Download PDF - Electronic Journal of Knowledge Management

At the Crossroads of Knowledge Management and Social Software Gabriela Avram Interaction Design Centre, University of Limerick, Ireland [email protected] Abstract: The growing phenomenon of Social Software seems to provide an opportunity to complement the top-down approach based on central knowledge repositories with tools that are simpler, smarter and more flexible. This article includes a brief description of the main categories of Social Software – weblogs, wikis and social networking sites - followed by an analysis of their utilisation in relation to the five core Knowledge Management activities of the Knowledge Management taxonomy proposed by Despres & Chauvel in 1999. Examples that illustrate the support Social Software could provide for knowledge management are presented. Finally, some of the problems that hinder the usage of Social Software tools, together with some of the latest developments and trends in the field are mentioned. Keywords: Social Software, weblog, wiki, social networks, knowledge acquisition, knowledge sharing.

– such as weblogs, wikis, social networking sites, social tagging, events management, geo tagging. Probably the specific tools, permanently evolving, are not as important as the social phenomenon generated around them – through spontaneous interaction, pro-active attitudes, enhanced knowledge creation, knowledge sharing and transfer.

1. Introduction Knowledge Management (KM) and collaboration are considered to be pre-requisites for more innovation and enhanced creativity. Until recently, most of the KM efforts were focused on the creation of central knowledge repositories, encouraging knowledge reuse and collaboration based on these repositories, in a typical top-down approach where knowledge was seen as a separate entity. The growing phenomenon of Social Software offers a chance to complement this approach with tools that are simpler and more flexible. This type of software is actually not new at all – software applications having similar traits have been in use for quite some time – but it is only recently that these have been labelled as “Social Software”. Social Software is the term used to designate, “the use of computing tools to support, extend, or derive added value from social activity - Including (but not limited to) weblogs, instant messaging, music and photo sharing, mailing lists and message boards, and online social networking tools” (Lawley 2004).

Section 3 contains an analysis of three important categories of Social Software (weblogs, wikis and social networking) in connection with the core KM activities included in the taxonomy of applied Knowledge Management proposed by Despres and Chauvel (1999). Several examples extracted from different sources presented in section 4 are meant to illustrate the various ways in which Social Software is able to support knowledge management. Section 5 considers some of the problems that hinder the usage of Social Software tools and points to some of the latest developments and trends in the field.

What could Social Software do for Knowledge Management? Knowledge emerges in conversations, actionable knowledge is mainly the result of collaboration, and more and more importance is given to social capital. Social Software provides the necessary support for conversations and collaboration, for knowledge creation, sharing and publication, for identifying experts and getting access to expert opinions worldwide. It leaves the control of knowledge with the individuals owning it. Each individual is able to maintain his own space for which he has complete control over the information he chooses to share. This creates a bottom-up style of information sharing and collaboration, rather than an imposed or corporate top-down strategy. (Fisher 2005)

2. Social software What types of software are actually included in the category of Social Software? There is a tendency to include email, discussion lists or message boards under the umbrella of Social Software. And to a certain degree, they are social interaction tools too. But there’s an important distinction between traditional communication software - forming people into groups with a top-down approach and assigning membership, as discussion lists and forums usually do, and Social Software - with its bottom-up approach, enabling people to organize themselves into a network based on their preferences (Boyd 2005). According to Stowe Boyd, social software encompasses one or more (though not necessarily all) of the following elements

The second section of this paper includes a brief description of the various types of Social Software ISSN 1479-4411

1

©Academic Conferences Ltd

Reference this paper as: Avram, G (2006) “At the Crossroads of Knowledge Management and Social Software” The Electronic Journal of Knowledge Management Volume 4 Issue 1, pp 1-10, available online at www.ejkm.com

Electronic Journal of Knowledge Management Volume 4 Issue 1 2006 (1-10)

Support for conversational interaction between people or groups. That includes real time conversations like instant messaging, and what Boyd calls “slow time” conversations that occur in collaborative virtual spaces. Support for social feedback. Reputation and trust are crucial in online interactions, as demonstrated by the importance placed by sites such as eBay on a seller’s rating and reputation. Support for social networks. Many Social Software applications create a digital layout of a person’s social network and facilitate adding new connections. (Kaplan 2005): The sudden popularity of social technologies is attributed to the increase in low-cost tools and the critical mass of millions of people who are now connected to the Internet (Boyd 2005), to the growing tendency of people to rely more on their own personal social networks than on traditional company structures (Nardi 2004), and to the people’s need to feel part of a community (Bryant 2003).

Jack Vinson depicts other characteristics of Social Software tools: they are extremely easy to use; they provide for networking and allow for selfforming networks; usually, readers are also contributors and vice versa, this virtual environment enforcing much less sense of hierarchy than in the real world; relationships become nothing but flows – one person can be at the same time part of several networks (Vinson 2005). The key areas of Social Software are considered to be the weblogs, the wikis, and the social network services of different kinds (Boyd 2003). Social network services range from some focused purely on networking, to others designed to share different types of resources, or meant for open coordination purposes (Figure 1). A strict classification is hard to derive, because the categories of Social Software tend to intertwine and to rely on each other.

Figure 1: Key areas of Social software permalink (permanent URL), which is usually automatically generated by all popular weblog systems.

2.1 Weblogs A weblog or simply a blog is a web application enabling periodic posts on a common webpage with public access. These posts are usually in reverse chronological order.

Most of weblogging tools not only generate HTML pages but also encode the post content in a format derived from XML known as RSS (Really Simple Syndication, or Rich Site Summary). The RSS format can be read by news aggregators, a type of software who checks automatically the weblog feeds for updates and display their content. These enable readers to keep up with many weblogs (and an increasing number of other websites), without navigating the actual web pages. (Efimova 2004)

Editing a weblog does not require any special training, enabling anyone to publish content on the web. As with any other website, the public can use any HTML browser to visit its pages. Weblogs range from personal diaries meant for family and friends and lists of visited links seasoned with short comments to personal knowledge repositories maintained by professionals, learning journals or networking instruments.

Besides weblog editing and publishing tools and news aggregators, weblog search tools give the users the chance to find weblog posts or connections between them.

The entries - called posts - are usually short. The most recent ones are displayed on the weblog homepage, while old posts can be retrieved from archives ordered chronologically (and possibly on topics). A post can be as short as a link to an online article, or as long as to contain an essay. Many posts link to interesting on-line articles, earlier discussions or related readings. They enable readers and other weblog authors to add comments or link back to a particular post using its

www.ejkm.com

Unlike an official web site, a weblog is highly subjective, reflecting the thoughts, opinions and preferences of its author(s). Most weblogs are written by individuals (also known as webloggers, or bloggers). These coexist on the World Wide Web with group weblogs, project weblogs and organisational weblogs.

2


Gabriela Avram

online encyclopaedia created and maintained collaboratively (Kaplan 2004).

