5.1 Context of the enquiry
I Summary overview
​
The selection of literature reviewed above (Section 2), demonstrated that recent research interest in academic confidence in adult learners, especially those in higher education, has indicated the importance of understanding more about how students at university engage with their studies. It has been shown that those who are able to tackle with confidence, the curriculum content and assessment processes that are core features of their learning experiences at university, are not only likely to gain higher academic rewards than their less confident peers, but will have also enjoyed their time at university. They will have been consistently well-motivated, will have considered their achievements to be worthy recompense for their investments, both in time and financially, and that the complete experience will be of lifetime benefit. Theoretical perspectives were presented which supported the view that academic confidence can be considered as a sub-construct of academic self-efficacy, itself one component from a plethora of attributes that together, serve to regulate human behaviour. Such self-regulation is argued to be derived from multifactorial influences stemming from interactions between intrinsic personal characteristics, the local and wider environment that serves as the domain for the behaviour, and the behaviour itself. This understanding of the ways in which individuals function co-operatively and independently is enshrined in Bandura's Social Cognitive Theory, a model which has been adapted to the domain of education and learning, and serves as one of the underpinning theories upon which this research project has been grounded.
​
This research study has been interested in relating this explanation of how we function as people not just to the domain of learning and teaching, but specifically to try to explore how the model accommodates the more atypical learning characteristics and behaviours of communities of learners from minority groups - in this case, students with dyslexia at university. The clearly stated aim was to find out more about how the particular attribute which casts them as 'different', 'othered', and 'labelled' might impact on their confidence when tackling the challenges of studying at university, an environment that remains steadfastly literacy-based, and in which they are inherently disadvantaged. No studies have been found to date which specifically focus on how academic confidence - as the mediating variable between students' inherent academic capabilities, their learning styles and preferences, and opportunities presented to them at university - is impacted by their feelings of 'otherness' attributable to being members of an atypical group of learners, notably those with dyslexic learning differences. For example, one student in this study captured these experiences, saying:
​
"I don't like feeling different because people start treating you differently if they
know you have dyslexia, and normally they don't want to work with you because of
this". (Respondent #85897154; ABC24=47.33; Dx16=850.90).
Drawing from prior anecdotal and practitioner-observed evidence in university learning development contexts, the objectives of the research were firstly to establish that students with dyslexia do, indeed, present lower levels of academic confidence in comparison to their non-dyslexic peers, and secondly, to determine whether evidence would emerge to support the conjecture that knowing about one's dyslexia might, in reality, compound the additional learning demands relatable to the dyslexia itself. Were this to be established, thence to argue that one remedial course of action may be to adopt an alternative, counter-intuitive approach to current norms for equalizing the learning playing field. At present, the approach is to firstly establish the dyslexia through a 'diagnosis', and subsequently design and attempt to implement a programme of individualized learning support, grounded in the application of 'reasonable adjustments' to teaching and learning engagements for that individual. Paradoxically, those approaches appear to increase the workload of some students rather than ease their pathways through the rigours of academic study. An outwardly supportive regimen of less self-regulated, more externally directed and supervised learning, although well-intentioned, can generate additional demands on time-management, organization and scheduling which may increase learning pressures rather than ease them. Resonating with many comments in this current study, this issue was succinctly put by a dyslexic student from the previous Masters dissertation who said:
​
"Going for help with my studies takes up more of my time when I'm already
struggling with too much work and not enough time; and it rarely helps as I can't explain why I'm struggling - otherwise I would have just done it on my own in the
first place". (Dykes, 2008, Respondent #28);
...and in this current study, a student with a moderate level of dyslexia-ness but with a high level of academic confidence suggested that:
​
"[Support] should not just be for one type or group of people such as those
with particularly learning difficulties. I think that puts many people off as soon
as they see the term 'learning difficulties'". (Respondent #71712644; ABC24=86.6;
Dx16=542.92).
