Statistics
Common terms and levels of measurement
Categorical variable Measurement that classes people, object or places into groups. There are two types: nominal and ordinal
Continuous variable Numerical measurement. There are two types: interval and ratio
Dependent variable Phenomenon being studied, outcomes
Descriptive research A study to describe a particular phenomenon as it occurs in a sample
Empirical Something measurable, observable and tangible
Empirical relationship Causal requirement that the independent and dependent variables have an observed relationship with one another
Exploratory research The study of issues that have seldom been examined by prior research and may be lacking in firm theory
Evaluation research The study of results or programmes or interventions to determine their effectiveness
Hypothesis A single proposition, deduced from a theory, that must hold true for a theory to be considered valid
Independent variable Factor or characteristic being used to attempt an explanation or prediction of a dependent variable
Nominal variable Classes people, objects or places into different groups according to a characteristic that cannot be ranked (i.e. self-defined ethnicity, sex)
Nonspurious Causal requirement that the relationship between an independent variable and dependent variable are not the result of a third variable omitted from the analysis
Ordinal variable Classes people, objects or places into group according to characteristics that can be ranked in terms of quantity (i.e. income brackets)
Population People, objects or places under study
Sample A subset taken from a population using people, objects or place, used as a way to generalise the population
Temporal order Causal requirement that an independent variable must precede a dependent variable (direction of relationship)
Theory Proposed explanations about reality that are bound by logic and evidence
Unit of analysis The object or target of study - individuals, crime incidents, micro-meso-macro geographical units
Descriptive statistics
Bivariate Analysis composed of two variables, often one is designated an independent variable and the other a dependent variable
Classes Categories or groups within nominal or ordinal variables
Contingency table Shows overlap between two variables
Cumulative Frequency, proportion or percentage derived from adding a unit to all those below it
Deviation score Distance between mean of a data set and any given raw value
Dispersion Spread of variability among the values in a distribution
Frequency Number of times a particular unit is counted in a data set
Longitudinal variable Variables measured repeatedly over time
Mean Arithmetic average of a set of data
Measures of central tendency Denote where the values in a data set cluster, i.e., the mode, median and mean
Median The value that cuts a distribution in half 50/50
Mode Most frequently occurring category or value
Negative skew A set of values that cluster to the right
Normal distribution A set of values that cluster in the centre and tail off to the left (negative) and right (positive), a bell curve
Percentage Standardisation of a frequency ranging from 0.00 to 100.00
Positive skew A set of values that cluster to the left
Proportion Standardisation of a frequency ranging from 0.00 to 1.00
Standard deviation Measure of dispersion that is the mean of the deviation scores
Trend Patterns indicating a direction of travel, increasing, decreasing or remaining stable over time
Univariate Involving one variable
probability
Binomial Trial with two possible outcomes
Empirical outcome Numerical result from a sample, also known as observed outcome
Probability distribution Table or graph showing entire set of probabilities associated with every possible empirical or observed outcome
Probability Likelihood an event will occur, representing a prediction
Theoretical prediction Prediction grounded in logic