Assessment beliefs of higher education staff developers

This research focuses on the assessment literacy, that is, the understandings of assessment terminologies and how they relate to each other, in academic staff developers in the UK, collected via questionnaires and semi-structured interviews. Academic staff developers have been trained and certified to support new higher education lecturers in learning, teaching, and assessment practices, and provide continuing professional development for more experienced staff. Results showed inconsistent and differing understandings between and within individuals. These inconsistencies may reflect the lack of consistency of terminology in the literature. This lack of common understanding has far-reaching implications and needs reconciling to enhance personal and collective assessment literacies, particularly since our respondents have responsibility for training the next generation of academics.


Introduction
Recent work (Taras, 2008a;Taras and Davies, 2014) on assessment literacy has examined university lecturers' beliefs and opinions relating to assessment in university departments of education and science in the UK in order to explore their understanding of assessment issues. This paper reports on a similar study investigating whether understandings and opinions of assessment are clear, cogent, coherent, and shared among a group of UK academic staff developers. Most higher education providers in the UK require new staff to undergo training in learning, teaching, and assessment practices and subsequently to have regular continuing professional development. Academic staff developers are experienced lecturers in HE who have responsibility within their respective institutions in relation to teaching and the support of learning (i.e. as staff developers). All the respondents in this study had a national profile through involvement in the external recognition of taught programmes for junior academic staff in learning and teaching, and all had broad and long teaching experience in higher education and had established institutional practice.
Each participant contributes to development of practice in relation to the training of academic staff in learning and teaching. Staff developers have an important influence in framing assessment practices directly through their own views and experiences, and also indirectly through the literature they utilize and recommend.
We deliberately chose the term 'beliefs', as opposed to 'knowledge', as the focus here because available understandings of formative assessment (FA) and summative assessment (SA) show inconsistencies, a paucity of alignment, and a lack of clarity (see, for example, Taras and Davies, 2014). Furthermore, beliefs are linked to knowledge, but also to past experience, practice, and one's own thinking, all of which we wish to explore. and beliefs about the relationships between SA and FA, but since this research aims to explore opinion from staff developers, we do not consider it appropriate to superimpose our own beliefs on this exploratory paper. Our beliefs can be found in Taras (2012b) and Taras and Davies (2014). Readers are encouraged to engage directly with Scriven (1967) in order to form their own understanding of his work.
The following research forms part of a body of work that has evidenced that tutors and students all have very individualistic and uncoordinated assessment literacies. Maclellan (2001) researched tutors' and students' perceptions and understanding of assessment by asking them to complete a questionnaire on assessment. The focus was on their views of principles and concepts of assessment, and the questionnaire examined purposes, mode, content, timing, and marking. Inconsistencies were evident between students and tutors, and within each group. Also, tutors' beliefs and principles were not reflected in their practice. Similarly, Taras (2008a) and Taras and Davies (2013;2014) demonstrated the lack of consistency within staff groups, whether in education or science departments, through a questionnaire identical to that employed here, focusing on the relationships between aspects of assessment. In the compulsory sector, Hargreaves (2005) asked teachers about their concepts of assessment for learning and found these contradictory. Again, beliefs and principles were not reflected in practice.
The present article explores how disparate interpretations in the literature impact on individual understandings of assessment terminology. The questionnaire was designed to disentangle elements of theoretical frameworks as understood by the respondents, and balanced questions between different aspects of assessment: process, product, functions, formative, summative, and self-assessment. The present study focuses on staff developers, who play diverse national roles in UK HE and present their views and understandings to future generations of educators. We hypothesized that since these staff were responsible for guiding policy development in both their own institutions and more widely in the UK, there would be a high degree of congruence in their understandings of assessment terminology and relationships between various assessment processes.

Method
The theoretical basis for the development of the present survey was examined at length in Taras (2008a) and is therefore not repeated here. The questionnaire is well-established, having already been used in a body of work (Taras 2008a;Taras and Davies, 2013;Taras and Davies, 2014). Our adoption of that questionnaire facilitates comparability.

