GB2445997A

GB2445997A - Comparative reading speed test (CRST)

Info

Publication number: GB2445997A
Application number: GB0624428A
Authority: GB
Inventors: Geraint W Griffiths
Original assignee: Individual
Current assignee: Individual
Priority date: 2006-12-07
Filing date: 2006-12-07
Publication date: 2008-07-30
Also published as: GB0624428D0

Abstract

The test uses letter recognition rather than sentences or individual words or numbers. It uses a chart consisting of two paragraphs of N10 sized letters each 13 lines long and single-spaced, printed on a sheet of A4 paper. One paragraph is printed with a serif (e.g. Times) font in black and the other is printed with a non-serif (e.g. Gill Sans) font in teal (a blue green hue) giving a lower contrast with the white background. Each line contains 60 characters, which are arranged to look like words in groups of between one and seven. It has been found that the rate of character recognition varies according to the font being read and the contrast of the letters against their background. When reading speed is measured in this way indications of reading difficulty such as dyslexia related to visual correction and problems with visual development such as amblyopia can be found.

Description

* -III, The Comparative Reading Speed test (CRST)

Description

The comparative reading test (CRST) is a test for reading-speed, which uses letter recognition rather than sentences or individual words or numbers. It compares reading speed with two different types of font; one at high contrast with a serif typeface and another with reduced contrast and a non-serif typeface.

It is printed on an A4 sheet of 8Ogms white paper. It consists of two paragraphs of characters arranged to look like words, which are in fact a random sequence of individual letters in groups varying in size from one single letter to groups of 7.

The size of the letters is Nb. There are 60 letters in each line and 13 lines single-spaced, in each paragraph. The top paragraph consists of a "Times" font and the bottom paragraph a "Gills Sans" font. The times font is printed in black to give a high contrast between the letters and their white background. The Gill Sans font is printed in Teal, which has a green blue hue and gives a reduced contrast between the letter and its background. The whole is laminated in a non reflective plastic sleeve to keep it clean and prevent reflections and also to make it easier to hold in one hand and keep steady while being read (See fig 1 best illustration -The comparative reading speed test CRST).

Using the card the subject is timed reading the 5th line down in each paragraph, indicated by a red letter at the beginning and end of each line. Reading in the middle of the paragraph increases the sensitivity of the test by inducing a crowding phenomenon, where poor visual development is present. The stopwatch is started as the first character is read and stopped when the last one on the line is read.

Typically times will be faster on the non-serif Gill Sans print when the high contrast serif typeface causes a pattern glare effect due to binocular instability.

Pattern glare is reduced by the less complex non-serif typeface. Less commonly speeds will be reduced on non-serif typeface when poor visual development has reduced visual acuity and the loss of contrast is felt more acutely. In faster normal readers loss of contrast may simply make things harder to see in a normal sense. In some subjects reading speed will be affected by a combination of pattern glare and poor visual development, the significance of the reading speed will then be interpreted in the light of other diagnostic tests.

Important signs of reading difficulties will also be noted at the time of measuring reading speed, including head postures, following text with the finger, heightened anxiety, excessive body movement and phonetic enunciation of characters.

Introduction

Optometric research in Sport Vision has identified two primary measures of visual performance Aiming and Anticipation (based on depth perception). It has been proposed that all sports and other occupations use these two primary skills to a greater or lesser extent.

In sport for example research showed that the occupation of tennis relies mainly on depth perception, which allows the player to judge how far the ball is away and therefore to anticipate its arrival at the racquet face. In reading depth perception is not so important and therefore aiming predominates. If a particular occupational task depends on aiming then the stability of the aiming or dominant eye is important, for instance words need to be read in sequence if the sentence is to make sense. An unstable dominant eye would cause the words to appear to jump on the page and the reader to loose the place on the page. A measurement of eye dominance was the key test introduced into the original sport vision research. 131.

