Theory+in+Community+Informatics

//**general note on CIRN articles**: If this article interests you please use the "DISCUSSION" tab. If you wish to collaborate please contact the author(s) before editing this page.//

=Theory in Community Informatics=
 * Author(s)** Larry Stillman


 * [this is being written, bit by bit]**


 * Human existence is packed with theory. Most of the time, we don't consciously theorise. We pick up a hammer, and by dint of experience or observation, we have a theory that if a hammer is applied to a thumb with some pressure, pain will result. We reflect upon, test, and improve on such experiences, and use them to extrapolate to other situations and particularly, to invent and teach others though the medium of communication. We try to create universal rules and specific rules on the basis of experiences, observations and insights. That is what distinguishes most of us from other species (though some primates can invent as well).**

In social research and practice of all sorts, we consciously aim to theorize about what we are doing (or we aim to do so) in a way that what we communicate has some applicability to problem solving with. The degrees to which theory represents or replicate reality is an issue that has been extensively discussed in the philosophy of the social sciences, and while taking a step back into philosophy may appear abstruse and irrelevant, in fact, getting back to basics helps with clarity about the nature of theory in community informatics. The experimental sciences, and social sciences, have been highly influenced by a school of philosophy known as logical positivism with its emphasis on the 'testable' verifiability of propositions and concepts. (A very useful discussion can be found in Passmore, 1972). While logical positivism was originally intended as a critique of science, it has had a strong cultural influence of the development of the social sciences, particularly in the United States, with an emphasis upon the development of strong theory and testing methods for rational decision-making

We also face the same problem in many areas of public policy under the thumb of the neo-liberal agenda --the constant calls for proof of outcomes, and accountability, even though reality (and the future), is not always predictable, and sometimes, there is no easy metric to demonstrate an outcome. We are involved in a struggle to develop a body of rigorous social concepts, theory, and method, that will be taken seriously in our interactions with those who control the funds.

Change can happen. As an example, in the field of development, where there is a constant desire for 'proof' of intervention effects, Amartya Sen eloquently speaks of 'informational exclusions' which because they effectively silence other conceptions of poverty, particularly those which are beyond a narrow econometric base, result in quite inadequate policy decisions and outcomes. His highly influential work has been devoted toward developing a new system of 'weights' to understand the different dimensions of poverty and development. For this, his insights he has won the Nobel Prize, but and writes:

//"There is an interesting choice here between 'technocracy and democracy'...a choice procedure that relies on a democratic search for agreement or consensus can be extremely messy, and many technocrats are sufficiently disgusted by its messiness to pine for some wonderful formula that would simply give us ready-made weights that are 'just right' "//. (Sen, 1999, p. 79)

This is familiar stuff from our experience, but what we can hope to due is at least have a theoretically strong base that reflects our ethical choices about the nature of Community Informatics.

The desire for strong forms of theory--that is theories that are built on clear and verifiable foundations, continues to bedevil discussions in community informatics.

This article has two broad components. First, it proposes that the quality of much theorizing and practice in community informatics (and by implication, development informatics), can be vastly improved by reference to key discussions about ontology and epistemology, and the nature of practice in the literature of program evaluation, since so much of CI work is about the implementation and assessment of projects in the public domain--as is much of the work covered in the practice and theory of Program Evaluation internationally.

Second, it takes up the key points drawn from the program evaluation literature in the context of recent discussions in IS about the nature of theory and its relationship to practice. Joining the insights of Program Evaluation to the insights of discussions about the relationship between CI and IS helps to provide to provide some answers -- or at least steps towards being clear about questions and possible answers - for the conundrums that consequently come up in community informatics concerning the relationship between social theories of change and theories and practices of technology intervention.

Part 1:Theory in Program Evaluation and its relevance to Community Informatics.
This takes us one step beyond merely 'doing' to steps involving describing, abstracting and judging what has been done and what will be in terms of particular criteria that hold particular value. In developing technical artefacts such as software or hard systems, we also base our work upon various bodies of theory (mathematics, physics, chemistry, metallurgy etc). The measure for the development of good theory is that it has good internal validity--that we can prove our case, based upon the explication of basic concepts, causes, processes, outcomes, and other effects (Campbell and Stanley, in Shadish et. al, 1991). That is not to say that all theories need to specify causation or process, but that whatever is proposed, is based upon a clear picture of reality (an ontology, for example, that there is a demonstrable reality, or another view, that there are multiple realities), and a way of understanding and demonstrating knowledge (an epistemology).

