Recruitment practice: towards equity in process and outcomes within research

Introduction

Creating a more diverse research environment is an important aspect of producing wide ranging, internationally leading research which will have an impact. Work to improve diversity in the energy research sector must be tackled both downstream, including within the student body, and upstream with the encouragement of, and creation of opportunities for, diverse teaching and research posts.

The purpose of this document is to highlight the ways in which individuals working as Principal Investigators (PIs) and team leaders can improve recruitment to encourage a more diverse talent pool in energy research. What follows is a brief walk through the entire recruitment life cycle for those with responsibility for recruitment into research teams, with some ideas on how to consciously consider ways of broadening the diversity of the workforce.

Identifying the role

There are certain jobs, areas of research and roles which are seen as essential in any team or work stream. Then there are those which are ‘nice to have’, ‘if funding allows’, or ‘in an ideal world’. Historically, matters pertaining to women, or marginalised groups tend to find themselves in this latter category. If an organisation wants to bring about a real and lasting diversification of its workforce, it is essential that the roles which are given priority are critically examined. In a research environment, there are likely to be areas of special interest which have been traditionally overlooked, and therefore given rise to less opportunities in those fields.

Related reading:

• With little research, and less data, women’s health remains overlookedOpens in a new tab | Eithne Duffy | The University Times
• Downside up: Science matters equally to the Global SouthOpens in a new tab | Regina R. Rodrigues | Nature

The first step towards a diversified research environment is to fund more diverse research.

Questions to consider:

• Are you creating the usual jobs for the usual people?
• Is there the scope to prioritise something different?

Suggested activities:

1. At the inception of your work package, carry out a gap analysis which includes looking specifically at the often overlooked or less glamorous areas of your specialisation. Look for opportunities to work on areas which may touch on less represented communities. Look for opportunities to bring different expertise into your team.

Creating the job description

When putting together the job description, it is important to examine essential elements of the role. Any items which do not have a material bearing on whether the applicant could perform as required should be removed.

For example, many roles within universities and on research projects, may be advertised as requiring a PhD. It is important to look closely at such a requirement and consider whether it is necessary. For example, are there other ways that a candidate could demonstrate deep expertise, data gathering and analytical skills, or intellectual authority?

It is important to stress that this does not mean a lowering of standards, but rather the adoption of more imaginative ways to discern whether those standards are met. This allows for the consideration of candidates who may have followed a different journey to reach their current place.

Questions to consider:

• Are you putting in requirements which are essential to the role? Can they be demonstrated by equivalent experience?
  

Suggested activities:

1. Consider the ways in which relevant equivalent experience could meet your requirements and include a statement that you will accept equivalent experience in the job description and advertisement.

Competitive or non-competitive recruitment

Non-competitive recruitment occurs within academic institutions and on large, funded projects with people being re-deployed from one role to another. This way of staffing helps to mitigate against the worst effects of precarious and short-term contracts which are a feature of academic life. One of the potential consequences of this, however, is to slow the process of diversification of the workforce, as a bottle neck is created for bringing in new people even if greater diversity exists in the talent pool.

Those in charge of staffing decisions need to strike a balance between competitive and non-competitive recruitment and ensure that sufficient roles are fully open. Collaboration and knowledge exchange naturally rely on networks and the building of long-standing relationships, but recruitment through the network can perpetuate exclusion and exacerbate existing homogenous environments.

Questions to consider:

• Does the team you have assembled consist mainly of people you have worked with before?

Suggested activities:

1. Take steps to broaden your team. Consider holding aside at least one position in every project for an open, competitive recruitment round.

Writing the advertisement

The job advertisement is first and foremost a marketing document designed to attract qualified candidates to your vacancy. Whilst the language should be precise in order to filter out unsuitable applicants, its aim should be to appeal widely so that candidates do not self-deselect.

Research has shown that the use of some styles of recruitment language can create exclusion which may be unintended.  (Gaucher, D., Friesen, J. and Kay, A.C. 2011. Evidence that gendered wording in job advertisements exists and sustains gender inequality, pdfOpens in a new tab | Journal of Personality and Social Psychology).

Certain language can lessen the number of female applicants, and formality or use of hyperbole in language can convey a culture which may cause people to opt-out.

For example, it has been shown that words which convey excellence over others, such as ‘world-class’ or ‘leader in the field’ will decrease the number of female applicants. Other ‘dominance’ toned language such as ‘drive-through a net-zero agenda’, or ‘be an evangelist for energy reduction’ may also be encoding masculinised attributes into the role which could be off-putting to some qualified candidates.

An overly formal style such as using the term ‘the successful applicant will have…’ rather than ‘you will have…’  can also be alienating and may not accurately reflect the working culture.