​
Hence, evidence has been assembled which firstly indicates an inverse relationship between academic confidence and dyslexia-ness in university students; and secondly, supports the polemic that from an academic confidence perspective at least, students with dyslexic learning differences studying at university may be likely to gain academic credentials more in line with their academic capability were they just left to get on with it as best they can. Hence there is an argument to support the focus of remedial activities at university being shifted towards changing the means of curriculum delivery and learning assessment, and away from changing the student into one who fits more readily into the more conventional and traditional interpretation of a university student in the UK.
​
As universities have opened their doors to a much broader spectrum of learners, notably through now well-established widening participation and alternative access schemes, it has become increasingly evident that all students would benefit from a better institutional-level understanding of the impact that individual differences can have on educational engagement, ownership of learning, and hence likely attainment. The learning environments and processes that generally prevail at university are not much informed by psychological knowledge.​ Hence, progress towards meeting the aspirations enshrined by UDL for example, has, to an extent at least, been stymied.
It is not unreasonable to suppose that a slowdown in the learning development initiatives at university that were beginning to bear witness to the value of greater adaptability and flexibility in curriculum delivery and assessment, are likely to have had the greatest impact on communities of learners who come to university with spectra of learning profiles and preferences that are atypical. Of these unconventional learners, some are labelled with difficult-to-define learning 'disabilities' which are likely to be dynamic in nature, and not necessarily an objective fact. Strong arguments have been presented to suggest that it is learning institutions that translate broad profiles of learning strengths and weaknesses into difficulties and disabilities, not least through the strongly literacy-based transmission of knowledge, and the traditional, inflexible assessments of how much of it has been retained. This is especially true for learners with dyslexia, which, it has been shown, is difficult to define and evaluate individually. This cohort of students comprise the most numerous amongst those deemed to be 'disabled' in our universities. Evidence from this current study readily documents some of the frustrations. Resonating strongly with the quotation above from the previous study conducted more than a decade prior to this current research, one participant from the dyslexic group said:
​
"... universities provide support with tutorials geared at helping the individual
with learning, but somehow they seem to expect that a person understands what
they find difficult ... [but] because they have been living with it their whole lives,
[they] can't see objectively what is 'wrong' ". (Respondent #87564798; ABC24=49.17;
Dx16=783.20);
With another respondent echoing experiences of poor levels of institutional understanding:
​
"I think there could be more support for students with learning difficulties. As yet,
the dyslexic team haven't been very helpful or supportive" (Respondent #61502858; ABC24=61.88; Dx16=629.90);
... and one of her non-dyslexic peers appeared to indicate unmet, or possibly unrealistic expectations about the learning experiences ahead of her, or at the very least a lack of understanding and preparedness for the challenges that university learning would present, and that these were considered not to have been met by an appropriate insitutional level response:
​
"Ways that studying at university can be improved is by far, to teach students
how to learn. We're always taught the content for a specific subject, but has
anyone ever taught a student on how to learn?" (Respondent #52289216; ABC24=56.88; Dx16=547.61).
These comments were neither uncommon, nor have been taken out of context, were all drawn from the dyslexic students' group in this current study, and they echoed much of the qualitative data collected in the preceding Masters dissertation. It is regretful that this appears to indicate, anecdotally, and for those students at least, that little, or at best, patchy progress has been made at institutional levels to demonstrate a deeper understanding about how to accommodate more readily the learning development needs of such students.
​
​
II Challenges
​
The challenges facing this research topic have been considerable, notably because the syndrome of dyslexia remains controversially defined. The research surrounding its aetiology and manifestation of the learning differences that are purported to be attributable to it, remain varied, sometimes co-morbid with other disadvantaging learning or physical characteristics, and apparently explainable from several different angles. Furthermore, the majority of research relates to dyslexia in children, not unreasonably stemming from an educationally desirable response to explain the slow uptake of reading and other literacy skills amongst a sizeable minority of early-years learners. Only relatively recently has the research focus widened to explore more about the nature of dyslexia in adults, more particularly in adult learners. The review of a wide range of selected literature in Section 2 concluded that taking dyslexia to be an information processing difference - that is, a multifactorial condition existing in many learners in varying degrees -would be the most neutral and unbiased perspective to adopt. The multifactorial approach fits well with the BDA's description of what it means to be dyslexic, notably where this adopts the idea of a continuum for cataloguing the characteristics and attributes of dyslexia. This acknowledges that due to its wide and varied nature, it remains difficult to adopt a binary approach to the existence of an individual's dyslexia. Hence in this current study, the continuous nature of dyslexia as a variable has given rise to the establishment of dyslexia-ness as the quantifier, and thus, The Dyslexia-ness Continuum.