Many weblogs also exhibit blogrolls, lists of weblogs that their authors read regularly. Through these lists, occasional readers can find trusted "sources" that influence the thinking and writing of a particular weblog author. These links are not only referrals to specific sources, but also signs of value and personal recommendation. In this sense, hyperlinks between weblogs fulfil a similar function like references in scholarly publications (Mortensen 2002).

2.3 Social network services The so-called “social networks” are circles in which people interact and connect with other people. They transcend strict delineation between personal and business (there's often overlap between the two), and tend to transcend organisational boundaries and hierarchies. Social networks can provide the essential context needed to make knowledge sharing possible, valuable, efficient and effective. (Pollard 2003)

2.2 Wikis A wiki is a website (or other hypertext document collection) that allows users to add content, as on an Internet forum, but also allows anyone to edit the content. "Wiki" also refers to the collaborative software used to create such a website.

2.3.1 Social networking sites The first online social networks started appearing in 2002, when the term was used to describe the means of networking in virtual communities, and became popular in 2003, with the advent of websites such as Friendster (www.friendster.com), TheHoosierWeb (www.thehoosierweb.com), Tribe.net (www.tribe.net) and LinkedIn (www.linkedin.com). The number of social networking sites currently available exceeds 300, and it is growing steadily. (Wikipedia 2005). Some of them are wide-ranging online social networking sites, such as Friendster and Orkut (www.orkut.com); others are dedicated to business networking, such as Ryze (www.ryze.com), OpenBC (www.openbc.com) and LinkedIn, or dedicated to location-based interaction, such as MeetUp (www.meetup.com), Plazes (www.plazes.com) and Tribe, and still others organised around business concepts, as in the case of ReferNet (www.refer.net) or Shortcut (www.shortcut.nu). Another category focuses almost exclusively on dating.

A wiki enables documents to be written collectively in a simple markup language using a web browser. A single page in a wiki is referred to as a “wiki page”, whiles the entire body of pages, which are usually highly interconnected via hyperlinks, is called "the wiki." A defining characteristic of wiki technology is the ease with which pages can be created and updated. Generally, there is no review before modifications are accepted, and most wikis are open to the general public — or at least to anyone who has access to the wiki server. In fact, even registration of a user account is not always required. Most wikis offer a title search, and some also provide full text search. WikiWikiWeb, the first wiki site that created the concept, defines a wiki as a “composition system, a discussion medium, a repository, a mail system, a chat room, and a tool for collaboration.” In the vision of Ross Mayfield of Social Text, wikis are tools for “transparent collaboration” (Kaplan 2004).

In these communities, an initial set of founders sends out messages inviting members of their own personal networks to join the site. New members repeat the process, growing the total number of members and links in the network. Sites then offer features such as automatic address book updates, viewable profiles, the ability to form new links through "introduction services," and other forms of online social connections.

There is a sort of safety clause in the wiki design: one page always lists recent changes to the page and enables users to revert to previous versions. That, explains WikiWikiWeb, allows people to correct mistakes, erase spam, and generally keep the content “meaningful”.

Some social networking sites are also facilitating music sharing (Kazaa) and photo sharing (Yahoo 360, Flickr, Phlog).

Two core assumptions are incorporated in the wiki mechanism. The first is that knowledge is transitory, not static. There's always some new piece of information to add, and some old piece to delete or revise. The second assumption is that the whole is greater than the sum of the parts. Through each individual’s contribution, the resulting product is made better and better. The most well known example of wiki usage is Wikipedia (http://wikipedia.org ), a free, multilingual

www.ejkm.com

In order to allow computers to link people to one another, a computer-readable social networking data format was created. FOAF (Friend-Of-AFriend) is a simple and easily extendable textbased data format defined using OWL (the Web Ontology Language).

3

ISSN 1479-4411


tagging events and sharing this kind of information. EVDB is another “events and venues database”, and has an associated web-based calendar service and search engine named Eventful http://eventful.com). There are a whole range of other web-based applications for handling events, location and calendaring currently under development. These could give individuals the chance of looking for other people who are going to attend the same event, or be in the same place at a specific moment in time.

2.3.2 Social tagging and folksonomies Social bookmarking is another type of online services, allowing users to save and categorise a personal collection of bookmarks. The individual bookmarks are public, so that anyone can see the bookmarks that have been saved by others and add them to their own collection, as well as to subscribe to other people’s feeds (Wikipedia 2005). There are several such sites, but the most well known seem to be CiteULike (www.citeulike.org), del.icio.us (http://del.icio.us), furl (www.furl.net), and spurl (www.spurl.net). Sharing and searching for bookmarks, photos and weblog content was made easier by the initiative to provide the users with the opportunity of adding tags to these types of content. Users are now able to categorise for themselves these various types of content as they wish. This is how social tagging was born, a sort of general taxonomy emerging from the individual tags. Social tagging was revealed to be a way to get some relatively reliable content classification out of a large number of people.

Another group of tools of this category are dedicated to geographic location management – the so-called geo tagging applications – making it possible to add geographic metatags (geotags) to web pages and to process RSS feeds. The GeoURL (http://www.geourl.info) is another popular service amongst bloggers. It offers a way to register a weblog in a directory for certain geographical coordinates. Furthermore, it also offers the possibility to obtain a list of other blogs near to a specific blog.

Folksonomy is a neologism for the practice of collaborative categorisation using freely chosen keywords. This feature began appearing in a variety of Social Software in 2004. There are currently several examples of online folksonomies: del.icio.us and Jots (http://jots.com) are bookmark sharing sites, Flickr (http://www.flickr.com) is meant for photo sharing, 43 Things (http://www.43things.com) for goal sharing, and Tagsurf (http://tagsurf.com) for tag-based discussions. (Wikipedia 2005)

Multimap (http://www.multimap.com/) and World kit (http://brainoff.com/worldkit/index.php) are both services allowing the visualization of websites geographically situated in a specific area. A series of search engines (such as Google Local, A2B Location Based Search Engine, GeoTags LocationBased Search Engine, RSS Weather) are able to search websites using geotags. Plazes (http://www.plazes.com) is a social networking site providing the possibility to register Internet access sites based on their IP number. Each member of the network is considered a discoverer of new Internet access points and can invite others to join. It mainly offers the possibility to find other people connected to the Internet in a specific area or to identify public access Internet access spots when travelling.

Folksonomies work best when a large number of users all describe the same piece of information. Discussions are taking place if folksonomies should be taken into account as a possible basis for building the Semantic Web.

2.3.3 Time and proximity management social tools

A constantly evolving list of social networking sites can be found at http://en.wikipedia.org/wiki/List_of_social_networki ng_sites.