There is also a tendency for job advertisements to contain hyperbolic descriptors for skills and attributes; ‘outstanding communication skills’, ‘superb attention to detail’, ‘exceptional leadership skills’. It is better to allow candidates the opportunity to give evidence of their experience and for you to evaluate whether they reach the required level.

Some institutions have started using software such as Recruitment marketing | Unlock your pipeline potential (textio.com)Opens in a new tab which aims to help remove the bias in language and job advertisements. These tools can be a useful addition to the process but should not be seen as a shortcut and time should still be taken to ensure that the text reflects the aims and values of the recruiting team.

Related reading:

• Towards diverse workforces: Transforming the language of exclusion and bias in recruitment, pdfOpens in a new tab | University of Nottingham

Including EDI statements

Containing a statement about the organisation’s commitment to EDI is now often standard practice. Whilst it is a step forward, it is often a set form of words tacked on to the end of the advertisement. As such, it can come across as tokenistic and of little value.

Positive statements to encourage applications from underrepresented groups are a good start, but it is far better to ensure that an organisation’s commitment to EDI is incorporated throughout the job advertisement.

Leadership roles can specifically mention a desire for the successful candidate to demonstrate an incorporation of EDI in their approach to team development, or leadership style.  The text should aim to tell the candidate about the expectations for them to be in the office or travel or work from home. The criteria should reflect how equivalent experience will be considered and given due weight. A willingness to design the hours around a candidate’s caring responsibilities could be mentioned. A real reason for wanting to diversify the team could be mentioned, rather than the HR mandated form of words.

These things can demonstrate a thoughtful and authentic attempt towards continuing improvement rather than ‘performative’ EDI.

Questions to consider:

• Are you using old templates for writing job advertisements? Look afresh at the language. If possible, get the opinions of a variety of people (a mix of genders, ages and job types). Is there anything in the language which appeals or repels them?

Suggested activities:

1. If you have developed and implemented an EDI policy successfully, mention it in the job text rather than just use a standardised EDI statement at the end.

Circulating the advertisement

Try to think of different places where the advertisement can be placed. Of course, these are specialist roles and will be directed at a very particular pool of people, but consider whether links can be made with organisations who will boost the range of people that see your vacancy. Groups such as Runnymede Trust, or Black British Academics, or groups which promote women in STEM, will often carry vacancies in newsletters and on websites. The vacancy will probably reach someone who has also seen it on your own website, but seeing it boosted by such organisations may help to underscore that your team is serious about increased diversity.

How long to advertise?

Institutions sometimes have set time periods for the length of time a vacancy should be left open. Consider whether you are advertising your vacancy for long enough. A two-week period for example will leave only one weekend for a candidate to complete the application which may be harder for those with caring responsibilities. There also needs to be genuine fairness if the process has both internal and external candidates involved. A short advertising schedule will improve the odds for the internal candidate but will continue to slow the progress of new and diverse applicants joining the organisation.

Suggested activities:

1. Try, where possible, to choose a minimum of 3 weeks for an open vacancy – 4 weeks if at all possible.

The shortlisting process

Time must be given to undertake a proper evaluation of the applicants. There are some process steps that can help to level the field for candidates at this stage, and your institution may have some of them in place already.

Anonymising of applicants

A simple way of anonymising applicants is to refer to them with a reference number. This can help to remove any potential for gender or racial bias and there is evidence to show that this can be effective.

Another helpful step can be the removal of the names of personal and job referees at the shortlisting stage which will help to anonymise the applicants. Other ways of doing this are the so called two-click processes where an initial screen is made on anonymised applicants but then further information is revealed at stage two.

Suggested activities:

1. Try to anonymise name and reference data from first screening.

Anonymising of other data

It can be more problematic to remove other details throughout the application. For example, it will be possible to remove a candidate’s name but if their educational details reveal a high school in Delhi and a first degree in Mumbai, it is likely that a conclusion can be drawn about ethnicity. Another complexity is around social mobility. There are some organisations attempting to remove the names of schools and universities from applications in order not to unconsciously favour those who attended ‘elite’ institutions. Furthermore, humans are adept at picking up large amounts of data about others based on the briefest interactions. Clues about socioeconomic background can be communicated in a few words. That in itself cannot be filtered out of any hiring process. What matters is whether other attributes (professionalism, competence, intelligence,) are then ascribed to the candidate based on that piece of knowledge alone).