This led to the most substantive challenge of the study which was to design a data collection instrument that could gauge dyslexia-ness sensitively, and across a sample of the complete university community. It had to reliably determine levels of dyslexia-ness but not be a dyslexia screener as this would have had ethical implications were one to have been covertly used to collect data. Drawing from a balance of theory taken from the multifactorial explanations of the dyslexic condition with an interpretation of the prevalence of characteristics (dimensions) of dyslexia noted in practical, university dyslexia-support situations, the Dyslexia Index (Dx) Profiler met its design objectives by providing a realistically dependable and reliable interpretation of research participants' levels of dyslexia-ness, and hence their location along the Dyslexia-ness Continuum. However, the Dx Profiler remains unique and untested as a data collection tool aside from its use in this current study which naturally limits the generalizability of its findings. The concepts upon which it is based are theoretically sound, and anticipated publication shortly relating its development and use to a wider research audience, may provide the opportunity to refine and strengthen both the Dx Profiler itself, and also the concept of dyslexia-ness which it aims to gauge.
​
Additional challenges arose through use of the existing but also recently developed Academic Behavioural Confidence (ABC) Scale, and the questionnaires that were designed to measure levels of ABC amongst university students. Although the Scale has been deployed effectively in a range of prior studies, and thus is some years ahead of the Dx Profiler along its development journey, it too warrants further refinement. For example, this current study has shown the usefulness of exploring the local nature of the factor structure of the ABC Scale. Sub-scales that emerged from the dataset-specific interpretation of reliability analysis and data reduction suggested that accepting some fluidity around its factorial composition should be an important component in understanding its determining characteristics, and hence, what the outcomes from the Scale(s) mean, both locally, and for that interpretation to be more widely generalizable.
But this is surely the nature of all measuring scales and instruments that are attempting to gauge tenuous, psychological and educational constructs that are not overtly measurable. It means that outcomes must be treated cautiously and that either repetitive use of the instruments in highly similar experimental circumstances, or otherwise careful meta-analyses of a number of studies should be conducted so that such scales can become more robust and dependable. In this current study, this led to outputs being generated from four versions of the ABC Scale, two being existing, apparently standardized versions, and two developed from a local interpretation of factor and reliability analysis. Far from this exposing weaknesses in the study design, it was considered as a strength since outcomes were all broadly similar, taken to indicate that whichever version of the ABC Scale was used, it would be possible to draw generalizable conclusions from the outcomes.
5.2 Summary of the outcomes
The data collected in this study were rich and varied, elicited from 80 dimensional statements that participants were requested to apply their best judgement to respond to. The questionnaire was designed to be concise, easy to navigate, interesting and accessible to engage with, and especially easy to record responses through use of innovative range-slider input fields. It has not been possible to estimate a response rate because the process used to publicize the questionnaire and hence recruit participants was twofold, and neither mechanism gauged how many students subsequently chose to take part as a proportion of the number who actively read the summary of the research purpose and the invitation to participate. However, of the 183 questionnaire datasets that were returned, only 17 were discarded through being less than 50% completed, reasonably determined as the critical value. Hence, the questionnaire design parameters were considered vindicated, and the size of the datapool deemed to be sufficient for meaningful outcomes to emerge.
I Composition of the datapool
​
The datapool comprised 98 students who did not declare a dyslexic learning difference and 68 who did. Overall, these 166 students were considered to be a reasonable reflection of the composition of university students more generally, as distributions according to study level, domicile and gender were unremarkable in comparison to the national trend (HESA, 2018), although overall, approximately twice as many female students as males took part. Since the analysis was not designed to look specifically for relationships between gender, dyslexia-ness and academic confidence, this bias was not deemed to be of consequence. These data were collected however, allowing scope for examining whether gender might have an impact on the relationships between academic confidence and dyslexia-ness in a subsequent study.