Several Social Networking services already contain tools (Calendar, Agenda) for managing community events. One of the most well-known services of this type is MeetUp, meant for intermediating and managing face-to-face meetings on different topics in locations placed all over the world. Some of these tools are focused on time management and events information sharing. For example, RSSCalendar (www.rsscalendar.com) is a sort of new way for individuals and organisations to share their calendars with family, friends, and co-workers – making use of the latest developments in RSS technology and including RSS channel creation and aggregation. More complex than RSSCalendar, events (http://www.events.org/) is a service for

www.ejkm.com

3. Social software in the service of knowledge management The success of Social software is regarded as being based on the availability of these low-cost, high bandwidth tools, coupled with a critical mass of millions of self-motivated, gregarious and eager users (Boyd 2002). This kind of tool gives individuals the chance to network in online versions of real world social systems. Social Software is transforming group interaction and has a notable impact

4


Gabriela Avram

nowadays on how businesses reach their markets, changing the way collaboration and communications are managed within and across businesses.

tomers and have an influence on the public image of the company (as in the famous case of Robert Scoble from Microsoft).

3.1 The KM taxonomy by Despres and Chauvel

News aggregators make subscribing to specific searches and monitoring the content of an extended number of blogs regularly much easier. In order to be able to adapt rapidly, businesses have to know who is speaking about them and their products, and in what terms.

In their paper titled “Knowledge management(s)”, Despres and Chauvel proposed a framework for categorising Knowledge Management (KM) regions of practice, taking into account five types of processes (activities) and three different contexts (individual, group and organisation) (Despres 1999). The KM activities mentioned in their paper are:

Wikis can also be seen as an emerging source of information. More and more people are using Wikipedia today, and there are other wikis on specific topics coming to public attention. These sources of information have the advantage that they are updated almost in real time, and bring together the contributions of thousands of people.

1. Scan/map - pointing to the world of business intelligence, perception; 2. Acquire/capture/create – associated with the world of research, development and creation; 3. Package/codification/representation/storing – related to the world of databases, information and knowledge bases, organisational memory; 4. Apply/share/transfer – related to the world of competencies, teamwork, intranets and cross border sharing; 5. Reuse/innovate/evolve/transform – associated to the world of leverage, intellectual assets and innovation.

Browsing other people’s tags, bookmarks, photos especially if these persons are known to share the same interests - can save hours of work and is an effective alternative to Google and catalogue searches. These other people invested time in collecting those items, and agreed to make them public – so they are free resources that should be taken into consideration. Social networking sites may provide information on potential contacts, partners and customers; by using the available information, specific expertise can be located and potential job candidates screened. LinkedIn, for example, is famous for bringing together employers, job seekers and multi-level marketing salespeople.

3.2 Social Software as support for knowledge activities We will now attempt to analyse the three categories of Social Software mentioned above (weblogs, wikis and social networking) using the framework of these core KM activities. Deliberately, we chose not to include the three different contexts (individual, group and organisation) of the Despres and Chauvel taxonomy, for an almost obvious reason: Social Software is meant for individuals to enhance their social interaction in groups, organisations and across them, so that separating these three different contexts would not make much sense in this case.

The time and proximity management social tools can be used for scanning events (fairs, conferences, workshops) in a particular field of interest, or for scanning places in search of people with specific expertise. Plazes, for example, can give the chance of finding local peers when travelling, or online acquaintances happening to be in the same town or even in the same hotel, creating the opportunity to meet in person.

3.2.2 Acquire/capture/create:

Let us now review some of the uses of Social Software from the perspective of the activities included in the Despres and Chauvel taxonomy:

Blog authors use this type of tool for capturing their own ideas and those of other people during research work, project development or simply during regular work. It is a method for exposing work-inprogress and for getting feedback, for storing drafts that could later develop into some form of deliverables (articles, reports, books) and for commenting on other people’s ideas.

3.2.1 Scan/map: For business intelligence, blogs reading proves to be an excellent way of collecting information on markets, competitors and latest innovation, and also of locating experts both inside and outside an organisation; for marketing, blogs monitoring is a new opportunity for examining customer opinions; weblogs run by individuals known as working for famous companies also attract feedback from cus-

www.ejkm.com

The open editing facility provided by wikis enables individuals to capture knowledge and afterwards jointly participate in its refinement. Knowledge can be restructured and reorganised at any time due to

5

ISSN 1479-4411


the remarkable flexibility offered by this category of tools. Inside organisations, wikis can be used for coordination purposes and for acquiring a common understanding of the concepts and procedures.

Weblog posts usually reflect personal knowledge and competencies in their original context, and their content can be used as log, illustration or source of inspiration for the author’s own work, but - at the same time - by peers and novices in the field. Distributed teams can use group blogs for coordination and information exchange. Wikis can be used as repositories for more structured knowledge by teams, communities of practice, or networks of various types.

Browsing social network sites may result in the acquisition of new customers, employees or consultants. It is mainly focused on discovering the needs, expertise and offers of other people and indirectly of other organisations. Bookmarking interesting pages found during the web browsing activity is an excellent method for organizing and storing links; bookmarks, photos, audio and video files can be stored on specialised sites providing the respective services, their owner retaining the right to decide if he wants to make them public or not. The opportunity to tag these artefacts enables their classification and organisation.

When people are sharing their links and photos by storing them on specialised sites, they first think of the advantage of re-using them from wherever they are, whenever they need them. Using other people’s links and photos is a secondary, but not less important, purpose. The tag search facility enables people to search blogs, link and photo repositories for information on a particular topic, making the search a lot easier and providing access to human-filtered information.

3.2.3 Package/codification/representation/ storing: In order that it is available for a large category of the public, Social Software is typically very easy to use and intuitive - at least at a basic level. The packaging, codification and storing activities are usually transparent for the users, and are frequently associated with acquire/capture/create activities: in the case of social networking sites, information is discovered, acquired and stored immediately, without intermediaries, just by pushing a button; blogs make web publication extremely easy and wikis facilitate collaborative editing without requiring any previous knowledge. The representation aspects differ from one tool to another. Some (like those dedicated to photo sharing or bookmarking) do not allow for much personalisation and innovation. Others – especially blogging software- allow more advanced users to intervene on the representation of the stored information. Together with the text entry, the aspect and form of a blog, the images, audio sequences and links can contribute to a great extent to the delivery of the intended message to its public.

3.2.5 Reuse/innovate/evolve/transform: Regular reading of weblogs incites reflection and instigates weblog writing. It can be the source of controversies and encourages synergy and creativity. Bloggers read each other’s posts and start conversations across weblogs and these conversations can become starting points for forming social networks, based on joint interests. This phenomenon occurs across teams, professions, organisations, countries and continents. The pace of knowledge transformation is amazingly fast: a few hours after an idea was born, there are people all over the world who have already retained it and adapted it to their own purposes, shaping their own domains. Wiki pages are edited and improved, sometimes reorganized; new knowledge emerges; open editing stirs up discussions, concepts and meanings are often vividly negotiated, definitions are continuously polished and facts updated. Synergy arises from collaboration, ideas exchange, and the amalgamation of knowledge from different domains.

