Perspectives

Information overload

I suppose that it is inevitable that as an information scientist I should have a keen interest in information overload. It is rare that a month does not go by without me dipping into the books by Wright (2007) and Blair (2010) for a glimpse into life before the digital age. Every age has had to cope with the problems of an information overload caused to some extent by a change in technology, be it printing, the telegraph or the CD-ROM. The term itself dates only from 1960 in an article by Grier Miller in the American Journal of Psychiatry entitled ‘Information Input Overload and Psychopathology’. This is just one of the many facets of information overload presented by Levine (2017) in a very thoughful and polished article. In his article, Levine takes us on a journey from Vannevar Bush to George Miller, Katz and Kahn and Linscott, ending up to the work of Daniel Levetin.

The focus of the article is on the way in which the term was used, and mis-used, in both the popular and academic press in America. Levin notes that what has happened to the term is that it has been appropriated by social scientists and psychologists as a way of describing a limited capacity communications channel for bits of information, which is some way from the approach taken in the 1960s and 1970s. Levine considers whether the use of the term is now being used to explain social conditions and human interaction in a way that is now quite radically different from the initial use of the term. This article was an element of the author’s submission for an M.Phil. degree and as a result the bibliography is excellent, even if your interests are wider than the use of the term in America.

Big data and small data

Big data is another term that has taken on wider social and economic implications over the last few years, but small data was something new to me. ‘Small data’ refers to data collected through front-line employees in their interactions and relationships with customers. Son K. Lam et al. (2017) discuss the integration of big data and small data using the construct of absorptive capacity to examine routines by which a firm acquires, assimilates, transforms and applies big data knowledge to create a dynamic marketing capability. This absorptive capacity defines the the ability of a firm to recognize the value of new, external information, assimilate it and apply it to commercial ends. It is driven by the capability of combining and integrating big data and small data knowledge at both the firm and the front-line employee levels.

As boundary spanners front-line employees can gather unique and context-specific data and knowledge about customer needs, problems in service delivery, ways to improve service quality and also indications of customer sentiment and preferences. The authors consider both the ways in which front-line employees can benefit from big data availability and also the ways in which front-line employees can contribute to big data collections. Big data insights do not guarantee improvement in service quality and reduction in service costs; rather, it is the change at both the organizational and the front-line employees levels that produces these positive outcomes. The article has an interesting conclusion that extracting knowledge from big data does not discount the role of knowledge derived from small data. In fact, without small data, big data may be more big costs than useful benefits.

This is a conceptual article, based on a review of the literature, and no case studies are presented. Nevertheless, I think that the approach taken in the article is an important one in putting big data into an organization-wide context. It is interesting that the hype around big data has been much reduced over the last few years, as the anticipated benefits of big data collection and distribution did not generate a return on the very considerable IT investments. Information professionals work much nearer to front-line employees, and this article provides a very good basis for discussions about how best to use the information collected by front-line employees (who might well work for a call centre) to make sense and built on big data applications.

Research data and academic librarians

To an academic librarian, institutional collections of research data certainly bring them face-to-face with the challenges of managing large-scale data repositories. However, most academic IT departments are more than happy to pass this responsibility on to someone else while they look after the hardware and software. The immediate challenge for the librarian is to gain, at some speed, information about this very new and novel form of information resource. Members of the Association of College and Research Libraries (ACRL) in the US have responded to these changes through a series of developments, aimed at providing sustainable professional development support to librarians in the area of research data management (RDM).

As a result, a roadshow concept for training was developed by the ACRL, and this article by Conrad et al. (2017) provides a fascinating glimpse into the origins of these roadshows (http://www.ala.org/acrl/rdmroadshow). Acting on the recommendation to the ACRL Board, the Digital Curation Interest Group (DCIG) Executive Committee designed a professional development needs survey for its members in 2014 that achieved a high response level. The outcomes of the 13-question survey are presented in detail.

Three learning outcomes were identified

Define data management as it relates to data information literacy in order to build upon existing information literacy pedagogy.

