228 terms

Thomas_How to do Your Research Project


Terms in this set (...)

a brief summary (usually 150-350 words) of a research project and its findings.
the freedom or opportunity to approach and communicate with the people, documents, institutions, etc. the researcher is interested in.
a record - written, or audio- or video-recorded - from a respondent.
Action research
inquiry, conducted by a practitioner, or involving collaboration between a practitioner and a researcher, where the principal purpose is the solution of a problem or the development of practice.
the removal of names and other identifying features from research participants.
ANOVA (analysis of variance)
a method of statistical analysis used to determine the significance of differences among the means of two or more groups on a variable.
a relationship between objects or variables.
Attitude scale
a way of measuring attitudes involving a ranking of a respondent's views on a subject. Examples are the Likert scale, Guttman scale and the semantic differential scale.
characteristics of persons or things.
the rate at which participants drop out of a study. If particular types of participant drop out faster than other types, bias may be introduced.
a single value representing the typical or middle value of a set of data. There are three different forms of average: mode, median and mean.
Axial coding
reanalysis of the results of open coding in grounded theory, aimed at identifying the important, general concepts.
a statement, widely accepted, of an established or self-evident truth.
Bar chart
see histogram.
a control measurement carried out before an experimental treatment.
a school of psychology which asserts that the scientific study of people's behaviour should restrict itself to the data obtained from observable events - and not concern itself with phenomena such as thoughts or emotions.
ideas that are accepted as true on grounds wherein supporting evidence is not immediately manifest.
Bell-shaped curve
see normal distribution.
a loss of accuracy arising from poor sampling, design or analysis.
a list of books and articles of interest. Not to be confused with references, which are all of the books and articles referred to in a research report.
Bimodal distribution
a distribution in which two scores are the most frequently occurring scores.
Biographical method
study emphasising the important events of individuals' lives.
short for 'web log': a diary or series of observations kept online and open for others to read.
Case studies
analyses of persons, events, decisions, periods, projects, policies, institutions, countries or other systems which are studied holistically by one or more methods. The case that is the subject of the inquiry will normally illuminate and explicate some broader theoretical theme.
Categorical data
data representing categories (such as gender or marital status) rather than numbers. The categories can be given numerical codes (e.g. male, 1; female, 2), but these cannot be ranked, added, multiplied or measured against each other. Also referred to as 'nominal data'.
Causal model
a model which represents a causal relationship between variables.
the act of giving rise to a phenomenon or a state of affairs.
the collection of data from all members, rather than a sample, of a population.
Central tendency
a measure of central tendency is any way of describing or representing typical, average, or common values in some distribution.
Chi-square test
a statistic used to determine the significance of observed differences between the values of categorical, or non-numeric, variables. Works by comparing an expected proportion or ratio to an observed proportion or ratio. Also written as χ2 ; pronounced as 'kye square'.
a statement which is made in response to the research question and is supported (or rejected) with evidence based on research.
ordering of related phenomena into categories, groups, or systems according to characteristics or attributes.
Closed question
a question in which the respondent is asked to select an answer from among a list provided by the researcher.
in quantitative research: numeric values assigned to levels of the variable in question. In qualitative research: labels or statements that summarise the meaning of elements of the data.
in quantitative research: the process of assigning values, typically numeric values, to the different levels of a variable. In qualitative research: the assigning of labels or statements to elements or aspects of the data.
a group of people, often with a shared demographic profile or shared experience, who may be observed or questioned over time.
Comparative research
The examination of societies (or other social units) in comparison with one another. In educational research, often used to mean cross-country comparison.
Concept mapping
see construct (or theme) mapping.
a state of affairs in which no one except the researcher knows the identities of the participants in a study. It also refers to the treatment of information that a participant has disclosed to the researcher with the expectation that it will not be revealed to others in ways that violate the informed consent agreement.
Confounding variable
a variable that is not of interest, but which distorts the results if the researcher does not control for it in the analysis.
see informed consent.
a value that stays the same for all the units of an analysis.
Constant comparative method
an analytical technique in qualitative research in which observations are compared with one another and with the evolving theory.
something that exists theoretically but is not directly observable. Or a concept developed (constructed) for describing relations among phenomena or for other research purposes.