3.2.4 Apply/share/transfer: The sharing facility is the essential feature of any Social Software application. Weblogs, wikis and social network services have as their core purpose knowledge sharing, and RSS feeds have made sharing even easier. By subscribing to RSS feeds, users can stay up-to-date with the latest developments on a specific topic. Information travels around the world, across professions and organisational boundaries, and is accessible to anyone, apprentice or scholar.

www.ejkm.com

What does Social Software offer more than the traditional communication tools? E-mail and instant messaging, discussion lists and forums involve messages sent to a person or a group. It is usually short-lived communication (synchronous or asynchronous) and destined to a specific, already known, public. The content of weblogs and wikis, the profiles, the tags and the comments left on social networking sites remain available for a longer period of time(if not forever) and they could be meant for everybody (in the case of public access)

6


Gabriela Avram

initiative of Denham Grey, which became in time an excellent collective repository of KM references (KmWiki 2005).

or for all the members of a group (in case of limited access). The most interesting thing about Social Software is the fact that it is social 6. In the way it is conceived – a networked approach to fitting connected tools around users 7. In its purpose - augments and extends online and offline social interaction to promote mutual understanding, and 8. In the way it behaves - it adapts to the user, instead of forcing the user to adapt to it; becomes part of the user’s means of representation, and augments human interaction, instead of narrowing it down (Bryant 2004).

Another example is the Knowledge Management Summer Camp wiki (http://wikifarm.roell.net/kmsc/KnowledgeManagem entSummerCamp). The 2004 edition was organised in Portugal with the support of the Knowledge Board The wiki cited was used for participants’ registration, for preparing the actual event, exchanging ideas between the online and on-the-site participants, for storing references and publishing ideas that emerged during the camp.

4. Selected examples of Social software usage

The following is an example of how a wiki can be used for collaborative work inside a company, encountered in Alexander O’Neill’s weblog, The Hallway (O’Neill 2005): At my company the development team is currently making heavy use of Wiki technology to allow us to easily share and edit documents with each other. A wiki essentially lets you edit content directly on the web server without having to upload HTML files or keep track of a local file tree. So, for example, you can go to a site like WikiPedia, click the 'edit' link, and suddenly you can make whatever changes you like to the page. The wiki software also helpfully keeps track of revisions people make, so if some unthinking soul deletes everything you just have to click the 'rollback to previous revision' button and the damage is undone. We’re finding this model very useful for editing and fixing up each other's work. We also don't seem to have any trouble with people feeling like they 'own' a piece of material. Everyone works on everyone else's stuff, to improve it and look for mistakes, and everyone also then has a better understanding of the overall project.“

We present here a few examples to illustrate how Social Software tools are used to support knowledge management activities.

4.1 Reasons for using weblogs The first example is a reflection on weblogs usage, extracted from a post made by Lilia Efimova in her weblog “Mathemagenic” (Efimova 2005): There are two sides of it, reading and writing. Reading weblogs as a way for prevention, preparation, relation and expertise building. It's like everyday exercise to stay fit - knowing what is going on, what are the trends, who are the people. It may feel as not very important in everyday scale, but every time when I face a new big challenge I appreciate it - like appreciating everyday exercises and being fit if time comes to run for your life. Reading is also about taking time to develop ideas (I often think of "being pregnant with ideas"), having time to explore, bit by bit, creating a space for emergent connections and associations. This is where writing comes into play* as well. For me writing is about catching ideas on the fly, growing and connecting. (Here I can go into a body of research on how artefacts support thinking and knowledge creation, but I wouldn't) Somehow the process of articulation is largely the process of idea development as well. Like a sculpture that exists only in a head of sculptor and needs to be moulded into physical shape to get a life, writing gives shape and life to fuzzy ideas in my head.

4.3 del.icio.us – A social network service for bookmarks management del.icio.us is a Social Software web service for sharing web bookmarks. It was developed by Joshua Schachter and “…is meant for people who are keeping track of their URLs for themselves, but who are willing to share globally a view of what they're doing, creating an aggregate view of all users' bookmarks, as well as a personal view for each user”. (Shirky 2005)

4.2 Wiki usage A wiki can be useful for different collaborative activities, such as jointly writing a paper, coordinating a project or preparing an event. A brilliant example is the KmWiki (http://kmwiki.wikispaces.com), an

www.ejkm.com

The next example is taken from a weblog post of Stephen Spaeth, who works for the Centre for

7

ISSN 1479-4411


down and bottom-up metadata and controlled vocabularies.

Teaching, Learning and Technology (CTLT) at Washington State University (Spaeth 2005): At CTLT, we have been trying to find ways to build collaborative communities around web resources. We have been exploring the intersection between two tools: delicious and wikalong. Nils discovered a small community (four people as of 2005-01-14) that have posted Wikalong's homepage to delicious. Nils recognized that by virtue of registering a common interest in wikalong at delicious the four are an incipient interest group. He created a wikalong for that page in order to provide the group with a common resource on which to build their interest. While four people is a start, that number seems too small. There must be others who have this common interest. I added the term wikalong to the tags for my post to delicious and discovered a larger community of eight citations using the term wikalong. Some of those pointed to the page which a much larger number of users (129 as of 2005-01-14) had noted. In creating the link to the delicious home page, I discovered another convergence of interest in delicious and wikalong. That community is starting to identify themselves in the wikalong for that page.” (Note: Wikalong is a Firefox Extension that embeds a wiki in the Sidebar of the Firefox browser, indexed off the URL of the current page.)

The use of connected individual and group weblogs was central to this process, not only in terms of the final product itself, but also in the way it was developed. Building such a system was considered a more than technical exercise, requiring an organisational commitment to building a knowledgesharing culture, and involving various communities in both on- and off-line activities. Headshift created a network ecology of individual and group weblogs (including other Social Software tools) for an initial population of more than 10,000 users and linked these together using common top-down metadata and bottom-up terms and categories to create a genuinely joined up knowledge sharing environment where every node, group and category is syndicated both within the network and outside to other agencies via web services and XML. (Bryant 2004)

5. Unsolved problems and future trends Social Software is rapidly evolving; new features are being thought of and made available almost every day. Probably the most interesting trend is the participation of users in the development of new features and the speed of developers trying to bring in new applications. Users with programming skills provide add-ons and make them available to the public. Some of them are rapidly adopted on a large scale, while others remain little used. The social phenomenon generated around the development and the use of Social Software tools is continually evolving.

4.4 Social Software tools integration The various Social Software tools can be easily combined and aggregated to suit the needs of a particular community and to support its culture. An interesting example in this direction is the system designed by Headshift for the National Institute for Mental Health in England (Bryant 2004). The system (http://kc.nimhe.org.uk) was designed for creating joint knowledge and for promoting collaboration and understanding to bridge organisational divides (e.g. local health services, charities, professional bodies), occupational divides (e.g. clinician, policy maker, academic) and different perspectives (e.g. service user, carer, researcher, etc.) within the Mental Health field.