A one-day pre-conference workshop was then piloted at the 2015 ACRL Annual Conference with considerable success. The pre-conference served as a test ground for future ACRL professional development opportunities on RDM. Multiple modules were presented with topics such as data and scholarly communication, data management instruction strategies, engaging the campus community, creating individual action plans and discussing future roles for subject liaisons in regard to RDM services. There are of course initiatives in many other countries around RDM by academic librarians, and these are noted in the article. I was impressed with the commitment of ACRL to the topic in supporting the survey and the pre-conference event, leading to a successful programme of educational roadshows.

RDM Quality Management

Another aspect of RDM is quality management. With the emergence of the National Science Foundation requirement for data management plans, academic librarians have increasingly aided researchers in developing these plans and disseminating research data. Van Loon et al. (2017) at Wayne University decided to determine the overall quality of data management plans at Wayne State University by arranging for the the Library System’s Research Data Services team to evaluate the content of 119 plans from National Science Foundation grant proposals submitted between 2012 and 2014. The results of the survey indicated that, while most researchers understood the need to share data, many data management plans fail to adequately describe the data generated by the project, how data will be managed during the project or how data will be preserved and shared after the completion of the project.

They found that 51 per cent of Data Management Plans (DMPs) did not identify the individual(s) responsible for data management, which may be problematic for proposals involving multiple principal investigators or cross-institutional collaboration or for labs with high turnover rates for graduate students and research staff. Most DMPs (92 per cent) did not provide an estimate of the total amount or expected rate of data generation, which is important for choosing the most appropriate data storage and preservation methods. Of the DMPs, 57 per cent did not specify the duration that data would be preserved after the project or policies governing how other researchers might reuse or redistribute their data, suggesting that researchers often do not carefully think about the lifespan of the data beyond the active period of the project. Furthermore, a majority of DMPs (62 per cent) did not mention specific metadata standards or methods of data description methods, indicating that the data might not be easily discoverable by or other researchers in the long term.

I was especially interested in a side-note to the main analysis in which the authors noted that while it is certainly expected that the results of research (i.e. interpreted, summary data in graphs and tables) would be shared through journal articles and conference presentations, these are not valid avenues of sharing the actual data underlying those results (i.e. uninterpreted, individual-level data in a variety of file formats). The authors considered that this may stem from a tendency for researchers to use the terms ‘data’ and ‘results’ interchangeably, which suggests that researchers could benefit from greater awareness of the NSF and Office of Management and Budget definitions of ‘research data’.

Visualization for data quality management

Moving even farther into data quality a data quality assessment process (DQAp) provides relevant and practical inputs to choose the most suitable alternative through a data defect mapping. A data defect denotes a nonconformity between a data instance and its contextual meaning which may arise at any point of the data life cycle. However, the DQAp strongly depends on data context knowledge since it is impossible to confirm or refute a defect based only on data. The context determines the structure of meaning between data and the environment (e.g. an organization department). Hence, human supervision is essential throughout this process, and visualization approaches can be of considerable assistance in this process. In this article, Josko and Ferreira (2017) consider which and how visualization system properties may facilitate the data quality visual assessment of defects that require high human supervision. I’m not going to consider this article in depth as it is highly technical, but in the light of the two articles referred to above, the need to be able to verify data quality in a context without the need to validate each data point is a very important requirement. For information professionals in data-rich environments (such as financial services and large-scale research data), I would suggest that it is of value to know about this approach and to add it to your tool box.

Selecting a content management system

Content management systems (CMS) are widely used to manage web sites and intranets but specifying and selecting a CMS is a challenging task requiring a team with IT and business requirements skills to spend some considerable time on the process. Books by Tate (2015) and Barker (2016) are good sources of advice, but there are always corporate issues around resources and budget that outweigh the selection of the optimum CMS. It is not uncommon for a CMS to have a very long service life. In most organizations, only the web and intranet teams will be familiar with CMS technology and products, but web applications are also used for more specialized solutions.

Clinical practice guidelines are used in a very wide range of organizations. In Australia, guidelines have been developed specifically for clinical support in areas where the nearest hospital may be hundreds of miles away. They include a set of Remote Primary Health Care Manuals (RPHCM). In 2007, a CMS was introduced to manage the production and distribution of these manuals. In 2013, the decision was taken to define and select a new CMS, and this article by Reddy, Herring and Gray (2017) tells the story behind the process. An important requirement was to be able to export content from the CMS directly to print-ready documents, an essential feature for the mostly hard-copy-driven RPHCM publication.

