RESEARCH TERMINOLOGY
Abscissa: horizontal line in a bar chart or X axis
Absolutism: a researcher may never tell a lie
Absolute zero: lowest possible point
Abbreviations: shorted a word or a phrase. Includes jargon, slang.
Acronyms: a word made up of the initial letters of other words
Accessible population: or study population is the population that the researcher can reach
Agents of action: people or institutions that take active steps to implement policies or plans
Ambiguous: vague and imprecise questions that lead to vague and imprecise responses
Analysis technique: process in which certain procedures, are used to work through the data that have been collected. (e.g. descriptive statistic technique)
Analysis triangulation: use of 2 or more analytical techniques to analyse the same set of data
Analytic technique: phenomenological analysis that focuses on private experiences & subjective perceptive
Anonymity: not known by name
Antinomianism: a researcher needs to observe moral laws
Assumption: a basic premise that we believe is true
Asymmetrical distribution: positively skewed: larger frequencies concentrated towards the lower end and negatively skewed: larger frequencies concentrated toward the high end of the variable
Attrition loss of participants over time
Attribute: example women and men are attributes of the concept “sex”
Assumption: a basic premise that we believe is true
Audience ethnography: describes how audiences experience radio or television programmes in the social context of everyday life
Authority: accept something as being true because of the status of the person who discovered this truth
Authenticity: genuineness
Bar chart: frequency distribution of categorical data represented in a graph
Bias: Prejudice or distortion
Bias in selection: due to faulty technique, may or may not be deliberate
Bipolar: having two poles or extremes.
Calculate % for distribution: frequency divided by number of cases multiply with 100 = (round off)
Case-study research: or one-shot case-study; in-depth investigation into various characteristics of a small number of cases over a specific period of time
Causality: one thing leads to another: a positive relationship is a causal association between two variables where, if one increases, the other one also increases; statement linking cause & effect
Chance factors: one element included rather than another-can be calculated statistically
Classification: arrange in groups; assign persons & behaviours to particular categories
Closed-ended question: fixed number of answers, used in limited way. E.g. inventory question. Also called standardised, fixed-alternative and highly structured questions
Cluster sampling: population divided in heterogeneous clusters
Code of conduct: rule/s or law/s that regulate behaviour
Codes: are tags or labels assigned to meaning units. shorthand method or way of identifying the theme/category in a transcription; typically takes the form of strings of letter and/or symbols; coding and analysis is not synonyms; coding is a crucial aspect of analysis
Coercion: forcing
Cognitive: knowing or mentally perceiving
Cohort study: same as panel study. Observational study. Compare with cross-sectional study. Group of people who share common characteristics
Coherent: unity of thought or purpose
Colonialism: the practice of gaining control over other countries & occupying them with settlers
Common sense: ordinary reasoning; intuition
Communalism: sharing scientific knowledge with other; research described in detail & scrutinized by other researchers.
Competency: legal capacity to deal with a matter
Complex question: lengthy, obscure question or statement causing confusion about what is being asked.
Concealed measurement: hidden recording devices used in non-reactive observation
Conceptualisation: the process of fully specifying and defining concepts used in a study.
Concurrent validity: is estimated by administering the attitude scale & criterion measure at approximately the same time
Confidentiality: undertaking not to publicly link specific response or behaviour to research participant
Confounding or nuisance or extraneous variables are variants that confuse the interpretation of research. Variables not considered but could influence the result of the study.
Consent: give permission
Consistency: giving consistent results; results that remain the same if circumstances change, the measuring instrument should reflect this
Constant comparison method: qualitative data analysis, code of a meaning unit that relates to a feeling experienced or an emotional reaction
Construct validity: focuses on WHAT is being measured. It is concerned with the nature of reality & the nature of properties being measured; estimation of the extent to which subjects posses the characteristics presumed to be reflected by a particular scale or test.
Contemporary: modern in style;
Content validity: to what degree the content of items in the measuring scale corresponds with the content of the domain being measured.
Context: the total situation in which an event exists or occurs
Contingency question: applies to some respondents only; includes clear instructions
Control group: group not receiving treatment
Convenience sampling: nonprobability sampling; accidental, selects only elements that’s accessible
Conventions: a general agreement regarding certain rules
Cooking: researcher retains or reports only findings that fit hypothesis or research question.
Correlate: association between 2 or more variables, determined statistically; measuring the match between on thing & another
Correlation coefficient: index of the extent of the linear relationship between two variables
Correlational research: systematic investigation of relationship between 2 or more variable to explain the nature of the relationship in the world and NOT to examine cause and effect.