Questionnaire
A questionnaire of 44 questions (see Appendix 1) was issued in 2011 at a single event to 12 UKbased higher education staff developers with a responsibility for academic staff, and in 2012 to one further participant. Individuals were selected opportunistically as they gathered for an event on learning and teaching practice. All had a national profile through involvement in the external recognition of taught programmes for junior academic staff in learning and teaching; six ran or had run such programmes; four were or had been head of their institution's academic development unit or equivalent; all had Fellowship of the UK's Higher Education Academy -some were Senior Fellows and some were Principal Fellows; at least six were consultants in learning and teaching at HE providers in the UK and beyond; and all had broad and long teaching experience in higher education and had established institutional practice. Some were educationalists but the majority were, at least in the initial phase of their careers, from other disciplines. They were each from different institutions, had different subject backgrounds, and were from a broad range of types of provider. They were told that they may complete the questionnaire anonymously, but that if they gave their name, this would indicate that they were willing to be contacted for a discussion of their responses. They were also told to answer the questions in order, not to go back to questions, and not to confer. Questionnaires were collected immediately on completion and no respondent took more than 15 minutes. The questionnaire was not piloted because it was used by Taras (2008a), who had already undertaken a piloting exercise.
Most questions required a yes/no response but some were qualitative in that they required a written comment: questions 1 and 3 asked for a rough definition of summative and formative assessment respectively; questions 2 and 4 asked for examples of summative and formative tasks respectively; question 10 asked for an example, if summative and formative tasks were being conflated; question 16 asked how formative work is related to summative work; and the final questions, 43 and 44, asked again for definitions of summative and formative assessment. The responses to these questions required analysis and interpretation before quantifying. Key words were selected and analysed (a 'semantic' interpretation, that is, based on the primary, literal meaning) to find the general trends that appeared from repetition of words and ideas. In addition, where possible the distinctions between assessment 'of' learning and 'for' learning were made. We asked for definitions twice to examine the impact of completing the questionnaire on participants' thinking. We have termed the initial definitions 'first definitions' and those asked for at the end of the questionnaire 'second definitions'.

Semi-structured interviews
The understandings of seven of the respondents were further explored by recorded and transcribed audio interview, at least two months after the questionnaire was completed. Respondents were all those who agreed to participate in interviews following completion of the questionnaire. We deliberately left a time gap so that the influence of completing the questionnaire on the interview responses would be minimal. All interviewees were asked the same questions (see Appendix 2). For some participants, we planned to ask additional participant-specific questions to clarify responses given in the questionnaire. In most cases, however, the response to generic questions was such that specific questions were superfluous. The questions focused on contentious issues in the literature in an attempt to explore and expose the extent of diversity between individuals who constitute a cadre of specialists in learning and teaching in higher education.
The transcripts were initially analysed to capture the essence of the content relevant to the questions asked. A second phase involved classifying the content (opinions) as they related to SA, FA, or both. The classifications enabled the exploration of participants' beliefs and understandings of these central aspects of assessment and how other factors related to them. The process of analysis was carried out by the two authors together, using the notions that SA is linked to final graded work used for validation and accreditation, while FA is considered to be assessment that supports student learning, and most categorizations of FA and SA were expressed clearly. In the cases where the data were less immediately clear to us, further contextual clues were used to shed light, and these generally sufficed. Having two researchers to discuss and reach a consensus greatly helped to avoid misrepresentations.

Results
In order to minimize the researchers' personal views from interfering with the reporting of the data, the discussion and interpretation of the data will take place in the 'Discussion' section, which will also summarize the key ideas.
The results are presented under the headings of 'Understandings of SA', 'Understandings of FA', and 'Understandings of the relationship between SA and FA', and interview data are used to comment and expand on questionnaire data. Within that framework, the results of the questionnaires are considered question-by-question in the order in which they were asked, apart from the first and second definitions, which were analysed together (see Appendix 1). The proportions quoted are based on the number of responses to each question. Since the number of respondents was not always the same, the denominator varies.