Related reading:  

• Evidence for the reproduction of social class in brief speechOpens in a new tab | Michael W. Kraus, Brittany Torrez, Jun Won Park, and Fariba Ghayebi | PNAS

In reality, a pragmatic balance needs to be struck. For some job roles, complete anonymisation is possible, and the well-known case study of orchestras holding auditions from behind a screenOpens in a new tab is compelling evidence to show that the removal of bias can open up opportunities. However, recruiting into an academic research team will necessitate disclosure of a person’s educational background, their publications and other distinguishing factors. Nor is it implicitly biased to give more weight to some achievements than others. What is helpful, however, is for those screening the applications to be aware that they must try to evaluate only the evidence in the application, rather than ascribe characteristics because someone has attended a particular institution.

There is an increasing move to incorporate anonymisation into academic recruitment, but it is difficult to evaluate its effectiveness. Whilst it might, on the face of it, result in the emergence of candidates from less elite institutions or those with less experience, it is not yet clear if these results are yielded through a fair and anonymised process or, simply through the operation of a different set of unconscious biases.

Perhaps the most important thing is to focus on fairness and transparency in the process itself.

Suggested activities:

If a form of anonymisation is to be used;

1. Clearly explain the type of supporting statement or material that a candidate can submit.
2. Be prepared to offer guidance on how to describe and evidence previous work and experience.
3. If queries are sent before the deadline, share the answers with all applicants and make them widely available through a website or newsletter so that all candidates have access to the same information.
4. Clearly explain how applicants can evidence their expertise if not through the disclosure of their publications.
5. Provide sufficient information on what material will be removed at the screening stage before the application reaches the panel.
6. Provide information on the weighting given to different elements of the application or proposal.
7. Ensure that sufficient time and expertise is allocated to the editing stage.
8. Supply the applicant with a copy of the submission which is finally put in front of the panel.

Extensive anonymisation is still fairly new in academic recruitment and it is therefore important to be as considered and open as possible to ensure that there is confidence and faith in the process.

Evaluate the equivalent experience

If you have told candidates that equivalent experience is valued, ensure your screening score sheet properly incorporates this. Be prepared to spend a bit more time understanding how the candidate came to their current position and what skills they have acquired in both formal and informal settings.

Questions to consider:

• Are you using a standardised score sheet? Does it give weighting to equivalent experience? How are you going to evaluate it?
• Are the people on the hiring panel briefed or trained about assessing the application in front of them and not assuming suitability based on preconceived ideas?

Suggested activities:

1. Consider unconscious bias training which has been specifically designed for recruitment.
2. Make sure your scoring sheet reflects the criteria which you stated in the job advertisement.

The interview process

Ask for HR support in managing the interviewing process. The HR team should, as standard practice, check with the candidates if any accommodations should be made, and needs must be addressed before embarking on the interview. This means any tests or tasks you may be asking the candidate to do must be adapted to take into account a candidate’s disability as disclosed.

Ensure your recruitment process does not contain a ‘one of us’ test. Some recruitment processes can follow well designed principles but fall at the last hurdle when a candidate is invited to meet the team, particularly at a dinner, or social event. These type of recruitment tests are, on the face of it, valuable for seeing how the candidate will fit with other team members. However, they inevitably favour the status quo. A candidate who is most similar to all your existing team members is likely to pass with flying colours. The failing candidate will be a puzzle because they looked perfect on paper, and performed well at interview but there was just ‘something which didn’t fit’. Diversity means difference, and different does not always, seamlessly fit.

Monitoring and evaluating

Recruitment is just one of the levers which can be pulled to encourage better diversity in research. It is useful to review and report on the composition of the workforce, allowing issues of disparity to be discussed openly and benchmarked with a view to improvement.

A survey carried out for CREDS in 2020 across the participating members found that the gender split of research posts was 80/20 M/F for competitively recruited posts and 61/39 for non-competitive recruitment. Keeping an eye on the statistics throughout the life of the project is an important exercise.

Carry out self -reflection after your recruitment

After each recruitment activity, it is worth reflecting on whether the process yielded the quality and diversity of candidates that you hoped for. Even if you are delighted with your new recruit (as you hopefully will be), the process and the pool of candidates should still be reflected upon.

For example, a role at CREDS received a range of applications from around the world, but not a single application from a British born and educated black or, brown person even though the subject of research was one which touched upon social and racial justice. It would be valuable to reflect on why that may be the case.

Questions to consider:

• What is the perception of the recruiting institution with these groups?
• Did the advertisement reach them?
• What were the barriers to application?

Summary

In summary, the lack of diversity in energy research is not simple to overcome, and time will be needed to see more, and more suitably qualified individuals move through the pipeline. However, the small steps outlined above, which are in the control of PIs and researchers doing recruitment can start to make a difference.

Banner photo credit: Clay Banks on Unsplash