A new software application is helping to shatter the glass ceiling in STEM academia. The app was developed by Lero, the Irish Software Research Centre, as part of an EU-funded gender equality project. Prof Ita Richardson, Dr Clare O'Hagan and Dr John Noll report
Tech

The FESTA Strategic Career Manager (FESTA-SCM) is a decision support system for junior to mid-level academics and researchers. It provides individuals with a personalised career-development profile based on their responses to questions.

While this system has emerged from research with a focus on women and on science, technology, engineering and maths (STEM), it is useful to both women and men, and to academics and researchers from disciplines outside STEM.

FESTA-SCM was developed in collaboration with researchers from Lero (the Irish Software Research Centre) during an EU-funded action-research project, Female Empowerment in Science and Technology in Academia (FESTA).

FESTA is concerned with women’s under-representation at senior levels in STEM disciplines internationally. Undertaking FESTA research, we identified gender differences in the career trajectories of male and female academics and researchers. We then structured output from the project into a usable on-line format, which is now available in FESTA-SCM.

FESTA-SCM will increase the awareness of academics and researchers regarding factors that influence their work and careers. These include institutional procedures and politics, social expectations, career requirements and work-life balance.

FESTA research focuses primarily on cultural, structural and organisational factors that impede women’s progress in STEM. Sub-projects within the project are primarily concerned with organisational processes and practices, for example: Perceptions of Excellence in Hiring Processes; Gendering Decision Making and Communications Processes; PhD Supervision; Gendering Meeting Cultures; and Gendered Quantitative Indicators and Organisational Awareness Raising.

Because gender also operates at individual, as well as organisational levels, one sub-project, Individual Awareness Raising, was undertaken to raise awareness of the differences of the ways gender operates for individual men and women in STEM, thus affecting their career trajectories differently. This sub-project was led by the University of Limerick.

Collecting and analysing gender knowledge


festa-1

CLICK TO ENLARGE Fig 1: Recommendation Promote your work structured into Pattern format and implemented in FESTA-SCM

Four case studies were undertaken in the University of Limerick, Bulgaria’s South-West University, University of Southern Denmark and Istanbul Technical University. We specifically investigated critical incidents, positive and negative factors that influenced careers and individual and organisational supports.

Documentary review was also conducted to examine policies regarding hiring and promotion which can advance careers in the universities. Adopting a critical realist approach, the career trajectories of men and women at early, mid and senior levels were investigated to determine factors that advance or impede a career in academia.

The qualitative methodology is both processual and reflexive, in the grounded theory tradition. By exploring men’s and women’s career trajectories in the case-study institutions, we are enabled to explain women’s under-representation at senior level.

The questions driving this study are: What are the differences between men’s and women’s careers? What are the critical points in a career trajectory? What organisational and individual factors advance or impede men’s and women’s academic careers? The research methodology was designed to provide answers to these broad questions, as well as to provide sufficient knowledge to generate theory regarding gender issues within STEM careers.

The key questions were developed into wider concepts which included ambition, mobility, mentoring, collaboration, research, individual and institutional support, work/life balance and family. Interview guides were designed by the University of Limerick in collaboration with partners to ensure a fit to local contexts. Guides contained a combination of open and closed questions, as well as reflections on critical incidents.

The research sample included both men and women, based on their positions and their gender. It was expected to include a purposive sample of 50/50 men and women, but the research team agreed not to compromise by including alternative positions (higher or lower) where there was no female comparator, in order to highlight women’s under-representation in STEM.

Positions at early, mid and senior levels were selected at points where there appeared to be critical thresholds, i.e. the points at which the numbers of women increase/reduce dramatically. Interviews were conducted with men and women at early- mid- and senior levels in each the four universities.

Overall, the sample includes 106 (57 male, 49 female) academics and researchers across the four organisations. Content analysis was used to analyse the data, ultimately providing categories, clusters and themes.

