User:CameronDowning/sandbox

Dyslexia screening tests (also known as dyslexia screeners) are short, evidence based, assessments of cognitive ability that indicate how likely an individual is to have dyslexia. They facilitate mass identification of dyslexia quickly and easily in children and adults. Screeners are commonly used in the United Kingdom and fit into a wider dyslexia assessment and intervention cycle. There are a number of different types of screeners developed by researchers according to their views on the etiology and manifestations of dyslexia. However, providing measurement using screeners is imperfect and alternative methods of identifying dyslexia are used in other countries. Alternate methods include the 'response to intervention' approach used in the United States.

Screening Tests and Assessment
In the United Kingdom, screening tests form part of a bigger assessment cycle. Full assessment administered by an educational psychologist can be costly, thus prohibitive. The most cost effective way of assessing dyslexia is to use a two-stage process. This process involves firstly screening using a low cost, objective test administered by a lay professional (i.e., specialist teacher), which will identify if further assessment is required. A further, more comprehensive, assessment of an individual's abilities/additional needs can then be delivered by a qualified psychologist who will diagnose dyslexia and identify areas for intervention and remedial teaching. Such a process is now common place in education settings.

Psychometric Properties of Screeners
Psychometric properties refer to how well the screener assesses the test-taker's ability and distinguishes those who have dyslexia from those who do not. This result must be consistent. The main properties a screening test must have is good; validity, reliability, sensitivity, specificity and be well normed. The sensitivity, or hit rate, of a screener is the ability to identify those with dyslexia. Whereas specificity, or false-alarm rate, refers to the screener’s ability to identify those without dyslexia. The minimum acceptable levels of sensitivity and sensitivity for dyslexia screeners are 80% and 90% respectively. Norms are values that represent average age scores in a population. They are used to compare individual's scores with the rest of the age-matched population, this identifies if the individual's results are within the expected range for someone of their age.

Types of Screeners
A number of different dyslexia screeners have been produced by researchers and companies working in the field of learning difficulties. Each research group or company's screening test is influenced by different theories of dyslexia's etiology and manifestations. Below is an overview of the main dyslexia screening tests used for screening dyslexia in the United Kingdom.

Bangor Dyslexia Test (BDT)
The Bangor Dyslexia Test (BDT) is a paper based screener produced by Miles (1983). The test was originally produced to provide a brief assessment of dyslexia rather than as a screening test. Test items were produced based on anecdotal evidence of what children with dyslexia found more difficult than children without dyslexia. The test has no theoretical underpinnings as it was produced from anecdotal work, but it does measures cognitive abilities including: short-term memory, working memory and spatial awareness across eight subtests and two questions. The eight subtests include: left and right awareness, subtraction, repeating polysyllabic words, multiplication tables, reciting months of the year (forward and reversed), and digit span tasks (forward and reversed). Two questions examine common confusions and genetic factors. The test has now been translated and normed in several languages including Russian, Welsh and Japanese.

Scoring is based on the total of each subtest's scores as well as observations made during the screening by the test administrator. Each subtest is given either a plus (+) indicating a dyslexic positive result, a minus (-) indicating a dyslexic negative result or a zero (0) indicating an ambiguous result. The + or - for each subtest is then summed to provide an indication of dyslexic tendencies (zeros are counted as a half plus when summing all the subtest's scores). The more positive the score the more dyslexic tendencies the individual has. The test was normed using 12 905 children that were taking part in a separate longitudinal study examining health and educational development. There has been very little published independent research examining the BDT. One publication by Nichols, McLeod, Holder and McLeod (2008), compared the tutor screening battery or TSB (a hybrid of the BDT and the DAST) with the LADS in the screening of 3 specific learning difficulties (dyslexia, dyspraxia and Meares-Irlen syndrome). The TSB contained all of the BDT's subtests apart from the digit span tasks. The authors found that the TSB had good sensitivity (91%) but poor specificity (79%). Caution should be employed in judging the BDT's specificity based solely on this evidence, as the BDT subtests were used in conjunction with another screening test. However, further investigation of the use of the BDT in an adult population may be warranted.

Dyslexia Adult Screening Test (DAST)
The Dyslexia Adult Screening Test (DAST) is based on the phonological deficit and the cerebellar theory of dyslexia. The test takes 30 minutes to complete 11 subtests and gives an ‘at risk’ indication. The 11 subtests include; speeded spelling, reading and writing, postural stability, phonemic segmentation, reverse digit span, nonsense reading, rapid naming, non-verbal reasoning, verbal and semantic fluency. Unlike other screeners, nine subtests measure weaknesses whilst two subtests measure relative strengths (semantic fluency and non-verbal reasoning). The cognitive skills these subtests tap include: speed of lexical access, balance, fluency, reasoning, phonological skill and working memory. Each subtest is scored individually and converted into age normed standard scores ranging from zero to three. A score of zero indicates a normal score whereas a score of three indicates a score 3 standard deviations below the mean. The scores of each subtest are summed and divided by nine to give an ‘at risk quotient’ or ARQ. Further assessment is recommended when the ARQ exceeds 1. The test was normed using 1100 adults nationwide. A study completed by the DAST's authors compared the DAST with a previously established screener (Adult Dyslexia Index; ADI) on detecting dyslexia in a sample made up of 15 people who had dyslexia and 150 controls. The DAST was found to have a sensitivity rate of 94% and a specificity rate of 100%. The DAST was also validated on a Canadian sample of 117 individuals with specific learning disabilities and 121 controls aged 17-41 years old. The authors found that the DAST had sensitivity of 74%, below the recommended guideline of 80% and a specificity of 84%, also below the recommended guideline of 90%. As previously mentioned, another study examined a hybrid of the DAST and the BDT (named the tutor screening battery; TSB) in relation to the LADS. The DAST's subtests that were included in TSB comprised of; nonsense reading, speeded spelling, reading and writing, phonemic segmentation, digit span and rapid naming. The sensitivity of the TSB in predicting specific learning disabilities was 91%, however its specificity was 79% and below the recommended range for a screening test. These results do echo those of Harrison and Nichols (2005) who also found the DAST had a low specificity. Although, caution should be exercised in generalising these findings to the DAST, as the TSB also comprised of subtests taken from the BDT.