Understandings of SA
For the first definition of SA in the questionnaire, a semantic analysis showed that five respondents indicated a notion of 'end' or 'final' and ten indicated that in SA marks awarded should 'count' and be used in validation of learning, although three used both the ideas of final and validation. One respondent used the word 'formal'. One response indicated that feedback to students was the focus of SA. Twelve responses were classified as 'of' learning (two explicitly using the term), and one 'for' learning. The latter stated, 'Feedback (written and/or verbal) against given criteria often with a numeric grade to indicate level of achiever.' For the second definition of SA, semantic analysis showed that eight respondents used a notion of 'end' or 'final' and eight a notion of validation of learning, with five mentioning both. The word 'formal' was not used. Two used the notion of feedback to learners, one indicating that this defined SA. Nine responses were classified as 'of' learning (one explicitly using the term), one as both 'of' and 'for', one respondent did not answer this question, and one was classified as unrelated (the respondent wrote, 'As per Q1'). The respondent giving an answer to the first definition of SA classified as 'for' learning did so again: 'Verbal, written or self-reflection on progress with a specific learning task which might aid enhanced achievement.' When asked to give an example of an SA task, seven questionnaire respondents mentioned 'exam' or 'examination', two 'portfolio', one 'essay', and one 'presentation'. Five used the terms 'end' or 'final'. Nine respondents gave an example classified as 'of' learning and four an example classified as both 'of' and 'for' learning.
Two interviewees referred to the role of learners in SA: one felt that it was not a good idea because it would raise quality assurance issues; the other, contrariwise, thought it would help learners understand the grades assigned by tutors.

Understandings of FA
For the first definition of FA in the questionnaire, a semantic analysis showed that seven respondents mentioned 'feedback' or 'feed-forward', and for a further four, feedback to students on their performance, including self-assessment, was implicit in the response. One equated FA with feedback. Three noted explicitly that marks were not for validation of learning. There was a strong link (ten respondents) between FA and its developmental purpose. Five defined FA as something that will improve performance, that is, is used by students, and five that it offers an opportunity to improve performance, implying or indicating that it might not necessarily be used as such. One indicated that FA does not involve the issuing of marks to students. Eleven responses were classified as 'for' learning (one explicitly using the term), one as both 'of' and 'for' learning, and one was not linkable to either meaning: 'Test/exam/quiz/etc. where mark doesn't count'.
For the second definition of FA, semantic analysis showed that one respondent equated FA with feedback (a different respondent to the one who had done so in giving a first definition of FA), four more respondents mentioned 'feedback' and for a further four, feedback to improve performance was implicit in the response. There was not as strong a link (six respondents) between FA and its developmental purpose as in the first definition of FA. Two defined FA as something that will improve performance and four that it offers a development opportunity that may or may not be taken up by students. One noted that FA 'contributes to overall grade', and one noted explicitly that marks were not for validation of learning. One respondent indicated that FA functions 'with a view to … adapting teaching/learning methods'. Nine responses were classified as 'for' learning (two explicitly using the term), two as both 'of' and 'for' learning, and two were unrelated: 'As per Q1', and 'Where the mark doesn't count'.
FA was recognized by one interviewee as developmental for both students and tutors, enabling the latter to help the former by understanding weaknesses in their learning.
One indicated that FA is much more important because of the feedback it provides: 'formative assessment is the far more important aspect of assessment. It's the build-up of feedback throughout the period of a module, a project, a year, whatever, towards a summative assessment point'. Five interviewees recognized the link between FA and feedback, and four of those emphasized the importance of feedback in FA.
When asked to give an example of an FA task, two questionnaire respondents mentioned 'presentation', three 'essay', and three 'draft'. Two gave observing students performing an activity as an example, one mentioned peer assessment and two 'feedback'. One mentioned 'test', but in the context given below. Six respondents gave an example classified as 'for' learning and six as both 'of' and 'for' learning. One was classified as 'of' learning: 'An in-class test which is designed to assess knowledge of fundamentals.' Regarding informing students, 12/13 questionnaire respondents either always or sometimes told students that a task was formative; 1/13 did not tell his/her students. One responded that informing students that a task is formative 'can be detrimental to effort'. However, all questionnaire respondents explained to students how a task was formative. Some (5/13) indicated that formative work was marked, although few (3/13) indicated that it was graded; 4/13 indicated that formative work was sometimes marked and 5/13 indicated that it was sometimes graded.
While two interviewees stated that it would be useful to include students in FA, one interviewee noted a discrepancy between theory and practice, in that the literature advocates not grading FA, whereas in reality staff do grade FA to encourage student engagement. This interviewee also noted that while FA generally encourages experimentation and risk-taking by students, this is diminished when FA is graded.