Construct (or theme) mapping
graphic display of concepts and their interrelationships.
Construct validity
the degree to which a test, questionnaire, or instrument measures the theoretical concept that the researcher hopes to measure.
the idea that reality is socially constructed. It is the view that reality cannot be understood outside the way humans interact: knowledge is constructed out of these interactions.
Content analysis
a procedure for organising narrative, qualitative data into themes and concepts.
Content validity
the extent to which a measurement reflects the specific intended domain of content.
the idea that reality is socially constructed. It is the view that reality cannot be understood outside the way humans interact: knowledge is constructed out of these interactions.
Content analysis
a procedure for organising narrative, qualitative data into themes and concepts.
Content validity
the extent to which a measurement reflects the specific intended domain of content.
Context effects
changes in the dependent variable resulting not directly from changes in the independent variable but rather from the influence of the research environment.
Continuous variable
variable that may have fractional values, e.g. height, weight and time.
a duplicate group, treated identically to the experimental group except for the variable being manipulated (i.e. the independent variable), which remains unmanipulated in the control.
Controlled experiment
controlled experiments must contain a control group, which receives no treatment, alongside the experimental group, which receives 'treatment'. Inferences can then be made about any changes observed in the outcome variable of the experimental group, based on comparison with the control group. See also randomised controlled trial.
Convenience sampling
a sampling strategy that uses the most easily accessible people to participate in a study.
the degree to which two variables are associated.
Correlation coefficient
a measure of the degree to which two variables are related. A correlation coefficient is always between -1 and +1. If the correlation coefficient is between 0 and +1 (e.g. 0.9) then the variables are positively correlated (i.e. 'go together'), and the higher the number, the stronger the relationship. If the correlation coefficient is between 0 and -1 (e.g. -0.3) then the variables are negatively (or 'inversely') correlated (i.e. as the values of one rise, the values of the other tend to fall).
Criterion-related validity
the degree to which a measure relates to some external criterion.
Critical theory
a critical-analytic approach to social science. Its goal is to promote emancipatory forces and to expose ideas and systems that impede them.
Cross-sectional study
a study based on observations made at a single point in time.
a display of the relationship between two categorical variables. A table is created with the values of one variable across the top and the values of the second variable down the side. The number of observations that correspond to each cell of the table are indicated in each of the table cells.
raw information, usually verbal, observational or numerical, used as a basis for reasoning, discussion or calculation.
Data collection
the observation, measurement and recording of information in a research study.
Deductive reasoning
a form of reasoning in which conclusions are arrived at from premises which are assumed to be true.
Dependent variable
the outcome variable. In experimental research, it is expected to depend on a predictor (or independent) variable.
Descriptive statistics
simple statistics (such as bar charts) used to describe and summarise data.
a systematic record of thoughts, feelings, conversations, activities, etc.
Discourse analysis
the study of language in social use, and associated procedures for organising language into themes and concepts.
the spread of a variable's values. Statistics that describe dispersion include range, variance, standard deviation, and skew.
the range of values of a particular variable.
Double blind experiment
a research design where both the experimenter and the subjects are unaware of which is the treatment group and which is the control.
Ecological fallacy
the error of assuming that one can infer something about an individual from data collected about groups.
Ecological validity
the extent to which the findings of a research study can be generalized to real-life settings.
Effect size
a way of quantifying the size of the difference between two groups. It is the standardised mean difference between the two groups, calculated as (Coe, 2002): [(mean of experimental group) - (mean of control group)]/standard deviation.
Emancipatory research
research conducted for the purpose of empowering or otherwise benefiting disadvantaged groups.
Empirical knowledge
knowledge based on observation or direct experience (rather than on authority or tradition, for example).
the study of how we can know: the study of the procedures via which we can discover knowledge.
principles of conduct about what is right and wrong.
study involving a researcher's immersion into a culture or group. The aim is to understand the group/culture 'from the inside' by participating in its practices and routines, and to allow meanings to emerge from the ethnographic encounter rather than imposing explanations from existing theory.
Evaluation research
the use of research to monitor change and to assess its effectiveness.