In the writer's view, a host of issues of consist concern to CI: around the nature of knowledge, measurement, truth, valuing, causality, proof, reporting, the politics of implementation, and a host of other means by which information and knowledge are recorded and assessed in CI are extraordinarily familiar in Program Evaluation. In fact, we appear to go around in circles, ignorant of a parallel wealth of knowledge in a practice field which is cross-disciplinary, in which significant breakthroughs and insights have been made.

There are now significant scholarly works which cover nearly 50 years of experience in Program Evaluation, which developed as a means of assessing, in particular, the impact of mid-century Great Society Programs in the USA in which billions of dollars of public funds were invested. There are a number of authors whose work has been fundamental to the field, and their work will be reviewed for this article. Of course, there are other authors, and a host of related articles, but these are ripe for further investigation become these indicative remarks.

Part 2: Community Informatics and ideas in Information Science.
In Community Informatics (CI), by and large, the emphasis is upon the design or implementation of solutions for human-technology interaction, as distinct from the 'hard' end of computing science. CI work emphasizes human and social technical, rather than technical work alone in the development of solutions to particular problems or cases and this leads to a weak approach to thinking systematically and theoretically( de Moor 2009, Stillman and Linger 2009). Furthermore, the world of CI is not static, and does not end with the design, or implementation of a product, but it is one of response in the interaction between people and technology. CIs agenda is thus very complex. With insights drawn from structuration theory and elsewhere, we can make the claim that the interaction between people and technology is a duality; that the relationship between the two is such that they both enable and constrain behaviour and outcomes, and particularly, that:

//[T]he concept of technology in use can be understood as the social structure (the set of rules and resources) mobilised by actors in the ongoing and situated use of a particular technological artifact. In this framing, technology-in-use is both the medium and the outcome of situated human action. (Orlikowski 1995: 3)//

This is a view also shared by de Moor, who suggests that:

" //communities and their technologies co-evolve in that the technologies both afford and constrain the behaviour of their communities, and, in turn, these communities shape the technologies as they are being applied in messy practice" (de Moor, 2009)//.

An additional nuance to these views is that significantly, indicative of the power of ICTs in this era, is that these human-technical relationships are conducted in and through time and space, allowing the production, storage, and communication of information in extraordinarily creative ways and at the community level, an enormous reservoir of power is now available to increase social capital (a complex term in itself), as well as physically related capital, including jobs, skills, and networks. ..

We can see such things happening in how mobile technologies are used and the explosion in SMS use, despite the constraints of short messaging. Time and space can be seemingly dissolved, through instant communication (SMS, Twitter), but equally, moments and actions can be stored out 'there', whether as an email or a stored document or video. Human creativity has taken advantage of a simple yet constrained system to engage in seemingly unrestrained communication. Without a doubt, new applications will be taken advantage of in ways that were never intended by their creators.

One of the major difficulties that many of us face, however, is that under pressure of time and funding, the capacity for reflective and critical theorising about the nature of the relationship between people and communities and technologies and more particularly, theorizing and designing system design and implementation is constrained, even though, in the long term, better theorizing and reflection may lead to better practice and the creation of a body of theory that can help to empower communities through working with them in the design and implementation of bottom-up solutions. All this is of course, easier said than done.

Gurstein has suggested a broad conception of the CI (community informatics) agenda, which includes:

//“… a commitment to universality of technology-enabled opportunity including to the disadvantaged; a recognition that the “lived physical community” is at the very center of individual and family well-being – economic, political, and cultural; a belief that this can be enhanced through the judicious use of ICT; a sophisticated user-focused understanding of Information Technology; and applied social leadership, entrepreneurship and creativity" ( Gurstein 2008: 12).//

Much of the research in CI is thus devoted to enabling ‘effective use’ (as Gurstein has put in) in a ‘lived’ communities setting for the purpose of innovative social-technical activity. Such activities include capturing community memory for community empowerment (Stillman and Johanson 2007), and in developing countries, poverty amelioration, literacy development, or AIDS information (Heeks 2008). Consequently, CI sees lived-in and situated communities not as passive recipients of technological opportunities, but as actors engaged in the comprehension and ‘doing’ of community problem-solving directed to social progress.