Understandings of the relationship between SA and FA
In response to the two questions that asked for a definition of summative assessment, 9/13 respondents gave answers consistently classified as 'of' learning and 1/13 was consistently classified as 'for' learning. In addition, 3/13 participants gave answers classified differently between the first and second asking of the question: 1/13 moved from 'of' learning to both 'of' and 'for'. The remaining 2/13 participants either gave a blank answer or one that was unrelated.
Similarly, for the questions that asked for a definition of formative assessment, 7/13 respondents gave answers consistently classified as 'for' learning, 1/13 was consistently unrelated, and 1/13 gave a blank answer. In addition, 3/13 participants gave answers classified differently between the first and second asking of the question: 2/13 moved from 'for' learning to both 'of' and 'for', and 1/13 from both to 'for' (see Table 1). For SA, 9/13 respondents gave both a definition and an example that were classified as 'of' learning, 3/13 gave a definition classified as 'of' learning but an example classified as both, and one gave a definition classified as 'for' learning but an example classified as both. For FA, 6/13 gave a definition and an example classified as 'of' learning, and 1/13 gave responses consistently classified as both 'of' and 'for' learning. However, 4/13 gave a definition classified as 'for' learning but an example classified as both, 1/13 gave a definition classified as 'for' learning but an example classified as 'of' learning, and 1/13 was unrelated (see Table 2). Of seven interviewees, five explicitly asserted that FA comes before, leads to, feeds into, or supports SA and can improve grades through feedback; for example, 'formative assessment comes at any point prior to that summative assessment. It could even be as near in time as an hour before the summative.' Two noted that the timing of an assessment within a programme or course dictates or influences whether it is FA or SA, and one that SA occurs at the end and does not provide opportunity for students to improve. Despite this last notion, it was further explained that SA can be formative for the next learning stage.
One interviewee also noted that SA, although providing a grade, repeats previous FA processes and tasks, that is, repeated similar FA leads to a plateau in learning such that feedback on SA is unnecessary and the associated mark is no surprise. One interviewee indicated a strong link between SA and mark, although another recognized that FA (in addition to SA) must be linked to learning outcomes and assessment criteria, thereby formalizing the focus of FA feedback and not making it an ad hoc disembodied guess. Another was uncertain about the relationship between assessment, whether FA or SA, and learning outcomes, not recognizing the latter as threshold concepts: 'how can you set learning objectives that are sufficiently open and give space for the person who just barely scrapes a pass and the person who's sailing through to a first?' All questionnaire respondents used FA tasks with students. Almost all used these tasks in class (12/13) and for homework (11/12). Regarding the distinction between FA and SA, 4/13 stated that they kept FA and SA separate, 7/13 did not, and the remainder (2/13) did sometimes; 8/11 indicated that they conflated FA and SA at least sometimes and each gave an example. However, there was little congruence and commonality in the examples: three concerned learners using feedback, two saw conflation as the provision of feedback and the students using that feedback, one concerned the tutor providing feedback, one the tutor adding marks together from different tasks, and one simply described assessment tasks. Four examples involved the notion of generating feedback for students from SA, although there was no consistency in the process of linking feedback with SA. Most questionnaire respondents (10/13) used some student self-assessment and, at least on occasion, used it formatively. Fewer (3/13) used student selfassessment in a summative way; 5/12 thought student self-assessment uses both SA and FA. As a general concept for assessment, one interviewee thought learner inclusion crucial and the other six thought it either useful or desirable. Of these six, three mentioned peer-assessment, one selfassessment, two that it helped learners understand the grades assigned by tutors, and one that it was important to develop group discussions.
Most (11/13) questionnaire respondents thought theory was important to teachers, although interestingly one disagreed, indicating: 'Do you mean research outcomes (yes) or "theory" (probably no).' All questionnaire respondents noted that SA can be used for end-of-course grades, but fewer (4/13) thought that FA can be used for end-of-course grades. One respondent noted 'End of "course" grades aren't formative unlike mid-course grades'. Most (10/12) thought that SA can be used for mid-course grades, and 6/13 thought FA can be used for mid-course grades. All questionnaire respondents who answered indicated that SA assesses for validation at least some of the time, although two did not answer the question and indicated that they did not understand it. Most (7/11) thought that FA assesses for validation at least some of the time. All thought that SA assesses for learning, although three qualified this with 'sometimes' and one with 'a little perhaps', and all thought that FA assesses for learning, although again one indicated 'sometimes'.