Process patterns


In developing FESTA-SCM, we built on the existing Lero’s aSPIRE system for the Global Teaming model [1]. Therefore, the approach is based on the use of process patterns. Many readers of this journal will be familiar with process patterns, as they were first conceived as a way to document building architecture design solutions [2].

Since then, patterns have been used by software developers in the form of design patterns, organisational patterns and process patterns. In general, a pattern starts with a description of a specific problem (design, organisational or process) that the pattern addresses.

Our goal was to develop a useful product for people interested in career development, while allowing us to structure the knowledge collected and analysed. The use of process patterns allowed us to this. In each pattern, background is included that explains the problem, the context in which it occurs, and other information to help the reader to understand the problem and its solution.

The key pattern component is the solution, which describes, as a succinct set of specific tasks or interventions, how to address the problem. References to related patterns that provide alternate or complementary solutions are included.

Finally, we have added references to the research literature from which the pattern was derived, to provide both international research on the topic in question as well as evidence of the empirical grounding of the pattern.

By adopting a pattern-based approach, we were able to realise several benefits. First, patterns provide FESTA-SCM with the necessary depth to be useful in practice, without overwhelming the user who is looking to develop career-development plans.

Second, the context and provenance of the patterns’ recommendations are presented along with the recommendations themselves allowing users to verify both the applicability and validity of the recommendations. Finally, a pattern-based approach significantly reduces the overhead of translating research results into a formal notation usable by a decision support system.

System description


FESTA-SCM is deployed as a web application. The Knowledge Base contains triggers and rules for the inference engine. Patterns are contained in a semantic wiki that allows individual researchers to add and edit patterns using a web browser. Users can browse the pattern library in the same fashion as any static web site. Further information on the inference engine is available in [1].

In the FESTA-SCM user interface is a set of interview questions which captures the characteristics of the user, as each user individually completes their personal profile. This results in them receiving personalised recommendations which directs them to focus on particular actions to suit themselves, presented in a ranked order of importance. The User Interface is detailed below.

The Knowledge Base enables users to be presented with recommendations based on their personal profiles. The Knowledge Base was developed in five steps:

  1. Structure each recommendation for career development into a pattern format;
  2. Group recommendations into categories;
  3. Devise career-stage Interview questions;
  4. Allocate recommendations depending on the answer to each question;
  5. Rank recommendations in order of importance.

1. Structure recommendations into pattern format
For FESTA-SCM, patterns are the result of distilling empirical career-development management research, developed during FESTA, into concise descriptions of a problem or goal and its solution. A process pattern comprises the following main components (see Figure 1, above):

  • Recommendation title (What): a pattern’s name is important because it can convey the purpose of the pattern and the actions or practices it recommends.
  • Importance: all recommendations are ranked. These can be used as guide to help decide the order in which users will implement their personal recommendations.
  • Problem or goal (Why): this describes, in one or two sentences, the problem that will be solved, or the threat to success that will be averted if the recommendations are implemented.
  • Evidence/proof: this section presents the evidence from the research and the international literature which demonstrates why/how this recommendation has worked.
  • Solution (ways to get there): short list of specific actions to take to implement the recommendation.
  • Related patterns: some patterns have strong relationships with other patterns – this section can capture these relationships.
  • References: provides links to the literature and the empirical FESTA research from which the pattern is derived, if any.

Fifty recommendations were identified, developed into a pattern structure and implemented into FESTA-SCM.

2. Group recommendations into categories
In addition to identifying recommendations, the research team grouped recommendations into categories and sub-categories, presenting succinct results to the user. The system hosts three categories, eight sub-categories and fifty recommendations. Figure 2 (below) shows detail for the Career Recognition category.

