zoom controls
dimensions graphic
page iconlink to results page 1link to results page 2link to results page 3link to page 4
Dimensions of dyslexia:
enquiry results
link to site maplink to home pagelink to enquiry formlink to blog
INTRODUCTION & DISPLAY 1: Data collected by the QNR has been summarized in Excel to generate correlation coefficients between pairs of datasets, also displayed as scatter plots.

dimensionsThe QNR collected data about the prevalence and frequency of attributes, that is, dimensions of dyslexia encountered by dyslexia support professionals in their interactions with students. A link to the QNR was sent by e-mail to all UK Higher Education institutions listed on the Universities UK database. The e-mail was directed to each university's respective student service for students with dyslexia where this could be established from universities' webpages (which was most of them) or otherwise to a more general university enquiries e-mail address. In total, 120 e-mail requests for participation were sent out. 30 replies were received upon which the analysis below if founded.

The QNR listed 18 dimensions and these were established by drawing guidance from two widely used Adult Dyslexia Checklists (Vinegrad, 1994 and British Dyslexia Association, 2002) and adapting these into this QNR where appropriate, based on the researcher's own experience in working with students with dyslexia in HE. The comparison of referral items used in dyslexia screening tests (Rice & Brooks, 2004) has also been consulted and so it was felt that the resulting list of 18 dimensions in this QNR would present a good cross-section of attributes that dyslexia support professionals are likely to encounter in their university contexts. It is acknowledged however, that the list is unlikely to be exhaustive and to this end, a free-writing section was included in the QNR form which invited respondents to list any other attributes that they may have regularly encountered in their interactions with students that were absent from the QNR's list. The results of this are dicussed below the results table.

The analysis and results of this small-scale preliminary enquiry will contribute to the design of the final section of the project's main research QNR.


checklistA good deal of research from the last two decades that discusses dyslexia in adults appears to rely on the Adult Dyslexia Checklist developed by Vinegrad (1994), although the original source of the research is proving hard to find. All that is known so far is that the checklist was constructed from 679 questionnaire responses from adults in the age range 18 - 68 of which more than half were undergraduate students. The research sample included 32 adults who were known to have dyslexia.

A similar checklist for adult dyslexia developed and published by the British Dyslexia Association seems to rely on research evidence from Smythe & Everatt (2001) but the research study from which this was developed also seems to be out of the accessible domain. These checklists seem widely used so it is surprising that this is happening almost on trust that the research that underpins them both is robust. At the time of writing, attempts are being made to contact the authors of each checklist to obtain copies of their original research studies. This point is also discussed in the StudyBlog post that reports more fully the results and analysis of this QNR.


dimensions graphicIn the QNR, each stem statement refers to one dimension of dyslexia. 18 dimensions were presented and respondents were requested to judge the frequency that each dimension was encountered in interactions with dyslexic students as a percentage of all interactions with dyslexic students. For example in the stem statement: "students show evidence of being disorganized most of the time" a respondent who judged that they 'see' this dimension in 80% of all their dyslexic student interactions would return '80%' as their response to this stem statement.

The table below presents the complete set of 18 dimensions explored in the QNR. Mouseover each icon to reveal the stem statement that the icon represents. The value in each cell is the Pearson Product-Moment Correlation Coefficient: r, between the respective dimensions' datasets with these values being calculated in an Excel spreadsheet summary of the data, rounded to 3 significant figures. So for example, the value of 0.469 in the first cell is the correlation coefficient between the dimensions "spelling is generally poor" and "shows anxiety when asked to read aloud". (The data is located in a protected folder - request access here).


scatterplot abstractThe correlation coefficient displayed in each table cell is also hyperlinked to the respective scatter diagram showing the relationship between the two dimensions graphically. This is useful because it presents the actual data points used in the calculation for the correlation coefficients but the scatter diagram also easily shows whether the datasets contained significant outliers. In all cases, the scatter diagrams were created from the complete datasets without outliers being removed if they occured. In some cases significant outliers did arise and where this occured the outliers were removed to explore the impact that this had on the correlation coeffiecients. Where this was judged to be worth reporting, the adjusted correlation coefficient is included in the lower quadrant of the table with the outlier(s)-removed value asterisked * and the original-data value in parentheses( ). It can be seen that in several cases, removing the outliers has produced a re-calculated correlation coefficient that is substantially different.

