Education: Elearning virtual
campus:
Definitions
These definitions are used in our courses.
A  B  C  D
 E  F  G  H
 I  J  K  L
 M  N  O  P
 Q  R  S  T
 U  V  W  X
 Y  Z
A,
 Action research
'Smallscale interventions in the functioning of the real world and a
close examination of the effects of such interventions' — Halsey (1972).
 Aim
An aim is a description of intention or broad purpose of research.
 Alternative hypothesis (H_{1})
A prediction that there is a difference between the groups of data being compared.
The alternative hypothesis is often the working hypothesis, or research question,
in a study.
 Association
The tendency for two things to occur together.
B
 Bias
A tendency to misrepresent. The term bias is used in statistics to refer
to how far the average statistic lies from the parameter it is estimating,
that is, the error that arises when estimating a quantity. Errors from chance
will cancel each other out in the long run, those from bias will not.
 Binomial distribution
Binomial distributions model (some) discrete random variables. When a coin
is flipped, the outcome is either a head or a tail ie there are two mutually
exclusive possible outcomes. For convenience, one of the outcomes can be termed
'success' and the other outcome termed 'failure'. If a coin is flipped N times,
then the binomial distribution can be used to determine the probability of
obtaining exactly r successes in the N outcomes. The formula that is used
assumes that the events: fall into only two categories (ie are dichotomous);
are mutually exclusive; are independent and are randomly selected.
 Blinding
A lack of awareness of a subject's treatment group. Doubleblinding is
blinding of assessor as well as subject.
C
 Care organisation
The organisation(s) that are responsible for providing care to patients
(and/or users) and carers who are participating in the research study in question.
 Casecontrol study
A type of study that involves identifying patients who have the outcome
of interest (cases) and patients who have not (controls), and looking back
to see if they had the exposure of interest.
 Caseseries
A report on a series of patients who have the outcome of interest. No
control group is included. Contrast with casecontrol
study.
 Categorical variable
A variable that can take values that represent distinct categories.
 Causality
The act of one thing actively causing another.
 Central limit theorem (CLT)
The central limit theorem demonstrates that in large enough samples, the distribution
of a sample mean approximates a normal curve, amazingly, regardless of the
shape of the distribution from which it is sampled. The larger the value of
the sample size (n) the better the approximation to the normal.
 Clinical effectiveness
A measure of how well an intervention, pharmaceutical, health technology,
etc actually achieves its aim in improving health status.
 Clinical trial
Any planned therapeutic, diagnostic or preventive study involving people and
comparing concurrently one intervention with another or with a placebo or
with no intervention to determine their relative safety and efficacy.
 Cohort
A group.
 Cohort study
A type of study that involves the identification of two groups (cohorts)
of patients, one of which received the exposure of interest, and one of which
did not. Both groups are followed forward over time and studied for the outcome
of interest.
 Confidence interval
A confidence interval is an interval used to estimate the likely size
of a population parameter. It gives an estimated range of values (calculated
from a given set of sample data) that has a specified probability of containing
the parameter being estimated. Most commonly used are the 95% and 99% confidence
intervals that have .95 and .99 probabilities respectively of containing the
parameter. The width of the confidence interval gives some indication about
how uncertain we are about the unknown population parameter. Confidence intervals
are more informative than the simple results of hypothesis tests (where we
decide 'reject the null hypothesis' or 'don't reject the null hypothesis')
because they provide a range of plausible values for the unknown parameter.
 Confidence interval for a mean
A confidence interval for a mean specifies a range of values within which
the unknown population parameter (in this case, the mean) might lie. The width
of the confidence interval provides some indication about how uncertain we
are about the unknown population parameter, in this case the mean. We calculate
these intervals for different confidence levels, depending on how precise
we want to be. We interpret a 95% confidence interval as 'we are 95% confident
that the interval contains the true population mean'.
 Confounding
A measured effect attributed to a variable that is actually due to an
unmeasured covariable.
 Content analysis
A method of analysis used in qualitative research in which text (notes)
are systematically examined by identifying and grouping themes and coding,
classifying and developing categories.
 Continuous variable
A variable that can take any value that represents a measurement.