Probably the most important characteristic of this category of tools is their extreme simplicity (anyone can use them) and the fact that they involve social interaction and a fun factor. Adding new features can cause difficulties for users. Keeping the balance between user-friendliness and new appealing features is not an easy task for the developers. Making the use of these kinds of tools compulsory in companies will probably take the fun factor away – and this is a problem that needs attention. There are several visible trends at this point in time. Upon users’ request, weblogs and wiki merged into Bliki (blog+wiki) type tools, the spontaneity of the weblogs marrying the structure of wikis.

Headshift envisaged the deployment of simple, usable Social Software to create new and interconnected opportunities for informal knowledge sharing between key stakeholders in the Mental Health field. Rather than seek to mediate these different perspectives and produce a single ‘official’ version of events, they decided to promote selfrepresentation by encouraging individual ‘voices’ from their network to stimulate informal knowledge sharing within an innovative framework of top-

www.ejkm.com

Some sites dedicated to weblog hosting, such as Live Journal and Greatest Journal, encourage the interconnection of weblogs forming social networks. Further evolution of this idea is the Semantic Social Network, which interconnects both peo-

8


Gabriela Avram

Social Software tools for individual knowledge creation and sharing are already highly visible. Professionals sharing the same interests (sometimes having very different backgrounds) find them extremely useful for locating expertise worldwide, keeping up-to-date with the latest developments in multiple fields, and for connecting to each other.

ple and weblogs, such as Stumble Upon and Funchain. Real Time social networking is the name given to a hybrid of web-based social networks and instant messaging technologies emerged recently and gaining popularity. Another current trend is collaborative real time editing, referring to the process of editing a text or media file by different participants to an event from different internet accounts (Wikipedia 2005).

The approach supported by this category of tools is informal, innovative and flexible giving enhanced support to the user-centric perspective – because it empowers users, bringing the tools to them, and not the opposite.

The typical social networking sites, storing individual profiles and facilitating contacts, are blamed for excessive data centralisation and their lack of standardisation. Each time an individual registers to a new social networking service, almost the same information has to be filled in again and again. Then, the respective information is stored on that site, instead of being held on a personal site and having the social networking sites link to it. After a boom when hundreds of thousands of people subscribed to such social networking sites, the current trend shows many of them are currently opting out. In spite of some widely publicised success stories, membership in most of these networks has failed to prove its loudly claimed advantages. Engestrom argues that what causes the failure of many of the social networking sites is the lack of a shared object. Social networks are not just made up of people – they consist of people who are connected by a shared object. (Engestrom 2005)

A sort of reluctance still hinders the usage of these tools on a large scale in organisations. Possible causes can be the control they are giving to the individual on his own generated content – as opposed to the hierarchic control on central knowledge repositories, and their bottom-up approach as opposed to the classic top-down one. Our paper attempted to provide a succinct presentation of Social Software, followed by an endeavour to describe its utility from the perspective of the five core KM activities included in the Despres & Chauvel taxonomy. The four examples of utilisation were meant to give an idea on the possible contributions of Social Software to Knowledge Management. In our perspective, far from being a substitute, Social Software tools could provide a useful complement to existing central knowledge repositories (Ras 2005). It is noteworthy that the results of Social Software deployment consist not only in the generated content, but also in the social interaction triggered and in a shared understanding of concepts and facts, as basis for joint actions.

An interesting initiative regarding the social networking phenomenon is The Augmented Social Network (http://asn.planetwork.net), a public initiative meant to build identity and trust into the architecture of Internet, in the public interest, in order to facilitate introductions between people who share affinities or complimentary capabilities across social networks. Another interesting development is the possible usage of folksonomies (collaborative categorisation using freely chosen keywords) as a basis for building the Semantic Web.

The huge number of available tools and features and the rapid pace of innovation in the field bring the advantage of a wide choice, adapted to the users’ needs and continually evolving to serve them better. This kind of flexible and rapidly evolving tools in the hands of innovative users will be probably one of the major sources of competitive advantage in the Knowledge Economy of the future.

6. Conclusions The bottom-up approach of Social Software encourages responsibility and content ownership, and at the same time opens wide opportunities for collaboration and interaction. The benefits of using

References Angeles, M.(2004) “Supporting enterprise knowledge management with weblogs: A weblog services roadmap”,[online],URLGreyHot , http://urlgreyhot.com/personal/cil2004?PHPSESSID=81c712287cc7519dec6d65ffc54067e5 Boyd, D. (2005) “The significance of social software”,[online], apophenia, http://www.zephoria.org/thoughts/archives/2005/05/08/the_significance_of_social_software.html

www.ejkm.com

9

ISSN 1479-4411

Electronic Journal of Knowledge Management Volume x Issue xx 2005 (xx-xx)

Boyd, S. (2003) “Are you Ready for Social Software?”,[online], Darwin Magazine, http://www.darwinmag.com/read/050103/social.html Bryant, L. (2003) “Smarter, Simpler, Social” [online], Headshift, http://www.headshift.com/moments/archive/sss2.html Bryant, L. (2004) “Informal, joined up knowledge sharing using connected weblogs in pursuit of Mental Health service improvement” [online], Headshift, http://headshift.com/archives/blogtalk/blogtalk_web.htm Despres, C., Chauvel, D. (1999) “Knowledge management(s)”, Journal of Knowledge Management, MCB University Press, Vol.3, No.2 , pp 110-120 Efimova, L., Fiedler, S. (2004) “Learning webs: Learning in Weblog Networks”, Paper presented at the Web-based communities 2004, 24-26 March 2004, Lisbon, Portugal, [online], Telematica Instituut, https://doc.telin.nl/dscgi/ds.py/Get/File-35344 Efimova, L. (2005) “Blogging as creating space for important”, [online], Mathemagenic, http://blog.mathemagenic.com/2005/02/21.html#a1499 Engestrom, J. (2005) “Why some social network services work and others don't — Or: the case for object-centered sociality”, [online], Zengestrom http://www.zengestrom.com/blog/2005/04/why_some_social.html Fisher, T. (2005),” Social Software in the Enterprise”[online], Java Developers Journal, http://timothyfisher.javadevelopersjournal.com/read/1247839.htm Kaplan-Leiserson, E. (2004) “We Learning: Social Software and E-Learning, Part II”, [online], Learning Circuits, http://www.learningcircuits.org/2004/jan2004/kaplan2.htm KmWiki (2005) [online], http://kmwiki.wikispaces.com KMSC04 (2004) [online], http://wikifarm.roell.net/kmsc/KnowledgeManagementSummerCamp Mortensen, T., Walker, J., (2002) “Blogging thoughts: personal publication as an online research tool. Researching ICTs in context”, [online], eg-no, http://blogs.intermedia.uib.no/happy/archives/var/www/html/happy/Researching_ICTs_incontext-Ch11-MortensenWalker.pdf Nardi, B. (2000)“It’s Not What You Know, It’s Who You Know”, [online],First Monday,http://www.firstmonday.org/issues/issue5_5/nardi/index.html O’Neil, “A. (2005) Using Wiki Software for Small Scale Collaboration“ [online], The Hallway”, http://tvt.blogspot.com/2005/01/using-wiki-software-for-small-scale.html Pollard, D. (2003) ”Social Networking, Social Software and the Future of Knowledge Management”,[online], How to save the world, http://blogs.salon.com/0002007/2003/05/28.html Ras, E., Avram, G., Waterson, P., Weibelzahl, S. (2005) “Using Weblogs for Knowledge Sharing and Learning in Information Spaces”, Journal of Universal Computer Science, Volume 11, issue 3, ISSN 0948-6968 [online] http://www.jucs.org/jucs_11_3/using_weblogs_for_knowledge Shirky, Clay (2005) “Ontology is Overrated: Categories, Links and Tags”, [online], Clay Shirky's Writings About the Internet, http://shirky.com/writings/ontology_overrated.html Spaeth, S. (2005) “Using Del.icio.us and Wikalong to find and build communities of interest”, [online], Transforming Learning, http://pbj.ctlt.wsu.edu/spaeth/archive/2005/01/14/1476.aspx Vinson, J. (2005) “There is something about social software” [online], Knowledge Jolt with Jack, at http://blog.jackvinson.com/archives/2005/04/29/there_is_something_about_social_software.html Wikipedia (2005) Wikipedia, the Free Encyclopaedia, [online], http://en.wikipedia.org/wiki/

