Understanding Research (45 page)

Action research
  This approach takes the full immersion model of ethnographic work one step further by the researcher engaging their research subjects, or community (e.g. a government or company department, or a professional group such as health professionals) in the design, execution, and analysis of the project. The project design is in this respect goal-directed and requires the full cooperation of all parties. See also
Participatory action research
.

ARPAnet
  An abbreviation for the precursor of the internet as we know it today, the US-based Advanced Research Projects Agency Network.

Avatar
  Sanskrit for manifestation or appearance, this term has been adopted for web-based and computer-mediated communications whereby a user nominates or designs an image to represent them online; e.g. a comic figure on your social networking site home page, your own creation or character through which you take part in a computer game or virtual world.

Behavioural
  A designation for social science research that studies how human actors behave by gathering data based on either direct or indirect observation, in controlled experimental or semi-experimental settings, or through surveys. Behavioural research findings are predominately rendered in quantitative forms although not exclusively. Behaviouralist project designs focus on the individual as the unit of analysis, as opposed to groups, communities, or institutions.

Bias
  In statistical terms, bias is a systematic error in the results. More generally the term is used to designate how ‘objective’ research projects, based on the gathering of empirical evidence to test a hypothesis, strive to avoid distortions of any kind, e.g. observational or cognitive interference, or expressing or revealing their
personal views or involvement during and after carrying out the research. In this view bias is something to be eliminated as far as possible if not accounted for in formal terms; e.g. establishing the
margin of error
, or avoiding the first-person pronoun (‘I’) in academic writing.

Case study
  A research topic that concentrates on a specific issue, group, organization, event, or individual within a defined time and place is called a case study. Comparative case studies involve two or more instances. A case study will engage a researcher in more than one approach to data-gathering and analysis. The aim and benefit of case-study work is the focus, depth of detail and richness of evidential material. Just how far any conclusions drawn can be generalized to other instances is a question methodologists debate, and students can find themselves being corrected on.

Coding
  A general term for setting up ways to ascertain patterns in written – and visual – texts in order to draw inferences about significance by analysis of recurring instances. Coding schemes can be quantitative (how many times a word or phrase appears) and qualitative (the position of a term in a larger text). The term tends to be employed in quantitative modes of content analysis based on standardized coding schemes. However, qualitative research projects can also code a text based on interpreting the position or recurrence of words, phrases, or themes.

Coefficient
  In mathematics this shorthand refers to the number used to multiply a variant; e.g. ‘2T’ means ‘2 [H11003] T’ (‘T’ stands for the thing you are multiplying – the variant). Put it another way, whatever represents the variant in the equation, the coefficient is the multiplying number before it. When applied to a research situation, a coefficient indicates the size of an effect; e.g. age and its effect on the probability of voting. In this case the effect multiplied by the age indicates how much age contributes to the overall probability of that person voting; the coefficient indicates the ‘size of that effect’. See also
correlation coefficient
.

Confidence interval
  In statistical terms this refers to the estimation of the highest and lowest statistical values for the observed data; a confidence interval should be high (usually around 95 per cent).

Constructivism
  A move with various inflections in social research that takes a different stance to the relationship between the researcher (as observer or analyst) and the object of research. In contrast to research that aims to consciously eliminate all forms of
bias
(see above), constructivist approaches maintain that researchers and that which they observe are mutually dependent. Stronger forms of constructivism argue that the social (and for some even the physical and biological) world does not exist outside human society and history.

Content analysis
  A form of textual analysis where the manifest content is collected and analysed according to a pre-designed coding scheme (see
coding
).

Control group
  In experimental research this is the group not subjected to the actions under investigation (e.g. a medical treatment or media effect) in order to permit comparison with those groups who are subject to the intervention.

Conversational analysis
  A form of linguistics in which conversations can be analysed not just in terms of the manifest content (see
content analysis
) but also in terms of speech rhythms, pauses and their meaning, inflections, and idiomatic phrases within a linguistic or cultural group.

Correlation/correlation coefficient
  When two variables are interdependent a change in one means a change in the other. A correlation is the extent to which this association can be measured; see
coefficient
). Independent variables do not correlate. The term in general usage refers to some sort of relationship between two or more values, or phenomena. However, in statistics, a correlation is a mathematical calculation (programs like SPSS do this for you) that establishes either a positive or negative value between measurements; minus one ([H11002]1) is negative, zero (0) means no outcome, and one (1) is a positive correlation. As M. Davies (2007: 261) notes, in real research scenarios the calculations usually lie somewhere in between these two values, the significance of which needs to be ascertained in turn. See Rummel (1976) for an extended, and helpful discussion in his guide to the uninitiated.

Critical discourse analysis
  This approach is usually identified with the work of Norman Fairclough and colleagues which combines linguistics (see
semiotics
) with
discourse analysis
techniques.

Data
  For many schools of thought, data are, by definition, quantitative; facts and figures gained by way of observational means (direct and indirect) that are then rendered in numerical values for statistical analysis. Notwithstanding debates about what does, or should, count as data or not, any sort of observation that is recorded (e.g. from field notes, quantities or tasks listed in a spreadsheet) could be regarded as a data point.