The concerns of CI can be considered within the tripartate framework developed by Giddens for considering how social systems (and people and technology constitute a social-technical system) can be analysed. There are three main dimensions to such a system that are brought into being through the //agency// of people and artefacts, and de Moor and colleagues have developed a related schema, in which 'cases' are the subjects of community-based research (see de Moor 2009). In structuration, there are three main dimensions:


 * Interpretive Schemes: the forms of communication used, whether language or media
 * Facilities: including types of technology and system, and processes and behaviours for conduct.
 * Norms: the values (and sanctions) for particular forms of behaviour or methods and processes (an obvious example is netiquette, but a commitment to Open Source solutions is another)

Adapting what Day and colleagues (Day, Farenden et al. 2007), have suggested, CI can be seen as an emergent normative framework, and de Moor refers to these as 'context and values'. This framework seeks to use ICTs and contingent networks and the facilities they offer in a particular way focusing on participative community change though particular used of ICT resources and processes (Facilities) and communication (the Interpretive dimension), somewhat in opposition to assumptions about a natural alignment between governance, rationality, and technology.

The outcome of this tripartite relationships poised at a nexus of practice, policy and research about and for ‘lived’, bedrock communities or at least, problem solving with and for communities (Stoecker 2005). CI has strongly emphasized the centrality of ‘people and place’ in community development, together with the importance of what Chester Barnard identified seven decades ago as ‘the informal organization’, the things that provide meaning and solidarity within formal organizations and by association, within that 'social being thing' called community as affected by ICTs. As de Moor has suggested, however, process and methodology is both a strength and weakness in CI in that because their is a surfeit of experience and particular theory, there is a lack of a parsimonious intellectual cohesion by which to consider the 'situatedness' (echoing both de Moor and Suchman) of particular community informatics initiatives.

Thus, once we begin to drill down into analyzing the specifics of cases, we see how important is the development of some form of generalized theory to cope with a surfeit of infinitely variable practice. For example, the term community is a controversial one in the social sciences, and continues to be constantly discussed in community informatics. Literacy has many dimensions, as does poverty amelioration, community based research, or the very idea of community action is itself based upon certain assumptions about how groups and communities work, more from a consensus than conflict perspective of society. The same applies to the important concept of 'effective use' as developed by Gurstein. Just what is meant by effective? What is meant by 'use'? Is it a universal etc. As Weber has suggested, there is much rhetoric but little rigour (2009) in our research and definitional and theoretical activity, and the weakness of 'constructs' in developing theory ultimately leads to weak theorization.

This is not the place to get into a discussion of the ins and outs of the philosophical assumptions about the nature of theory and the critiques of conceptions of theory, causation, or theory-building for which there is ample literature, including foundation work of Popper, but as well, the mid-range by very sophisticate work on evaluative theorising by Chen (for example, Chen 1990 and later, and Mjoset 2004). The emphasis in this article is upon the use of theory for theory-building and problem solving in community informatics.

CI is not concerned with the highest levels of abstract theory, though it can draw upon such theories and the classic theories are classic because they have such power. They should not be ignored our work even though it is more concerned with practical, bread-and-butter solutions and modest theoretical findings to the design and application of technology for problem solving in group settings, particularly community settings with high levels of social-economic deprivation. However, the importance of such theoretical work cannot be underestimated, since it can underpin important social effects, and be used to analyse, describe, and improve upon them to different audiences. This kind of theory work is what Merton referred to as:

//" Theories of the middle range (sic): theories intermediate to the minor working hypotheses evolved in abundance during the day-by-day routine of research, and the all-inclusive speculations comprising a master conceptual scheme. (Merton 1968: 5)."//

Additionally, and this is a high-level, rather than practical challenge, there is a need to develop CI theorizing so that it can productively interact with //information systems problem-solving//, in addition to the //community problem-solving// orientation of CI. This requires an exploration of how critical thinking in IS could be exploited by CI. What is needed is a way of adapting and communicating the theory frame of IS to strengthen the conceptual, theoretical and practical activity of CI while also evolving the IS theoretical framework to accommodate the social, and in particular the community, agenda (Stillman and Linger, 2009).

Gregor (2009) has an extensive discussion of types of theory in IS, and while this should not be regarded as an end point for the classification or development of theory types, it bears much in common with the theory challenges to CI. Her Table 2 has been modified with observations pertinent to community informatics.

Highly desired for accountability purposes. Often uses theories and techniques drawn from community work, social sciences, management.
 * || **Type** || **Attributes** || **Observations for community informatics** ||
 * I || **Analysis** || Says what is. Theory does not articulate relationships amongst constructs nor provide the basis for predictions || Often project descriptions and process narratives ||
 * **II** || **Explanation** || Says what is, how, why, when, where. Explains the behaviour of phenomena || Practical field work: qualitative and quantitative/mixed forms of reporting and observation.

The domain of much social theory, including feminist, race, development, colonialist theory || 4) ICT unpredictability (e.g. unknown initial effects of SMS)
 * III || **Prediction** || Says what is and what will be. The theory provides predictions and has testable propositions but does not have well-developed causal explanations. || Highly desired but extraordinarily difficult to achieve in practice due 1) time needs for effects 2) social reality of community interventions 3) community and institutional politics

Social planning theory;community development theory ad hoc community informatics project theories ||
 * **IV** || **Explanation and Prediction** || Says what is, how, why, when, where and what will be.

