Interpreting LADS Plus results
The Reports screen
In the LADS Plus Software User’s Guide (available on the LADS Plus CD) explains how the Administrator can access the Reports screen. The Administrator’s password will be required. This is not only to prevent persons being tested from accessing other people’s results, but also to ensure that a person being tested does not access their own results and misunderstand them.
The report is composed of an upper left-hand white panel showing the person’s results on each of the three dyslexia-sensitive tests, the scores for which can range from 1 to 9. The categories used are as follows:
These are depicted on the reports screen in graphical form: a red bar signals strong indication of dyslexia on that test, an amber bar signals weak indication of dyslexia on that test, and a green bar signals no indication of dyslexia on that test. In other words, the higher the score on each assessment module in LADS Plus, the higher the probability that the person has dyslexia. If a test has not been completed, no bar will be shown against the name of that test.
To the right there is another white panel showing the person’s result on the two reasoning tests, which gives a fair estimate of the person’s verbal and nonverbal intelligence. This is also depicted in graphical form (a blue bar for nonverbal reasoning and green bar for verbal reasoning) on a five-point scale as follows: ‘low’ (bottom 10% of adult population); ‘below average’ (next 15%); ‘average’ (middle 50%); ‘above average’ (next 15%); and ‘high’ (top 10%).
At the bottom of the reports screen is a third white panel containing the overall classification in terms of the probability that the person has dyslexia. (‘high’, ‘moderate’, ‘borderline’ or ‘low’) together with a brief description of the results. Note the classification and description will only be shown if the person has completed all five of the LADS Plus tests. In cases where it has not been possible for the individual complete all five tests see Section 4.2.7 for advice on interpretation.
The Administrator may add his or her own comments to a report and these can be included in the lower half of the individual print-out for that each person. Clicking on the Print button takes the user to the LADS Plus Print Preview screen, through which a print-out may be obtained for each individual. Clicking on the Testing progress button on the Administration menu will bring up a simple spreadsheet depicting which of the assessments each registered person has completed. This can also be printed out as a useful record. For further information on printing see the LADS Plus Software User’s Guide.
Understanding the overall classification
The overall classification, in terms of the probability that the person has dyslexia (‘high’, ‘moderate’, ‘borderline’ or ‘low’), is shown on the reports screen and printed report. This classification will only be given if the person has completed all five of the LADS Plus tests; otherwise, it will ‘Report unavailable as assessment incomplete.’ The classification algorithm is based on the pattern of LADS Plus scores, which have been calibrated against known dyslexic and non-dyslexic cases. Although research has shown that LADS Plus achieves a high degree of classification accuracy, Administrators are advised to check the classification against the individual’s profile on the three tests. This is explained in Section 4.2.5.
LADS Plus has been developed using extremely careful psychometric and statistical analyses to make it as accurate as possible. This scientific development process is described in detail in Chapter 2. However, it is important that those administering LADS Plus or interpreting its results appreciate that no screening tool can be 100% accurate, and consequently occasional misclassifications can occur. It is helpful, therefore for administrators to understand the probability of the test making a misclassification.
In developing screening tests there tends to be a trade-off of false negatives (i.e. dyslexics who are wrongly classified as ‘not dyslexic’) against false positives (i.e. individuals who are not dyslexic who are wrongly classified as ‘dyslexic’). This has already been outlined in Section 2.2.1; for further discussion see Singleton (1997a). In LADS Plus, both false negatives and false positives are well within acceptable levels. There comes a point in test development at which it is not usually possible to improve the measures in the test further in order to reduce one type of classification error without at the same time increasing the other. However, since LADS Plus has been designed to help institutions and organisations identify adults with dyslexia so that they can be provided with appropriate support, it was sensible to try to minimise false negatives rather than false positives, so that the smallest possible number of dyslexics are overlooked.
Probability of dyslexia explained
If a person’s overall classification places them in the ‘Low probability of dyslexia’ category, there is a 95% probability that they are not dyslexic.13 If a person’s overall classification places them in the ‘High probability of dyslexia’ category, there is a 95% probability that they are dyslexic. If the person’s overall classification places them in the ‘Moderate probability of dyslexia’ category, there is a 90% probability that they are dyslexic. These individuals appear to experience less extreme dyslexic symptoms, or are very well compensated, and might be regarded as having milder dyslexia.