All questionnaire respondents regarded SA as assessing product and FA as assessing process. Almost all (12/13) regarded SA as assessing process and most (10/13) thought that FA assesses product. One questionnaire respondent noted that there were 'some discipline differences here'. One interviewee did not discuss this aspect specifically and a further five gave, at least in part, unaligned rationales, two of which did not relate to assessment. One that did address assessment invented new terms to describe the relationship, for example 'semi-formative' and 'semi-summative', although the link to process and product was not made. In three cases, product was stated as being an artefact that is assessed and in two of these cases process constitutes 'workings-out', as is done, for example, in mathematics. One equated SA with product and FA with process. This interviewee stated that SA assesses learning that has happened and FA assesses learning as it happens, and that the development of assessing product occurred in parallel with the development of the phenomenon of learning outcomes.
All questionnaire respondents thought that FA provides useful feedback, and almost all (12/13) thought SA did, at least 'sometimes' (2/13), although one qualified the response with 'but too late' and another with 'limited'. Two interviewees spoke generally about feedback but did not link either FA or SA to feedback. Another interviewee believed that verbal feedback is more powerful than written feedback because students must 'pay some attention' to it. Four interviewees did not link SA to feedback. Of the others, one first indicated that feedback has little or no place in SA, but then changed this by acknowledging that SA shapes learning, so feedback from it is important in learning. Another two believed SA is subsumed within FA and therefore embodies feedback. Of these two, one interestingly clarified that feedback justifies the SA grade, the other that using the feedback from FA for a final SA grade protects the tutor from self-contradiction.
Most (10/13) questionnaire respondents regarded formative work as related to summative work, and 11/13 regarded SA and FA as similar processes, but 5/12 thought SA and FA were different processes. One thought SA and FA were different processes 'sometimes'. Most (9/11) respondents indicated that they were sure how summative and formative assessment relate to each other, although one qualified this with the word 'reasonably'. Two questionnaire respondents were not sure and two declined to respond, which perhaps indicates 'not sure'. Four interviewees showed degrees of individual inconsistency and there were many examples of contradiction, especially in relation to the functions of assessment and feedback. Two used individually created terms, such as 'interim SA' (which has the same function as FA but does not occur at the end of a course) or 'semi-formative' (which was not defined), to explain their stance when the coherence of their argument was challenged. When probed, these terms were exposed to be either meaningless or not understood by the interviewee, or both. Much of this discourse was framed by the assertion that all assessment is the same process and that SA and FA should be aligned, though this was contradicted by one interviewee who stated there could be no learning in SA unless it contained an FA element and who was explicit that FA leads to learning while SA does not. Four interviewees exhibited lack of coherence in the discourses distinguishing FA and SA. While recognizing diversity of interpretations and understandings, one signalled that FA must be a dialogue between students and tutors and also practice for SA, but that there was almost no overlap between FA and SA (because FA is classroom interaction and SA represents tests and examinations). On being asked whether FA and SA assess learning differently, one interviewee did not recognize the difference and all others did not answer the question, merely giving a list of features that can be linked to either SA or FA. For example, SA is a product, more rigorous, impartial, and considered than FA, whereas FA is a process that is more 'slapdash', more immediate, and includes discussions in class.
Almost all (12/13) questionnaire respondents thought that students understood SA, though of those 2/13 indicated 'sometimes'. Fewer (8/13) thought that students understood FA, although 3/13 indicated 'sometimes'. All respondents thought students focused on SA, but only 1/13 indicated firmly that students focus on FA and a further 2/13 offered 'sometimes'.

