One-way tables

Individuals and variables

The simplest kind of information we’ll work with in this course is a set of
individuals with one or more properties, called variables. The individuals
are the items in the data set and can be cases, things, people, etc.

In this table for “heights of people in the car,” the individuals are the
people, and their heights are a variable because height is a property of
each individual.

Person Height

Wendy 5’6’’

Michael 5’9’’

Rachael 5’3’’

Allen 5’11’’

Together, individuals and variables are called data. When we organize

data into a table like this one, we call it a data table. Here’s an example of
a data table about ice cream:

1
 Flavor Scoops sold Contains chocolate? Smooth or chunky?

Vanilla 300 No Smooth

Chocolate 450 Yes Smooth

Cookies & Cream 275 Yes Chunky

Mint Chocolate Chip 315 Yes Chunky

Fudge Brownie 375 Yes Chunky

Rocky Road 250 Yes Chunky

The individuals are the flavors: Vanilla, Chocolate, Cookies & Cream, etc.,
and the variables are their properties: Scoops sold, Contains chocolate,
and Smooth or chunky.

Variables can be categorical or quantitative. In the table for “heights of

people in the car,” there’s one quantitative variable: the height. In the table
for “ice cream shop data for July,” there’s one quantitative variable
(scoops), and two categorical variables (contains chocolate and smooth or
chunky).

• Categorical variables are non-numerical variables. Categorical

variables are also called “qualitative” variables. Their values aren’t
represented with numbers. Whether or not the ice cream contains
chocolate, and whether the ice cream is smooth or chunky are
categorical variables. This is because the values there are words,
not numbers.

• Quantitative variables are numerical variables. Their values are

numbers. The height of the people in the car and the number of

2
 scoops of ice cream sold are quantitative variables, because the
values there are numbers.

We can divide quantitative variables into discrete variables and

continuous variables.

• Discrete variables are those that we can obtain by counting.

Therefore, they can take on only certain numerical values. For
example, the number of scoops of ice cream sold is a discrete
quantitative variable, because we can’t really sell 8.3 or 5.23 scoops
of ice cream.

• On the other hand, continuous variables may include data as

decimals, fractions, or irrational numbers. For example, the height
of the people in the car is a continuous variable.

Levels of measurement
When we work with data, we need to understand the level of
measurement, because not every statistical test can be used with every
type of data set. There are four levels of data measurement: nominal,
ordinal, interval, and ratio.

• Things like favorite food, colors, names, and “yes” or “no”

responses have a nominal scale of measurement. Only categorical
data can be measured with a nominal scale.

• Categorial data can also be ordinal. This type of data can be

ordered. The top three national parks in California is an example of

3
 ordinal data. Or, for example, when we ask a group of people
about how they liked their trip, we may get responses like
“awesome,” “good,” “satisfactory,” or “terrible,” which follow an
order from best to worst.

• Data measured using an interval scale can be ordered like ordinal

data. But interval data also gives us a known interval between
measurements. For example, temperature is measured using an
interval scale, because we can understand the exact interval of
difference between 50 and 60 degrees.

• Data measured using a ratio scale is just like interval scale data,
except that ratio scale data has a starting point, or absolute zero.
Whereas interval data like temperature can have negative and
positive values, things like time, height, and weight are examples
of ratio scale data because those measures can never be negative.

When we have a single individual and a categorical or quantitative variable

assigned to it, we can construct a one-way table. These kinds of tables are
called “one-way” because they usually represent data for one individual
only.

Constructing one-way tables

When we construct a table, we want to think about whether we have more
individuals or more variables. In the two tables we looked at, we had the
individuals listed down the left-hand side, and the variables listed across
the top.

4
But if we have lots of variables but only a few individuals, it can be helpful
to list the individuals across the top and the variables down the left side.
For example, if we’re comparing lots of information about two houses for
sale in the same neighborhood, we might make a table like this:

317 Spruce Rd 819 Lilac St

Price $299,000 $349,000

Square footage 3652 3812

Price per square foot $82 $92

Bedrooms 4 5

Bathrooms 2.5 3

Stories 3 2

Basement Finished Unfinished

Garage spaces 3 2

Lot acres 0.36 0.31

Grass backyard Yes yes

Year built 1974 2001

Property tax $2,356 $2,595

Payment $1,120 $1,045

The individuals are the addresses listed across the top of the table, 317
Spruce Road and 819 Lilac Street. All the variables are listed down the left
side of the table. Since there are so many more variables than individuals,
listing the variables vertically makes the table fit better on paper than if we
had tried to list all the variables horizontally.