On the other hand, if a person’s overall classification places them in the ‘Borderline’ category, then there is a 3:1 chance that they will not be dyslexic. In fact, in most cases of this type the individual will have mainly green or amber scores for the LADS Plus tests, but an occasional red score (on the Working Memory test) may occur from time-to-time. Although this implies that is would be safer to advise the person that they do not have dyslexia, such action would mean overlooking a substantial number of dyslexic cases that might otherwise have been helped. In such cases, it is strongly advised that the Administrator checks the classification against the individual’s profile on the three tests before giving feedback to the person who was screened. This is explained in section 4.2.4.
13 Please note that the probabilities given in this chapter are based on data obtained in Validation Study A (Section 2.4.1). The calculations were not based on the LADS Plus composite score, but on an algorithm that took into account the pattern of red, amber and green scores, which proved to be more robust in the context of clients of very low or very high general ability.
The 'borderline' category
The ‘Borderline’ category is used by LADS Plus when the person being screened has revealed in the LADS Plus tests some difficulties that could be due to dyslexia, but the LADS Plus profile overall was not marked enough to make a clear categorisation as ‘dyslexic’. Basically, this is a signal to the Administrator to look more carefully at (a) the results on the separate LADS Plus tests, and (b) the person being tested. In the Validation Studies reported in Chapter 2, about 75% of borderline cases were found not to be dyslexic.
When an individual has been rated as ‘Borderline’ by LADS Plus, the Administrator should first inspect the results of the separate tests. The most typical reason for an individual being given a ‘Borderline’ rating is because they have performed poorly on the Working Memory test but satisfactorily on the Word Recognition and Word Construction tests. This will show up as a green score for Word Recognition and Word Construction and a red score for Working Memory. Some highly compensated dyslexic adults (particularly if they have received specialist tuition or if they read a great deal) can show this profile (see Section 2.2.2 for discuss of ‘compensation’). Also, brighter dyslexics are usually able to develop more compensatory strategies, so if the Reasoning score is above average or high, then it would be appropriate to suspect that this is the case. Note also that LADS Plus does not use the ‘Borderline’ rating if the Reasoning score falls into the ‘below average’ or ‘low’ category. This because a proportion of people with below average or low reasoning ability tend to experience slight difficulties on the other LADS Plus tests even though they do not have dyslexia. To use the ‘Borderline’ rating in such cases would introduce unreliability into the decision process and create unnecessary problems for Administrators. In some cases it may be possible to resolve the problem by investigating the person’s educational history. Evidence of difficulties in literacy when at school would support a conclusion that the person was probably dyslexic. Similarly, if the person being screened on LADS Plus has a child (or children) formally diagnosed as having dyslexia or specific learning difficulties, then there is a greater likelihood of the ‘Borderline’ rating indicating dyslexia.
Secondly, when making judgements about individuals who fall into the ‘Borderline’ category, accuracy can be enhanced by utilising any relevant additional information about the person. In cases where the person has been referred for screening or assessment specifically because of difficulties in literacy and/or studying, then if that person is classified as ‘Borderline’ by LADS Plus it will usually be safe to assume that the person does have dyslexia, and take action accordingly. On the other hand, if the LADS Plus results derive from a general screening of one or more unselected individuals (e.g. on-entry screening to college) unless there is additional evidence that would point to dyslexia, it would be prudent to assume that individuals classified by LADS Plus as ‘Borderline’ do not have dyslexia. There are a number of ways in which helpful additional evidence may be sought, including use of adult tests of word reading and spelling (e.g. WRAT3 Reading, WRAT3 Spelling14) or by using the Adult Dyslexia Checklist, a copy of which is provided in Section 7.2. If the person is found to have significant difficulties in literacy and/or a large number of problems of a dyslexic nature revealed by the Adult Dyslexia Checklist, it will usually be safe to conclude that they probably have dyslexia. Note, however, that because responses on the check list are highly subjective and it is vulnerable to falsification it is not recommended that the check list is used as the sole means of identifying adults with dyslexia.