Discussion
In defining SA, most questionnaire respondents recognized SA as contributing to validating learning and classified it as 'of' learning, though there was considerable variation in the semantics between the first and second definitions offered. Most unexpected was that one respondent consistently defined SA in terms of 'for' learning, going against the grain of convention. Although respondents did not identify feedback as a defining feature of SA, most respondents acknowledged that it does provide useful feedback. In the questionnaires there were strong links between FA, the notion of 'for' learning, and the provision of feedback to students on their performance, indicative of a generally shared perception of FA, even though there were again considerable semantic differences between definitions. Interviewees readily linked feedback with FA but found it difficult to exclude feedback from SA, and therefore found it extremely difficult to differentiate between the two. As with most of the distinctions between FA and SA, this led immediately to internal inconsistencies. From the questionnaire, SA was generally understood more clearly than FA. In the interviews, SA again provided an anchor since respondents associated it with grades, or end of learning or assessment cycles. There was a general inconsistency attached to the meaning of FA, especially between grades and grading and, overall, inconsistency in distinguishing between FA and SA.
There appeared to be some degree of inconsistency within individuals about the functions of assessment, given that questionnaire definitions of SA that were consistently 'of' learning were given by 9/13 of the respondents and 8/13 defined FA consistently in terms of 'for' learning (see Table 1).
This and similar evidence, such as that more examples given of SA were classified as 'of' learning than examples of FA classified as 'for' learning, and that there was greater consistency between definition and examples of SA than FA, indicate that SA is understood more precisely than FA, which seems to be a more nebulous concept and one that is more difficult to grasp, even for educational developers. In the Assessment for Learning literature generally (Harlen, 2006;James, 2006;Black and Wiliam, 2009;Berry and Adamson, 2011), the separation of SA and FA has resulted in the belief that they are unrelated. In addition, they are rarely seen as processes (Taras, 2009;Taras, 2012a;Taras, 2012b). This would go a long way to explaining why FA has become a nebulous concept, even in HE (Taras, 2007), where the work of Black and Wiliam is cited even though the focus of their work is in the compulsory sector.
In general, there was agreement on the types of assessment-related activity and methods of presentation that the educational developers used with their students, many of whom will be academic staff in the early stages of their careers. However, there was disagreement on how the activities could be used to promote learning, for example in marking and grading formative work, and in whether or not to keep FA and SA separate. About half our sample did not keep them separate, and in this group there was little common ground in approach.
Given the centrality of students in the learning process, we were not surprised that most questionnaire respondents were using student self-assessments, although we had expected more than just under a quarter to be using student self-assessment in summative work. Their reluctance might be deep-rooted in notions relating to the traditional role of the teacher in marking and grading, and issues of power (Taras, 2008b;Tan, 2004). In the interviews, there was a general lack of enthusiasm for the inclusion of learners in assessment, despite a strong body of literature mandating for its use (see Taras, 2010), suggesting a lack of engagement with current issues. Peer-assessment was favoured over self-assessment.
Although respondents had in the main a shared understanding of the relationships between SA and FA and assessment of process and product -with the deviation from unanimity being in the small number (3) who did not agree that FA assesses product -in general there was a high degree of differentiation about the uses to which SA and FA are put. In particular, for some respondents use appeared to be dependent upon when, during a course, the assessment was made, which is a curious notion, although not without precedence in the literature (Bloom et al., 1971;Black and Wiliam, 2009;Wiliam, 2009). Also, although most recognized SA and FA as similar processes, half also recognized them as different, and almost all stated they were sure about how SA and FA relate to each other. Thus, many respondents are sure that SA and FA are not different, yet at some points all respondents identified how SA and FA are different from each other. This is clearly problematic, and has already been demonstrated in academic staff (Taras, 2008a;Taras and Davies, 2014).
Most questionnaire respondents agreed that SA and FA are similar processes, yet about half tried to explain considerable differences between the two. In most interviews the inconsistencies were much more marked than they had been in the questionnaires. The most rational aspect was that product was the artefact that is assessed impartially, and in a more considered and rigorous manner than FA. FA, on the other hand, is a process that is more informal and more immediate, and that includes classroom discussions. The above links directly to much of the literature on 'Assessment for Learning', as discussed by Black and Wiliam (2009) and Wiliam (2007;. The general perception that students do not focus on FA represents a challenge for the sector, one that reflects the negative perceptions of SA, which is blamed for dominating students' priorities (Broadfoot and Black, 2004;Broadfoot, 2008;Berry and Adamson, 2011). Further, just one of our educational developers had managed to create an environment where her/ his students focused on FA. Hargreaves (2005) examined teachers' perceptions of functions of assessment and noted that most prioritized assessment for validation purposes. This could also link to tutors prioritizing SA over FA unconsciously and unwittingly, even if they attribute this focus to students.

