» Statistics are used to describe numerical outcomes and the probabilities derived from
calculations on raw data.
» There are two types of statistics: descriptive and inferential. Descriptive statistics explain
characteristics of variables found in a sample. Inferential statistics are the basis for predictions
about the phenomenon.
» Univariate analysis involves only one variable at a time and includes frequency distributions,
measures of central tendency, shape of the distribution, and measures of variability.
» Bivariate analysis is conducted to describe relationships between two variables.
» Frequencies may be grouped or ungrouped data describing how often a variable is found to
occur. Line charts, bar graphs, pie charts, histograms, and scattergrams are ways frequencies
can be depicted.
» The mode, median, and mean are measures of central tendency. In a normal distribution,
they are all the same value.
» A normal distribution is a bell-shaped curve that is symmetric around the mean. Asymmetric
distributions are known as skewed and have either positive or negative directions.
» Data with little variability are considered to be homogenous, whereas data with wide varia-
tions are considered to be heterogeneous.
» Measures of variability include the range, semiquartile range, percentile, standard deviation,
z scores, and coefficient of variation.
» The Rule of 68–95–99.7 is a way of describing the percentage of scores falling within specific
standard deviations of the mean.
» Correlation coefficients are used to describe the relationship between two variables.
» Inferential statistics are used to estimate population parameters and test hypotheses.
» CIs describe the probability of being correct.
» When selecting which statistical tests to use, researchers must consider the type of question
being asked and the level of measurement. The null hypothesis is statistically tested and is
rejected when significant findings occur. Research hypotheses are supported but never proven.
» Probability is the likelihood of a frequency of an event in similar trials under similar conditions
and indicates how probable it is that the results were obtained by chance. Sampling error
is the tendency for statistics to fluctuate from one sample to another and forms the basis of
probability.
RAPID REVIEW
13.10 Keeping It Ethical 375