This reading hypothesis has been tested in the Moreton Project (Report attached) along with several others based on a series of diagnostic tests indicated by the research in sport (reprint attached). The Moreton project took place at a school in V\klverhampton in the UK where 95 year 7 students were screened using a battery of tests including the Comparative rate of reading test to determine the predisposition of this group to dyslexia and its associated disabilities (Dyspraxia, Dysgraphia and dyscalculi) Measurement of reading speed is not taught or routinely measured in optometric practice. However reading speed is beginning to be used as a measure of the effect of tints for example. It is often reported by patients that print is easier to see at high contrast or in different colours and backgrounds and it is anecdotally common for diagnosed dyslexics to say that uncomplicated type faces like Gill Sans and Anel are easier to see.

This suggests that reading speed is likely to be an important test in the standard eye examination for children (and adults), but what is actually read may be equally important. Real sentences may measure ability to extrapolate and guess rather than an ability to see and individual words can be recognised without seeing the individual letters that make them up (See Fig 2 Typoglycaemia).

Certain typefaces and font colours may suit some people more than others. A test for reading ability should not be measuring reading speed but the speed of individual character recognition and these characters should be letters rather than numbers, to reduce the chance of guessing (26 letters to choose from as opposed to 9 numbers).

To cover the range of reading skills, two reading or character recognition tasks are proposed which vary contrast and type face at a normally comfortable angular subtense (font size): * Times (a serif N 10 optoype high contrast black on white) * Gill Sans (non serif N10 low contrast teal (a blue green colour), optotype -4.

Method Reading speed is measured in two stages. First the time is recorded to read one line of 60 characters imbedded in the top paragraph (the 5th tine down); this is the high contrast serif optotype (Times). Then the subject reads a second line in the bottom paragraph (5th tine down) and the time is recorded; this is the low contrast teal colour non-sent optoype (Gill Sans).

In addition to recording the times any distinctive behavioural traits (clinically significant signs) should be noted, these might include: * Following the text with the finger or thumb * Marking the beginning of the line with the thumb * Reading characters inaccurately * Missing out characters * Re-reading or loosing the place * Head postures (eg head tilt or chin elevation) * Heightened anxiety, excessive body movement * Indications of tension in body language or breathing * Gripping and re-gripping the test card * Facial contortions * Voice or body tremor * Saying the characters phonetically * Saying the characters phonetically for half the line * Inability to read any or more than a few characters * Test card held too close, (closer that 10cm) * Test care held "too far" away (greater that 30cm) * Rapid fatiguing * Big sigh when reading finished Reading Speed A comparison of reading speeds in this group of 11 year old children confirms that the typeface and or contrast has a significant effect (see table 1 Comparison of reading speed with two different optoypes at high and low contrast).

Difference Between High and Low Contrast The hypothesis was that a simple typeface like Gill Sans would always be easier to read for children with reading difficulties. In fact 12 (12.9 %) subjects read the simpler low contrast teal typeface more slowly. 17 (18.3%) showed no change and the majority 64 (68.8%) read it faster (See Table 2 Increase and decrease in reading speed at low contrast) The mean high contrast reading speed (black uTimesn typeface) of students who times were worse or stayed the same at low contrast, was 33.1 seconds. When compared with the high contrast speed (39.14) of those who were faster at low contrast they were significantly different (p=0.0023).

Distance Vision Distance vision was another test in the Morton study, which showed a close relation to reading speed. This was measured using the Balie Lovie LogMar high and low contrast charts (see fig 3 L0gMAR high and low contrast charts). Unlike the reading chart the low contrast is achieved in the distance charts by fading the black characters to grey to a measured contrast difference between the letters and their background of 10%. Research has shown that vision measured at high contrast becomes disproportionately worse at low contrast in subjects with visual deficiency.

When the following two near reading groups were isolated: 1. All with a slower reading time on the teal Gill Sans type 2. Increased speed of greater than 5 seconds on the teal Gill Sans type face, The significance of the difference between high and low contrast logMAR visions is greater with group one (p=O.00025) than group two (p=0.00041). The relatively poorer visual performance at low contrast with the first group was against expectations in a generally faster reading group.

Key Stage 2 Perhaps the most important part of the new test of comparative reading speed is its relation to key stage results (in this case to stage 2).

