 |
Narrative literature review:see literature review.
|
|
|
|
GLOSSARY
Abolitionism: a theory, a strategy and a social movement. As a theoretical perspective, it is informed by an ethical and political critique of the violence and harm of prison. As a strategy, it calls for the end of the use of imprisonment and the building of communities and societies shaped by the principles of social justice. As a social movement, abolitionism is a collective organization working directly through research, policy interventions and direct action for progressive social change.
Algorithm: a finite sequence of instructions, to perform a task, solve a problem or answer a question. For example, long division is an algorithm that converts inputs (the dividend and divisor) into the output (here, quotient). Machine learning relies on algorithms.
Autobiography: an account of one’s own personal life as described by the person hold- ing such experience (i. e. the person themselves).
Benchmarks: data that are collected at a particular time and in relation to which future comparisons are made.
Big Data: generally defined as data having high volume, high velocity (speed of produc- tion and processing) and high variety (of data types and sources). The term sometimes also incorporates analytic techniques employed to extract information from such data. However, there are different definitions (Chan and Bennett Moses, 2016), including a ‘new epistemological approach for making sense of the world’ (Kitchin, 2014: 2) and ‘a cultural, technological and scholarly phenomenon’ involving technology, analysis and mythology (boyd and Crawford, 2012: 663).
Biography: an account of the experiences of another person, as described by someone else (i. e. an author, biographer or social scientist).
‘Brief and charge’: the totality of intentions, instructions and restrictions adopted or generated by the maker(s) of images (e. g. artists), whether consciously or subconsciously, before and during the image production process.
Case: a spatially and temporally bounded unit, which is observed at one point in time (synchronic) or over time (diachronic); it is the phenomenon you want to explain.
Case study: an intensive study of one case to understand a larger set of cases.
Category saturation: in qualitative analysis, the point at which your interviews/observations add nothing to what you already know about a category, its properties, and its relationship to the core category. When this occurs you cease coding for that category.
Causal relationship: something that exists when one phenomenon exerts an influ- ence on another.
Chain referral sampling (also known as snowball sampling ): a technique by which researchers use existing participants to recruit further respondents from their circle of acquaintances. Since the respondents are recruited solely because they are known to other participants, the sample cannot be probabilistic or taken as representative of the population as a whole.
|
|
|
Clean/dirty data: involves the notion that many data sets contain flaws. Cleaning data involves identifying those flaws and removing them. Like hygiene, we can remove many known flaws but can never know that we have removed all of them.
Coining: an offence that involved the making and passing off of counterfeit coins. The offence was particularly associated with women, who were thought to be able to pass counterfeit currency in shops, although the number of women who were prosecuted at the Old Bailey and elsewhere suggests that this was actually a very risky form of offending for women.
Comparative methodology: widely used to test hypotheses by examining whether simi- larities and differences between two or more cases accord with theoretical expectations.
Complainant: the person who brings the case to court, not necessarily the victim of the crime. It may be a police officer or a parent who brings a case on behalf of another person.
Complementarity: used in mixed methods research where two or more methods are used to investigate distinct, albeit often overlapping, aspects of a phenomenon in order to produce rich, deep understanding.
Conceptual themes: themes that represent underlying ideas or principles emerging from your data. A conceptual theme or themes are findings in their own right because they are commonalities.
Constructivism: a philosophical notion of science that is critical of positivism’s and postpositivism’s belief in objective reality, and the proposition that it can be explored using the human senses. Constructivists instead consider that there are people and groups of people who construct images of reality. These are the images that should be the subject of scientific research.
Content analysis: a research method based on the study of documents or other forms of communication. The content of such communications can take the form of text, images, audio and video recordings, animations and even computer games. The aim is to examine such content so as to uncover patterns or tendencies in communi- cations or representations, and it can be undertaken using both quantitative and qualitative techniques.
Continuous evaluation: the efforts by evaluators to generate feedback and improvement in, and as part of, the evaluation process itself.
Correlation: occurs where two variables tend to fluctuate together, either positively or negatively. The demonstration of correlation can be distinguished from the dem- onstration of causation, the latter implying that a change in one variable is what leads to a change in the other.