It can be seen that the ‘Borderline’ category serves the purpose of drawing the Administrator’s attention more closely to that individual and their results, and in most cases it should not be inordinately difficult to make a judgement about whether or not that individual is likely to have dyslexia. If in doubt, the Administrator should seek advice from someone who has more knowledge of dyslexia and/or experience of working with dyslexic adults.
14 These tests are available from the Psychological Corporation (see Section 7 for address).
Checking the profile of scores from individual tests
In addition to providing an overall classification in terms of the probability that the person has dyslexia, the LADS Plus report screen and print-out also gives a description of the results. In most cases this will be more than adequate to enable the Administrator to decide on the most appropriate course of action. However, there may be circumstances in which it will be helpful (or even imperative) for the Administrator to check the profile of scores for individual LADS Plus tests. Such circumstances are likely to include the following:
- If the overall classification places the person in the ‘Borderline’ category (see Section 4.2.4).
- If the Administrator has reason to suspect that the overall classification and/or description is incorrect.
- If the Administrator believes that the overall classification and/or description do not tally with what they already know about the person’s capabilities.
- If the person is known to have very poor literacy skills, possibly as a result of a very disadvantaged background and disrupted schooling. Individuals conforming to this description may often be encountered imprisons or young offender institutions. See Section 5 for further information on screening and interpreting LADS Plus results from adults in this category.
Generally speaking, two red scores on individual LADS Plus tests (or one red Plus one or two amber scores), will give a strong indication that the person has dyslexia, regardless of the third score. Likewise, one red score on either Word Recognition or Word Construction, or two to three scores at the top end of the amber range (score 6) will indicate dyslexia, but less strongly. Examples would include dyslexic individuals who have developed very good strategies for coping with sequential short-term memory tasks, and so can perform surprisingly well on the LADS Plus Working Memory test, but who still experience difficulties with the Word Recognition and Word Construction tests.
Two green scores on individual LADS Plus tests are generally a safe indication that the person does not have dyslexia. This is particularly the case if the highest score is Working Memory, as scores on this test show the greatest variance (of the three LADS Plus tests) amongst non-dyslexic individuals. This means that some non-dyslexic adults perform rather poorly on the LADS Plus Working Memory test. The most difficult cases to interpret are those with a mixture of score types (red, green and amber) or ones where low amber scores (range 4– 5) predominate. In these cases a more qualitative approach to interpretation is called for. As a general rule, if the highest score is for Working Memory, and the other two scores are not higher than 4, it is probably safe to conclude that the person does not have dyslexia, but simply has weak (or unpractised) memory skills. But if the highest score is for Word Recognition or Word Construction, and the other two scores are higher than 3, then it is more likely that the person does have dyslexia. If in doubt, the Administrator should seek advice from someone who has more knowledge of dyslexia and/or experience of working with dyslexic adults. With increased experience in using LADS Plus, it is expected that Administrators will be able to deal more confidently with rare and unusual cases.
A note on intelligence
The verbal and nonverbal reasoning tests in LADS Plus enable the Administrator to gain a fairly good indication of the person’s intelligence. This section is intended to shed some light on this frequently misunderstood and sometimes maligned concept.
What is intelligence? In 1997, 52 international experts on intelligence all subscribed to the following definition: “Intelligence is a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience” (Gottfredson, 1997).
Lay people (and some scientists, too) often get confused or upset about whether intelligence actually exists – whether it has a physical reality. This misses the whole point. Intelligence is a psychological construct – in other words, a name given to a hypothesised characteristic that underpins many human cognitive activities that are valued in society at a given time and place. What is meant by ‘intelligence’ may vary from culture to culture. What enables us to propose the concept of intelligence is the overwhelming evidence that all human cognitive abilities show a high positive correlation – in other words, they all tend to measure something in common. The ‘something’ that they measure in common is generally referred to as ‘general intelligence’, and it corresponds to what most lay people are thinking of when they use the term ‘intelligence’. Put this way, it is clear that intelligence is no more (or less) than the common ground between a wide variety of different cognitive skills, and hence measures of intelligence enable us to predict (to some extent) how well somebody is likely to deal with tasks involving those cognitive skills.