The students were given a grade according to their SSATs results in English, Maths and Science. There were five levels from B the lowest to 5 the highest Levels 3 to 5 are subdivided into low, medium and high. To make this data statistically useful, each grade was given a score and the scores for each subject were totalled (see Table 3 Key stage grades and scores).

Wien reading speed with the serif optotype (high contrast) was compared with key stage 2 scores, the negative correlation was highly significant (r = -0.408, p 0.0005, N=62). At low contrast with the non-serif typeface (teal low contrast), correlation is still significant but less so (r =0.298 p = < 0.025) Eye Examinations Full eye examination were carried out on 14 subjects identified by their reading speed and other tests, to be at risk of reading problems related to visual deficiency.

Following the tests and during the initial period of spectacle wear, mean reading speed almost doubled (See Table 4 Measures of reading speed over the testing period).

One student achieved an increase of over 500% in his reading speed from being barely able to read at all, to one of the fastest. His general health, skin colour and demeanour (physical tremor) were very much improved. His behaviour also improved beyond recognition, from a very disruptive influence and a real problem for the school, to head student potential. The reading speed of all the children given spectacles improved from the initial screening.

A number of signs amenable to optical correction were identified and could be considered to be predisposing factors in learning difficulty (see table 6 Optical signs which predispose to learning difficulties)

Conclusion

Reading speed is not routinely measured in practice and it is difficult to say why, other than previously held views that optometric interventions could not affect it or that it was sufficient to measure vision at near without any attention to speed and accuracy. It may also have been, as this submission suggests that the reading task and its direct relation to academic achievement had not been fully specified The results show that reading speed using the Times optotype is highly correlated to Key stage two results (p=O.0005, N=62). Intuition supports the importance of this result -how can you be good at English if you can't read, how can you add up figures if they won't stay still. Perhaps over reliance on science can sometimes be a barrier to common sense. It may be that the only thing holding up many of the tower achievers is the ability to read comfortably and accurately.

The results in this study show that reading speed is affected by contrast and typeface (p=O.0021), even before visual deficiency is taken into account. Any measurement of speed therefore has to take into account what is being read and levels of illumination.

A precursor of a decrease in reading speed using the low contrast teal Gill Sans typeface, is a relatively fast reading speed using the "Times" high contrast black on white typeface. It may be that the "Times" type causes more pattern glare when there is binocular instability. When good readers are slower reading Gill Sans it is just because of loss of contrast in the teat Gill Sans type-face.

Individual inspection of the results however shows that there were also some poor readers whose times got worse at low contrast. The test may be identifying another characteristic, visual deprivation, which has induced a degree of amblyopia, in the presence of good binocularity. This could explain why the poorly performing students are more susceptible to loss of contrast.

The use of different typefaces at high and low contrast may distinguish four separate groups: 1. Visual deprivation and amblyopia in the absence of reduced binocularity Readers slower that 46 seconds on the uTimes high contrast type face who get even slower on teal low contrast "Gill Sans" type face 2. Pattern glare induced by poor binocularity Readers slower that 46 seconds on the "Times" high contrast type face, faster by over 5 seconds, reading teal low contrast Gill Sans.

3. Routine follow-up Readers with times at high contrast between 27 and 46 seconds 4. High Achievers Readers faster than 27 seconds at high contrast who are very likely to be achieving well in school This categonsation does not definitely exclude any or another form of deficiency from any group. Decisions about the management of a particular pupil would be made with reference to the other screening tests.