A second, notable exception to the generality of the datapool was indicated by the disproportionally low number of non-UK students in the dyslexic group even though the ratio of home to non-UK students in the complete datapool was approximately representative of the national distribution. Rather than this being an indication of a low proportion of non-UK students with dyslexia studying at UK universities more widely - clearly an inappropriate conclusion to draw based on the relatively small datapool in this current study - it may be more likely accounted for due to non-UK students with dyslexia being not known to the university's Disability and Dyslexia Service, and hence were not on the Service's e-mail distribution list that was used for recruitment. Access to the Service's advice, support, guidance and resources would be most likely driven by acknowledging students with dyslexia who had applied for the DSA through the university. So it is of no surprise that international students with dyslexia were largely invisible to the Service.
​
II Prevalence of dyslexia in this current study
The prevalence of dyslexia amongst university students in the UK (and elsewhere) is notoriously difficult to determine for several reasons. For example, the true proportion may be obfuscated through a reluctance to disclose by some dyslexic students due to perceptions of the likely consequences outweighing the possibly marginal benefits; but also, that the student community is likely to include a significant proportion of unidentified dyslexia. Taken with issues about how dyslexia is defined and thus, how this raises challenges for measuring it, the proportion of quasi-dyslexic students identified in this current study was of no surprise. Indeed, this was exactly the outcome that was anticipated in the foundation stages of designing the enquiry, and is consistent with evidence from earlier studies that dyslexia amongst university students may be under-reported (Richardson & Wydell, 2003; Stampoltzis & Polychronopoulou, 2008), or that a substantial number of dyslexic students in UK HE are only identified after their enrolment at university (Singleton, et.al.,1999; Singleton & Aisbitt, 2001). With all participants levels of dyslexia-ness determined by the outputs of the Dx Profilers, it was heartening to that the small sub-group of 18 (Dx20) or 19(Dx16) quasi-dyslexic students were identifiable from the group of participants who had declared no dyslexic learning differences. At close to 20% of the non-dyslexic group, this sub-group represented a substantial minority although in absolute terms was still a small sample.
The Dx Profiler was not designed to be a dyslexia screener. However its composition did include a breadth of gauging parameters that would not be misplaced in a screener. So, it is not unreasonable to suppose that of the eighteen or nineteen students comprising this quasi-dyslexic sub-group, a proportion, possibly a significant proportion, may be unidentified with dyslexia, such that this would be determinable using existing dyslexia screening tools. To put this in context the most recent data available from HESA (2016/17) indicated that students in UK HE institutions who disclosed a learning disability accounted for 4.8% of the student population overall. This was the incidence of all defined learning disabilities, which, in addition to dyslexia, includes dyspraxia, ADHD, and Asperger's Syndrome. Although there is currently no mechanism in place in their data collection process for discriminating students with dyslexia as a discrete subgroup of those disclosing learning disabilities, it was indicated that HESA views dyslexia as the most represented subgroup. However, it is important that this apparently significant incidence of quasi-dyslexia in this study is not considered as a 'dyslexia fact'. Accurately gauging the prevalence of dyslexia amongst university students remains flawed, with the 'dyslexia debate' continuing to be controversial and to a point, contested (e.g.: Elliott & Grigorenko, 2014). Equally this is a result not to be ignored, nor set aside as a data analysis glitch, because the Dx Profiler which revealed this subgroup of students, demonstrated good, if not strong sensitivity when applied to students in the known dyslexic group, subsequently used as the control.
III The relationships between dyslexia-ness and academic confidence
​
This study set out to explore the extent to which students with dyslexia at university are less confident about their academic studies than their non-dyslexic peers. The nature of academic confidence as a distinct, but related construct to academic self-efficacy has been extensively discussed earlier, validating its theoretical underpinnings. Academic confidence, as a construct that is measurable, emerged from an interest in explaining differences in learning preferences between students on similar academic pathways. Early studies led to the reimagining of the metric to acknowledge a more keenly focused interest on study actions and behaviours as the measurable indicators of students' confidence for tackling the rigours and challenges of university study.