Provides predictions and has both testable propositions and causal explanations || As above and the danger of applying scientific-experimental models to complex human situations. Technical-rational governance in the face of reality (Habermas)

? ||
 * **V** || **Theory for design and action** || Says how to do something. The theory gives explicit predictions, e.g., methods, techniques, principles of form, function for constructing an artifact || Applicable for the most part to strict artefactual design, much more complex for people-machine networks and social-technical artefacts.

IS theory; sophisticated forms of community development theory ||


 * || Weber (2009), at least,has suggested that for Development Informatics (DI), has suggested that **Type II** theories 'have been proposed primarily by researchers who have undertaken field-study analyses of development projects associated with the implementation of ICT in a particular community", and this is certainly the case with many, if not most CI project work. As Weber notes, while the explanatory theory developed by researchers is of most interest to policy makers, they cannot be classified as //predictive// theories insight as actor network, institutional, social network, or social capital theory. However, Weber considers these theories as not rigorous because they do not provide a 'normative' foundation for policy development (and probably too, for implementation or adaptation by others).

On this point, Weber can be criticized for perhaps expecting too much of practical on-the run theorizing that is often done in a situation of client dependency. But on the other hand, he does make a very strong point there, that frequently, there is a lack of rigour, that argumentation is not strong, and boundaries are not carefully defined. It is probably in this Type II area that there is a need to methodological training about field work that will result in an increased confidence and capacity of people to develop theory. ||
 * || Weber also suggests that **Type III** theories as they apply to DI 'have been proposed primarily by researchers who have used statistical models to investigate relationships among data they have collected via surveys obtained or from secondary sources'. He argues that because these theories successfully pass empirical tests for reliability etc. they can be applied in different countries and situations. However, a problem with Weber's way of thinking is that much work in CI (and DI) is now testable through traditional empirical tests for several reasons 1) such tests don't exist 2) the nature of the intervention cannot be measured through a traditional test 3) field conditions make it impossible to conduct such traditional forms of testing. A correction to Weber would be so say that rigorous qualitative measures and mixed methods can develop equally significant findings and are often far more suited to 1) the skills of people doing the work 2) the resources available 3) the field conditions. To be fair to Weber, however, he recognises that many tests are in fact proxy measures for other outcomes or effects that cannot be otherwise measured.

Weber suggests that **Type IV** models are a distinct possibilty and are the desired theories of the middle range. However, he also recognizes the difficulty of developing associations between what he calls 'focal constructs' (core concepts) that are rigorous, and compelling (for example, between 'community capacity' and ICT capacity). He recommends for ICT4D, and this equally applies to CI, that it develop a deeper and more precise understanding of its focal constructs as a key to theory building which could have normative relevance not just on a higher theoretical level, but, taking a bottom up approach, in improving the capacity to work with communities to develop ICT initiatives that engage them as the major stakeholder in activity. Being able to develop explanations and predictions in conjunction with communities might in fact lessen the dangers of project failure and risk aversion as well.


 * Type V models** appear to be rare, at this stage, in community informatics. A recent and sophisticated example is that developed by Aldo de Moor for Collaboration Patterns in Community Informatics (2009). In this case, the artefact to be constructed is a tool for decision-making based upon particular 'social-technical lessons learnt in optimizing the effectiveness and efficiency of collaboration processes. The tool itself is based upon carefully delineated knowlege in particular domains strucutred in such a way that collaboration is enhanced around information, communication and coordination purposes. Social and technical questions are queried an answered, thus providing a construction map or template of what is to be built. Thus, de Moor's model provides a road map for all the neccessary conditions as outlined by Gregor (see the diagram above). The one flaw with de Moor's model--and he admits that it is operating at the highest level-- is that it makes an assumption that people will actually participate equally in the discussion and documentation process to create a rational map of what is to be constructed. of course, in reality, particularly in many community informatics enviroments in the field, realities of literacy, access, communication confidence, competency, and institutional politics and problems will interfere with the creation of a perfect match. But his model is an important step forward in delineating a stystem that brings together, as Hirschheim and others have decribed, the Technical, Langugage and Organization Domains of Information Systems --each of which tends to have non-communicating specializations--to consider how the more instrumental or 'construction' aspects of Information Systems design can be integrated with the better Sense-Making and Argumentation aspects of Community Informatics. ||

Conclusions Suggest further modifications