Corroborated: confirmed
Credibility: having a believable status
Criterion: principle or standard according to which something is judged
Criterion validity: collective term used for predictive and concurrent validity
Critical approach: emphasises the need to uncover hidden processes & structures within society
Cross-sectional: observation of some subset of a population of items all at the same time
Cross-sectional research: all the information on a specific topic is collected at the same time and no identical project will be done after a specific period of time
Cumulative frequency (cf): number of scores below or above a certain value; highest class interval is equal to the total number of cases
Data: information collected
Data condensation/Data distillation: eventual outcome of a qualitative analysis: the body of data becomes smaller & more manageable because there is less to deal with; is the result of interpretation & organisation. See de-contextualisation and re-contextualising
Data saturation: when multiple interviews have been conducted and/or observations made & patterns & themes start recurring or no new information emerges, the data achieved saturation
Data triangulation: use of more than one data source
Deception: to be false, misleading or deceiving
De-contextualise: to remove from a context; consider something separate from its context
Deduction implies going from abstract ideas or principles to concrete and specific manifestationsFrom general principles to specific
Dependent variable: variable that is influenced or changed; affect the independent variable
Depersonalisation: losing one’s sense of identity
Descriptive research: gives specific details of a situation, social environment or relationship
Descriptive statistics: mathematical techniques used to see underlying patterns of data; used to organise, summarise and visualise the data.
Design coherence: decisions about research should fit together logically
Deviation score: subtract the mean of the sample from each raw score in the sample; this score indicates the extent to which each raw score deviates from the mean
Dimension: magnitude or extend
Dialectic: method of argument - dialogue between two persons or more
Discrete: separate, distinct, isolated
Disinterestedness: neutral, impartial, respective & open to unexpected observations or new ideas
Disk-by-mail survey: self-administered questionnaire filled in on personal computer
Domain: problem, area, field or discipline being researched
Double-barrelled: a questioning that contains 2 parts; a question that is 2 questions in one
Eclectic: selecting what appears to be best in various doctrines, methods or styles; composed of elements drawn from various sources
Ego involvement: when your ego stands in they way of being objective, causes errors in human inquiry
Element: the unit or case from a defined population
Eligibility criteria: same as inclusion criteria or distinguishing descriptors-specific stipulation of the criteria to be included in the population
Epistemology: how we can get to know social reality / nature of knowledge
Equivalence: equal in meaning
Equivalent-forms method: or alternate-forms method, involves compiling 2 forms or measures that contain items which measure the same phenomenon. Administered to the same group and the 2 sets of scores correlated to estimate level of reliability
Ethics: standard according to which a particular group /community agree to regulate behaviour
Ethnographic research: specific type of participant-observer research. Aim is to describe a particular group’s way of life, from the group’s point of view in its own cultural setting
Evidence: facts indicating whether something is true or not
Exhaustive: includes all possible responses; enough intervals to include all the data
Experimental group: receiving group (treatment)
Explanatory research: aimed at gaining insight into a situation, phenomenon, community or person. Necessary because of a shortage of research for basic information about a new field of interest
Expert-jury validity: to minimise the degree of subjectivity, we use several experts of the particular domain to judge the content of the scale independently
Exploratory research: research in a particular field or topic in social sciences on which no research has been done or of which mention has been made. Used to formulate questions & hypotheses & to clarify ideas
Fact is an empirically verifiable observation
Feminist research: gender sensitivity in research, particularly acknowledging the role of women
Field research: observing events & people in their natural setting, eg a prison, hospital etc.
Filter questions: also called a screening question. Group of related question that apply to some people, not to others, used to exclude respondents from answering irrelevant questions
First-level coding: or open coding is a combination of identifying meaning units, fitting them into themes/categories and assigning codes to the themes categories
Fixed-alternative question: same as closed-ended question, consists of two or more responses
Flexibility: how adaptable the research process is
Focus group interview: semi-structured or totally unstructured, guided by discussion with leader or moderator. Group interview, consist of 6 – 12 people interviewed together at the same time.
Follow-up questions: open question, unique to interviews; follows from respondent’s answers to a closed-ended or open question
Follow-up study: longitudinal study; similar to panel study, usually undertaken to determine the subsequent development of individuals who have a specific condition or received specific intervention
Free-narration questions: open question; respondent tells his or her own story
Frequency distribution: table or graph indicating how observation are distributed
Frequency of score or category: number of cases; indicated with symbol ƒ
Frequency polygon: graph in which the frequencies of class intervals are connected by straight lines
Gratifications: rewards, pleasure, enjoyment, satisfaction
Group administration: questionnaires administered directly to the sampled individuals as a group, during one session
Grouped frequency table: frequency distribution table with a limited number of categories: used especially in case of interval & ration measurements
Hawthorne effect: effect on participants knowing they are being researched
Heterogeneous: dissimilar
Highly structured questions: same as closed-ended question. Consist of fixed number of responses.