The colour coding of the cells in the table provides an 'at a glance' indication of the level of correlation between the respective pairs of dimensions. There appears to be no definitive criteria for setting correlation level boundaries for indications of strength although 'rule of thumb' guidelines suggest that r > 0.6 indicates a 'strong' positive correlation, values between 0.4-ish to 0.6 suggest moderate strength of association and values between 0.3 (or even 0.2 in some sources) and 0.4 indicating weak correlation between the variables. With this in mind, boundaries for the correlation coefficient, r, set for these datasets are more finely defined so that the data table below is easier to understand:

0.3 - 0.39... 0.4 - 0.49... 0.5 - 0.59... 0.6 - 0.69... 0.7 - 1
Dimension spelling bee gantt speaking disorganized words clock text writing mindmap memory compass think big problem solving lists lefts and rights confused writing systematic reading aloud
reading aloud 0.469 0.503 0.111 0.271 0.341 - 0.038 0.139 0.323 0.000 0.136 0.013 0.088 - 0.046 0.302 0.193 0.552 0.382  
systematic 0.317 0.482 0.190 0.541 0.532 0.211 0.369 0.571 0.213 0.407 0.369 0.241 0.315 0.547 0.238 0.310    
confused writing 0.437 0.211 - 0.012 0.419 0.451 0.314 0.079 0.352 0.213 0.258 0.236 - 0.007 0.029 0.336 0.420      
lefts and rights 0.073 0.342 0.245 0.591 0.299 0.554 0.302 0.130 0.454 0.383 0.538 0.320 0.523 0.386   0.515*
lists 0.365 0.062 0.115 0.350 0.543 0.249 0.058 0.413 0.146 0.133 0.222 0.368 0.268       0.677*
problem solving - 0.018 0.292 0.508 0.418 0.104 0.219 - 0.047 0.339 0.469 0.483 0.638 0.363         0.464*
think big 0.071 0.230 0.469 0.375 0.329 0.361 - 0.070 0.001 0.028 0.102 0.439   0.523*
compass - 0.044 0.207 0.511 0.485 0.238 0.496 - 0.117 0.137 0.495 0.376   0.572*
memory 0.070 0.356 0.159 0.389 0.108 0.112 0.196 0.370 0.333       0.529*
mindmap 0.079 0.079 0.169 0.165 0.163 0.335 - 0.012 0.348   0.197*
writing 0.572 0.290 0.107 0.436 0.261 0.184 0.431   0.459*
text 0.221 0.337 - 0.081 0.452 0.247 0.232                        
clock 0.197 0.284 0.226 0.733 0.263       0.430*
words 0.230 0.164 0.231 0.382                     0.543*
disorganized 0.314 0.613 0.305     0.779*
speaking 0.002 0.204                   0.684*
gantt 0.417     0.696*
spelling bee 0.761*

A commentary on the dimensional relationships that may be indicated is available on the project StudyBlog which includes a reflection on how the analysis and results of this small-scale enquiry may impact on the development of the main research QNR, which is scheduled for deployment during this academic year.