 Consequences
The results (measured in some way) of employing a healthcare intervention
(or pharmaceutical, etc).
 Costbenefit analysis (CBA)
A type of economic evaluation in which both the costs and consequences of
different interventions are expressed in monetary units.
 Costconsequences analysis
This term can be used for a type of costeffectiveness analysis in which
cost implications are presented against a range of different output measures.
The decisionmaker then decides the relative importance of these.
 Costeffectiveness analysis (CEA)
A form of economic evaluation in which consequences of different interventions
might vary but can be measured in identical 'natural' healthrelated units
(eg lives saved, cases detected); relative inputs are then costed. Competing
interventions are compared in terms of cost per unit of consequence.
 Costeffectiveness ratio
The total cost of a programme divided by the effectiveness measure (eg
cost per lifeyear gained) used in costeffectiveness analysis to select among
programmes.
 Costminimisation analysis (CMA)
A form of economic evaluation in which consequences of different interventions
are the same and in which only the costs are taken into consideration. The
aim is to decide the cheapest way of achieving the same result.
 Costutility analysis (CUA)
A form of economic evaluation in which interventions that produce different
consequences in terms of both quantity and quality of life are expressed as
utilities (eg QALYs). Different interventions are compared in terms of cost
per utility.
 Costs
The resource implications of undertaking an activity. What is actually measured
in the costs will depend upon the type of activity, and also from which perspective
the analysis is being taken.
 Critical appraisal
The fair assessment of research literature, judging the validity of results
and conclusions, the contribution to scientific knowledge and the populations
to which the conclusions can be applied. A basic tool to aid evidencebased
health care.
 Crossover study design
The administration of two or more experimental therapies in sequence in a
specified or random order to the same group of patients.
 Crosssectional
A 'snapshot' eg the UK census.
 Crosssectional study
A type of study that involves the observation of a defined population
at a single point in time or time interval. Exposure and outcome are determined
simultaneously (a 'snapshot').
D
 Decision analysis
An explicit, quantitative approach for prescribing decisions under conditions
of uncertainty.
 Degree of freedom (DF)
A mathematical concept that indicates the number of observations or values
in a distribution that are independent of each other or are free to vary.
They are used with various measures such as ttests to refine the results
of treatments of probability or chance in determining statistical significance.
For example, if you have a distribution of three numbers that could vary but
the sum of those numbers has to equal 100, then although you could select
three numbers, in reality you only have to select two because the third number
is determined by the first two numbers selected. In this case, there are two
independent values, or two degrees of freedom.
 Delphi technique
This technique is essentially a survey method whereby individual members of
a panel are asked for written information, which is then fed back to the panel
as a whole for further comment with the intention of reaching a consensus.
 Descriptive studies
These studies describe the state of nature at a point in time.
 Diagnosis Related Groups (DRGs)
The Diagnosis Related Group system classifies patients into groups based
on the principal diagnosis, type of surgical procedure, presence or absence
of significant comorbidities or complications, and other relevant criteria.
Diagnosis Related Groups are intended to categorise patients into groups that
are clinically meaningful and homogeneous with respect to resource use.
 Direct costs
These are the costs that occur as a result of organising and running the
intervention or programme that is being evaluated.
 Disclosure
The conscientious effort to ensure that participants are given full information
pertaining to a research study.
 Discounting
This is a technique that allows the calculation of present values of inputs
and benefits that accrue in the future.
 Dispersion
The data values in a sample are not all the same: this variation between values
is called dispersion. When the dispersion is large, the values are widely
scattered; when it is small they are tightly clustered. There are several
measures of dispersion eg the standard deviation.
 Distribution
Measurements on any variable, even the same variable on the same subject will
vary. The pattern of variation of a variable is called its distribution, which
can be described both mathematically and graphically. In essence, the distribution
records all possible numerical values of a variable and how often each value
occurs (its frequency). The most well known example of a distribution is the
bellshaped curve.
E
 Economic evaluation
The process used to relate the cost implications of an intervention (or
pharmaceutical product, etc) to its consequences. It covers a number of methods,
which vary mainly in the ways the effectiveness of an intervention is measured.