Covert observation: a social research method that involves data collection through observation of people’s social practices in the natural settings in which they occur, and without those individuals’ knowledge that they are being observed and studied.
Covert research: conducted by a researcher in the absence of the participants’ knowledge. Once common, it is now regarded by many universities as a form of deception and therefore unethical, especially when it coincides with a high safety risk for the researcher.
Crime concentration: the number of crimes experienced by each victim on average.
Critical research: that which connects personal problems to wider social issues, and exposes how structural relations of power impact on understandings and experiences of violence, harms and justice.
|
|
|
Cross-sectional methodology: involves researchers studying a sample or a popula- tion at a single, specific moment in time, in contrast to longitudinal research.
Cross-sectional studies: rely on multiple observations of different cases from the same point in time. Such studies are generally cheaper than longitudinal studies and can often involve more cases or a larger sample size. Nevertheless, they present what is essentially a ‘static’ snapshot of what may be a dynamic and quickly changing phenomenon.
Data analytics: a general term to describe analysis that is conducted on data in order to extract useful information or inferences, including approaches based on machine learning. Related terms include data analysis, data science and data mining.
Data signature: describes the available evidence patterns that follow from the operation of specific causes in specific contexts. The more detailed the better for testing hypotheses.
Decision making: the process of adjudicating between alternative courses of action and of making choices based on such adjudication. One way of viewing the research process is as a series of decisions taken about the topic of research, the way in which research problems are formulated, the form of case selection, data collection and data analysis. The validity of research conclusions is the outcome of such decisions.
Deductive: refers to the relationship between theory and research, where the research is conducted based on a hypothesis that utilizes existing theory (see inductive for a contrasting approach). In simple terms, deductive is an approach to research that seeks to test a theory or hypothesis with observations or data.
Deductive research: involves the use of qualitative or quantitative research to test the validity of hypotheses derived from existing theories or data sets.
Descriptive research: aims to provide information about the unit of analysis at the centre of the research question. The research can be qualitative (e. g. the physical layout of a courtroom) or quantitative (e. g. the number of offences recorded by the police across police force areas). Although the aim of the research is not to look for causal relationships (see explanatory research ), it is possible for descriptive research to be deeply analytic.
Development: a technique of mixed methods research involving the employment of a sequential design, with the inferences drawn from the first component used to help inform the development of the second component. This approach is employed to increase the robustness of the findings and any concepts generated as a result.
Developmental and life-course criminology: in contrast to traditional criminology, has its focus on the similarities and differences in crime between or within individu- als measured at the same point in time or at repeated points in time. The field focuses on understanding the onset, persistence and desistance of crime.
Diachronic: an approach to research that seeks to determine whether there is change over a period of time in the phenomenon of interest.
Digital corpus linguistic methods: put simply, where we take a large amount of text and count the number of times that particular words or phrases appear. This allows the researcher, for example, to count how many times words like ‘lawyer’ or ‘victim’ are used each year.
Discourse analysis: a broad term that refers to the analysis of written and spoken communication. It generally focuses on what is being said, and the implicit ideas, assumptions and ideologies that may underpin this. It is concerned with the implica- tions of the construction of knowledge and ideas. In the criminal justice and criminological setting, it may be applied, for example, to seek to know how crime or victimization or responsibility are understood.
|
|
|
Emotionality: involves reflecting on how research processes and practices generate (sometimes challenging) emotional responses, and how these impact on the collec- tion and use of research.
Environment assessment: analysis of the context within which research or evalua- tion is to occur, such as auditing the human and material resources needed for a project and examining the policies that might affect its development.
Epistemology: the theory of knowledge, concerned with the question of what counts as valid knowledge. Broadly, epistemology is constructive or positivistic. It concerns how we go about evidencing our beliefs and claims to legitimate knowl- edge. It refers to what and who is included/excluded from our conception of social reality.
Ethnography: a research method that originates in social and cultural anthropology, but which has been widely adopted and adapted across the social sciences. A qualitative
approach, ethnography focuses on the cultural practices and interactions that take place in a particular locale, setting or community, and seeks to elicit a detailed and fine-grained understanding of the lives of those studied.
Evaluation: a process of assessing what we are doing, understanding why we are doing it and evaluating how we can make improvements in the future.