Expectancy effect
any unconscious or conscious cues that convey to the participant in a study how the researcher wants them to respond.
a systematic procedure conducted to support or refute a hypothesis by demonstrating what outcome occurs when a particular variable is manipulated. Experiments should rely on repeatable procedure and objective analysis of the results.
Experimenter effects
unwanted influences on the recorded data, occasioned by the actions or mere presence of the researcher.
External validity
the degree to which the results of a study can be generalised beyond the study sample to a larger population.
Extraneous variable
a variable that interferes with the relationship between the independent and dependent variables and which therefore needs to be controlled for in some way.
predicting the value of unknown data points by projecting beyond the range of known data points.
Face validity
the extent to which a survey or a test appears to a respondent to measure what the researcher claims it measures.
Factor analysis
in this statistical technique correlations among variables (from responses on an attitude scale, for example) are explored for potentially lower numbers of previously unidentified variables called 'factors', which are taken in some way to be influential on the findings. By inspection of the possible source of these 'factors' the researcher may make inferences about the nature of the factors' influence and label them accordingly.
Field notes
a text document that details behaviours, conversations, or setting characteristics.
data collection in the research environment.
Focus groups
discussion groups of 4-12 participants for examining specific topics. The organiser will give the group focus materials - newspaper articles, pictures, audio clips, video clips, etc. - to help stimulate discussion. Moderators may assist in keeping the discussion flowing.
Frequency distribution
the frequency with which values of a variable occur in a sample or a population.
the extent to which research findings and conclusions from a study conducted on a sample population can be applied to the population at large.
Gini coefficient
a measure of inequality in a group of values (e.g. income distribution in a population). The larger the coefficient the greater the dispersion.
Grounded theory
the development of explanations (or theory) from the analysis of raw qualitative data. The process is characterised by continual reinspections of the data (see constant comparative method) until 'themes' emerge.
Hawthorne effect
a change in people's behaviour during a research study, occasioned merely because an interest is being taken in them.
the degree of dissimilarity among cases with respect to a particular characteristic.
a visual presentation of data that shows the frequencies with which each value of a variable occurs. Each value of a variable typically is displayed along the bottom of a histogram, and a vertical bar is drawn for each value. The height of the bar corresponds to the frequency with which that value occurs. A histogram is also called a 'bar chart'.
the assumption that research is best conducted by looking at the whole situation and its context, rather than at particular variables and the relationships between them.
a statement that predicts a relationship between an independent (causal) and a dependent (outcome) variable.
an approach to studying individual people, institutions or phenomena in detail, for insight and understanding about how they work.
Independent variable
the conditions of an experiment that are manipulated by the researcher.
In-depth interviewing
a research method in which unstructured interviews are conducted using open-ended questions to explore topics in depth.
Inductive method
a term used in qualitative research to describe the process of beginning with specific observations, from which regularities or themes are detected. These regularities or themes lead to the formulation of tentative hypotheses, and ultimately to the development of general conclusions or theories.
Inductive reasoning
a form of reasoning in which a generalised conclusion is formulated from the observation of (usually a large number of) particular instances.
a person who knows about the situation, person, people or other phenomenon under study and who is willing to disclose what he or she knows about it.
Informed consent
the provision of full information about the data-gathering process and the disclosure, reporting, and use of data and results of a research study in such a way that potential participants can make an informed decision about whether they want to participate in the study.
Interaction effect
a situation where the effect of the independent variable on the dependent variable varies depending on the value of another, different variable.
Internal validity
the rigour with which a study is conducted, including considerations about design, the care taken to conduct measurements, and decisions about what is and is not measured. Also refers to the extent to which the designers of a study have taken into account alternative explanations for any causal relationships they infer.
the paradigm wherein the assumption is that the social world is interpreted and constructed differently by each person. Studying the social world therefore depends on researchers understanding these interpretations and constructions.
Inter-rater reliability
a measure of the consistency between the ratings or values assigned to a behaviour that is being rated or observed; usually expressed as a percentage of agreement between two raters/observers, or as a coefficient of agreement which can be stated as a probability.
Interval data
numerical data where the distance between any two adjacent units of measurement is the same.
a research instrument in which one person (an interviewer) asks questions of another person (a respondent).