 Effectiveness
The benefit of an intervention when deployed in the field.
 Efficacy
The benefit of an intervention under ideal conditions.
 Electronic database
Information, usually consisting of linked data fields, that can be accessed
through an electronic medium (eg via a personal computer that has access to
this data). Most electronic databases have a search facility, which allows
the reader to look for individual topic components within large sets of indexed
data.
 Employing organisation(s)
The organisation(s) that employ the principal investigator and/or other
researchers taking part in the study. The organisation employing the principal
investigator will normally hold the contract(s) with the funder(s) of the
study. Organisations holding contracts with funders are responsible for the
management of the funds provided.
 Estimate
An estimate is an indication of the value of an unknown quantity based on
observed data.
 Estimation
Estimation is the process by which sample data are used to indicate the value
of an unknown quantity in a population. Results of estimation can be expressed
as a single value, known as a point estimate; or a range of values, known
as a confidence interval.
 Ethics
A generic term for various ways of understanding and examining the moral conduct
of human behaviour and actions. Some approaches are normative (ie they set
standards of right of good action) others are descriptive (ie they report
on what people believe and how they act).
 Ethnography
'The study of people in their natural settings; a descriptive account of social
life and culture in a defined social system, based on qualitative methods
(eg detailed observations, unstructured interviews, analysis of documents)'
 Bowling (1997).
 Experimental research
A research technique in which variables of interest (such as drug treatment
or diet) are deliberately manipulated by the researcher.
F
 Face validity
This is the extent to which we logically believe that the survey is measuring
the proper areas. For example, we would logically measure attitudes towards
health with questions about health.
 Focus groups
A method of collective (or group) interview, which explicitly uses the interaction
within the group to generate data.
 Frequency distribution
A frequency distribution shows the number of observations falling into each
of several ranges of values. Frequency distributions are portrayed as frequency
tables, histograms or polygons.
 Funder(s)
Organisation(s) that provide funding for the study through contracts, grants
or donations to an authorised member of either the employing and/or care organisation.
G
 'Grey literature'
Unpublished or unindexed reports. These can include conference proceedings,
nonindexed journals, internal reports, pharmaceutical industry reports and
student dissertations and theses.
H
 Hawthorne effect
The impact of the researcher on the research subjects or setting, notably
in changing their behaviour.
 Healthcare Resource Groups
A data analysis tool for reflecting at aggregate level the interventions
that are undertaken and the resources that are required in providing healthcare
activities. These groups provide a standard and comparable description of
healthcare provision, which is useful for a range of NHS purposes.
 Health services research
Investigation of the health needs of the community and the effectiveness
and efficiency of the provision of services to meet those needs in order to
provide information relevant to the planning, management and development of
services'.
 Hypothesis
A hypothesis is a statement of the expected relationship between or among
the things being studied.
 Hypothesis testing
Hypothesis testing is one of the two main branches of inference. Setting up
and testing hypotheses is an essential part of statistical inference. In order
to formulate such a test, usually some theory has been put forward, either
because it is believed to be true or because it is to be used as a basis for
argument, but has not been proved. For each problem considered the question
of interest is simplified into two competing claims/hypotheses between which
we have a choice: the null hypothesis (denoted H_{0}) against the
alternative hypothesis (denoted H_{1}). Special consideration is given
to the null hypothesis.
I
 Incremental costs
These are the additional costs that one programme or intervention imposes
over another. An analysis of the incremental costs (and effects) tells us
how much extra we would have to pay for the extra effect.
 Indepth interviews
A facetoface conversation conducted to explore issues or topics in detail.
Preset questions are not used, but a defined topic or issues are explored.
 Indirect costs
These include any costs that are borne by patients or their families as
a result of taking part in a healthcare intervention that is being evaluated.
These costs are mainly productivity losses.
 Informed consent
A process in which the risks, benefits, and requirements of a research study
are explained to people invited to take part in the study. Before entering
the study a participant should sign an informed consent form, which should
contain in writing the benefits, risks and basic structure of the study.
 Interview
A social experience, depending on one person, the interviewer, asking
a series of questions of another person or group of people.