===Lucid Adult Dyslexia Screening Test (LADS) ===

The Lucid Adult Dyslexia Screening Test (LADS) is a computer based screener developed by Lucid research Ltd. The rationale behind the LADS is not based upon a singular theory of dyslexia but what are considered to be core cognitive difficulties that dyslexics typically display: phonological processing, lexical access, working memory and speed of processing. The LADS comprises of five subtests. Three of which are speeded and measure the aforementioned core deficits: word recognition, word construction and working memory. The remaining two subtests measure reasoning which gives an indication of intellectual ability. Intellectual ability is measured as it has been found to moderate scores of cognitive tasks typically presented in screeners. This means that intellectual ability may produce false positives or false negatives in screeners (i.e., an intelligent dyslexic will use compensatory mechanisms, thus making diagnosis harder). Each subtest has several items of varying difficulty with discrete discontinuation rules. Item scores are calculated by a process known as algorithmic fractionation’ which categorizes the individual into one of 3 specific groups (no indications, weak indications, strong indications of dyslexia). The higher the score the individual achieves, the stronger the indication of dyslexia.

The LADS was validated in three discrete studies by the test's authors. The first study examined 140 (71 dyslexic) adults across two institutions. The authors report that each of the subtests measuring cognitive skills significantly differentiated between dyslexia and normal readers. The sensitivity was found to be 92%, well within the accepted criteria for a screening test, however specificity was not reported. In the second study, the authors compared the scores of the cognitive skills based tasks with other validated measures of reading ability (Woodcock-Johnson Word Attack, WRAT 3, WAIS-3) in 48 normal readers. The LADS subtests were found to be independent from each other (no inter-correlations) and had good concurrent validity with the established measures. In the third study, 38 students (19 dyslexic) completed the LADS, significant differences between the scores of the dyslexic and non dyslexic groups were found, however the specificity and sensitivity was not reported. An independent study compared the LADS with the TSB (hybrid of BDT and the DAST) on screening 74 participants. The sensitivity of the LADS was found to be 66%, below the minimum acceptability range and it’s specificity was reported as 90%, which is adequate.

Limitations of Screeners
Like all psychological tests dyslexia screeners are not perfect. This fact has been recognised earlier in relation to specific screeners. More general limitations applicable to all dyslexia screening tests include:
 * False positives; where the screener identifies the individual as likely to have dyslexia when they do not, such an outcome can cause unnecessary stress for the individual.
 * False negatives; where the screener identifies the individual as not being likely to have dyslexia when they do, such an outcome would mean the individual does not gain access to further, more comprehensive testing and suitable help (i.e., intervention or remedial teaching).
 * Delay in diagnosis; the individual must fail sufficiently to meet the criteria of the screener in order to be identified as at risk of having dyslexia. However, early intervention is crucial for the successful treatment of dyslexia . Waiting to meet the criteria means the individual could miss that crucial early intervention stage.
 * Static measures of ability; screeners only measure the individual’s ability during that specific test period and so do not provide an ongoing measure of progress, which would provide more relevant information to practitioners providing interventions/remedial teaching

Alternative Approaches to Screening
Early identification and intervention is advocated in tackling dyslexia. Screening tests have enabled the mass testing of those at risk, but at the cost of making individual predictions much harder without further specialist assessment which is often long and costly. One alternative approach to screening is the more dynamic ‘response to intervention’ (RTI) system developed in the US. RTI involves a stepped series of interventions based on failure to respond to high quality teaching/previous intervention. This allows those having reading difficulties to be identified early, and given access to intervention/additional help more quickly. A similar approach was recommended by Sir Jim Rose (2009) from his review of identifying and teaching those with dyslexia and literacy difficulties in England. Rose (2009) recommends a graduated, or wave, approach whereby young people identified as being behind would receive high quality main stream education adapted for slow learning in the first instance. Failure to respond to this results in the individual receiving a brief small group or one to one remedial intervention delivered by a teaching assistant. If no improvement follows, the young person would gain an individualised intervention with specialist external consultation/assistance (i.e., specialist teacher/educational psychologist/speech and language therapist). Where necessary the individual would gain a statement of special educational needs. Such an approach removes some of the limitations presented by dyslexia screening tests. Mainly the graduated system allows those who are struggling to gain access to additional help early on, rather than waiting to meet the diagnostic criteria of a screener. Such a approach allows continuous measurement of progress rather than relying on a static screening measure.