www.ejkm.com

10


Intellectual Capital and Value Creation: Evidence from the Portuguese Banking Industry Maria do Rosário Cabrita and Jorge Landeiro Vaz Technical University, Institute of Economics and Business Administration, Portugal [email protected] [email protected] Abstract: Intellectual capital has been described as intangible assets that may be used as a source of sustainable competitive advantage. However, intellectual capital components have to interact, to create value. Previous studies demonstrate that intellectual capital is positively and significantly associated with organizational performance. Our aim is to consolidate these findings, examining the inter-relationships and the interaction effects among intellectual capital components and organizational performance, in the Portuguese banking context. Keywords: Intellectual capital, human capital, relational capital, structural capital, value creation.

tual capital is actually, at critical cross-roads with increased emphasis on developing theoretical concepts and testing relationships guided by such concepts. It is vital to consolidate some findings, namely arrive at a set of operational measures that meet minimal criteria of measurement.

1. Introduction Wealth and growth in today’s economy are primarily driven by intangible (intellectual) assets. The rise of new economy has highlighted the fact that the value created depends far less on their physical assets than on their intangible ones. These assets, often described as intellectual capital, are being recognised as the foundation of individual, organizational and national competitiveness in the twenty-first century (Wigg, 1997; Bounfour and Edvinsson, 2005). As noticed by Pike et al. (2002:659), “as the business society is developed, the key step in value creation has ascended an intellectual staircase”.

Some authors (Churchill, 1979; Venkatraman, 1989; Straub, 1989) claim that the linkage between theoretical definitions and their corresponding measures has been generally weak, despite “the process of construct development and measurement is at the core of theory construction” (Venkatraman, 1989:944). Linking theory construction (exploratory) to theory testing (confirmatory) is a sine qua non condition for the management theory development (Hughes et al., 1986) and comparing findings in different settings is an important tool that serves that purpose.

Intellectual capital has been identified as a set of intangibles (resources, capabilities and competences) that drives the organizational performance and value creation (Roos and Roos, 1997; Bontis, 1998; Bontis et al., 2000). This suggests causal relationships between intellectual capital and organizational value creation (Marr and Roos, 2005). However, intangible assets seldom affect performance directly. Instead, they work indirectly through relationships of cause and effect (Kaplan and Norton, 2004).

Previous studies (Bontis, 1998; Bontis et al., 2000) demonstrate that intellectual capital is positively and significantly associated with organizational performance. The purpose of our study is: (i) to validate a set of operational measures, which compared with other studies, may result in a measurement instrument for financial sector; (ii) to examine interrelationships among intellectual capital components and organizational performance and; (iii) to study interaction effects among intellectual capital components and organizational performance.

From the management point of view it is essential to recognise that none of the elements of that set of intangibles is per se sufficient for a successful performance. These key elements need to be combined to generate value. In this context, intellectual capital is a phenomenon of interactions, transformations and complementarities, meaning that a resource’s productivity may improve through the investments in other resources.

2. Reviewing the literature 2.1 Intellectual capital

Despite the tremendous theoretical improvement during the last years, intellectual capital phenomenon requires theory and research methodology that enhances the integration of theory construction and theory testing. Research in intellecISSN 1479-4411

There is no widely accepted definition of intellectual capital. However, the literature revision point out that intellectual capital is essentially related to “knowledge that can be converted into value” (Ed11


Reference this paper as: Cabrita, M and Vaz, J (2006) “Intellectual Capital and Value Creation: Evidence from the Portuguese Banking Industry” The Electronic Journal of Knowledge Management Volume 4 Issue 1, pp 11-20, available online at www.ejkm.com


vinsson and Sullivan, 1996:361). Moreover, at least three elements are common in almost all definitions: (i) intangibility; (ii) knowledge that creates value and; (iii) effect of collective practice. This means that are excluded all irrelevant intangibles that have no function over the firm’s future potential and it is assumed that competitive advantage depends on how efficient the firm is in building, sharing, leveraging and using its knowledge.

extrinsic motivation is an important and hard-toimitate competitive advantage (Osterloh and Frey, 2000).

2.1.2 Structural capital Structural capital represents the organization’s capabilities to meet its internal and external challenges. It includes infrastructures, information systems, routines, procedures and organizational culture. Structural capital is the skeleton and the glue of an organization because it provides the tools (management philosophy, processes, culture) for retaining, package and move knowledge.

The most important challenge for researchers is to prove that intellectual capital creates value (MERITUM, 2001). Our focus is on intellectual capital value drivers and the way its different components interact to generate value. Which component is most valuable may have different answers depending on internal and external organizational variables. The literature presents a great number of classification schemes for intellectual capital. However, a convergent taxonomy emerged, categorising intellectual capital onto three components: (i) human capital; (ii) structural capital and; (iii) relational capital.

Banking industry scenario has recently changed. Globalization, deregulation and internationalization create new business challenges. In the past, banks sought to improve their balance sheet and asset growth, increasing profitability. But, since the Basle Accord, the emphasis is on assets productivity, capital efficiency and revenue growth. Information and communication technology has been largely used in a variety of ways to reduce costs, increase efficiency and accelerate innovation, drivers of today’s banking performance.

2.1.1 Human capital

2.1.3 Relational capital

Human capital is the primary component of intellectual capital (Edvinsson and Malone, 1997; Stewart, 1997; Bontis, 1998; Choo and Bontis, 2002), because human interaction is the critical source of intangible value in the intellectual age (O’Donnell et al., 2003).