Histogram: graph representing the frequency distribution of successive scores or class intervals
History: uncontrolled events that influence the outcome of the research
Homogeneous: similar characteristics
Honesty: general cultural norm; science demand honesty in all research; dishonesty or cheating is a major taboo
Horizontal line: bar chart its the X axis or abscissa
Hypothesis: a tentative statement about relationships between phenomena; can be tested
Ideal type: a mental construct based on ideas of what it entails & exaggerates essential characteristics: is a mental construct; has to do with ideas rather than ideals; exaggerates essential characteristics.
Identity: people’s understanding of who they are, what they are like: their self concept
Impartiality: not favouring one idea or person above another
Inaccurate observation: carelessness about observing everyday events
Independent variable: the variable that influences or changes the dependant variable. The CAUSE. The intervention or the treatment. Researcher can control the independent variable/s.
In-depth interview: intensive, unstructured, conversational, ethnographic & focused interviews
Indicator: an observable measure of a variable or concept
Induction: implies moving from concrete and specific finding towards abstract and general principles; Use number of specific observations to formulate general principles.
Inductive process: moving from observations or data towards generalisations, hypotheses or theory
Inferences: making deductions based on, for example, what is being observed
Inferential statistics: method used to make inferences about the attributes of the population
Instrumentation: threat to validity of a study if the measuring instrument deteriorates or improves over time
Integrity: Honesty & trustworthiness
Intercoder reliability: “the level of agreement” examines the extent to which different interviewers, observers, or coders use the same instrument or measure to get equivalent results
Internal validity: extent to which confirms the existence of cause-effect relationship
Interpretation: a productive process that sets forth the multiple meaning of an event, object, experience or test. It is transformation; it illuminates, throws light on experience; brings out; refines; clarifies the meaning that can be sifted from text, object or slice of experience
Interpretivism: emphasises the importance of insiders’ viewpoint to understanding social realities
Interrupted time-series design: researcher studies a single group but outcomes are measured at regular time intervals before the treatment is introduced. Results are then measured at time intervals after the treatment
Interval: the difference between two points on a scale
Interval level of measurement: interval scale measure the interval or distance between 2 points on a quantitative instrument. Categories used are the same as nominal & ordinal levels of measurement. This scale does not have absolute zero point.
Intervening variable is a variable mediating between the initial cause and the final effect. The effect of one variable and the cause of another variable
Interview guide: or interview schedule, used in structured or semi-structured interviews. Help the researcher about what to ask, in what sequence, how to pose questions & follow-ups
Inventory question: closed-ended question, respondent not limited to selection only one option. Purpose is to get comprehensive overview of all possible options that could apply to the respondent.
Justify: provide reasonable grounds
Justice: to treat fairly
Kinesics: use of facial expressions, gestures & bodily position as nonverbal communication
Kurtosis: flatness or peakedness of the distribution. Symmetrical bell-shape distribution is a normal distribution = mesokurtic. A more peaked distribution is a leptokurtic; a flatter distribution is a platykurtic
Legitimacy: acceptance of the rightfulness & responsibleness of research credibility; having a believable status
Levels of independent variable: value or category of the independent variable determine the treatment condition, eg independent variable is counselling, one level gets counselling, another not.
Linguistic insulation: separating oneself from others by means of language
Longitudinal: repeated observations of the same items over a long period of time, often decades
Longitudinal research: done over an extended period of time.
Leading question: loaded questions lead to leading questions or statements & pressurise the respondent to agree or disagree
Leptokurtic: a more peaked distribution (kurtosis)
Likert scale: attitude scaling / summated scale. A rating scale in which the respondent indicates the extend to which he agrees or disagrees: “strongly agree, Agree, Neutral, Disagree, Strongly disagree”
Loaded language: phrases that contain misleading implications. Conveying a position for or against a topic.