Expansion: refers to mixed methods research that provides breadth and depth to the exploration of a particular phenomenon. This approach reflects the philosophi- cally pragmatic notion of selecting the most appropriate tool for the job at hand.
Experience: the sensory awareness of the world during encounters with objects in it. At this sensory level, experience is largely pre-conceptual, pre-language even.
Explanatory research: about looking for patterns and relationships between varia- bles, thus providing a causal explanation for a phenomenon.
Fieldwork: primary data collection done in the field, which is a time-limited and space-limited setting in which a particular ethnographic research project is being conducted. Field notes and audio or visual recordings are the standard methods used in fieldwork.
Formative evaluation: forward-looking evaluation, the purpose of which is to iden- tify needs, clarify rationales and improve implementation in the future.
Generalizability: the extent to which data and analyses produced in one specific research setting can be claimed to be relevant to similar research settings in other places and time periods.
Grounded theory: associated originally with Glaser and Strauss (1967) and is an inductive approach which starts with data collection, and then, as data emerge, con- stant comparisons are made between data and theory.
Hard to reach: a term which refers to a group of people who are not typically represented in survey data due to their marginalized, isolated, socially excluded or highly mobile nature. Examples include rough sleepers, chaotic drug users and undocumented migrants. They may also be termed ‘seldom heard’.
Hot spots: locations where crime rates are relatively high. Hot spot policing focuses police efforts on hot spot locations, for example by increasing police deploy- ments in those areas or focusing problem-solving efforts on those areas.
Hypothesis: a logical conjecture about the possible relationship between two or more variables expressed in the form of a testable statement.
Hypothesis testing: refers to the collection and analysis of data whose patterns are expected in the light of theory. The more specific the expected pattern and the more closely high-quality data patterns accord with it, the better corroborated the theory.
|
|
|
Incidence rates: incidence rates for crime describe the number of incidents in rela- tion to the population at risk. It is often expressed as crimes per 1, 000 or per 100, 000. The denominator refers to the population at risk, which depends on the crime in question.
Inductive: an orientation to research inquiry that involves reasoning from evidence or ‘inference’ and often involves generalizing from something that has been studied or observed to other cases that have not.
Inductive research: a broad approach in which researchers start with as few precon- ceptions as possible and build theory from the ground as they gather and reflexively analyse their data.
Inference quality: the evaluative term used in mixed methods research to judge the value of the conclusions and interpretations that stem from the integration of find- ings from each component of the research. It is the mixed methods equivalent of the evaluative term ‘validity’.
Inferences and meta-inferences: the conclusions and interpretations from each sin- gle component (quantitative or qualitative) of a mixed methods research project, whilst meta-inferences are the conclusions and interpretations drawn from across all of the quantitative and qualitative components.
Initiation: mixed methods research where new perspectives or paradoxes emerge. This may not have been the purpose of the mixed methods design, but the inferences generated from each component of the research allow for further analysis to be undertaken to create new knowledge and ideas.
Inquest: a judicial inquiry which is tasked with ascertaining the facts regarding the circumstances of a given incident. When a person dies in prison (or indeed any other form of state custody), an inquest is conducted by a coroner with the assistance of a jury in a specially convened ‘coroners court’.
Intelligence-led policing: (which includes predictive policing): a ‘strategic, future- oriented and targeted approach to crime control, focusing upon the identification, analysis and “management” of persisting and developing “problems” or “risks” … rather than on the reactive investigation and detection of individual crime’ (Maguire, 2000: 315).
Interviews: a method of data collection, information or opinion gathering that specifically involves asking a series of questions.
Inverse trend: a trend where an increase in one is associated with a fall in another.
Life-course study: in general, focuses on tracing, understanding and explaining within-individual continuity and/or changes in attitudes, thought patterns and/or behaviour over time.
Life history interview: a form of interview where the interviewee narrates their life in a more or less structured form.
Literature review: an evaluative overview, synthesis or survey of the main published work in a given topic or field. It provides an overview of the state of academic knowledge on a research topic and is increasingly known as a narrative literature review.
Longitudinal research: a form of research involving repeated observations (through measures such as surveys, register data, tests or interviews) of the same unit of analysis (such as a variable, person or place) over time.