Relational capital is the knowledge embedded in the relationships with any stakeholder that influences the organization’s life. The literature defends that relationships with stakeholders are the necessary condition for building, maintaining and renewing resources, structures and processes over time, because through external relationships firms can access critical and complementary resources. Recently, some authors (Prahalad and Ramaswamy, 2000) suggest that customers become a new source of competence for the organization because they renew the overall competence of the organization and rejuvenate the knowledge base preventing it from the obsolescence in a turbulent environment (Gibbert et al., 2001).

A macroeconomic perspective recognizes human capital as the driver of national economic activity, competitiveness and prosperity (OECD, 1996). On individual level, human capital is defined as a combination of four elements: (i) genetic inheritances; (ii) education; (iii) experience and; (iv) attitudes about life and business (Hudson, 1993). The organizational perspective refers to human capital as “the source of innovation and strategic renewal” (Bontis, 1998). Gupta and Roos (2001) added that “core intellectual capital”, comprising competence, intellectual agility and attitude, are the potential of synergies for the value creation.

Relational capital can be measured as a function of longevity (Bontis, 2002), while marketing relationship literature argues that long lasting relationships are a source of competitive advantage (Håkansson and Snehota, 1995)

Knowledge generation and transfer is an essential source of firm’s sustainable competitive advantage, but it entirely depends on the individuals’ willingness. As such, if the human capital can suggest the economic potential of individuals within a firm, it is also true that the outcomes are intimately connected to motivation.

There is evidence of how employees’ satisfaction, motivation and commitment have positive influence in customer satisfaction, loyalty and retention, leading to higher firm’s productivity (Kaplan and Norton, 1996, 2004).

Although not a goal itself, motivation should serve to support the organization’s goal. Thus, managing motivation, especially balancing intrinsic and

www.ejkm.com

12


Maria do Rosário Cabrita and Jorge Landeiro Vaz

is at the heart of strategic management and the rationale of intellectual capital is its ability to create value. Thus, intellectual capital and strategy are intricately woven. In this sense, a perspective based on the intellectual capital provides a more holistic view of the firm and its value, driving and nurturing the strategy. Nevertheless, given the uniqueness of each firm’s configuration of knowledge characteristics and the idiosyncrasies of the firm’s history, it does suggest that there are a variety of routes to success.

2.2 Organizational performance Organizational performance is a recurrent theme in various domains of management, becoming an important concept in strategic management because performance improvement is the time test of any strategy (Schendel and Hofer, 1979). Based on the perspective of organizational effectiveness, Venkatraman and Ramanujam (1986) circumscribed the concept of organizational performance. According to the authors organizational performance is a subset of organizational effectiveness. The narrowest conception of organizational performance considers the use of financial indicators (e.g., sales growth, return on investment and return on equity) while the broader concept of organizational performance includes emphasis on indicators of operational performance (i.e., non-financial). We consider in our study both aspects of performance (i.e., financial and operational indicators).

Intellectual capital is a matter of creating and supporting connectivity between all sets of expertise, experience and competences inside and outside the organization. The “value platform” model explains in an illustrative way the importance of a balanced intersection between the three dimensions. The contribution of this model is to show that: (i) the organizational value is created in the interaction of the three dimensions and; (ii) the intersection area increases, as the three dimensions interact.

2.3 Value creation Intellectual capital refers to the intellectual assets from a strategic and global perspective (Viedma, 2002). We argue that, from a strategic perspective, intellectual capital is used to create and apply knowledge to enhance firm value. Value creation

3. Research model and hypotheses Based on the literature revision, the model supporting our research is depicted in figure 1.

HC*SC* R C

H6

H1 SC OP

H4

HC

H3 H5 RC

_________ Main effec ts _ _ _ _ _ _ Interac tion effec ts

H2 Figure 1: Research model Hypotheses to be tested are: H1: Human capital is positively associated with structural capital; H2: Human capital is positively associated with relational capital; H3: Structural capital is positively associated with relational capital; H4: Structural capital is positively associated with organizational performance; H5: Relational capital is positively associated with organizational performance; H6: The relationship between human capital and organizational performance is positively moder-

www.ejkm.com

ated by the interaction between structural capital and relational capital.

4. Research methodology 4.1 Measurement instrument We used the original questionnaire developed by Bontis (1997), administered in Canada and Malaysia, with eight more items extending the concept of relational capital. A copy of the questionnaire can be requested to the authors. Following Churchill’s (1979) recommendations, the 63 original items were validated again. Independent assessment of validity enhances the quality of

13

ISSN 1479-4411


measures. New items were included after being submitted to the recommendation of the author’s original questionnaire.

ness” may affect organizational performance. Other studies (Chen et al., 1993; Kumar et al., 1993) provide encouraging evidence about the utility of the “key informant” approach. The informants were chosen not on a random basis but because they have special qualifications such as status, number of years in the profession, or specialized knowledge.

With a total of 71 items and a cover letter explaining the concept of intellectual capital, the questionnaire was administered with a letter from the President of Portuguese Institute of Banking Management explaining the aims of the study (academic purpose) and assuring confidentiality. Literature encourages the use of some tactics to stimulate response rate: (i) the existence of a sponsoring organization; (ii) the status of person signing letter accompanying questionnaire and; (iii) the use of follow-ups.

Content validity was attempted through the depth of literature search and expert opinions (Bontis, 1998; Darroch and McNaughton, 2002; Chin, 1998). The measurement instrument was pre-tested through personal interviews with eight banking managers, aiming: (i) to correct weaknesses and ambiguities in the questionnaire; (ii) to identify the most knowledgeable people about the subject and; (iii) to grasp the sector dynamics.

All questions are perceptual. The subjective approach has been used extensively in empirical studies, based on executive’s perceptions, having been justified by several authors. Venkatraman and Ramanujam (1986) and Dess and Robinson (1984) have found consistency between executive’s perceptions of performance and objective measures.

To validate the instrument, a pilot test was carried out at a convenience sample of 178 members (including first, second, third and fourth-levels executives). The 151 returned questionnaires, helped us to purify the measures and to refine the sample design. Pilot study indicated that: (i) strategic awareness cannot be assumed to exist, at all high levels in the organization; (ii) strategic awareness declines as we move downward in the organization; (iii) strategic awareness mostly depends on the informant’s position and status in the organization.

Measures studied are borrowed from other disciplines. Intellectual capital is interdisciplinary and borrowing measures can be expected because the constructs studied are embedded in theories from other disciplines. Moreover, as Peter (1981:138) explains: “the availability of multi-item scales from other disciplines substantially increases the probability of a validation study”. The instructions in the questionnaire were altered to replace the words “organization”, “industry” and “transaction” with “bank”, “sector” and “operation”, respectively. The ten performance items were reworded in accordance with banking system accounting plan, reflecting a more familiar financial language. Respondents were asked to state how their bank’s performance compares to that of their competitors.

Given the literature insights and our empirical findings, we applied the final test only to the chief executives and to the first and second-levels executives. An analytical sample resulting in a response rate of at least 150, as recommended by Chin (1998) was sought to ensure sufficient statistical power. To attain a sample of 150 observations, 430 executives (chief, first and second levels) were drawn from a list of 1081.