Macro level behaviour of individual linked to social structures & processes on a large scale
Maturation: changes within the subjects themselves over time, impacts on internal validity
Mean: is the sum of a sample of scores divided by the number of scores in the sample; arithmetic midpoint of the scores and represents all the scores in the sample
Meaning unit: segments or chunks of information that are the building blocks of a classification scheme
Measurement: giving data numerical values
Measures of central tendency: the score that is central to a distribution and can be used to summarise the entire distribution
Measuring instrument: observers, questionaires, interview, analysis of existing documents, any standard method of data collection
Median: is the value which falls right in the middle of the list; value or score such that half the observations fall above it and half below it
Mesokurtic: symmetrical bell-shaped distribution or a normal distribution (kurtosis)
Methodological triangulation: use of multiple methods to study a single topic eg combine quantitative and qualitative methods in a single study
Mode: is the score value with the highest frequency
Modernity: belief in rationality & progress associated with confidence in science & technology
Mortality: subjects that do not continue through out a study but drop out or die before it is completed
Multiple-choice questions: respondent selects one of the options from those given. Often used to collect demographic data.
Multiple methods: triangulation; using two or more methods simultaneously
Multistage sampling: stages of cluster sampling, eg cities, then residential blocks, then individuals
Mutually exclusive: choosing one category automatically excludes other categories; a case cannot be classified in more than one category
Mysticism: (religion) knowledge based upon the authority of sacred texts or some supernatural force
Narrative: the story one tells about oneself
Negative relationship: changes in different directions
Negative items: eg “not” , leads to misleading answers cause easy to miss the negative term.
Nominal level of measurement: Discrete, mutually exclusive & exhaustive; distinct & seperate, all possible responses are included and it can not be measured by other categories
Nonexperimental research (quantitative): study of the difference between existing groups that have been classified according to fixed levels of an independent variable. Eg survey, ethnographic
Nonparticipant observation: the researcher fulfils the role as observer & is entirely removed from the social interaction he is observing.
Nonprobability sample: relates to the size of the population
Nonprobability sampling: procedure where we do not know if we included each element of population
Non-reactive observation: observation that does not influence the observed behaviour
Non-response error: element of the population does not respond to a measurement instrument
Non-zero chance: equal chance of selection
Norms: rules of conduct
Number of cases: frequency of score or category; uses symbol ƒ
One logical answer: words like “can” or “might” creates questions with one logical answer. Yes or no.
Objectivity: refer to something abstract; for positivists this means impartiality
Ontology: the nature of social reality
Overgeneralisation: arriving at a general conclusion about a particular thing when we have only observed a few cases of that thing
Overtly: done in an open and unconcealed manner
Open questions: encourages respondents to freely express ideas in their own words. 3 types of open questions - free-narration, follow-up an role-playing questions
Operationalisation is the process of making abstract concepts empirically measurable
Ordinal level of measurement: used when researcher wants to put observations or data in certain order of importance. Can include categories of nominal measurement BUT also allows data to be ranked.
Ordinate: Y axis in a bar chart; vertical line
Paired-comparison question: questions phrased in a way to make respondent think about the options and to compare their meanings; selects only one option
Panel studies: longitudinal research; the same people, group or organisation is investigated again over various periods of time
Paradigm: example or model of how things should be done.
Paralanguage: the vocal (not verbal) characteristics, such as the rate of speaking, the volume & rhythm & vocalisations such as “uh-uh”, “mmmmm” or “shhhh”
Parameter: characteristics of the elements of a population
Parsimony: using no more than is strictly needed
Partial observation: or partial concealment, the status of the researcher is known to the group but the researcher conceals what or who is being observed
Participant observation: or fieldwork or field observation. The researcher, as observer) joins & becomes involved with the group being observed
Percentage: divide the frequency (f) by the total number of cases (n) and then multiply by 100 (100% represents the whole sample)
Phenomena: an event or experience
Phenomenology: aims at understanding & interpreting essence of the meaning subjects give to their lives.
Placebo: medicine given for psychological & not physical effects
Plagiarism: pretend that the thoughts, writings & inventions of others are one’s own
Platykurtic: a flatter distribution (kurtosis)
Plausible rival hypothesis: alternative explanations for cause>effect relationship-threatens internal validity
Population: the entire group of persons or objects & events of interest to the researcher. Also called the target population or universe
Positive relationship: both variables change in the same direction
Positivism: systematic way of doing research, emphasises importance of observable facts
Post-experimental: after an experiment has been conducted
Postmodern: no absolute “truth”, no theory or method is better than another to determine truth
Postulation declaration of something self evident; assumed as basis for argument
Prediction: statement that tells us of a future outcome
Predictive validity: ability of an instrument’ to measure and predict, eg IQ test is valid if it accurately predicts which scholars will perform well
Pre-experimental design: may or may not involve manipulation of treatment conditions but they do not include randomisation; only one research group & subjects not assigned on a random basis if two or more groups are used.