Additional dimensions:
  • The QNR provided an option for respondents to list other dimensions of dyslexia that they may have encountered but which weren't listed in the questionnaire. The summary table below lists the additional dimension reported together with the % of dyslexic students presenting that additional dimension. Some additional dimensions were reported by more than one respondent so the respective %s are indicated in the table. Some respondents reported other dimensions but provided no % prevalence. Where more than one respondents reported the same or similar dimensions but used slighly different descriptors the gist of that dimension is given:

    Additional dimension reported % of students reported as presenting this additional dimension

    poor confidence in performing routine tasks 90 85 80 no % reported
    slow reading 100 80 no % reported  
    low self-esteem 85 45    
    anxiety related to academic achievement 80 60    
    pronunciation difficulties / pronunciation of unfamiliar vocabulary 75 70    
    finding the correct word when speaking 75 50    
    difficulties taking notes and absorbing information simultaneously 75 no % reported    
    getting ideas from 'in my head' to 'on the paper' 60 no % reported    
    trouble concentrating when listening 80      
    difficulties proof-reading 80      
    difficulties ordering thoughts 75      
    difficulties remembering what they wanted to say 75      
    poor grasp of a range of academic skills 75      
    not being able to keep up with note-taking 75      
    getting lost in lectures 75      
    remembering what's been read 70      
    difficulties choosing the correct word from a spellchecker 60      
    meeting deadlines 60      
    focusing on detail before looking at the 'big picture' 60      
    difficulties writing a sentence that makes sense 50      
    handwriting legibility 50      
    being highly organized in deference to 'getting things done' 25      
    having to re-read several times to understand meaning no % reported      
    profound lack of awareness of their own academic difficulties no % reported      

Dimensions' relative rankings

  • The graphic below shows the relative rankings of all 18 dimensions according to the prevalence of each dimension being presented.
    So for example: taking the dimension 'students show evidence of having difficulty putting their writing ideas into a sensible order' the mean average prevalence of this dimension reported by the 30 respondents who replied to the QNR was 75.7%. This is being interpreted to mean that on average, this dimension is seen in 75.7% of dyslexic student interactions.
  • The number of student interactions, n, in each dimension has been backward-calculated from the mean average % prevalence of that dimension reported from the 30 respondents who returned replies to the QNR. So this has been extrapolated to a theoretical total number of interactions with students with dyslexia, n = 3000, this being 30 QNR respondents x 100 student interactions each, which for the dimension 'students show evidence of having difficulty putting their writing ideas into a sensible order' generates the value of n = 2271 interactions for this dimension.
    That is, were the 30 respondents to engage in 3000 interactions with students with dyslexia altogether and between them, on the basis of their QNR replies we might expect 2271 of these interactions to present students who 'show evidence of having difficulty putting their writing ideas into a sensible order'.
  • One point that should be clearly noted is that any one interaction, more than one dimension of dyslexia may be presented so the numerical extrapolation outlined above should be treated guardedly. The aim is to add an element of interpretive meaning to the results and to the graphic shown below. It should also be noted that 'student interactions' is not the same as 'number of students'. So we cannot necessarily say that, in the example used here, '75.7% of students 'show evidence of having difficulty putting their writing ideas into a sensible order''. This might be the case, but these results to not enable us to confidently draw that conclusion, not the least because any one student may have requested help on any number of occasions.
  • We should note that this QNR has provided data about the prevalence of these 18 dimensions of dyslexia, not from a self-reporting process amongst dyslexic students, but on the observation of these dimensions occuring in interactions between professional colleagues working with dyslexic students in HE institutions across the UK. The QNR did not ask respondents to state the number of student interactions on which their estimates of the prevalence of dimensions were based over any particular time period, but it might be safe to assume that the total number of interactions on which respondents' estimates were based is likely to have been substantial. An alternative view of this data, presented using the animated bullet-chart format that will be used to display Academic Behavioural Confidence and Dyslexia Index on the main research QNR radar charts, is available here.

dimensions of dyslexia rankings

Rice, M., Brooks, G., 2004, Developmental dyslexia in adults: a research review
. London, National Research and Development Centre for Adult Literacy and Numeracy.
Smythe, I., Everatt, J., 2002, The Dyslexia Handbook, Reading, British Dyslexia Association