Thus, the Academic Behavioural Confidence Scale emerged as a practical means to operationalize and gauge levels of academic confidence amongst university students. Very few studies have been found to date which have used the scale to look at how disabilities impact on academic confidence - at least within the same framework that has defined academic confidence for this current study - and none have taken dyslexia as a specific focus. Whether or not dyslexia, as a difficult-to-define syndrome, should feature on the disability agenda remains a contested point.
In attempting to adopt a position of relative neutrality, this current study has taken the position that dyslexia is best considered as a multifactorial, information processing difference, but one where many individuals beset with a range of its dominant characteristics in varying degrees, persistently remain disabled by the broadly literacy-based learning environment within which they are striving to engage. It has been argued that this only serves to erode confidence for tackling their studies at university. This position has been grounded in evidence from the literature reviewed and discussed in Section 2. Hence, the research framework defined academic confidence as a sub-construct of academic self-efficacy; and considered dyslexia from the multifactorial perspective, for which a new metric, the Dyslexia Index Profiler, was introduced to operationalize the extent and levels of individuals' aggregation of characteristics or dimensions of dyslexia, their levels of 'dyslexia-ness'.
​
From the outset two, clear objectives were defined: firstly to establish that students with dyslexia, in whatever form it is defined, consistently show lower levels of academic confidence (as determined from the ABC Scale) than their non-dyslexic peers; and secondly, to determine whether quasi-dyslexic students' levels of academic confidence might be less severely depressed. Hence this would be to suggest that by studying without knowing that their pattern of challenges and difficulties are comparable with those within dyslexia envelope, the outcome is that this would appear to carry with it a less severe impact on their academic confidence.
Meeting the first of these objectives was relatively straightforward, subsequent to the development of a suitable data-gathering instrument. With two clearly demarcated groups of students in the datapool - those who had declared dyslexic learning differences, and those who had not - individual levels of academic confidence were gauged using the ABC Scale and mean average outcomes together with confidence interval estimates were constructed for each group. The results suggested an inverse relationship between academic confidence and dyslexia-ness. That is, higher levels of dyslexia-ness tended to be associated with lower levels of academic confidence. However, it cannot be said from this that dyslexia, or dyslexia-like learning and study characteristics cause reduced levels of academic confidence because firstly, the dataset collected was too small to support such a radical claim; and secondly, no attempt was made to control for other factors that may also have a bearing on levels of academic confidence in university students. These could be wide-ranging and possibly include social, cultural or ethnic influences on prior learning experiences and history, or intrinsic personal, psychological factors such as anxiety, self-esteem, predisposition towards academic procrastination, and so forth, or academic credentials on course-commencement, for example. But this datapool of university students was firstly, considered to be reasonably representative of students more generally, and secondly too small to permit analysis at the micro level required were groups and subgroups reduced in sample size to accommodate the distributions of these other factors. Hence this is to recommend further studies be undertaken to enable this level of analysis.
In the event, however, and for this representative sample from the university community, those students with dyslexia presented levels of ABC that were on average nine percentage points lower than their non-dyslexic peers, representing 12-13% lower absolute levels of academic confidence. This was a consistent outcome across the four versions of the ABC Scale. Adjusting values to account for sample sizes and distribution variances, the differences also consistently generated moderate effect sizes, with average ABC values shown to be significantly lower for the dyslexic group. When students were further sifted according to their levels of dyslexia-ness, as determined from outcomes from either of the Dx Profilers used for the data analysis, students from the non-dyslexic group with low levels of dyslexia-ness showed substantially higher levels of ABC in comparison to the control group of dyslexic students with relatively high levels of dyslexia-ness, where the greatest difference of 14.96 percentage points emerged from analysis using the ABC 17-dimension Scale and the 16-dimension version of the Dx Profiler. This appeared to indicate that students with low levels of dyslexia-ness were presenting levels of academic confidence some 20% higher on average than their strongly dyslexic peers (Tables 15, 16; sub-section 4.3/IV.II). At a factorial level, where dimension reduction had indicated two factors relating to the academic processes of studying with the third concerned with how these processes were planned and organized, outcomes showed that differences were most notable at the academic rather than the planning levels. Indeed, the data showed that all students in the datapool had similar levels of confidence in their capability to plan and organize their studies to meet their academic challenges, and that dyslexic students in the Control subgroup presented similar average ABC levels to their non-dyslexic peers with low levels of dyslexia-ness in the Base subgroup. For the other two factors of Study Efficacy, and Academic Engagement, differences between these subgroups were significant, however, with the non-dyslexic students showing much higher levels of ABC (Tables 30-32, sub-section 4.5/V). Hence it emerged that these two 'academic' factors were the strongest contributors to reduced levels of ABC overall for the dyslexic group.