Interview schedule
a guide for an interviewer conducting a semi-structured interview as to the subjects to be covered during the interview.
Life history
a record - written, or audio- or video-recorded - of an event or events in a respondent's life told from his/her own perspective.
Likert scale
a scale on which survey respondents can indicate their level of agreement or disagreement with statements presented to them.
Linear regression
a statistical technique used in seeking a linear relationship between an outcome variable and potential predictor (independent) variables.
Literature review
an organised review of books, articles and published research on a specific topic to discover themes, gaps, inconsistencies and agreements.
Longitudinal study
the study of a group of individuals or cases over an extended period (involving at least two data collection points) examining those individuals or cases for the effect of time on the variable or variables of interest.
Margin of error
the maximum expected difference between an actual finding in the population and a finding calculated from a sample taken from that population.
Matched samples
two samples in which the members are paired or matched explicitly by the researcher on specific attributes, such as IQ or income.
the procedure whereby pairs of subjects are matched on the basis of their similarities on one or more variables, and one member of the pair is assigned to an experimental group and the other to a control group.
a measure of central tendency. To calculate the mean, all the values of a variable are added and then the sum is divided by the number of values.
A measure of central tendency. The median is the value that lies in the middle of a set of values: 50 per cent of the values lie above the median, and 50 per cent lie below the median.
Member checking
during open-ended interviews, the practice of a researcher restating, summarising, or paraphrasing the information received from a respondent to ensure that what was heard or written down is in fact correct.
the process of writing notes during the analysis of qualitative data. The notes may be about any aspect of analysis, and will usually include initial ideas about associations, themes and theoretical formulations.
a statistical technique that combines and analyses data across multiple studies on a topic.
the study of research method. The word is also used to mean the general strategy for research and the data collection methods and analytical techniques adopted in a particular research study.
systematic approaches to the collection and/or analysis of data.
Mixed methods
using two or more methods in the same study from different research traditions.
a measure of central tendency: the value that occurs most frequently in the data.
Narrative inquiry
a qualitative research approach based on a researcher's narrative account of the subject under study.
Naturalistic inquiry
study of social phenomena in natural settings without experimental manipulation or other forms of deliberate intervention.
Network analysis
a diagram showing the relationships among ideas; particularly useful for showing how subsidiary ideas branch from a central idea (a trunk).
in NVivo, a node is a mid-stage categorisation, broadly equivalent to the axial coding of grounded theory.
Nominal data
see categorical data.
an approach to research in which the aim is to discover laws about human activity via the study of large groups from which generalisations can be drawn.
Non-probability sampling
any sampling technique not drawing on probability theory. There is, in other words, no assumption of representativeness in the sample.
in statistics, the average or usual performance.
Normal curve
the bell-shaped curve that is formed when data with a normal distribution are plotted.
Normal distribution
an arrangement of a dataset in which most values bunch in the middle of the range and the rest diminish symmetrically towards either extreme. On a graph, a normal distribution is bell-shaped.
Null hypothesis
this hypothesis states that there is no difference between groups on a variable of interest. The alternative hypothesis states that there is some real difference between the groups.
a software package that helps you to code data - documents, PDFs, audio, video - and sort these data into themes.
the attempt to see or understand the things of our research interest without personal bias, interpretation or emotion.
the study of 'what there is'. The examination of the nature of the subject that we are researching.
Open coding
in grounded theory, the initial classification of concepts based on the researcher's examination of the data.
Open-ended question
questions for which respondents are asked to provide their own answers.
Ordinal data
numerical data based on rank order along some dimension: first, second, third, etc.
an observation or phenomenon that is very different from most other observations or phenomena.
Panel study
a study in which data are collected from the same set of people (the sample or panel) at different points in time.
a set of ontological and epistemological assumptions which frame and shape the way that research is conducted. Or a set of beliefs about the world we study, which determine the way that we study that world.
Participant observation
a research method involving the researcher taking a full part in the situation being studied and taking full account of the insights gained from that participation - which is to say, not attempting to be a dispassionate, objective observer.
Pearson's correlation coefficient
a statistical measure of the degree to which two variables are associated. Usually denoted by r.