Longitudinal studies: studies that rely on repeated observations of the same exam- ples/cases over time. They are often time-consuming and costly but bring the advantage of being able to demonstrate the dynamic nature of a given phenomenon – for exam- ple, the development of the impact of crime on victims over time.
Machine learning: occurs where a computer optimizes parameters in a model drawing on data in a training set. Examples of machine learning techniques include neural networks, decision trees, random forests and Bayesian learning. Machine learning can be supervised (where a training set with known classifications is employed) or unsupervised (where the algorithm itself identifies clusters or pat- terns in data).
Method: the tools used to collect data on a particular phenomenon, broadly cate- gorized as quantitative or qualitative.
Methodological tools: the data collection and analysis techniques that are used in research.
Methodology: the principles and ideas on which researchers base their procedures and strategies, shaped by ontological and epistemological positions. Methodology is a process or system.
Mixed methods design: the element of a mixed methods study where decisions on the priority question (is there a dominant component or are the components equal in status? ) and the sequence question (do the components occur simultaneously or sequentially? ) are made.
|
|
|
Multiple realities: an ontological position often associated with mixed methods research. It suggests that there is no single ‘correct’ ontological understanding of the social world, and that instead there are multiple understandings of reality and each has some validity.
Narrative literature review: see literature review.
Natural experiment: occurs where changes that can be used to test a hypothesis are made, but not through an experimenter’s intervention.
Non-participant observation: a social research method that involves data collec- tion through the observation of people’s social practices in the natural settings in which they occur, but in which the researcher him or herself does not participate. The researcher typically maintains a distance from the activities and phenomena being studied.
Observation(s): the basic unit of empirical inquiry; the ‘lowest-level units in an analysis’ (Gerring, 2017: xxviii).
Offender risk assessment: tools that attempt to quantify the risk that a person charged with or convicted of a crime will commit an offence while on bail, comply with bail conditions, re-offend once released or released on bail, and/or commit a violent offence while on bail or parole. These tools are used in some jurisdictions by those tasked with making decisions about bail, parole and/or sentencing.
Old Bailey (or The ‘Bailey’): London’s central criminal court which dealt with seri- ous offences, and still does today.
Ontology: a way of conceiving of the social world around us. It reflects certain assumptions about the nature of society and the way in which society is likely to develop in the future. It is the theory of ‘reality’ – whether reality exists independent of society, or whether it is constructed by people.
Operationalization: this refers to the laying down of rules which stipulate when instances of a concept have occurred. Operational rules link abstract concepts to observations. Such observations are sometimes also known as indicators. The extent to which observations are truly indicating instances of a concept is the extent to which an operationalisation has measurement validity.
Overt research: research conducted openly by a researcher who supplies the par- ticipants with full and transparent information about the research project and the purpose and destination of the findings. The information is usually provided in writ- ing and accompanied by consent forms for participants to sign.
Participant observation: ethnography’s principal research method. It is a social research method that involves data collection through the observation of people’s social practices in the natural settings in which they occur, and in which the researcher is an active participant. It entails the immersion of the researcher in the research setting and the systematic recording of data regarding events, cultural prac- tices, artefacts and symbolic meanings ready for analysis and theorization. Most closely associated with ethnography, participant observation typically involves the researcher in a close and extended interaction with those under study, taking part in the various activities that are the focus of analysis.
Performance indicators: measures designed to indicate present performance in rela- tion to stated goals.
Performance targets: identification of the end point goals of a particular pro- gramme or project.
Persistence and desistance in crime: in criminology, persistence is used to describe an individual’s continuity in crime; desistance means the cessation, or ending, of the criminal career.
Perspective: a particular attitude towards or way of regarding something; a dedi- cated point of view.
Pilot study: is an exploratory investigation that attempts to outline the area of study prior to the development of questionnaires. “If the main study is going to employ mainly closed questions, open questions can be asked in the pilot to generate the fixed-choice answers” (Bryman, 2001: 155).
Piloting: is a term used to describe the process of ‘trying-out’ or practicing your research instruments (interview/ questionnaire) on a population that matches your target population. Piloting reveals which parts of an interview/questionnaire work and which don’t. Research tools can then be adjusted accordingly.
