice’ and coaching needs. On average, most readers need 5 of these, but a small minority only need 3 or 4, while another small minority may need 6 or 7. However, we are finding that further coaching sessions are leading to continued improvements.

3. Coaching in how to move on from using the overlays and keep improving their silent reading effectiveness after the coaching ends.

a) This is handled by the final coaching session (usually 20 minutes)

Consequently, the total coaching time needed is, on average, less than 2 hours. For ‘practice’, readers are asked to read independently for 10 minutes a day, using the overlays. During this whole process, readers are encouraged to read passionate interest texts to acquire advanced reading skills before applying them to more difficult texts. On average, reading excellence is achieved with around 6 hours ‘practice’. Without ‘practice’, there is no progress. It is very much all about learning by doing.

We have been capturing the weekly reading speed improvements (in the context of improved comprehension) for most of our neurodivergent readers since October 2021, presented in Table: 1 and Graph: 1 below.

Table 1

Standard deviation calculations are based on postgraduate readers (non-dyslexic), when reading for meaning (n=355). The ‘average range’ of reading for meaning speeds was calculated as 131-248 words per minute (wpm). This means that 1 Standard Deviation is represented by 59 wpm, and the mean reading speed is 190wpm. At that speed, postgraduates were able to answer, on average, 53% of straightforward questions correctly. The texts used were suitable for 12-year-old children, so they were not challenging for postgraduates. The standardization process demonstrated that postgraduates very rarely use any metacognitive reading strategies, even though they knew they would be tested on their comprehension, and their reading in general, was not as effective as is generally assumed.

Graph 1

We recognise that standardising reading tests is problematic because reading comprehension, in particular, is dependent on a wide variety of factors, not least of which is prior knowledge. Nevertheless, this gives us a baseline framework to work with.

Features of Reading Progress

The first important feature is the large gains achieved by all the dyslexic readers in very little time. After just one week of ‘practice’, the mean and median improvement is 1.6 and 1.5 Standard Deviations (SD) respectively and has already achieved high statistical significance (p<.0000). Very few educational interventions, particularly with reading, achieve 1.5 SD progress in total. The final improvement for median scores is 6.2 SD.

Table 1 and Graph 1 above show progress across 6 categories. The first is the mean progress achieved. We can see that this is very similar to the median progress at the beginning, and then the mean gradually separates from the median progress until it overtakes the mean score at the end. This is explained by some individuals making very dramatic progress early on, inflating the mean score achieved. This can be seen, for example, with the individual student (B), who made the most progress in this cohort.

Consequently, median scores represent a better reflection of progress for most students since the median is the score achieved by the individual in the middle of the cohort. The apparent progress achieved is slightly less dramatic at first but better reflects the experience of most students.

The 4th category is the progress achieved by the individual (A), who made the least progress in this cohort. This allows us to compare the highest achiever with the lowest, but also to identify some of the issues experienced by those who benefit least in terms of reading speed.

Student (A) did not start with a particularly slow reading speed (154 wpm- which is very close to the mean score for this cohort- 150 wpm). The problem was that she had learnt to read at some speed without understanding what she was reading. This is very common among a subgroup of dyslexic students who can use phonics fluently and can appear to rattle off reading, but all their working memory is discharged in that mechanical process, leaving very little for understanding. They have often not learned any strategies at all for monitoring meaning. Indeed, another student in the sample started ‘reading’ at 400wpm, without any idea at all about what they had just ‘read’. Consequently, these students need to learn to slow up their reading and develop better monitoring for meaning strategies and enable reading to become a more visual process. This can be seen by the drop (-0.5 SD) from the first measure of student A’s reading speed and one week later. This particular student, for whom English was not her first language, made 2.1 SD progress overall in reading speed, although most of her real progress involved developing a very large improvement in comprehension, which is not captured by her reading speed score of 273 wpm (SS 122), or top 7% of postgraduate readers.

Weekly Progress

1st Week

After just one week, the median progress is 1.4 SD. This is due to a 55% increase in reading speed (with improved comprehension). In real terms, an initial median of 146 wpm has increased to 229 wpm (standardised scores [SS]: from 89 to 110, or from 23rd to the 75th percentile), while comprehension has improved. This is a shift from towards the bottom of the average range to near the top. The correlation between starting speed and improvement is low (r=0.465) and can be interpreted as explaining just 22% of the improvement.