One of the main issues faced in the RPHCM CMS review was a lack of relevant and detailed advice to instruct guideline developers in choosing an appropriate CMS. The RPHCM project involves end users of its clinical practice guidelines in its review process along with subject experts spread across Australia. The reviewer involvement is entirely voluntary in nature. Any CMS chosen would, therefore, have to present minimal impediments for access and a friendly interface that does not deter less computer literate users. The review identified that the alternative CMS products they explored, while providing some enhanced editing, graphics and project management features, largely failed in this area. For the project, retaining the reviewer’s confidence and buy-in outweighed any additional editing and export functionalities they would gain by replacing the current CMS. It was also determined that there were editing features present in the current CMS that could not be provided by a replacement CMS, as these features were the result of customization over years.

As a result, the decision was made not to replace the CMS, but out of the very thorough exercise came a much better understanding of the CMS requirements for these publications and an appreciation of the time and effort that needs to be taken to ensure that the best decision is taken.

Virtual team skills

The dynamics of virtual teams fascinate me, as you will gather from my article in BIR (White 2014). I did have an early introduction to virtual teams. In 1974, I ran a library and information service that integrated offices in London, Brussels, New Delhi, Melbourne and New York. No e-mail and only Group 1 fax. Those were the days! There have been many research papers on the management of virtual teams and they feature in the new Cambridge Handbook of Meetings Science. (Allen 2015). However, there have been few papers that look in depth at what are often referred to as the knowledge, skills, abilities and other characteristics (KSAOs) that are needed to contribute successfully to a virtual team. (In passing, KSAOs is not a pronounceable acronym.)

Schulze and Krumm (2017) is a very comprehensive and critical review of the literature on virtual team skills, based on around 200 research papers. I thought I knew the research literature quite well in this area but was pleasantly surprised to find quite a number of papers that I was not aware of. The bibliography alone makes this 30 pp paper well worth acquiring.

The analysis is based on three perspectives. These are virtuality related challenges (technology use, cultural differences and geographic dispersion), distal challenges (personality and experience) and proximal challenges (knowledge, skills and motivation). This is a useful model and enables the authors to correlate a significant amount of research with a good degree of clarity. The authors consider both theoretical and practical implications of their work and highlight areas where further research is needed. To give an example from the section on practical implications, the authors suggest that their model and analysis can serve as a flexible tool for member selection and training programs. For instance, in a virtual team that works under conditions of high cultural diversity, high reliance on (multiple) technologies and geographically dispersed locations, many proximal and distal characteristics are likely to play a role to handle the challenges imposed by such a work setting. Yet in a virtual team that is homogenous with regard to the cultural backgrounds and cultural orientations of the team members, selection and training is not appropriate for the KSAOs associated with this dimension.

The authors acknowledge that they have developed a KSAO model that needs to be validated in practice, and there are some parameters from the research literature that they were not able to incorporate into their model. Nevertheless, this is an article that any manager of a virtual team would benefit from reading, especially if they have responsibility for training and team selection.

Cultural impacts on team performance

A few days before writing this column, I was taking part in a meeting with a client team in Berlin. All four German members of the team had excellent English skills, and most of the meeting was conducted in high-speed colloquial English. But from time to time, team members found themselves at a lost to determine what the appropriate English word was for a concept that might have been a single word in German but for which there was not a direct English equivalent. They would then have a brief discussion in German and offer me some English options. This can work both ways. ‘A bold step’ is virtually untranslatable into German without conveying entirely the wrong impression.

This article by Ahmad (2017) considers two issues

What are the challenges of knowledge sharing in a non-native language?

One of the interesting elements of this article is that the author grew up in Pakistan but his studies have been in Finland, which has one of the most impenetrable languages in Europe for a non-speaker. He notes that since language is a means of knowledge sharing, any variances, discrepancies and limitations in participants’ language proficiency will have considerable impact on the quality of knowledge sharing. Due to the varying proficiencies and thought processes involved when people with different linguistic backgrounds speak a non-native language, knowledge sharing in this context is a challenging experience.