Positivism: the scientific approach to studying human life. It is a philosophical notion of science, where one assumes that an objective reality exists and can be stud- ied systematically and numerically (e. g. using quantitative data) and without interference from the researcher’s subjectivity. Positivists believe that the methods and methodologies of the natural sciences can be applied to the study of human societies.
Pragmatism: an epistemological position often adopted in mixed methods research. Pragmatism seeks to bypass the debates around positivism and interpretivism by tak- ing a practical stance that focuses on the research question at hand, utilizing the ‘best’ research tools available to provide the ‘best’ outcome.
Praxis: how theory connects to practice, particularly in terms of the research con- tribution to transforming unequal or oppressive social conditions.
Prediction: refers, in Chapter 11, to statements or inferences as to the probability that particular events will take place in the future, such as statements as to the most likely locations of future crime or statements as to the probability that a particular individual will commit, or be the victim of, an offence.
Predictive policing: a term applied to a range of policing practices linked by their claimed ability to ‘forecast where and when the next crime or series of crimes will take place’ (Uchida, 2014: 3871) in order to ‘change outcomes’ (Beck and McCue, 2009). Predictive policing is a prediction-led business process consisting of a cycle of activities and decision points: data collection, analysis, police operations, criminal response and back to data collection (Perry et al., 2013: 128).
Prevalence rates: prevalence rates for crime describe the proportion of a population at risk that has experienced one or more incidents. It is normally expressed as a per- centage. The denominator refers to the population at risk, which depends on the crime in question.
Primary data this refers to the data that is derived from a form of inquiry and analysis whereby an investigator collects the data at first hand.
Probability sampling: a method by which all elements of a given population have a chance of being included in the sample studied. Unlike a purposive sample, a proba- bilistic sample uses random selection so as to generate, as far as possible, a sample that is representative of the broader population. Using probability sampling enables researchers to generalize from their findings, thereby deriving conclusions about a far larger group of people to which the sample belongs.
Process analysis: assessment of progressive changes over time that may include actions directed at a particular result or impact.
Prosopography: collecting key characteristics of the lives of a group of people (e. g. doctors, prisoners, railway train drivers; it can be any group) and aggregating them together to form a group ‘pattern’ of their lives.
Prospective design: a research design consisting of repeated observations of the same group of individuals through time, as they age. Prospective essentially means ‘looking forward’.
Purposive sampling (also known as nonprobability sampling): organized around a non-randomly derived subset of a population. Convenience sampling and Chain referral sampling are commonplace instances of such an approach. Given its non- random nature, a purposively derived sample cannot be taken as representative of the wider population from which it has been drawn.
Qualitative: qualitative methodology is concerned with exploring the behaviour, opin- ions or perspectives, feelings and experiences of people as individuals or groups. It lies in the interpretive approach to social reality: reality does not exist independently of people, but rather is socially constructed and meaning is developed through experience.
Qualitative research: research that investigates aspects of social life which are not amenable to quantitative measurement. Associated with a variety of theoretical per- spectives, qualitative research uses a range of methods to focus on meanings and interpretations of social phenomena and social processes in the particular contexts in which they occur. Qualitative data includes words, sounds and images.
Quality assurance: ensuring that what is done is of the highest standard through constructive critical review.
Quantitative: quantitative methodology is concerned with measuring or test- ing existing ideas or theories (hypotheses) and generally assumes that reality
exists independently of human construction and experience. It is associated with
positivism.
Quantitative research: involves the collection of data in numerical form for quan- titative analysis. The numerical data can be durations, scores, counts of incidents, ratings or scales. Quantitative data can be collected in either controlled or naturalis- tic environments, in laboratories or field studies, from special populations or from samples of the general population. The defining factor is that numbers result from the process, whether the initial data collection produced numerical values, or whether non-numerical values were subsequently converted to numbers as part of the analysis process, as in content analysis (Garwood, 2006).
Questionnaire: a structured set of questions designed to gather comparable information from research respondents. Most of the questions will have a fixed choice of answers.