Intelligence is influenced by both genetic and environmental factors. The inherited component in intelligence is substantial – approximately half the variability in IQ scores can be attributed to genetic factors (see Plomin & Petrill, 1997). However, the high heritability of IQ certainly does not mean that environment or education has insignificant impact on intelligence. For some children, the environmental or educational impact will be massive, for others less so. However, education (in its widest sense) is likely to increase intelligence not only by equipping individuals with new knowledge and new strategies for thinking, problem solving and remembering, but also by promoting behaviours and attitudes (such as working hard and concentrating when under pressure) that are likely to enhance test-taking performance.)
How can we assess intelligence? To some extent, all measures which involve cognitive abilities – memory, general knowledge, speed of information processing, etc., as well as any tests of educational attainment – are also measures of general intelligence. However, tests that are specifically designed to do the job (i.e. intelligence tests) give us much better measures. But at the end of the day all tests can only give estimates of intelligence – they do not measure intelligence directly. During the last 100 years of psychological investigation into intelligence, progress in statistical techniques has enabled researchers to look more closely at the components of intelligence. Researchers have come to recognise that of the various components, the two most important – at least for educational and most occupational purposes – are verbal ability and nonverbal ability. Most tests of intelligence – whether individually administered or group administered – now comprise measures of verbal and nonverbal ability. There is obviously a wide range of tasks that could be used to assess these two core factors. Writing an essay or giving a speech both draw upon verbal ability. Working out a route using a map or assembling an item of flat-pack furniture both require nonverbal ability. But research has shown that tasks involving understanding of verbal concepts (verbal reasoning), and tasks involving solving mental problems involving pattern, shape or spatial orientation (nonverbal reasoning), are highly suitable for giving us stable and reliable measures of verbal, and nonverbal ability, respectively. Consequently, the reasoning measures in LADS Plus are tests of verbal concepts and matrix reasoning.
In the past there has been controversy regarding whether intelligence tests are culturally biased. Extensive research has concluded that properly constructed and correctly administered IQ tests are not biased against different social, economic, ethnic or racial groups within an advanced culture in which education is universal. Of course, if individuals do not speak the language used in a verbal ability test, or if they do not come from an advanced culture or have not received any schooling, such tests would not be a fair or appropriate measure of their abilities.
Verbal intelligence is generally a better predictor of educational and academic attainment, while nonverbal intelligence is usually a safer guide to the level of an individual’s practical skills. In most individuals verbal and nonverbal reasoning skills are broadly similar and fall within the average range. Some individuals may excel in both of these aspects of intelligence, and others may have below average or low abilities in both. However, a few people may show more substantial differences between their verbal and nonverbal reasoning skills, or even be high in one and low in the other. There is evidence that nonverbal ability is less affected by cultural and educational factors and verbal ability is more affected cultural and educational factors, so it follows that people from disadvantaged backgrounds or who have not had adequate education are more likely to have higher nonverbal ability than verbal ability. Some people who have dyslexia are also in this category; they often have good practical and visual thinking skills and are less adept at thinking with language. Occasionally, people display good verbal skills but poor nonverbal skills; people with dyspraxia (developmental coordination disorder) tend to fall into this category. In cases where a person has verbal and nonverbal abilities in the same broad area, it is appropriate to average these and to treat the average as an indicator of their general intelligence. However, cases where a person’s verbal and nonverbal abilities are quite different it is not appropriate to average these as this will not properly reflect their intelligence. Instead, they should be regarded as having different capabilities in the two areas of intelligence.
For further information on intelligence, see Deary (2001) and Mackintosh (1998).
Cases where not all the tests have been completed
It is strongly recommended that all five tests in LADS Plus should be completed whenever possible. However, in exceptional circumstances the person may not have completed all five tests, e.g. because they became unwell or because of excessive anxiety. In such cases, the program will not be able to give an automatic categorization in terms of probability of dyslexia, although partial guidance on interpretation will be automatically provided. In order to interpret the findings of the screening the Administrator will need to refer to the results of the individual tests that have been completed (see Section 4.2.5). Sometimes it will be possible to supplement the information gained from LADS Plus with other information available about the person, e.g. gained from interview, results of other tests or examples of the person’s work. This may enable the Administrator to reach a judgement about the likelihood of dyslexia, but special care should be taken in drawing conclusions in these circumstances and it is recommended that advice be sought from an appropriate psychologist or dyslexia specialist.