J, K,
L
 Likert scale
This refers to a widely used questionnaire format named for its developer,
Rensis Likert. Respondents of questionnaires are asked to choose from several
responses in a range such as 'strongly agree', 'agree', 'undecided'. 'disagree',
and 'strongly disagree'. Each response receives a number rating from 05.
The fivepoint Likert scale is most common.
M
 Marginal costs
This is the additional cost of producing one extra unit of output (eg
treating one extra patient).
 Mean
The average of a set of values.
 MEDLINE
The US National Library of Medicine's database that contains more than
11 million references to journal articles in the health sciences.
 Metaanalysis
An overview in which quantitative methods are used to summarise the results
of several studies on a single topic. A metaanalysis is used in an attempt
to gain greater objectivity, generalisability and precision by including all
the available high quality evidence from randomised controlled
trials carried out on a specified topic.
N
 Nonexperimental research
Research in which the researcher does not have control over all the variables
in the study, eg the researcher cannot assign subjects to interventions.
 Normal distribution
Normal distributions are a family of distributions that are characterised
by a bellshaped, symmetric curve, with scores more concentrated in the middle
than in the tails. They are defined by two parameters: the mean (m,
mu) and the standard deviation (s, sigma).
Many kinds of data are approximated well by the normal distribution. Many
statistical tests assume a normal distribution. Most of these tests work well
even if the distribution is only approximately normal and in many cases as
long as it does not deviate greatly from normality. The normal distribution
plays a vital role in inference.
 Null hypothesis
The null hypothesis is used in experimental research. It asserts arbitrarily
that there is no relationship among the variables
being studied. Then statistical tests are used to determine if any relationship
shown by the research data is due to chance alone or to alternative hypotheses.
 Numerical data
Data that consist of counts, or measurements that can take any value within
a given range (eg age, length of hospital stay, blood pressure).
 NHS research
National Health Service (NHS) research can be defined as a planned activity
to generate new knowledge that is generalisable and can be accessed through
the search of published articles and the so called 'grey'
literature.
O
 Objectives
Objectives are specific outcomes that can be measured using agreed criteria.
 Observational
A nonexperimental research technique in which observations are made without
any intervention on the part of the researcher.
 Observer bias
A systematic difference between a true situation and that observed owing to
observer variation in perception (ie interpretation).
 Onesample ttest
A onesample ttest is a hypothesis test for answering questions about the
mean where the data are a random sample of independent observations from an
underlying normal distribution. The sample has been drawn from a population
of given mean and unknown variance (which therefore has to be estimated from
the sample).
 Onesided test
A onesided test is a statistical hypothesis test in which the values for
which we can reject the null hypothesis, H_{0} are located entirely
in one tail of the probability distribution. The choice between a onesided
and a twosided test is determined by the purpose of the investigation or
prior reasons for using a onesided test.
 Overhead costs
This is an accounting term that refers to resources that cover more than
one department, centre or programme in an organisation.
P
 Parameters
Numerical characteristics of populations ie parameters are values, usually
unknown (and which therefore have to be estimated), used to represent certain
population characteristics. Within a population, a parameter is a fixed value,
which does not vary.
 Participants (formerly known as subjects)
Patients, users, relatives of the deceased, professional carers or members
of the public who agree to take part in a study.
 Phenomenology
'The philosophical belief that, unlike matter, humans have a consciousness.
They interpret and experience the world in terms of meanings and actively
construct an individual social reality'  Bowling (1997).
 Pilot study
A trial study carried out before a research design is finalised in order to
assist in defining the research question or to test the feasibility, reliability
and validity of the proposed study design.
 Placebo
An inert substance or intervention designed to appear the same as the experimental
substance or intervention, but which has no physiological effect.
 Point estimate
Results of estimation can be expressed as a single value, known as a point
estimate.
 Poisson distribution
Poisson distributions model (some) discrete random variables. Typically, a
Poisson random variable is a count of the number of events that occur in a
certain time interval or spatial area. For example, the number of cars passing
a fixed point in a five minute interval.
 Population
The entirety of subjects ie it is any entire collection (eg of people
or things) from which we might collect data and that we wish to describe or
draw conclusions about.
