Glossary

 

A


Alternative hypothesis
A precise statement relating to the research question to be tested, expressed in terms which assume a relationship (association) or difference between variables. Used in conjunction with a suitable null hypothesis. (See the issues of quantitative analysis for more information.)


Audit trial
Confirms the quality of qualitative research


Auditable
Able to audit the qualitative research process


Average
A general term given to a descriptive statistic, which gives a measure of central tendency of sample data, eg mean, median, mode.

B


Bias
An unplanned effect on the data collection in research, which may influence results. For example 'non response bias' in the return of postal questionnaires.

C


Categorical data
Data at non-measurement level, grouped into categories. For example, nominal – gender, or ordinal – income band


Confidence interval
A range of values, within which we are fairly sure the true value of the parameter being investigated lies. A common confidence interval (CI) is 95%. Thus, for example, we can be 95% certain that the true population mean lies approximately within the interval calculated from the sample mean ± 2 x standard error of the mean. 2 is an approximation, dependent on sample size.


Contingency table
A contingency table is a two-dimensional table of counts, usually showing frequencies of two variables, displayed in rows and columns respectively.


Continuous data
This data measured at least at interval level. It is as precise as measuring instruments will allow.


Critical appraisal
Interpreting the strengths and weaknesses of the research process and applying judgements to practice

D


Data
Information gathered in the course of a research study. It may be quantitative or qualitative.


Data analysis
Processing, interpretation and analysis of findings


Deductive paradigm
The testing/application of theories


Dependent variable
The variable which is assumed to respond to the values of the independent (explanatory) variable. For example, blood pressure could be deemed to respond to changes in age.


Discrete data
Data measured at least at interval level, but only as whole numbers (integers). For example, household size, or number of siblings.

E


Ethical committee
A committee of members who judge the appropriateness and merit of proposed research

F


Focus groups
A group composed of between six and twelve individuals who meet to discuss a research problem

G

H


Hypothesis
A precise statement relating to the research question to be tested.

I


Independent variable
The variable which is assumed to determine the values of the dependent (response) variable. For example, blood pressure could be deemed to respond to changes in age.


Inductive paradigm
The development of theories from observation


Interquartile range (IQR)
This is the difference between the upper (Q3) and lower (Q1) quartiles. It is less sensitive to extreme outliers than the range, as a measure of spread of data.


Interval/ratio data
This is data recorded on a scale with equal distances between points. Data can be continuous or discrete. Data at ratio level has the additional quality of an 'absolute zero'. Thus temperature (in °C or °F) is measured at interval level. Weight, height, etc are at ratio level.

J

K

L


Linearity
To calculate Pearson's coefficient of correlation to measure the level of association between 2 variables, it is necessary for the data to be related following a straight line. Thus a check for linearity is obtained by plotting a scatter diagram of the two variables.


Literature review
Appraisal of previous research or literature on a subject

M


Mean
The arithmetic mean is a descriptive statistic, which is a measure of central tendency, or average, around which the data clusters. All data in a sample is used. It is appropriate for data measured at least at interval level.


Median
The median is a descriptive statistic, which is a measure of central tendency, or average, around which the data clusters. It is the middle value when data in a sample is arranged in order. It is appropriate for data measured at least at ordinal level.


Mode
The mode is a descriptive statistic, which is a measure of central tendency, or average, around which the data clusters. It is the most frequently occurring value in a sample. It is appropriate for categorical data.

N


Nominal data
Categorical data gathered into groups, with no order attached to them. For example, ethnicity.


Non-probability sampling
Use of random selection in obtaining the sample


Normal distribution
Data following a normal distribution, such that a bell-shaped curve.


Null hypothesis
A precise statement relating to the research question to be tested, expressed in terms which assume no relationship (association) or difference between variables. Used in conjunction with a suitable alternate hypothesis. (See the issues of quantitative analysis for more information.)

O


Ordinal data
Categorical data gathered into groups, with order attached to them. For example, job grade, age group.


Outlier
An extreme, or atypical, data value(s) in a sample. They should be considered carefully, before exclusion from analysis. For example, data values maybe recorded erroneously, and hence they may be corrected. However, in other cases they may just be surprisingly different, but not necessarily 'wrong'.

P


Parameter
This is a property of a population, eg the mean or standard deviation, which is being estimated from the sample data.


Percentiles
Percentiles split the sample data into hundredths. For example, the 25th percentile is equivalent to the lower quartile, and the 50th percentile is the same as the median.


Population
The group of individuals, or items, to be studied is called the population.
For example, men aged 21 and over; pregnant women; households in Bristol; houses in Bristol. The subset of this population that is measured or observed is called the sample.


Probability sampling
Sampling techniques that do not use random selection


P-value
The probability of an observed result happening by chance under the null hypothesis.

Q


Qualitative analysis
Interpretation of words and text


Quantitative analysis
Interpretation of numerical data


Quartiles
The lower (Q1) quartile is the value below which the bottom 25% of the sample data lie, and the upper (Q3) quartile is the value above which the upper 25% lie.
NB. The middle quartile (Q2) corresponds to the median.

R


Random sample
Every member of a population has an equally-likely chance of being in a random sample. It is representative of the population being studied.


Range
A descriptive statistic equal to the maximum less the minimum value in a data set. It is a crude measure of variation (spread) of the data.


Rank
Sample values are ordered or ranked.


Raw data
Rows and columns containing numbers representing the collected data. Numbers may be values or category codes. Rows relate to subjects/cases, and columns relate to individual variables.


Representative
The extent to which sample data or members reflect accurately the characteristics of the population from which they are drawn.


Research process
The process undertaken by researchers to answer research questions/hypotheses.


Research question
A specific question that guides the research process

S


Sample
A subset (n) of the entire population (N). Those people, objects or events selected from the population for inclusion in the study.


Scatter
A scatter plot is drawn between 2 variables at interval/ratio level to check for a linear relationship, prior to calculating Pearson's coefficient of correlation. The independent (explanatory) variable is plotted on the horizontal (x) axis, and the dependent (response) variable is plotted on the vertical (y) axis.


Semi-interquartile range (SIQR)
This is half the interquartile range (IQR).


Significance level
Set as the p-value.


Standard deviation
The standard deviation is a descriptive statistic, which is a measure of dispersion, or spread, of sample data around the mean. All data in a sample is used. It is appropriate for data measured at least at interval level.


Standard error of the mean
A measure of the accuracy of the sample mean as an estimate of the population mean.
Equal to sample standard deviation divided by the square root of the sample size


Statistically significant
If a result is 'statistically significant', it implies a statistical test has been carried-out, and the probability of obtaining the observed data (or more extreme) under the null hypothesis, is small – typically less than 0.05.


Summary or descriptive statistics
These are a set of calculated terms from the sample data to describe the sample data to the reader. They include sample size, maximum & minimum values, averages (mean, median, mode), measures of variability (range, interquartile range, standard deviation).
Note that it is important to use the correct statistic depending on the level of measurement of the data.


Systematic review
Review of the literature based on a scientific design

T


Triangulation
A research design that includes two or more approaches to data collection or analysis


Trustworthiness
A description of the credibility, transferability, dependability and confirmability of qualitative research

U

V


Variable
Is a term ascribed to the characteristic(s) being investigated, and can take any value in a reasonable range. For example, blood group, blood pressure, age of patients being studied.


Verbatim extracts
Exact word for word and punctuated account of speech

W

X

Y

Z