Before starting Visual Reading, 45% of neurodivergent readers are below the average reading for meaning speed of postgraduate readers (bottom 15%). 9% are above the mean, and just 2% were above the average range (top 15%) of postgraduate readers. We can see the weekly progress of all the neurodivergent readers against key benchmarks in Chart 2 below.

11 of these readers are dyslexic children (8 to 16 years old), and the pattern of scores and progress are strikingly similar to neurodivergent adults. 45% are below the average range, and 9% (n=1) were above the mean before starting Visual Reading.

Chart 2

Weekly Progress

1st Week

After just one week with less than one hour coaching, and reading with the Structured Saccade Overlays for 10 minutes a day (about one hour in total), we can already see a dramatic improvement. Only 15% are still below the average range, 61% are above the mean, 39% are above the average range (top 15%), and 15% are in the top 1%.

14% of the children are still below the average range (bottom 15%), 71% are above the mean, and 14% are in the top 1%.

2nd Week

In the second week, we see a further improvement in median score, leading to a total of 2.2 SD. This represents an increase in median score to 273 wpm (SS 122), or top 7% of postgraduate readers. The correlation between starting speed and improvement is low (r=0.406) and can be interpreted as explaining just 16% of the improvement.

After two weeks with less than 1 hour coaching and around 2 hours of reading using the Structured Saccade Overlays, we see further dramatic improvement against key benchmarks. 0% of the readers are below the average range (bottom 15%), 84% are above the mean, 62% are above the average range, and 36% are in the top 1%.

Similarly for the children, 0% are in the below average range, 55% are above the mean, 27% are above the average range, and 18% are in the top 1%.

3rd Week

In the third week, we see a further improvement leading to a total of 2.6 SD. This represents an increase in median score to 300 wpm (SS 129 or top 3% of postgraduate readers). The correlation between starting speed and improvement is low (r=0.393) and can be interpreted as explaining just 15% of the improvement.

After 3 weeks, 94% are above the mean, 80% are above the average range, and 45% are in the top 1%.

Similarly for children, 71% are above the average range, and 43% are in the top 1%.

4th Week

In the 4th week, we see a further improvement leading to a total of 3.5 SD. This represents an increase in median score to 389 wpm (SS >150), or top 0.1% of postgraduate readers. The correlation between starting speed and improvement is low (r=0.287) and can be interpreted as explaining just 8% of the improvement.

After 4 weeks, 63% are in the top 1%.

For children, all are above the mean, 45% are in the top 1%.

5th Week

In the 5th week, we see a further improvement leading to a total of 4.3 SD. This represents an increase in median score to 400 wpm (SS 155), or top 0.0001% of postgraduate readers. The correlation between starting speed and improvement is low (r=0.279) and can be interpreted as explaining just 8% of the improvement.

After 5 weeks, 70% are in the top 1%.

For children, all are above the mean with 29% in the top 1%.

Final Week

When all the students have completed the course, we see a further improvement leading to a total of 6.2 SD. This represents an increase in median score to 500 wpm (SS 181), more than 3 times faster than at the beginning. The correlation between starting speed and improvement is low (r=0.298) and can be interpreted as explaining just 9% of the improvement.

In the final week, after receiving around 2 hours of coaching, and having read for a week without the need for the Structured Saccade Overlays, all the students are above the average range, 86% are in the top 1%.

Similarly for children, all are above the average range, 73% are in the top 1%.

Conclusion

The impact of Visual Reading with Structured Saccade Overlays is dramatic and consistent. While it can improve the reading effectiveness for any reader, changing a poor reader into a world class reader in less than 2 months is, we would suggest, unprecedented and potentially life-changing. Student feedback confirming individual experiences can be read on our website.

The progress across the group over time is surprisingly linear. This suggests that further coaching could increase reading for meaning speeds even further, and this will inform our work going forward. We have recently confirmed that with children, regular coaching sessions over months lead to improved outcomes and continual progress. However, the most striking feature of these results is just how easy it is for neurodivergent/dyslexic readers to acquire world-class reading skills in a short period of time with little effort. Most of this sample are in Higher Education, but it includes 11 children (8 to 16 years of age), five of whom made some of the most dramatic progress (two achieved 500 wpm, two achieved 800 wpm, and one achieved 1500 wpm).