Conclusions
There are two main limitations to this study. First, the use of yes/no answers in the questionnaire may restrict responses from participants, although this was partly mitigated by the use of the semi-structured interviews. Second, the small sample size may restrict generalizability, although we were fortunate that our sample was from a broad range of HE providers.
When asked to present their views orally, respondents readily displayed greater diversity and inconsistency, much more so than when answering the questionnaire. Further, in many cases oral contradictions came very swiftly, often within a few seconds. On the other hand, they were enthusiastic, even when they were obviously unsure of their argument and lacking in information, and two respondents invented phrases to justify their stance. Many interesting ideas were aired that did not fit in to the rest of their discourse.
These staff developers, who are likely to have a good command of general and specific educational literature, made almost no use of it in explaining or evidencing concepts or their practices or positions on assessment. From personal observation we note that students often do not recognize learning outcomes as thresholds, and this is perhaps because they are not presented as such by tutors. However, we had not expected this common misconception to be present in this context.
The variation in perceptions and attitudes in relation to assessment indicates a lack of accepted definitions for common concepts used in HE by those who are looked upon as experts. Until there is a framework of understanding that individual definitions can be inserted into, such that differences are acknowledged and indeed championed, it is unlikely that progress can be made in pursuing academic understandings of the nuances of assessment, how it works, and how decisions are arrived at and used. There seems little acknowledgement of the individual differences in such a fundamental aspect of HE, and this can lead to entrenchment and positions where individual beliefs are presented by teacher to student (or from educational developer to academic staff) without acknowledgement of the complexity of the concepts and the richness of their interpretation. If there is no shared understanding, or acknowledgement of complexity, by educational developers, who are looked upon as authority figures, what prospect is there for consensus of understanding among academic staff, those who bear the brunt of assessment and its design in HE? The implication is that a much greater sharing of assessment literacies should be an integral part of pedagogic development.
Thus, we call for the development of such a framework, perhaps led through active communities of practice that support congruent, rational, and logical interpretations of practice that align with theoretical perspectives of assessment. Greater dialogue, perhaps mediated by national HE bodies but that permeates within individual HE providers, would seem a way forward, perhaps even concurrently with activity in other sectors to promote a shared understanding of assessment theories and practices.
YES Thank you very much for your time and brain power.

Appendix 2 Generic audio interview questions
• How do formative assessment and summative assessment relate to each other? • Where does feedback fit into formative assessment and summative assessment? • Do you distinguish between assessing a process and a product? If so, how? • How do formative assessment and summative assessment assess learning differently? • Is it important to include learners in assessment? If so, how and why? • Do you want to comment on the questions asked or the process of enquiry?

Notes on the contributors
Maddalena Taras's research has focused on a range of assessment issues: self-assessment -developing an original framework and examining issues of power; institutional discrepancies and contradictions in assessment practices and discourses; constraints from language, culture, and power impacting on assessment perceptions and practices; and critiquing the notion of 'assessment for learning', particularly the theoretical framework.
Mark S. Davies is a biologist with a large portfolio of activities relating to learning and teaching in higher education. His research in education concerns formulating strategies for student retention and the use of computer simulations in teaching complex subjects.