At the very least, it is argued that these outcomes are cause for reflection about how students with dyslexia engage with their university learning experience, perhaps indicating that the academic tasks and challenges being faced could be more appropriately packaged and presented. A logical conclusion is that more could be done to enable equitable levels of academic confidence to be an outcome of initiatives designed to improve the 'lot' for students with dyslexia, especially if these can make them feel less othered, different and disabled. Such initiatives are likely to be of benefit to all students as a result of more accessible and flexible learning situations, that are individually adaptable so that academic performance and achievement are more accurate reflections of their capabilities and learning potential.
IV - Diagnosing dyslexia: does this impact significantly on academic confidence?
​
An additional focus of this study has been to consider whether there may be evidence that the process of identifying dyslexia in university students (or in their prior learning) may have had an impact on their academic confidence.
​
It has been argued earlier that the outcome of a 'disability diagnosis' may lead individuals with dyslexia to perceive themselves to be valued less by their peers, or even by society more generally; a characteristic typically associated with stigmatization, and an acknowledgement of the stigma that is reportedly associated with the dyslexia label (Morrise & Turnbill, 2007; Lisle & Wade, 2013). It has been suggested that the significant part of these feelings may stem from associating 'diagnosis' with curative treatments, and uncertainty about what they are for, or whether they will work. It has been shown in Section 2 that affective responses are known to influence compliance with remedial regimes constructed around the modification of behavioural intentions and actions. Translated into the dyslexia context, it is reasonable to suppose that individuals whose dyslexia is diagnosed to them as a learning disability are likely to experience aversive emotional responses to the fact - perhaps even perceiving dyslexia as an illness without a cure. The negative construction of disability is widespread in society by being frequently associated with a clinical condition (e.g.: Connor & Lynne, 2006, Phelan, 2010). A student's reaction to their new situation could be unpredictable, and may impact on their perceptions of their capacity to study, not least from any lasting issues associated with a reimagining of their self-identity and impact on their confidence more widely. This may be particularly notable when disability has been perceived as imposed, especially if this was unexpected. One student in this current study whose dyslexia was diagnosed as a disability said:​
​
"[Only] In the first year of my degree I found out I was dyslexic and I was massively
traumatised by it. I thought learning would never be the same again due to my
learning disabilities" (Respondent #89059542; ABC24=53.25; Dx16=811.71)
In Section 2 it was suggested that the process of identifying dyslexia, the dyslexic 'label' that is the outcome, and especially the manner through which the label is attached to the individual concerned, may be a critical factor in establishing whether the affective response to this knowledge is positive or not. One of the strands of this study is acknowledgement of the stigma that is reportedly associated with the dyslexia label. Notably characterized as the ‘dilemma of difference’, there is a persistent and unresolved debate about the value of attributing labels to individuals with atypical educational needs (e.g.: Norwich, 1999; Warnock, 2005; Terzi, 2005) as discussed earlier. At the time of designing this study, no literature had been found which considered exploring as a variable, the impact of the different ways in which dyslexia is communicated to an individual as a result of a screening or assessment at university. Some studies have examined other psychosocial experiences of receiving an identification of dyslexia, but none appear to have explored the possible significance of how individuals were told of their newly idenfitied situation. Dyslexia is persistenly referred to as being diagnosed. No studies were found to indicate that the term ‘diagnosis’ itself had been considered as a likely impactor on an individual's internalization of new knowledge about their dyslexia. Hence to attempt to gauge this impact through the lens of academic confidence appears to be unique.