Deconstruction
  The French philosopher Jacques Derrida pioneered this critical approach to the relationship between writing, meaning-production, and the transmission of knowledge in western philosophy. All meaning-making and its interpretation is problematic by this account in that the very act of writing and reading are complicit in the legacy of western imperialism and its worldview. Derrida then subjected key texts in the canon to a radical exposé (deconstruction) of their underlying assumptions and pretexts. See also
structuralism, semiotics, postcolonial research
.

Dependent variable
  This variable changes according to another – independent – variable. Dependent variables form the focus of research in hypothesis formulation and testing. See
independent variable
.

Discourse analysis
  In contrast to
content analysis
, this approach treats the written (and spoken) word as more than the sum of its parts (countable words or phrases). It takes the manifest content and the larger social, cultural, political or economic context together in order to present an interpretation of both its explicit and hidden meanings.

End-user license agreement (EULA)
  The long legal contract that users sign when downloading or installing any new software; no installation can take place without clicking the ‘yes’ box. As is the case with all legally binding contracts, beware of the small print. Computer game and Virtual World players have discovered many things to contest and reconsider in the respective EULAs of their game/virtual world of choice.

Epistemic community
  A phrase to designate scholars or experts working closely together or sharing a common set of assumptions about the form and substance of their respective research enterprise. An epistemic community can be geo-
graphically or philosophically delineated, large or small, inclusive or highly specialized.

Ethical codes of practice
  These codes, maintained and developed by the professional associations of respective disciplines, aim to codify and guide the development of ‘good practice’, deter researchers from engaging in forms of
academic misconduct
, or inadvertently being socially irresponsible or culturally insensitive. Codes of ethics differ in content from institution to institution, and between countries and disciplines. National research bodies and funding organizations also maintain their respective codes, to which funded research projects are bound.

Falsification
  This term, coined and codified by Karl Popper, is the understanding that the benchmark of scientific research is through the testing of hypotheses that must be able to be proven wrong – ‘falsified’. According to Popper, if a hypothesis is phrased in such a way that it can never be proven ‘wrong’ by countermanding (observational) evidence then it cannot purport to be scientific.

Feminist research
  Since the women’s rights movement of the 1960s, academic research across the board has been influenced, critiqued, and remodelled by various streams of feminist thought and politics. Feminist research varies in terms of the weight it gives to qualitative or quantitative modes of research from discipline to discipline (e.g. feminist economics, media studies), between different persuasions and generations of feminist scholarship (e.g. postmodern feminism and its critics), and political ideologies (e.g. liberal, Marxist, or socialist feminism). What all feminist research has in common is an emphasis on studying women (as a group) or gender (as an analytical or empirical category).

Formal theory
  In political science this term refers to theories based on the modelling of relationships, or social systems, that can elicit hypothesis-testing; e.g. game theory.

FTP
  ‘File Transfer Protocol’ allows the downloading and uploading of files from a remote computer system without encryption.

FTTP
  This is the abbreviation for ‘Fibre To The Premises’ (or to the home) whereby a carrier installs a fibreoptic cable; these form the internet’s transmission infrastructure, directly into the home, or office.

Gopher
  This is the name for an internet protocol that was a popular tool for searching for information on the internet in the early days of the world-wide web (the 1990s) before user-friendly browsers and powerful multimedia search engines took off. A gopher search result presents a ‘no-frills’ text-based index of ‘hits’.

Hermeneutics
  This term, ‘the art of interpretation’ and derived from Greek for interpreter (
hermeneus
) originally referred to the interpretation of Biblical scripture. It now refers to theories of interpretation more generally in which researchers aim to understand the complexity of human actions and their cultural production (e.g. literary and other sorts of ‘social texts’, institutions). It is closely allied to textual analysis traditions. Research falling under either rubric assumes that meaning is difficult to ascertain – hence the need for interpretation; observational evidence also by this account.

HTTP
  ‘HyperText Transfer Protocol’ is one of the fundamental protocols governing the inner-workings of the web. It mediates between computers, servers,
and users by processing requests and responses in a uniform way across different systems; e.g. by instructing a server or web-browser what to do in response to a user command (e.g. retrieve a web-page based on the web address command).

Hybrid research
  As the term suggests, this is research that synthesizes several approaches to data-gathering and analysis within a single project; often used as a synonym for
mixed method
research.

Ideal type
  This term, coined by the German sociologist Max Weber, refers to the distillation of the main characteristics of a particular social instance, or entity, in order for the researcher to extrapolate from the particular to the general and so make comparisons, e.g. between ‘traditional’ or ‘charismatic’ authority. Using the term in this way, according to Weber, is not putting the entity under discussion as an ideal in terms of a ‘better’ type. It is an analytical device rather than a normative evaluation.

Incommensurability
  This term, coined by Thomas Kuhn (1962), refers to the way in which research traditions, between generations or epochs even, do not communicate with each other because they cannot find common ground; not only are terms of reference foreign to one another but so are worldviews; e.g. the Ptolemaic view of the universe in ancient Greece and medieval Europe in which the sun revolved around the earth is incommensurable with the Copernican view that the earth revolves around the sun.

Independent variable
  This variable is the factor that explains or predicts the outcome of actions performed on a
dependent variable
. An independent variable therefore does not change; it induces change. Causal inferences are drawn based on the statistical measurements of the relationship between independent and dependent variables.