2/4/’25
Chapter 2: Basic Descriptive Statistics
(1)
Social Statistics (SOC 241-04) | Spring 2025
Instructor: Jisoo Cho
Descriptive Statistics
■Purposes
– Uses numbers, tables, or graphs to make large
amounts of data easier to understand →
Summarize data
– Shows different categories and how often they
appear → Organize data
– Helps identify trends, common values, and
differences in data → Find patterns
Discrete Variables Continuous Variables
■ Can only take whole numbers ■ Can be infinitely subdivided into
(no decimals or fractions). smaller units.
■ Examples: ■ Only interval-ratio level variables (not
nominal or ordinal) can be continuous
– Number of siblings (1, 3, but
not 1.5) ■ Examples:
– Number of students in a – Time in a race (10.7 sec, 10.732
classroom (25, 30, but not sec, etc.).
25.5) – Height of a person (5.8 feet,
– Number of pets owned (1, 2, 5.8123 feet, etc.)
3… but never 1.3) – Time spent studying (2.5 hours,
– Number of text messages sent 2.75 hours)
in a day – Temperature (98.6°F, 72.4°F)
– Number of cars in a parking lot – Distance traveled (10.3 miles,
– Number of goals scored in a 10.345 miles)
soccer match – Weight of a package (2.45 kg,
2.457 kg)
Variable Types & Descriptive Statistics
Variable Descriptive Statistics
Type
Discrete Frequency (count): Number of times a value
appears.
Proportion: Fraction of total cases in each
category.
Percentage: Proportion expressed as a
percentage.
Continuou Central Tendency: Mode, median, mean
s (measures of average).
Dispersion: Range, interquartile range (IQR),
■ Different variable types need different statistical methods.
variance, standard deviation (measures of spread).
■ Best way to Distribution:
describe discrete variables:
Skewness (shape of Frequency,
data
proportion, percentage,
distribution).etc.
Sample Size (N)
■ Total number of cases (e.g., people, objects, or
observations) included in a study
■ Why is Sample Size Important?
• A larger sample gives more reliable results.
• A smaller sample may lead to errors or biased
conclusions.
■ Examples:
• A study on university students → N = 500 students
surveyed.
Frequency (f)
■ Number of cases observed in a
category
■ Provides a quick overview of
data distribution.
■ Example: How many apples
would you usually buy in one
shopping trip? → Most people
buy 4 apples per trip, as it has
the highest frequency (9).
■ Interpretation: Small to
moderate numbers of apples (3-
5) are more common, while few
Proportion ()
𝑓
𝑝=
𝑁
■ f = Frequencies (number of cases in particular category)
■ N = Total number of cases
■ Provides a basis for judging relative size of frequency
count → useful when sample sizes differ
■ 0 ≤ Proportion ≤ 1
– When proportion is 0 → No cases fall into this category.
– When proportion is 1 → All cases belong to this category.
Example: What proportion of customers buy exactly
4 apples?
■ What is the proportion of
customers who buy 4
apples?
Total customers surveyed
= 26
Customers who bought 4
apples = 9
= = 0.35
Percentage (%)
𝑓
%= ∗100
■ f = Frequencies (number 𝑁
of cases in particular category)
■ N = Total number of cases
■ Percentage expresses a proportion as a value out of 100
for easy comparison.
■ 0% ≤ Percentage ≤ 100%
– 0% → No cases fall into this category.
– 100% → All cases belong to this category
Example: What percentage of customers buy exactly
4 apples?
■ What is the percentage of
customers who buy 4 apples?
Total customers surveyed =
26
Customers who bought 4
apples = 9
= = 34.62
Frequency vs. Proportion vs. Percentage
■ Frequency: The count of occurrences in a category. Cannot be
easily compared across groups without context.
■ Proportion or percentage: Expresses frequency relative to the
total sample size (N). quick and easy comparisons between
different groups/ categories.
■ Example: Are you a cat person? Yes/No
■ NY: 2000 (f) said yes / 5000 (N) = 0.4 = 40% are cat lovers.
■ NJ: 2000 (f) said yes / 2000 (N) = 1 = 100% are cat lovers.
■ Even though 2,000 respondents said "Yes" in both states,
the percentage (proportion, too) tells a very different story
about cat preference.
■ Percentage is easier to read than proportion. Why?
– Fewer decimal places make it more intuitive
Rate
𝑓 (𝑎𝑐𝑡𝑢𝑎𝑙)
R ate= ∗ 𝑠𝑜𝑚𝑒𝑚𝑢𝑙𝑡𝑖𝑝𝑙𝑒 𝑜𝑓 10
𝑁 (𝑝𝑜𝑠𝑠𝑖𝑏𝑙𝑒)
■ Actual occurrences of some phenomenon compared with
number of all possible occurrences.
■ To make rate easier to interpret, multiply by some multiple of
10 (e.g., 1000, 100,000, etc.)
■ Example:
– Crude death rate =
= For every 1000 people, there were
14.29 deaths during this particular year.
Example: Murder Rates
Murders Reported In 1996
New York (Total pop: 7,381,000) 989
Los Angeles (Total pop: 3,554,000) 721
Detroit (Total pop: 1,000,000) 427
Calculate the murder rate for each city.
■Murder rate:
Example: Murder Rates
Which city has the highest murder rate among
these three cities? Detroit → 42.7 Murders per
Ratio
𝑓1
Ratio=
𝑓2
■ = the number of cases in the first category
■ 𝑓2 = the number of cases in the second category
■ useful for comparing the relative size of different
categories of a variable
■ Examples:
– Number of males per female
– Number of females per male =
– Number of males per 100 females =
Frequency Distributions
■ A table that summarizes the distribution of a variable.
■ Shows the number of cases (counts) in each category of that
variable.
■ The first step in most statistical analyses.
■ Used to calculate proportions, percentages, and other statistics.
■ Valid percent = f/N (total valid cases, excluding missing data) X
100
■ Why do missing values occur?
– Skipped survey questions (e.g., someone refuses to answer).
– Invalid responses (e.g., entering "200" as an age).
Example: Frequency Distribution Table Using SPSS
RESPONDENTS LIVING STANDARD COMPARED TO PARENTS
Cumulative
Frequency Percent Valid Percent Percent
Valid BETTER 1215 42.9 64.7 64.7
SAME 408 14.4 21.7 86.4
WORSE 256 9.0 13.6 100.0
Total 1879 66.3 100.0
Missing DK 31 1.1
System 922 32.6
Total 953 33.7
Total 2832 100.0
DK = Don’t Know 1215
∗ 100=42.9 %
System Missing = Not Asked Question 2832
NA = No Answer 1215
∗ 100=64.7 %
NAP = Not Applicable 1879
Visualizing Frequency vs. Percentage
Calculations - Review
𝑓 Proportio
𝑁 n
Rate
𝑓 Percentag 𝑓 (𝑎𝑐𝑡𝑢𝑎𝑙)
∗ 100 ∗ 𝑠𝑜𝑚𝑒𝑚𝑢𝑙𝑡𝑖𝑝𝑙𝑒𝑜𝑓 10
𝑁 e
𝑁(𝑝𝑜𝑠𝑠𝑖𝑏𝑙𝑒)
𝑓 1 Ratio
𝑓 2
For Next Class:
■ Finish reading Chapter 2
■ Assignment #2 (due
2/7, 11:59pm)