Primary data: information collected at the original source, eg medical records, testimonials, annual report
Privacy: confidentiality
Probability: chance
Probability sampling: random sampling, every element in population has equal chance of being selected
Probing question: follow-up question used to elicit more information. When? Where? What? Why? How?
Properties: qualities, characteristics or attitudes
Proposition: an abstract statement of relationship between phenomena
Prospective study: watches for outcomes, such as development of disease, during study period
Proxemics: study of spatial communication & specifically distances between 2 or more people – intimate, personal, social and public distance
Qualitative research: Naturalism, interpretivism and humanism; broad, subjective and holistic
Qualitative data analysis: a non-numerical process of examining & interpreting data in order to elicit meaning, gain understanding & develop empirical knowledge; it’s like building a puzzle
Quantitative: values expressed in numbers
Quantitative research: logical positivism; cause-effect relationship; systematic & standardised
Quasi experiment : has some but not all the characteristics of an experiment-almost a true experiment
Quasi experimental design: resembles true experimental designs in that the researcher aims to establish a cause-effect relationship BUT does not include randomisation
Quota sampling: nonprobability sampling, similar to stratified sampling except NO final random selection
Random assignment: every subject has equal chance of being selected for treatment. NOT same as random selection.
Random-digit-dialing: used in survey research, use telephone to draw a sample
Random error: haphazard errors due to different factors, may cancel each other out after a number of cases/measurements. For example, respondent ill, equipment malfunction etc.
Random selection: each element in population has equal chance of being selected for the sample.
Range: the difference between the highest & lowest scores; a measure of variability of scores in a sample, it indicates the range of the distribution of scores from the lowest to the highest
Rank: arrange in a hierarchy
Ranking question: closed-ended question. Respondent ranks order of preferences from least to most
Ratio level of measurement: the highest level of measurement. Includes ALL characteristics of nominal, ordinal & interval levels of measurement BUT has an absolute zero point
Raw data: the scores of research findings
Reactive observation: observation that influences the behaviour being observed
Re-contextualise: extract texts, signs or meanings from original context (de-contextualise) and introduce it in another context; implies a change of meaning; the dynamic transfer-and-retransformation of something from one discourse/text-in-context to another
Regression: tendency of extreme scores to move towards the average upon retesting
Relationships: links
Relativism: belief that claims all different viewpoints are valid
Reliability: when identical investigations are repeated, similar research results are obtained; degree to which a scale yields consistent results or scores.
Replication: repetition of a test or experiment for control and verification purposes; reproduction of research by other researchers
Representative sample: a sample that resembles the population & which enables the researcher to accurately generalise the results-resembles the population in as many ways as possible
Research: making a careful and planned study of (social) events
Researcher triangulation: use more than one researcher in a single study to achieve inter-subjective agreement
Response bias: errors due to individual responses not being truthful
Response rate: number of questionnaires returned
Response set: respondents fall into the pattern of answering questions in a particular way
Retrospective: look back at events that already took place
Role-playing question: open question where respondent play particular role & respond from this perspective.
Sample: a subset of a population or a selected group of elements from a defined population
Sampling approach: sampling procedure, sampling method or sampling plan
Sampling bias: when there’s a difference between sample data & population data that can be attributed to an incorrect selection process; a threat to external validity of a study when subject not randomly selected; not representative of the population
Sampling error: differences between population parameters & sampling statistics
Sampling frame: comprehensive list of all units (elements) from which the sample is drawn. Population list.
Science: process of inquiry, a way of learning & knowing things about the world, using logic, observation and theory
Scientific research: systematic examination and thinking about a question
Scope: range of situations, techniques, and observations included in research
Scope of theory : degree of abstraction & generalisability. Degree to which studies can be generalised to other settings/people through an in-depth study or observation, analysis & interviews
Secondary data: information taken from a source that’s not involved in the original investigation
Selective observation: overgeneralistaion may lead to selective observation. Paying attention mainly to future situations that correspond with the way you see things
Self-administered questionnaire: respondents fill in a questionnaire without researcher’s assistance
Semantic differential scale: used to measure attitudes. A collection of 7 point rating scales & scale points on each end defined by opposing adjectives. Eg. Powerful ..... Powerless/hostile .... friendly. Also called bipolar ratings
Simple random: most basic probability sampling technique. Lottery or fishbowl or random number table
Single-case experiments: similar to time-series designs but involves only one sampling unit, often a single subject (human or animal)
Situationism: situated between 2 extremes. Certain ethical, rules & requirements acknowledged/accepted.
Skewness: the symmetry or asymmetry of the distribution-distributions differ in skewness.
Smoothed polygon: midpoints are linked by curved lines
Split-half method: the scale is split into 2 halves, each half scored separately.