 Population mean
The average value across the population of subjects.
 Power
The chance (given by %) that a predetermined effect in a population can
be detected as statistically significant in a sample.
 PreMEDLINE
In process bibliographic citations for MEDLINE.
 Primary research
A unique research study that produces original data.
 Principal Investigator
The person who is designated to take overall responsibility within the team
of researchers for the design, conduct and reporting of the study.
 Probability
A numeric scale measuring degree of chance from 0 to 1 (0% to 100%).
 Proposal
An application for funding that contains all the information that is necessary
to describe the project plans, staff capabilities and funds requested.
 Prospective
Going forwards in time.
 PubMed
The US National Centre for Biotechnology Information (NCBI) and US National
Library of Medicine's freely available online database and search facility.
It contains more than 11 million references to the many indexed biomedical
journal articles that are in MEDLINE, PreMEDLINE
and other related databases.
 Pvalue
The probability of observing in a sample an effect at least as unlikely
as that observed when in fact there is no effect in the population.
Q
 Qualitative research
Research that focuses on the experiences, interpretations, impressions or
motivations of an individual or individuals, and that seeks to describe how
people view things and why. It relates to beliefs, attitudes and changing
behaviour.
 Qualityadjusted lifeyear (QALY)
A single measure of outcome, combining the quantity of health improvement
and quality of life, which reflects people's preferences for health status.

Quantitative research
Research that focuses on measuring and counting facts and the relationships
among variables, and that seeks to describe observations through statistical
analysis of data. It includes experimental and nonexperimental research
and descriptive research (research that attempts to describe the characteristics
of a sample or population).
R
 Randomisation
The allocation to a treatment group that is performed at random.
 Randomised controlled trial (RCT)
A prospective research trial in which a group of individuals is randomised
(ie allocated at random) into one or more experimental groups or study groups
and a control group. The control group may receive a placebo or an existing
treatment and the treatment group(s) will receive the treatment(s) or intervention(s)
under investigation. All the groups are followed up for the variables/outcomes
of interest.
 Random sample
A fraction of the population that is selected at random (ie all individuals
have an equal chance of selection).
 Range
The range of a sample (or a data set) is a measure of the spread or the dispersion
of the observations. It is the difference between the largest and the smallest
observed value of some quantitative characteristic.
 Recall bias
Systematic error due to the differences in accuracy or completeness of recall
to memory of past events or experiences.
 Reliability
Consistency of measurement (ie something is reliable if you repeat
the intervention with the same subject and get a similar/equal finding).
 Research
Research is the empirical investigation of the relationships between or among
several variables.
 Researchers
Individual(s) conducting the study.
 Research Interview
An interview that is specifically intended to transfer information (be
it 'facts', opinions, attitudes, or knowledge) from one person, the subject
or participant, to another person, the interviewer.
 Research protocol
A detailed written plan that guides the conduct of a research study or investigation.
 Responsible care professional
The doctor, nurse or social worker who is formally responsible for the care
of the participant while they are taking part in the study.
 Retrospective
Looking backwards in time.
 Risk or risk factor
Risk is a term that is used by epidemiologists to quantify the likelihood
that something will occur. A risk factor is something, which either increases
or decreases an individual's risk of developing a disease. However, it does
not mean that, if exposed, an individual will definitely contract a particular
disease.
S
 Sample
A fraction, or subset, of a population. By studying the sample, it is
hoped to draw valid conclusions about the larger group, the population.
 Sample mean
The average value across the subjects in a sample. The sample mean is
an estimator available for estimating the population mean.
 Sample size
The sample size is simply the size of the sample.
 Sample variance
Sample variance is a measure of the spread of or dispersion within a set of
sample data.
 Sampling bias
The tendency for a sample to misrepresent its population for reasons not due
to chance.
 Sampling distribution
The sampling distribution describes probabilities associated with a statistic
when a random sample is drawn from a population.
 Sampling variability
Sampling variability refers to the different values that a given function
of the data takes when it is computed for two or more samples drawn from the
same population.
 Secondary research
Research that uses data that has been derived from a primary
research study and may include a systematic
review of primary research.