Category Sub-Categories Patterns
 

Career Recognition

 

Publication Strategy

-Achieve duly recognised research results

-Develop relationships with established academics and publish jointly

-Develop relationship with journal editors

-Publish strategically

-Target high impact journals

 

Professional (Visibility) and Networking

-Acquire international visibility

-Be mobile

-Extend your network

-Join research consortia

-Know your competitors

-Study abroad

Figure 2: Category Career Recognition, with sub-categories and recommendations
festa-3

CLICK TO ENLARGE Figure 3: Career Stage Interview as implemented in FESTA-SCM

3. Devise career-stage interview questions
The real value of a library of patterns is the ability to select those patterns that are relevant to a specific context. Each pattern, therefore, needs a rule or trigger that defines when FESTA-SCM should recommend the pattern. These triggers were developed by devising a set of questions in a focus group with experienced STEM researchers. This set of questions is the career-stage Interview through which the system develops a user’s personal profile.

4. Allocate recommendations depending on the answer to each question
The FESTA project team participated in a workshop during which, for each potential answer to the questions in the career-stage interview (Figure 3), we made a decision whether the recommendation was relevant or not relevant. Using the output from this stage, triggers were expressed in the system using FESTA-SCM’s rule definition syntax.

5. Rank recommendations in order of importance
When a researcher completes the career-stage interview, it is likely that they will be given in the region of twenty recommendations which they should follow. To make their development of an action plan easier, we have ranked all recommendations in order of importance. This ranking was decided by analysing answers to questionnaires completed by FESTA researchers in the four partner universities. To reduce bias, we circulated 26 different questionnaires with the recommendations presented in a different random order in each.

Using FESTA-SCM


festa-4

CLICK TO ENLARGE Figure 4: FESTA-SCM menu

FESTA-SCM is available for use by following this link. The user is normally a junior to mid-level academic or researcher who is interested in managing her/his personal career development. When you log into the system, the user guide is available on the left-hand side menu (Figure 4).

When a personal career-stage Interview is completed online, the system presents a number of Recommendations. As FESTA-SCM is a decision support system, it is not designed as a prescriptive system. However, it will support the individual in making career-development decisions. Therefore, FESTA-SCM can be used in different ways, for example:

  1. Following each recommendation in the ranked order suggested;
  2. Choosing the recommendations deemed by the user to be most important, without completing an interview;
  3. Changing the order of recommendations because this may better suit the current situation;
  4. Choosing recommendations from within a single category.

FESTA-SCM is available for you to use. We hope you get benefit from it and that it supports you in your career development. FESTA reports, which we encourage you to read, are available here.

This project has been supported by both FESTA, European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no 287526 and Science Foundation Ireland through Lero – the Irish Software Research Centre, grant 10/CE/I1855. Thanks to FESTA Principal Investigator Prof Pat O’Connor and to Lero researcher Seun Adebowale who worked on the development of FESTA-SCM.

Authors:
Prof Ita Richardson, Lero – the Irish Software Research Centre; Dr Clare O’HaganFESTA project manager, University of Limerick; and Dr John Noll, Lero – the Irish Software Research Centre

References:
[1] Noll, John, Sarah Beecham and Ita Richardson, Patternizing GSD Research: Maintainable Decision Support for Global Software Development, Ninth IEEE International Conference on Global Software Engineering (ICGSE2014), August 18-21, Shanghai, China. pp 110-115
[2] C. Alexander, S. Ishikawa, and M. Silverstein, A Pattern Language: Towns, Buildings, Construction. Oxford University Press, 1977.

http://www.engineersjournal.ie/wp-content/uploads/2016/09/FESTA-1024x683.jpghttp://www.engineersjournal.ie/wp-content/uploads/2016/09/FESTA-300x300.jpgDavid O'RiordanTechLero,software,technology,UL,women in engineering
The FESTA Strategic Career Manager (FESTA-SCM) is a decision support system for junior to mid-level academics and researchers. It provides individuals with a personalised career-development profile based on their responses to questions. While this system has emerged from research with a focus on women and on science, technology, engineering and...