4.2 Data collection Data were collected from a sample of 53 banks, all affiliated members of the Portuguese Bankers Association.

An extensive review of mail survey response studies indicates that some form of follow-ups can increase response rates. Thus, two follow-ups (letters, telephone calls and e-mails) were carried out. We defined a cut-off time of eight weeks and evaluated the “resistance factor” (Huxley, 1980). Consistent with previous studies (Huxley, 1980; Parasuraman, 1982), the resistance over time, rather than continuously increasing during the survey, it was relatively high at first, dropped for a short period even before the first follow-up could have any effect, then started to increase. Total answers (253), after 8 weeks, represent a response rate of 58,8%.

Intellectual capital is an organizational construct that requires “strategic awareness” from informants to answer the questionnaire. Thus, a range of key informants was sought, including chief executive officers, regional directors and the directors of functional areas. Despite the limitations of “key informant” methodology (Phillips, 1981), we used this method of data collection because the organizational characteristics we intend to measure are only known by a selected set of members. Hambrick (1981) evidences that “strategic aware-

www.ejkm.com

14


Maria do Rosário Cabrita and Jorge Landeiro Vaz

(i.e., items) used to measure them. We assessed the adequacy of the measurement, examining: (i) individual item reliabilities; (ii) convergent validity and; (iii) discriminant validity.

5. Data analysis To avoid the standard assumptions of multinormality and the necessity of a large sample size, Wold’s (1982) method of Partial Least Squares (PLS) was used for parameters estimation. PLS, considered a “second generation multivariate technique” (Fornell, 1987:408), is a powerful approach to analyse structural models, involving various constructs and multiple indicators. PLS model is analysed and interpreted sequentially in two stages: (i) the assessment of the measurement model (validity and reliability of measures), followed by; (ii) the assessment of the structural model. This sequence ensures that the researcher has reliable and valid measures before attempting to draw conclusions about the nature of the relationships among the constructs.

Table 1 reports the measurement model results. Individual items reliabilities were determined by examining the loadings of measures on their corresponding constructs. Individual factor loadings greater than 0,70, indicates a high degree of individual item reliability. Cronbach indicators are all greater than 0,94, exceeding Nunnally’s (1978) heuristics. Internal consistency, a measure recommended by Fornell and Larcker (1981), is similar to Cronbach’s alpha but preferred in this context because it estimates consistency based on actual construct loadings. Discriminant validity was assessed by examining the correlation matrix of the constructs. Satisfactory discriminant validity among constructs is obtained when the diagonal indicating the square root of the average variance extracted (AVE) is greater than all other entries in the corresponding rows and columns. This implies that the variance shared between any two constructs is less than the variance shared between a construct and its indicators, which is our case.

Pilot test results helped us to refine the measures and retain the reliable items. Cronbach’s alpha coefficients for each of the four constructs are fine, since the alpha values are greater than 0,93, exceeding the level of 0,7, considered good for exploratory research (Nunnally, 1978). To validate the intellectual capital dimensions established a priori, a principal components factor analysis (Varimax rotation) was performed. Following Hair et al.’s (1992) recommendations, items that loaded at least 0,50 in its corresponding construct were retained. To confirm our factor findings, we used the PLSGRAPH 3.0 to assess individual item reliabilities. Results are very similar in the two approaches. Thus, we retained items that simultaneously loaded: (i) 0,50 in its corresponding construct by the principal component analysis, and; (ii) 0,50 in the individual item reliabilities by the PLS analysis, considered acceptable at the early stage of theory development (Chin, 1998). As a result, 48 items, from the 71 original items, were used in our final test.

As the measurement model satisfies the criteria for convergent and discriminant validity, our next step was to evaluate the structural model.

6.2 Structural model We started by running the main effects model, presented in figure 2. To test the structural model included: (i) estimated path coefficients, interpreted as standardized beta weights in the regression analysis; (ii) t-statistics, using jackknifing procedure, a nonparametric test of significance, and; (iii) R2 for each endogenous construct, to assess the proportion of variance in the endogenous constructs which can be accounted for by the antecedents.

It is important to notice that comparing studies in the three different international contexts (Canada, Malaysia and Portugal) we found that from the 48 total items, 15 are simultaneously reliable in the three studies and 18 are reliable in, at least, two contexts. Once collected the final data, we assessed the measurement model (outer model) and the structural model (inner model).

Since PLS makes no distributional assumptions, traditional parametric methods of significance testing (e.g., 2) are not appropriate. Therefore, a jackknifing method, produced by the blindfolding algorithm provided by the PLSGRAPH 3.0, was used to ascertain the stability and significance of the parameter estimates.

6. Discussion of results 6.1 Measurement model The measurement model consists of the relationship between the constructs and the indicators

www.ejkm.com

15

ISSN 1479-4411


Table 1: Measurement model results Items Number Cronbach of items Alpha

Human Structural Relational Performance

Human

14 10 14 10

0,9505 0,9406 0,9501 0,9416

Internal consistency (Fornell and Larcker)

Discriminant validity (*) (Correlation of constructs) H S R P

0,9569 0,9498 0,9563 0,9507

0,783 0,755 0,809 0,697 0,700 0,782 0,568 0,634 0,592

0,812

R squared (%)

57,0 55,6 44,5

Loadings H1 H3 H5R H6 H7 H8 H9 H10 H11 0,7769 0,7848 0,7958 0,7893 0,7592 0,7768 0,7604 0,7538 0,8210 H12 H15R H17 H18 H20 0,7702 0,8127 0,7552 0,7766 0,8287

Structural

S2 S3 S6 S7 S8 S9 S10 S11 S12 S15 0,8389 0,8665 0,7954 0,7743 0,8483 0,8028 0,8469 0,7703 0,7901 0,7488

Relational

R6 R8 R9 R10 R11 R14 R16 R17 R18 0,8003 0,7646 0,7489 0,7280 0,8464 0,8433 0,8710 0,7608 0,7272 R19 R20 R21 R22 R23 0,7984 0,7214 0,7445 0,7658 0,8010

Performance

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 0,7897 0,8555 0,7593 0,7594 0,7986 0,8292 0,8165 0,8350 0,7795 0,8854

(*) Diagonal elements in the correlation of constructs matrix are the square roots of average variance extracted. 0 ,7 5 5 2 1 ,0 5 6 9 ) ***

0 ,4 3 1 ( 7 ,0 8 9 0 ) ***

SC R 2 = 57%

HC

0 ,4 0 5 ( 5 ,9 7 0 4 ) ***

0 ,3 9 1 ( 5 ,7 6 2 9 ) ***

RC R 2 = 5 5 ,6 %

OP R 2 = 4 4 ,5 %

0 ,2 9 1 ( 4 ,5 7 8 4 ) ***

Note: Top number is path, t-values in brackets, *** significant at p-value