EDI and Unconscious Bias in REF 2021 Facilitated

EDI and Unconscious Bias in REF 2021 Facilitated by Susan Lee April 2019 *Based on EDI and REF 2021 event by Advance. HE

Principles of respectful enquiry • • Honour confidentiality Be mindful of the impact of what we say Assume good intention in others Listen to understand

Workshop objectives = Understand the legislative and policy drivers and context for embedding consideration of equality and diversity in REF 2021 = Understand the concepts of conscious and unconscious bias and how these can impact on REF 2021 decision making = Discuss strategies individuals and the College can adopt to mitigate against bias in any decision in relation to REF; and embed equality and diversity

Why is equality important in REF 2021 • Funding bodies’ and institutions’ obligations under equality law - See Table 1: Summary of equality legislation in guidance on codes of practicewww. ref. ac. uk/media/1086/ref-2019_03 guidance-on-codes-of-practice. pdf • Ensure staff treated fairly - Better retention, performance and satisfaction • Evidence from previous exercises that some groups of staff are under-represented

Public Sector Equality Duty (PSED) of the Equality Act 2010 In the context of the REF, HEIs and funding bodies need to demonstrate ‘due regard to’ the need to: = Eliminate discrimination, harassment, victimisation and any other conduct that is prohibited by or under Equality Act = Advance equality of opportunity between persons who share a relevant protected characteristic and persons who do not share it = Foster good relations between persons who share a relevant protected characteristic and persons who do not share it

Individual characteristics recognised in REF = Age = Religion & belief = Disability = Sex (including breastfeeding) = Caring responsibilities (children, older, disabled people) = Sexual Orientation = Gender reassignment = Family related leave = Marriage & civil partnership = Part-time employment status = Political opinion (Northern Ireland only) = Early Career researcher = Pregnancy & Maternity = Race = Welsh Language (Wales only) = Junior clinical academics = = protected characteristic in Equality Act 2010

Intersectionality and the accumulation of disadvantage = Intersectionality – ‘the ways in which our many identities interact to produce effects that are not simply the sum of each of them’ (Crenshaw, 1991) = The accumulation of disadvantage – ‘very small differences in treatment (which can) result in large disparities in salary, promotion and prestige’ (Valian, 1999)

Unconscious / Conscious bias as related to REF 2021

Definition of Unconscious Bias Unconscious bias refers to a bias that we are unaware of, and which happens outside of our control. It is a bias that happens automatically and is triggered by our brain making quick judgments and assessments of people and situations, influenced by our background, cultural environment and personal experiences (ECU: 2013 Unconscious bias in higher education)

Psychological explanation = Our brains use two ‘thinking’ systems that operate simultaneously*. (Kahneman, 2011) = We evolved to instantly categorize people into in-group and out-group (Tajfel & Turner, 1979, Elias and Scotson 1994) = Categories are also formed through Unconscious X 200, 000 Norretranders 1998 association = Biases are wired within the brain. (Bodenhausen and Mc. Crae, 1998) = Unconscious people preferences are just as important

What is unconscious bias and where does it come from? What is unconscious bias and where does it come from http: //www. shirepro. co. uk/about. php

Unconscious Bias and Stereotypes Our biases lead to stereotypes = Particular types of knowledge structures or cognitive schemas that link group membership to certain traits; and = Influence the interpretation of others’ behavior …………… and behavior toward others = Also influences what/whose academic work we value Purkest et al, 2006

What biases might play out in REF decisions? Affinity bias (in-groups and outgroups) Stereotype threat Confirmation bias Stress, pressure Unconscious or implicit bias Micro behaviours

Stereotype threat and REF ‘The condition where we behave, or feel like we behave, in a way that confirms a negative stereotype about a group that we belong to. ’ It may lead to: = Poorer performance in assessments/other tasks = Acceptance of poor performance/distancing from the task = Distancing from stereotyped group = Redirection of aspirations and career paths Steele and Aronson, 1995

Micro-behaviours = Micro inequity – apparently small events which are often ephemeral and hard-to-prove, events which are covert, often unintentional, frequently unrecognized by the perpetrator, which occur wherever people are perceived to be ‘’different’’ = Micro affirmation – apparently small acts, which are often hard-to-see, events that are public and private, often unconscious but are very effective, which occur whenever people wish to help others succeed. Mary Rowe, MIT 1973

Bias in metrics = ‘Any system of assessment based on total citation number(such as h-index) was likely to favour more established researchers, as they would have had more time to produce articles and other outputs’ = ‘Individual researchers should be mindful of the limitations of particular indicators in the way they … evaluate the work of colleagues’ = ‘Quantitative indicators are as infused with human values as qualitative approaches’ = H indices also favour sciences and large disciplines; Google scholar and Alt metrics should also be used

Metrics and gender Productivity - men publish more papers (although the gap differs between fields and subfields) - men are around 50% more likely to be invited by journals to submit papers (Although most journal papers are responsive mode not invitation, see above point) Self-citation: men are 56% more likely to cite their own work, 84% in mathematics (Bergstrom et al) Gender gap in citations: both men and women are more likely to cite work by other men (Mitchell et al, Sugimoto et al, Fong and Wilhite)

Metrics and gender • Authorship: women are less often found in authorship positions usually associated with seniority (Holman, Stuart Fox and Hauser; West et al) • Reach: women's publication portfolios are more domestic than men’s (Sugimoto et al) • Culture: female authors 38% more likely than male to be ‘pressured’ to add authors to papers – usually by lab director or other senior academic (Fong and Wilhite ) mainly in science

Using metrics - ethnicity • In some areas of Social Sciences and Humanities, outputs are more likely to be produced in the relevant national language, which may not be English. • Outputs not produced in English are less likely to be included in certain bibliographic databases and indexed, and may be less likely to be captured in bibliometric indicators. The Metric Tide, 2015

Conditions that exacerbate biases Bias control is less effective when we are: = Cognitively or emotionally preoccupied = Stressed = Under-time pressure/rushing to make a quick decision = Under emotional load = Physically tired or low in glucose = Having conflicting priorities = Being challenged

Discussion In groups discuss: ‘’What overall strategies can you adopt to mitigate against bias in any decisionmaking in relation to REF? ’’

Embedding equality and mitigating bias in REF 2021

REF 2021 - key changes • HEIs must return all eligible staff with significant responsibility for research • Each submitting unit will return: – total number of outputs equal to 2. 5 times the summed FTE of the unit’s submitted staff – a minimum of one output and no more than five attributed to each staff member returned – in exceptional circumstances units can request reductions in outputs – it is completely up to individuals if they would like their circumstances to be considered • Institutional codes of practice should cover the processes for: – identifying staff with significant responsibility for research – determining research independence – selecting outputs, taking account of staff circumstances

Embedding equality and mitigating bias Critical in: 1. Determining: – Who has significant responsibility for research – Who is an independent researcher 2. Selecting outputs: – Determining the number of outputs individuals contribute to the pool – Assessing personal circumstances – Creating a culture and process supportive of disclosure of personal circumstances 3. Calculating reductions 4. Selecting impact case studies

Group Discussion

Equality is supported in REF through… = Codes of Practice = Equality Impact Assessments = Local and institutional environment statements = Appeals process

Summary = Legislative and policy driver and why equality is important in REF = Unconscious Bias in relation to REF 2021 = Explore ways of embedding equality and mitigating bias