 Selection bias
A bias in assignment or a confounding variable that arises from study design
rather than by chance. These can occur when the study and control groups are
chosen so that they differ from each other by one or more factors that might
affect the outcome of the study.
 Sensitivity analysis
This is a technique that reruns the comparison between inputs and consequences,
and varies the assumptions that underlie the estimates. The aim of sensitivity
analysis is to test the robustness of the conclusions by varying the items
about which there is uncertainty.
 Significance level
The level, set usually at 5% or 1%, below which the pvalue would need
to be in order to declare that a population effect exists.
 Sponsor
The organisation that takes primary responsibility for ensuring that the design
of the study meets appropriate standards, and that arrangements are in place
to ensure appropriate conduct and reporting. The sponsor is usually, but does
not have to be, the main funder.
 Standard deviation
Standard deviation is a measure of variability (spread) or dispersion
of a set of data. The more widely the values are spread out, the larger the
standard deviation.
 Standard error
Standard error is a measure of variability. The standard error of a statistic
is the standard deviation of the sampling distribution of that statistic.
Standard errors are important because they reflect how much sampling fluctuation
a statistic will show. The standard error of a statistic depends on the sample
size. In general, the larger the sample size the smaller the standard error.
 Statistic
A statistic is a quantity that is calculated from a sample of data. It is
used to give information about unknown values in the corresponding population.
It is possible to take more than one sample from the same population and the
value of a statistic will in general vary from sample to sample.
 Statistical hypotheses
Statements about population parameter values.
 Statistical inference
Statistical inference makes use of information from a sample to draw conclusions
(inferences) about the population from which the sample was taken ie it is
the act of drawing conclusions about a population from a sample.
 Statistics
The word 'statistics' is used in several different senses. For example it
can be used to refer to a range of techniques and procedures for analysing,
interpreting and displaying data. In a second usage, a 'statistic' is defined
as a numerical quantity (eg the mean) calculated in a sample ie a numerical
summary of sample data. Such statistics are used to estimate parameters.
 Survey
An investigation in which information is systematically collected, but
in which the experimental method is not used.
 Systematic review A review in
which evidence about a topic has been systematically identified, appraised
and summarised according to predetermined criteria.
T
 Test statistic
Measures the degree to which a sample is inconsistent with the null hypothesis.
 Triangulation
The use of three or more different research methods in combination to study
the same phenomena, (eg observational studies, indepth interviews and focus
groups). Triangulation is used mainly as a check of validity.
 Twosample ttest
A twosample ttest is a hypothesis test for answering questions about the
mean where the data are collected from two random samples of independent observations,
each from an underlying normal distribution. The null hypothesis for the twosample
ttest is: H_{0} = m_{1} =
m_{2}. That is, the two samples have
both been drawn from the same population.
 Twosided test
A twosided test is a statistical hypothesis test in which the values for
which we can reject the null hypothesis, H_{0} are located in both
tails of the probability distribution. The choice between a onesided and
a twosided test is determined by the purpose of the investigation or prior
reasons for using a onesided test.
 Type I error
In a hypothesis test, a type I error occurs when the null hypothesis is rejected
when it is, in fact, true ie H_{0} is wrongly rejected. A type I error
is usually considered to be more serious, and therefore more important to
avoid, than a type II error.
 Type II error
In a hypothesis test, a type II error occurs when the null hypothesis H_{0}
is not rejected when it is, in fact, false. A type II error is frequently
due to sample sizes being too small.
U, V
 Validity
True representation or the extent to which the value obtained represents the
object of interest in the absence of 'measurement error' (ie something is
valid if it measures what it is supposed to measure).
The internal validity of a study refers to the integrity of the
experimental design; the external validity of a study refers to
the appropriateness by which its results can be applied to nonstudy patients
or populations.
 Variable
A variable is an 'object' that is being investigated in a research study and
that is considered to be capable of varying. A variable can be (a) dependent
ie the effect or response to what the researcher does, or (b) independent
ie what the researcher does or controls.
W
X, Y, Z
© 2004 Cambridge Institute for Research, Education and Management; Contact
CIREM