Standardised question: same as closed-ended question, consist of two or more responses
Street ethnography: takes place in setting such as a street, shelter, hostel; aim is to describe the way of life of certain groups of people
Systematic error: bias that occurs constantly & distorts the measurement in particular direction
Size of sample: can data be generalised
Snowball sampling: nonprobability sampling method, starts with few respondents, more identified
Spuriousness implies an apparent causal relationship which is false because a third variable actually causes changes to the two variables thought to be related
Social law: the regularities or patterns in social reality are called social laws.
Social law: if there is substantial evidence confirming a hypothesis.
Stability: the quality of being constant
Stability component: test-retest method is suitable for determining stability component of reliability
Stakeholder: individuals or groups who have an interest in or concern with the research
Standardised: uniform
Standard deviation: index of variability tat is expressed in the same units as the original measures; the square root of the variance of a sample of scores
Stratified random sampling: population divided into different groups/subgroups called strata
Stratifying variables: includes gender, age, socio economic status & educational level
Summative scaling procedure: the numbers for different items are added together
Systematic sampling: interval sampling-draws every ƒth element from population
Structured: a prescribed way of approach
Survey research: systematic collection of specific information by asking sample respondents the same questions at a particular point of time
Symmetrical distribution: a symmetrical distribution can have same shape on both sides of the midpoint
Target population: same as population, also called “universe”
Tersiary data: interpretation based on secondary data
Testing effects: subjects become familiar with the testing procedure or more aware of certain issues after the pretest
Test-restest method: involves correlating one scale to the same group of respondents at different times, with intervals in between that vary from a week to a year
Theory: framework of ideas that provides an explanation of something
Theory: conceptual framework; theoretical propositions suggest relationships between concepts; structures & systematises our ideas on phenomena we study; suggests what we can expect in certain circumstances. This then deductively guides research. Identifies links between phenomena; provides and explanation; is abstract
Theory triangulation: use of multiple theories or perspectives to interpret a single set of data
Tied score: when several scores with the same numerical value occur near the median
Time-series designs: the researcher can observe the trend of the data before and after treatment
Tracer study: methodological middle-range approach; investigation to trace something that’s missing
Tradition: inherited body of information
Training in skills: knowing how to ask questions, how to listen, write notes & need to ask probing questions
Trend designs: investigation into changes in the general population in relation to a particular phenomenon
Trimming: when researcher smooths out irregularities in data to achieve better fit between actual & expected results.
True experimental design: research studies in which the researcher manipulates the treatment conditions; researcher decides who receives which treatment
Ungrouped frequency distribution tables: or tally; show number of times each score occurred within a range of scores; disadvantage is that it still contains large number of scores
Unit of analysis: element
Universalism: judging research on basis of scientific merit, irrespective of who conducted the study
Universe: the target population or population
Unobtrusive: not making oneself noticed
Unwarranted assumptions: not all people subscribe to same ideas about things. Avoid this in questionnaire by asking filter and contingency questions
Validity: methodological requirement for research methods as well as research results.
Value: can be used to mean measurement of amount quantity or degree
Value free: scientific enquiry should not be influenced by values of scientist
Value freedom: key tenet of positivism. Detachment & neutrality
Variability of scores: the degree to which scores in a sample differ; how spread out they are
Variable is an empirical phenomenon, can take on more than one value. Property that changes empirically
Variance: measure of variability based on the deviation of each score in a distribution from the mean of that distribution
Vertical line: in a bar chart is the Y axis or the ordinate
RESEARCH IN SOCIAL SCIENCE
Structure and outline
Sources of knowledgeo Traditiono Authorityo Mysticism and religiono Common senseo Media myths
Errors in human inquiry and how scientists try to avoid themo Inaccurate observationo Overgeneralisationo Selective observationo Ego involvement in understandingo Premature closure of inquiry
Norms of the scientific communityo Universalismo Organised scepticismo Disinterestednesso Communicationo Honesty
Stages in researcho Stage 1: define the problem
Research design or plan Research problem Literary review Theory Assumptions Hypothesis Research questions
o Stage 2: obtain information Sampling Data collection
o Stage 3: analyse and interpret information Qualitative data Quantitative data
o Stage 4: communicate results Report writing