Peer review
A system used by editors of scholarly journals to assess the quality of work submitted for publication. Submissions to a peer review journal will be assessed by two or more experts in a field, and on the basis of these assessments a journal editor will make a judgement about whether a submission is worthy of publication. Publication in a peer review journal is taken to be the most robust assurance of an article's quality.
the value below which a given percentage of observations in a set of observations fall. For example, the 70th percentile is the value below which 70 per cent of the observations may be found.
the study of the way in which individuals structure consciousness, examining the way meanings are given to the things of their experience.
Pilot study
a small-scale study that is conducted prior to the larger, final study. The pilot study gives researchers a chance to identify any problems with, or possible developments to, their proposed sampling, methodology, or data collection.
the entire set of a clearly defined group of people or objects. Samples may be drawn from the population, with the aim of generalising from the sample to the whole population.
an account of the researcher's personal and social background and the context for the research, provided by the researcher so that readers can assess the status of the interpretations being made by the researcher.
the paradigm which asserts that the social world can be studied objectively, largely adopting the assumptions of natural science, and using its methods (compare interpretivism).
the view that there can be no overarching explanatory theories about the social world, and that the intellectual structures that guide our thought about explanation and prediction are invariably misleading.
Predictive validity
the extent to which a score on a scale or test predicts performance on some criterion measure.
Prestige bias
the tendency for respondents to give answers that are socially desirable or acceptable, but which may not be accurate (also called social desirability bias).
the likely occurrence of a particular event. Usually expressed on a scale from 0 to 1; a rare event has a probability close to 0 (e.g. p = 0.001), while a very common event has a probability close to 1 (e.g. p = 0.9).
a non-directive phrase or question used in an interview to encourage an interviewee to elaborate on an answer.
Purposive sampling
a sampling strategy in which the researcher selects participants who are considered on an a priori basis to be typical of the wider population.
Qualitative research
research conducted in naturalistic settings which generates data largely from observations and interviews.
Quantitative research
research in which the researcher explores relationships between and among variables using numeric data.
the values that divide a list of numbers into quarters. The first quartile is the number below which lie the lowest 25 per cent of the data. The second quartile is the median value. The third quartile is the midpoint between the median and the highest value of the list.
an experiment in which individuals are not assigned randomly to groups; rather, some environmental factor (such as place of residence) determines who belongs to each group.
structured sets of questions on specified subjects that are used to gather information, attitudes, or opinions.
Random sampling
a sampling technique in which individuals are selected from a population at random. Each individual in the sample is selected entirely by chance.
Randomised controlled trial
a type of experiment in which the subjects are randomly allocated to one or other of the different groups under study. Randomisation minimises the possibility of allocation bias. Such trials must contain (as well as the experimental group(s)) a control group which receives no treatment, and/or a comparison group which receives a 'non-active' comparison treatment.
a measure of the dispersion of data. The range is calculated by subtracting the value of the lowest data point from that of the highest data point.
the allocation of ranks (1st, 2nd, 3rd, etc.) to values in a dataset.
a harmonious relationship in which people communicate well.
Rating scale
a measuring instrument for which judgements are made in order to rate a subject or case at a specified scale level with respect to an identified characteristic or characteristics.
the quotient of two amounts, showing the number of times one value contains (or is contained by) the other.
Raw score
a score from a test, survey, etc. that has not been converted to another type of score such as a standard score or ranking.
see randomised controlled trial.
the position that there is a reality independent of our research instruments and independent of our interpretation.
the methodological position that complex phenomena are best studied by describing and analysing them in terms of their simplest constituents.
all of the books and articles referred to in a research report, appearing in a list in a standard format (such as the Harvard system) at the end of the report.
Regression coefficient
indicates how much a dependent (outcome) variable will change, on average, with each change in an independent variable.
the consistency and dependability of a measure, procedure or instrument in gathering data.
repeating a research study to test the findings of the study being repeated.
Representative sample
sample in which the relevant characteristics of the participants closely match those of the whole population in question.
the orderly, fair and thorough investigation of a question, an issue or a phenomenon for the purpose of adding to knowledge. For formal, widely used definitions of experimental and applied research, search for 'Frascati Manual definition of research'.
Research question
a clear statement in the form of a question about the issue that a researcher wishes to study.
a person who responds to a request for information.