The multinational organization used in this study is one of the biggest companies in the marine and energy industry. It has operations in 70 countries around the world and has more than 15,000 employees in its workforce. It has a multilingual workforce, so English is used as a common corporate language. Organizational multilingualism was evident not only in the company’s subsidiaries operating in different countries, and therefore using different languages, but also as intra-subsidiary multilingualism. For example, in the company’s Finnish subsidiary, around half of the workforce is composed of non-Finns. Moreover, expatriation and staff movement across subsidiaries are quite common. Most of the work is done in the form of virtual teams consisting of members from different geographical locations.

The data analysis showed that employees identify different kinds of variations in the process of knowledge sharing when it is conducted in a non-native language. These variations are called adjustments and are of three different types: discourse adjustment, media adjustment and language adjustment. I’m not going to try to summarize these adjustments as each has many variables, but I would commend the article to any manager or employee whose work involves sharing knowledge in a non-native language. The author emphasizes that the article is based around a single case study and that for reasons of commercial sensitivity the participants in the study had to self-record their impressions of knowledge-sharing situations. Nevertheless, this article is an important contribution to identifying issues that could inhibit the sharing of knowledge in multilingual organizations.

Identifying experts in group work

Working as I do in the area of enterprise search the issues of how search can be used to identify expertise is a hot topic in organizations, although there is little evidence that existing channels are inadequate. This article by Treem and Leonardi (2017) takes an unusual perspective on expertise identification by considering the ways in which a group recognizes which members have expertise in a specific area. Employees are often reluctant to present themselves as experts, a mixture of social culture and the concern that someone else make out-expert them. The authors set out a number of hypotheses for how a group might recognize expertise within the group and then carried out a project in a financial services company to test these hypotheses. Most were not supported, which must have come as a surprise to the research team. One that did emerge was that expertise recognition did not come from broadcasting to the group but in asking questions of members of the group individually about the task of the group.

It would seem that this creates a personal two-way conversation channel that enables the advice giver to feel that they are important because of their role in the conversation and also enables them to judge (albeit indirectly) what the questioner’s areas of expertise are. It comes down to an issue of mutual trust. The implication of this mechanism is that recognition of expertise comes slowly as the group works on a project, and not as the result of an initial round-table disclosure about areas of expertise.

The other hypothesis that was confirmed was that the participation by someone with expertise in a communal communications platform, such as a blog or other social media, is an important factor in the expertise of the individual being recognized within the group. Although the authors do not comment on this, my experience would suggest that this is because a communal communication platform provides a sort of crowd-sourced peer review of the expertise. If the person comments on the basis of their knowledge and no one else challenges them, then at a group level there is a confidence in the level of expertise.

The authors emphasize that because of the results coming from a single case study and the multivariate nature of the data around the hypotheses, it would not be advisable to extrapolate the outcomes of the project to all organizations and group work situations. Nevertheless, I found the focus on group recognition factors to be novel and interesting and yet another tool for my consulting toolbox.

Business development in professional services firms

I’ve spent most of my career working for professional services firms, having been running my own for almost 20 years and have worked on major projects for three global consulting firms. No matter how much a firm promotes its ‘brand’ clients are buying people that they trust to give them the best independent advice possible. As a result, the people doing the work are also heavily involved in business development.

Rogans and Mor (2017) note that the extent to which these networks include relationships built using predominately individual rather than firm resources, which we refer to as individual ties. Research shows that professional service firms traditionally have relied on personalized interactions with clients when exploring for novel knowledge, ideas and business opportunities. Yet, as the firms grow, the establishment of formal structures and procedures reduces the reliance on personalized interactions with clients, and firms focus on the reuse of existing knowledge rather than the exploration for new knowledge. The situation is further complicated by the fact that some business development resources rest with individual managers and are accumulated through prior experience other resources rest with the firm, and the manager has access to these only via membership in the firm. Thus, when building and maintaining external networks, a manager has a choice about whether to invest more of the firm’s resources or more of his or her individual resources.

To assess the balance between individual and firm resources the authors interviewed 77 senior managers in a global professional services firm using a mixed methods approach with interviews and two surveys. The research took place over a two-year period. They found support for our main argument that managers who invest primarily individual resources in their professional relationships perform better in terms of new business and new knowledge growth. Three different mechanisms could explain this result; managers who invest primarily individual resources in their professional relationships are exposed to a greater diversity of information, they have greater autonomy vis-à-vis the firm, and their contacts are more willing to provide resources in return.