In conclusion the ability to read may the only thing that is holding up the majority of lower achievers in the school SSATs results tO Fig I Best Illustration Reading Speed w ivog ski qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe zlza or oah azlz efjkf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz lvptj anaxai xl) rzs agqo jfx iaxana jtpvl zb wr f lsns euouv qeb wtz juq kfc uz bhfrk xw ouju oqga szr w ivog ski qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe ziza or oah azlz efjkf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz Ivptj anaxai xlj rzs agqo jfx iaxana jtpvl zb wr f lsns euouv qeb wtz juq kfc uz bhlrk xw ouju oqga szrw ivog ski qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe ziza or oah azlz efjkl gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz Ivptj anaxai xl) rzs agqo jfx iaxana jtpvl zb wr f Isns euouv qeb wtz juq kfc uz bhfrk xw ouju oqga szrw ivog ski qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe ziza or oah azlz efjkf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz lvptj anaxai xfl rzs agqo jfx iaxana jtpvl zb wr f tsns euouv qeb wtz juq kfc uz bhfrk xw ouju oqga szr w ivog skr qgfo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe ziza or oah azlz efkf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz lvptj anaxal xi as agqo jfx iaxana jtpvl zb wr f Isns euouv qeb wtz juq kfc uz bhfrk xw ouju oqga szrw ivog skr qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe ziza or oah azlz efjkf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro ujuo wx kfrhb zu cfk quj ztw beq vuoue znsl f rw bz lvptj anaxai xfl rzs agqo jfx iaxana jtpvl zb wr f Isns euouv qeb wtz juq kfc uz bhfrk xw ouju oqga szrw ivog skr qglo ufdtnt atsj qcv okhlv xuzsqdy qa uxeq hyg fkjfe zlza or oah azlz ef)kf gyh qexu aq ydqszux vlhko vcq jsta tntdfu olgq rks govi w hao ro Record Time to read the top line * Black print * Blue print Phonetic Yes/no Fig 2 Typoglycaemia I cdnuolt biveiee taht I cluod aulaclty uesdnatnrd waht I was rdanieg The phaonmneal pweor of the hmuan maid Aoccdmig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer inwaht oredr the Itteers in a wrod are, the olny iprmoatnt tihng is taht! the frist and Isat Itteer be in the rghft pclae. The rset can be a taotl mses and you can situ raed it wouthit a porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey Iteter by istlef, but the wrod as a wlohe. Amzanig huh? and I awlyas thought sipeling was ipmorantt....

Table 1 Comparison of reading speed with two different optoypes Optotype Average Reading speed High contrast "Times" black print 37.35 (SD 9.4) Low contrast Gill Sans" teal print 33.65 p=O.0021, N=95 Table 2 Increase and decrease in reading speed at low contrast Number % Decrease in speed 12 12.9 No change 17 18.3 Increase in speed 64 68.8 ILt-Table 3 Key stage grades and scores Grade B N 2 3 4 5 L,M,H n/a n/a n/a L M H L M H L M H Scorei2 34 5 6 7 8 9 10 11 12 Is Table 4 Measures of reading speed over the testing period.

Time of measurement At screening 24.11.05 27.4.06 Reading speed 36.04 30.5 28.6 p 0.038 0.0079

Claims

Claims 1. A reading test, which measures the speed of character

recognition using two different fonts, serif (iimes) and non-serif (Gill Sans).
2. A test as claimed in claimed in claim 1 wherein each font has a different contrast
3. A test as dairned in claimed in claim I wherein low contrast is achieved by using a teal hue for the non-serif font
4. A test as claimed in claimed in claim I wherein the characters are letters not numbers
5. A test as claimed in claimed in claim I wherein there are two paragraphs of different optotype characters each containing 13 lines with 60 characters on each line
6. A test as claimed in claimed in claim 1 wherein the 5th line in each paragraph j identified by the first and last character being red.
7. A test as daimed in claimed in claim I wherein significant secondary clinical signs include finger following head posturing and phonetic enunciation of characters
8. A test as claimed in daimed in claim I wherein reduced reading speed due to visual deprivation and induced amblyopia is identified
9. A test as claimed in claimed in claim I wherein reduced reading speed due to binocular imbalance is identified
10. A test as claimed in claimed In claIm 1 wherein high achievers whose greater potential may be limited by visual deficiency, are identified
11. A test as claimed in claimed in daim I wherein two different type-faces are used to differentiate between reading difficulties caused by binocular instability and visual deprivation.
12. A test as claimed in claimed in daim I wherein poor SSAT's results and reduced key stage 2 achievement can be predicted
13. A test as claimed in claimed in daim I wherein distance vision is directly related to reading speed
14. A test as claimed in daimed in daim 1 wherein high achievers in key stage 2 children can be predicted