Analysis of results showed that students in this study whose dyslexia was diagnosed to them presented a substantially lower ABC when compared with students whose dyslexia was identified, described or disclosed to them. The outcomes were similar whether dyslexia was diagnosed as a disability or as a difficulty, and it was notable that nearly two-thirds of dyslexic students who participated in this study had been diagnosed with dyslexia, and of those, more than half recalled their dyslexia being described to them as a disability. The absolute difference in average ABC between the subgroup whose dyslexia was diagnosed as a disability and those others where dyslexia was identified, disclosed or described as a difference or a difficulty was more than ten percentage points. So amongst the students in this sample at least, the academic confidence of the diagnosed/disability subgroup was depressed by more than 14%. This was a statistically significant difference, and when adjusted to account for sample sizes and variances, showed a moderate-to-large effect size of g = 0.64. A deeper interpretation of the results revealed that students whose dyslexia was diagnosed, were, in particular, less confident about attending their lectures, seminars and other university teaching situations, and less confident about engaging with their peers or lecturers when they were there. Such behaviour appears consistent with observations in Cameron's (2016) study of the day-to-day learning lives of dyslexic students at university, which revealed that as members of a class, seminar or lecture in the company of other students, participants often found these learning experiences uncomfortable or threatening, reporting 'fear of speaking out in seminars or discussions' so as not to appear 'stupid or incompetent in some way'; that they all felt 'different from others', 'less able or intelligent' and that they 'didn't belong' in academic spaces. Similar feelings were reported by some dyslexic students in this current study. One participant said:
"I don't speak in class because I am not very confident in answering questions
in case I get them wrong and people laugh" (Respondent #85897154; ABC24=41.54; Dx16=862.79);
... with another reporting the lasting impact of feelings of difference stemming from earlier educational experiences:
​
"[In class] I do have to battle with elements of doubt ... particularly influenced
by bullying at primary and secondary school to do with 'stupidity' and 'slowness'
and my seemingly unrelated comments to topics at the time" (Respondent #87564798; ABC24=49.17; Dx16=783.20);
​
Another study, however, (Thompson et.al., 2015), established that the majority of participants indicated a greater alliance with the perception of dyslexia as differences in ability than with disability. This was despite the apparently contradictory finding that many felt encumbered by an identity of dyslexia as a disability in educational contexts. In the Thompson et al. study, three distinct identity personae emerged as distinctly identifiable: firstly, that of being learning-disabled, where the dyslexia was focused on impairments and deficits; secondly, of being differently-enabled, in which dyslexic individuals were able to focus on their strengths and celebrate their alternative ways of thinking and learning as an asset rather than a liability; and thirdly as a dyslexia-identity construction rooted in social-disablement, where individuals admitted to feeling disabled by the ways in which their conceptualized, diagnosed, disabling factors were transformed into barriers which they felt prevented them from conforming to the aspirations of a society focused on literacy as a marker of ability, achievement and normality. Evidence of these identity constructions also emerged from data in the current study, with multiple examples of all three being identified.
Hence, there appears to be a likelihood that the means by which dyslexic students are informed about their dyslexia may be a contributing factor to a measurable, negative impact on their ABC, and by logical deduction, on their academic confidence about approaching their studies at university. What seems clear is that the manner in which individuals make sense of their dyslexia and internalize it into their academic self-identity, and especially how this impacts on their engagement with their learning, is an under-researched area - no other studies were found that took this topic as the focus. It also seems reasonable to suggest that greater effort needs to be made firstly, to recognize dyslexia as a difference rather than as a disability; and secondly, that for definitions of the syndrome and the labelling of individuals with dyslexia to remain apposite in educational contexts, care must be taken about how the dyslexic label is communicated to students. Lastly, we are drawn back to the recurring strand that runs throughout this project: that were learning environments designed and structured in more genuinely inclusive ways, the impact of learning differences on academic confidence would be at least minimized, perhaps eliminated, as one of the factors that affect confidence in tackling academic studies. Taking this approach would likely enhance learning quality, and hence likely achievement for students whose learning styles, needs and preferences are atypical.