Response rate
in a survey, the percentage of questionnaires completed and returned out of the total distributed.
a group that is selected from a population. By studying the sample the researcher tries to draw conclusions about the whole population.
Sampling error
the degree to which the results from the sample deviate from those that would be obtained from the entire population.
in constant comparative method, the point at which analysis begins to reveal repetition and redundancy.
Scatter graph
a display of the relationship between two variables. Dots on the graph (also called a 'scatter plot') show the values of x and y for each case.
Selection bias
error due to differences in the characteristics of those who are selected for a study and those who are not.
Selective coding
the final part of the coding process in grounded theory in which the researcher systematically relates axial codes to each other, seeking to assess the validity of each code and to establish core themes.
Semi-structured interview
an interview in which interviewers use an interview schedule (rather than a fixed list of questions) which contains a list of topics that they intend to cover during the interview. The interviewer may conduct the interview flexibly, and may follow up on things the respondent says during the interview.
if there is only a very small probability that a relationship observed in a statistical analysis is due to chance, the results are said to reach statistical significance. This means that the researcher can conclude that there is some kind of meaningful relationship between the observed variables or a meaningful difference (i.e. a difference not due to chance) between groups.
Snowball sampling
a strategy for sampling in which study participants themselves refer the researcher to other individuals who fit the study criteria.
Social desirability bias
see prestige bias.
Social media
websites and applications that enable users to create and share information or to participate in social networks. Some of the most popular social media websites are Facebook, WhatsApp, Twitter, Pinterest, Google+, YouTube and Snapchat.
a diagram showing the relationships among individuals in a group. Individuals in the research group will give the names of usually two preferred or non-preferred individuals in the group, and arrowed lines among the individuals will represent these patterns of preference or non-preference.
Standard deviation
the average amount by which each score differs from the mean.
grouping a population into subgroups (or 'strata') by certain characteristics (e.g. level of highest qualification), and drawing from these strata in proportion to their size in the whole population. Stratification may improve the representativeness of a sample.
Structured interview
an interview in which the interviewer asks all respondents the same questions using a predetermined list of interview questions.
a person, usually a university member of staff, who oversees a student research project. They should be familiar with the student's area of research.
Survey research
a research approach designed to collect data systematically- usually attitudes or beliefs - about a particular population. Data are usually obtained through direct questioning of a representative sample of respondents.
Test-retest reliability
the degree to which a measure produces consistent results over several administrations.
Theme mapping
graphic display of concepts and their interrelationships.
Theoretical sampling
in a naturalistic research study, the selection of individuals based on emerging findings as the study progresses to ensure that key issues are adequately represented.
in social research, theory has many meanings. In interpretative social research, theory is a developing conceptual organisation that helps links to be made between ideas, and enables explanations to be offered about the findings of research. In experimental research, by contrast, it usually refers to a body of knowledge which provides the basis for testing an idea or a hypothesis via research: the findings of the research will then contribute to that body of knowledge.
Thick description
a report on the observations and other data made during qualitative research; to qualify as 'thick description' the report must include interpretation, conjecture, analysis and discussion.
Time series
an ordered sequence of observations.
using a combination of data collection methods or analytical techniques to gain different perspectives on a particular research problem or question.
t test
a statistical test used to determine if the scores of two groups differ significantly on a single variable.
a system of categorisation, often represented in tabular or diagrammatic form.
Unstructured interview
an interview in which the researcher asks open-ended questions and allows the respondent to talk freely on a topic and influence the direction and substance of the interview. By contrast with structured and semi-structured interviews, there is no pre-set plan (beyond the general topic) about the substance of the interview.
the degree to which data and findings are accurate reflections of reality.
an attribute of a person or phenomenon that varies within the population under investigation (e.g. age, height, gender, income).
in statistics, the extent to which a set of numbers is spread out from the mean of the set of numbers.
Weighted scores
scores or values which are modified by different multipliers to reflect their importance.
an emphasis made to particular parts of a sample to ensure that statistics produced from that sample are representative of the entire population.
the horizontal axis of a graph.
the vertical axis of a graph.
the tendency of questionnaire and interview respondents to say 'yes' when asked a question. See also prestige bias.