Statistics Glossary

Statistics Glossary
Average: Same as the mean. Add up all the values and divide by the
number of values.
Categorical variable: A variable whose values can be sorted into

categories. Ex: gender, fur color
Chi-square test: A statistical test used with categorical or nominal

variables to tell whether their distribution is different than expected,
i.e., there is something going on that is more than chance.
Continuous data: Data that can be measured along a continuum of

many values. This is distinct from discrete data, which is divided into
categories. Ex: weight, height, age
Correlation: A statistical test that describes the strength of a linear

relationship between the two variables. Correlation does not
necessarily mean there is a causal relationship.
Dependent variable: A variable which is dependent on the levels of

other variables for its value. Dependent variables are measured but
not manipulated.
Descriptive statistics: Statistics that describe the data. This includes

the mean, median, mode, standard deviation, and variability, among
others.
Discrete variable: This is a variable where the values are distinct,

separate (not on a continuum), and can be counted. Ex: # of kittens in
a litter, the breeds of puppies living in a particular subdivision
Experimental hypothesis: H1 speculates that there is a relationship

between your variables. This is the hypothesis you are trying to
support with your statistical test.
Independent variable: A variable that can be manipulated.
Inferential statistics: Statistics that analyze the data and allow you
to infer something about the population from the sample.
UT Southwestern Medical Center Library—October 2007

Interval variable: An interval variable can be ranked and the
intervals between the values can be compared. For statistical
purposes, interval and ratio variables are generally treated the same.
Mean: Same as the mean. Add up all the values and divide by the
number of values.
Median: The median is the middle number in a set of data values. The
median is important because unlike the mean, the median is not
affected by outliers.
Mode: The data value that appears most frequently in a set.
Nominal variable: This is essentially the same thing as categorical

data. Nominal data values cannot be ranked in any order; they can
only be placed in categories.
Null hypothesis: H0 speculates that there is no relationship between

your variables. You always should have both a null and an
experimental hypothesis when you are performing statistical tests.
Ordinal variable: Ordinal data values can be ranked, but the intervals
between them are essentially meaningless. Ex: Runners in a race—
Someone comes in first, but we don’t know how much faster the first
place winner was than the second place finisher.
Outlier: A data value that is significantly larger or smaller than the

other values measured for that variable.
Population: The set of all objects you are interested in studying. You
define what your population is.
Probability: How likely a value is to occur. Probability is important

because it helps us distinguish whether we are getting a value by
chance or because of the levels of other variables.
Ratio variable: A ratio variable has a true zero point where you can
have zero of whatever you are measuring. These are very rare. For the
purposes of statistics, interval and ratio variables are treated the
same.
Regression: A statistical test that predicts the level of the dependent

variable based on the level(s) of one or more independent variable(s).

Sample: A subgroup of the population. It is often difficult to measure
the entire population, so samples are measured instead and their
results are extrapolated to the population as a whole.
Significance: The fixed level of probability of incorrectly rejecting a

true null hypothesis. This level is normally 0.05 and is frequently
indicated by p. It means you have a 5% chance of saying your null
hypothesis is wrong when it is really true.
Standard deviation: The square root of the variance. This is a

measure of how much your data values vary from the mean.
Standard normal curve: A symmetrical bell-shaped curve with a

mean of 0, and a variance of 1 that represents the distributions of all
the sample groups that could be drawn from a population.
Statistics: The branch of mathematics that deals with describing and

interpreting data and drawing conclusions based on the data
Trend: The general direction of a relationship. In statistics, a trend is

noticeable but not statistically significant.
Variable: A factor that can change from one measurement to the next

Statistics Glossary

Uploaded by

Statistics Glossary

Uploaded by

Statistics Glossary

Categorical variable: A variable whose values can be sorted into

Chi-square test: A statistical test used with categorical or nominal

Continuous data: Data that can be measured along a continuum of

Correlation: A statistical test that describes the strength of a linear

Dependent variable: A variable which is dependent on the levels of

Descriptive statistics: Statistics that describe the data. This includes

Discrete variable: This is a variable where the values are distinct,

Experimental hypothesis: H1 speculates that there is a relationship

Independent variable: A variable that can be manipulated.

UT Southwestern Medical Center Library—October 2007

Mode: The data value that appears most frequently in a set.

Nominal variable: This is essentially the same thing as categorical

Null hypothesis: H0 speculates that there is no relationship between

Outlier: A data value that is significantly larger or smaller than the

Probability: How likely a value is to occur. Probability is important

Regression: A statistical test that predicts the level of the dependent

UT Southwestern Medical Center Library—October 2007

Significance: The fixed level of probability of incorrectly rejecting a

Standard deviation: The square root of the variance. This is a

Standard normal curve: A symmetrical bell-shaped curve with a

Statistics: The branch of mathematics that deals with describing and

Trend: The general direction of a relationship. In statistics, a trend is

UT Southwestern Medical Center Library—October 2007

You might also like