​
The final part of this discussion, however, is driven by results that appear to suggest that it may be best to leave students considered likely to be dyslexic to some degree or another, in ignorance of the fact. This is a radical conclusion. It is tempered by the small sample upon which the results are based, but detailed analysis and re-analysis using all permutations of the metrics used in this study, generated broadly comparable conclusions. If subsequent studies produce similar outcomes, it may raise difficult ethical issues relating to disclosure which will need to be resolved. The outcome is based on average levels of ABC from the subgroup of students categorized as 'quasi-dyslexic'. Whichever versions of the ABC Scale were used to determine these students' academic confidence, and whether their dyslexia-ness was assessed from the 20-item or the 16-item Dx Profiler, this subgroup of students showed higher levels of academic confidence than their dyslexia-identified peers, across twenty of the twenty-four ABC dimensions of the full scale. Of these, dimensions where the greatest effect sizes were noted, were confidence to 'attain good grades' (ABC24.#7), and to 'read the recommended background material' (ABC24.#14). We can relate these specific dimensional differences between the dyslexic and quasi-dyslexic students into SCT theory, and also into the practical realization of academic confidence as Sander's academic behavioural confidence. The logical strand that connects Bandura's (1995) broad statement about self-efficacy being an individual's context-specific beliefs about their capability to get something done with Sander's adaptation of this idea into academic (behavioural) confidence is illustrated by these confidence differences in 'attaining good grades,' for example. Identified dyslexic students expressed reduced confidence in this dimension, both measured against their non-dyslexic peers but particularly against quasi-dyslexic students where confidence remained substantially higher than the average ABC level of their dyslexic peers. In one study, Sander and Sanders observed that students' confidence in their capability to perform well academically was affected by later outcomes of their assessments, not least due to setting unrealistic expectations about their academic performance. Hence it is possible to surmise that dyslexic students may attribute both their poor performance and their unrealistic expectations of their ability, to their identified dyslexic learning challenges, whereas quasi-dyslexic students with likely similar learning challenges may not present similar attributions. Thus, their confidence levels are not as notably depressed when academic expectations are not matched by outcomes.
Effect sizes between quasi-dyslexic and dyslexic students in overall ABC were small - but not negligible - when the Test subgroup of quasi-dyslexic students was determined by outcomes from the full Dx20 Scale. When the reduced, Dx16 Scale was used as the subgroup sieve, effect sizes were higher, falling well into the 'moderate' category (Tables 15, 16; Section 4.3/IV(II)). Although significant differences were not detected, some outcomes were marginal. In terms of absolute mean ABC levels, values for this subgroup were located squarely between those for the Base subgroup of non-dyslexic students with low levels of dyslexia-ness and the Control subgroup of strongly dyslexic students. At the factorial level, it became clear that the major influences on effect size differences in ABC between the Test and Control subgroups emerged from the two academic process factors of Study Efficacy and Academic Engagement, with negligible differences noted for Organization and Planning.
Assuming (hypothetically at least) that these quasi-dyslexic students would emerge as dyslexic following a standardized dyslexia assessment, this outcome appears to suggest that knowing about being dyslexic is an impacting factor that depresses academic confidence. It is not likely to be the only impacting factor, however, as if so, it would be expected that these students' academic confidence would not be dissimilar to levels for the non-dyslexic group overall, which it was not. Indeed, the tentative regression analysis conducted more as a pilot for future studies than as a major contributor to the analysis outcomes for this current study also indicated that for the students in this datapool at least, the quasi-dyslexic subgroup presented higher than expected levels of academic confidence which although might be thought of as marginal, was nevertheless, not negligible or to be dismissed as a natural variation of the data.
These outcomes suggest firstly, that a more detailed analysis of individual differences in ABC levels at a dimensional level is worthy of investigation, but more so, that future studies are warranted so that the academic confidence of quasi-dyslexic students can be assessed, compared, and subsequently aggregated with the results established here. From this a clearer indication may emerge about the extent to which dyslexia is to be considered as a significant impactor on academic confidence, and about whether the identification of dyslexia in university students might be counter-productive.