Reflexivity: a broad concept but essentially encompasses a process whereby the researcher actively engages in self-assessment at all stages of the research process with a view to acknowledging and, where possible, addressing any biases or other short- comings in their work, especially those deriving from culturally held assumptions. It encourages researchers’ recognition and articulation of the research process and their own relationship to the research (including to participants, where relevant), and it offers an ongoing reflection on the development of understanding in the research process. Often, it is identified as a point of distinction from ‘hard’ science where the research is presumed to be objective; however, there are ongoing debates in this area on the meaning and relevance of reflexivity.
Repeat crime rates: the number of crimes experienced by each victim on average.
Research design: a detailed plan of the procedures for data collection and analysis.
Research ethics: institutionalized means of ensuring that research is carried out in ways that respect the rights of all researchers and participants. The three main issues associated with ethnographic research and governed by research ethics are deception, safety and confidentiality.
Research objectives: the purposes for which the research is being carried out. They can be basic – couched in terms of exploration, description, understanding, explana- tion or prediction – or they can be of an applied nature, for example to change, to evaluate or to assess social impacts.
Research proposal: a written document which describes the proposed research, including what it aims to do, how it will be undertaken and the anticipated outcome(s). A proposal also outlines why the proposed research is important and justifies the research design, including how it connects the research questions to the data.
Research question: the overarching question that defines the scope, scale and con- duct of a research project. It is an initial statement of the territory to be examined in the research inquiry.
Retrospective design: meaning ‘looking backward’, a research design in which a sample is selected based on some criteria. Having done so, the researcher traces their backgrounds and past experiences through some empirical material (such as register data or life history interviews).
Sampling there are different ways of and approaches to constructing samples in social and criminological research. Sometimes it is not possible, simple or neces- sary to generate a statistically representative sample (e. g. of all young adults living in a particular locality). Convenience sampling refers to the process of constructing a sample in a way that is convenient to the researcher, given lack of resources such as time (e. g. interviewing relatively easily accessible young adults in a particular locality, such as those attending local colleges). Theoretical sam- pling refers to the way that a sample is generated that reflects the theoretical concerns and research questions of the study. Random samples are intended to be representative of the population from which they are drawn and therefore the characteristics of the sample are assumed to be generalisable across the whole population. Members of the sample are drawn randomly using a sampling frame and everyone in the population stands the same chance of being included in the sample. Probability sampling is another name given to random sampling. Purposive samples are not intended to be generalisable to the population as a whole. Instead, they are used to focus on specific groups or categories and select units based on pre-defined characteristics (such as gender or victim status). They are particularly useful for populations that are hard to reach. Non probability sampling is another name given to purposive sampling.
Sampling error: essentially refers to false findings from statistical data caused by the use of a sample that is unrepresentative of the full population under study.
Secondary research/analysis is based upon existing sources of information which has been collected by someone other than the researcher and with some purpose other than the current research problem in mind. It can be distinguished from pri- mary research and analysis whereby an investigator collects the data at first hand. Examples of secondary sources include Census data and the Recorded Crime Statistics.
Secondary data: data that has already been collected by someone other than the researcher who is presently analysing it. Such data may well have been collected for a purpose other than that for which it is now being examined or re-used. Commonplace and widely used examples of secondary data include official statistics, large-scale surveys and census returns.
Self-inflicted deaths (SIDs): refer to people who have taken their own life and is a term now used by the prison service to describe all non-natural prisoner deaths. Although the injury or harmful activity that resulted in the death of the prisoner was inflicted by their own hands, the person who died may not necessarily have intended
to take their own life. The harmful activity may have been a cry of pain or a cry for help rather than an attempt to kill themselves.
Semiotics: the discipline and study of signs.
Semi-structured interviews: a popular form of interview in the social sciences, which requires the development of an interview schedule. This is not necessarily specified questions but key areas to address in an interview, which can be raised in any order. It is semi-structured because the loose structure allows the interviewer to explore new ideas and diversions or issues that arise in the course of the interview in depth, whilst also ensuring that they consistently address the questions outlined in the schedule.
Sensitive research: used to describe taboo or difficult topics, for example victimiza- tion or deviant behaviour. The term may be used where research elicits the views of vulnerable or ‘powerless’ groups such as children.
|
|
|