STRATEGIES OF DISCOVERY
Questions about the nature of the world and the nature of knowledge about the worldo Assumptions
Ontology Epistemology methodology
o Dominant approaches in research Positivist Interpretivist Critical
o Contemporary methodological challenges Feminism Post modern Postcolonialism
THE ROLE OF THEORY IN RESEARCH
Theoretical grounding of researcho Defining theoryo Theory as a conceptual frameworko Scope and level of abstraction of a theory
Range Attribution theory Psycho analysis & social constructionism Theoretical frameworks
Empirical generalisation Middle range theory Theoretical frameworks / perspectives
Micro level Macro level
o Reconsidering the relationship between theory and research Deduction Induction
The conceptual grounding of researcho Conceptso Explanation
Causal explanations Interpretive explanations
o Deciding on a topic to research
Social significance Scientific significance Broad factors influencing choice of topic
Personal interest Need to solve a problem Development of theory
o Conducting a literature review Journals, books, dissertations, research reports & conference papers Government documents, policy papers, newspaper articles Internet Libraries
o Specifying a research question Quantitative Qualitative
o Framing a hypothesis Dependable variables Independent variables Intervening variables
o Operationalising concepts conceptualising
Abstract ideas Indicator
SCIENTIFIC RESEARCH
1. Classification o Categorieso Methodologyo Aims o Time dimension
2. Types of research (Categories) Basic research Applied research
o Actiono Social impact studyo Evaluation
Evaluational research Formative evaluation Summative evaluation
3. Approaches that form the basis of research Quantitative research Qualitative research Multiple methods (triangulation)
4. Aims of research Exploratory research Descriptive research Explanatory research Correlational research
5. Time dimension of research Cross sectional Longitudinal
o Panel studieso Follow-up studieso Cohort studieso Time series studies (Tracer studies)
Trend designs Case study research
ETHICS IN RESEARCH
1. The meaning of research Situational factors
oMethods of collectionoPurpose or goal oPersonal motivesoConsequences
Regulated by behavioural normso Universalismo Communalityo Disinterestednesso Organised scepticismo Honestyo Respect
2. Ethical issues in research relating to human rights Do no harm
Obtain consent Ensuring privacy
3. Ethical issues relating to different research methods Quantitative research Qualitative research
4. Ethical issues relating to the researcher Absolutism vs antinomianism Situationism Integrity > the core: trustworthiness & honesty Ethical issues relating to various stages
oChoice of topico Samplingo Selection of appropriate methodo Treatment of respondentso Treatment of dataoDevelopment & application of measuring instruments
Internal validity External validity Reliability
oPolitical & organisational constraintsoPersonal constraintso Interpretation of data & report findings
RESEARCH DESIGN
1. Research design as part of the research process2. Design coherence3. Design validity in quatitative research
o Internal validity History Maturation Instrumentation Selection Mortality Regression Testing effect
o External validity
4. Research design involving quantitative datao True experimental design
Characteristics Manipulation of the treatment condition Random assignment
True experimental designs Posttest only control group design Pretest-posttest control group design
o Pre-experimental design Characteristics Pre-experimental designs
One-group posttest-only design or one-shot case study design One-group pretest-posttest design
o Quasi-experimental design Characteristics Quasi-experimental designs
Time-series designs Interrupted time-series design Single-case experiments
o Nonexperimental research involving quantitative data Survey research
Cross-sectional surveys Longitudinal study
5. Qualitative research designso Key issues
Boundaries / scope of the research Design flexibility Research relationship Intended outcome Applying principles of qualitative designs
Description Scope Flexibility Relationship Outcome
SAMPLING
1. Basic sampling concepts Population Element Parameter Sample Sampling frame A representative sample Sampling error
o Chance factorso Bias in selectiono Non-response error
Sampling bias2. The essentials of sampling
Homogeneous Heterogeneous
3. Sampling approaches Probability or random sampling
o Simple random samplingo Systematic samplingo Stratified random sampling
Proportional stratified samples Disproportional stratified sampling
o Cluster or multi-stage sampling Nonprobability sampling
o Convenience (or accidental) samplingo Quota samplingo Purposive or judgmental samplingo Snowball sampling
4. Sample size
DATA COLLECTION
1. Measurement Levels of measurement
o Nominal level of measuremento Ordinal level of measuremento Interval level of measuremento Ratio level of measurement
Measurement errors
2. Using existing data Three advantages to collect and use existing data
o Unobtrusiveo Practicalo Accuracy confirmed
Sources of existing datao Primary datao Secondary datao Tertiary data
Authenticity and credibility of existing datao 8 questions to consider authenticity of persono 4 question to consider authenticity of actual data or evento 8 questions to consider credibility of sources
3. Surveys versus field research Systematic collection of specific information Field research
4. Interviews Structure of interviews
o Highly structuredo Semi-structuredo Unstructured or completely open
Types of interviewso In-depth interviewso Focus group interviews
Focus group interviews combined with other data collection methodso Types of questions asked in interviews
Closed-ended questions Open questions
Free-narration questions Role-playing question Follow-up questions
o Conducting an interviewo Advantages and limitations of conducting interviews
Advantages of interviews Limitations of both face-to-face and telephone interviews Limitations of specifically telephone interviews
5. Questionnaires (in survey and field research) The wording of statements and questions
o Items or questions to avoid Double-barrelled Loaded language Leading questions One logical answer Negative term Incomplete question Ambiguous question: agents of action Lengthy and complex questions Unwarranted assumptions Abbreviations and acronyms
o Types of items included in questionnaires Paired-comparison question Contingency questions Ranking questions Inventory questions Multiple-choice questions Scales that measure variables
Likert scales Guttman and Thurstone scales Semantic differential scales
Ordering of questionnaire items Properties of scales
o The requirement of reliabilityo The test-retest methodo Split-half methodo Equivalent-forms method or alternate-forms method
Determining the stability componento Intercoder reliabilityo Requirements of validity
Predictive validity Concurrent validity Content validity Construct validity
Conducting a surveyo Broad guidelines when conducting a surveyo Guidelines for using self-administered questionnaires
Advantages and limitations of using questionnaireso Questionnaires distributed by mail or inserted in other publicationso Distribution via internet or electronic mailo Self-administered questionnaireso Group administration
6. Observational techniques (in field research) Five conditions to be met Three factors that influence the researchers’ observations
o Selective attentiono Biaso Intentions
Forms of field observationso Participant observation vs nonparticipant observationo Ethnographic research
audience ethnography street ethnography
o Reactive observations versus nonreactive observations Conducting field observations
o What is observed External or physical signs Kinesics Proxemics
Linguistic insulation Depersonalisation
Language behaviouro How are observations recorded
Classification Operational definitions
Select or devise instruments to observe Advantages and limitation of observational techniques
DESCRIBING AND INTERPRETING QUANTITATIVE DATA
1. Descriptive statistics
2. Tables and graphs Frequency distribution tables
o Frequency distributiono Ungrouped frequency distributiono Grouped frequency distribution
Percentages Graphic presentation of frequency distribution
o Bar chartso Histogramso Polygonso Pie charts
Skewness and kurtosiso Positively skewed (asymmetrical at low end)o Negatively skewed (asymmetrical at high endo Normal (symmetricalo Kurtosis
Leptokurtic – more curved Platykurtic – more peaked Mesokurtic – symmetrical bell shape
3. Measures of central tendency Mode Median Mean
4. Measures of variability Range Variance and standard deviation
5. Relationships
QUALITATIVE DATA ANALYSIS AND INTERPRETATION
1. Commencement of the activity of data analysis Describe qualitative data
6 characteristics in research that yield qualitative datao Social contexto Everyday eventso One/few caseso Grounded theoryo Meaningfulnesso Integrity
2. Recording qualitative data Text-based
o Jotted noteso Direct observation noteso Research inference noteso Analytic noteso Personal notes
3. Analysing qualitative data Label Indexing Grouping concepts Open-coding Axial coding Selective coding Cross-sectional coding Non-cross-sectional coding Step-wise format or plan Step-2ise format or pan (7 steps)
4. Interpreting qualitative data Successive approximation
5. Using computer Word processing packages
Text retriever systems Relational database management systems Code and retriever software Theory and network builders
WRITING AND EVALUATING RESEARCH REPORTS
1. Functions and characteristics of a research report 3 functions
o Communicates data to audienceo Contributes to general body of scientific knowledgeo Stimulates & directs further inquiry
2 requirementso Reporting o Accuracy
Characteristics
2. Six types of reports Written assignments Presentations Journal articles Commissioned research Dissertation and thesis Mass media
3. Organisation of report 4 broad areas to be included in report
o Define the problem The research problem The literature review
Accuracy Relevance
o Obtain information Data gathering
Sample size Sample characteristics
Method Reliability and validity
o Analyse & interpret information
Results llustrating your findings Description of your analysis Description of findings
Interpretation and discussion Summarise main finding Discuss implications of interpretations Limitations and practical problems Suggestions for further research
o Making known the results The introduction
State the problem Explain the theoretical framework Justify the research Indicate the aim of the research
The conclusion The abstract
4. Technicalities 5 stylistic characteristics of good scientific writing
o Write systematicallyo Write clearlyo Write preciselyo Write directly & simplyo Write concisely
5. Evaluating a report Definition of the problem Obtaining information Analysing and interpreting the information Technicalities