1.2: Populations and Samples
- Page ID
- 236810
\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)
\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)
\( \newcommand{\dsum}{\displaystyle\sum\limits} \)
\( \newcommand{\dint}{\displaystyle\int\limits} \)
\( \newcommand{\dlim}{\displaystyle\lim\limits} \)
\( \newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\)
( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\)
\( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)
\( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\)
\( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)
\( \newcommand{\Span}{\mathrm{span}}\)
\( \newcommand{\id}{\mathrm{id}}\)
\( \newcommand{\Span}{\mathrm{span}}\)
\( \newcommand{\kernel}{\mathrm{null}\,}\)
\( \newcommand{\range}{\mathrm{range}\,}\)
\( \newcommand{\RealPart}{\mathrm{Re}}\)
\( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)
\( \newcommand{\Argument}{\mathrm{Arg}}\)
\( \newcommand{\norm}[1]{\| #1 \|}\)
\( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)
\( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\AA}{\unicode[.8,0]{x212B}}\)
\( \newcommand{\vectorA}[1]{\vec{#1}} % arrow\)
\( \newcommand{\vectorAt}[1]{\vec{\text{#1}}} % arrow\)
\( \newcommand{\vectorB}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)
\( \newcommand{\vectorC}[1]{\textbf{#1}} \)
\( \newcommand{\vectorD}[1]{\overrightarrow{#1}} \)
\( \newcommand{\vectorDt}[1]{\overrightarrow{\text{#1}}} \)
\( \newcommand{\vectE}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash{\mathbf {#1}}}} \)
\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)
\(\newcommand{\longvect}{\overrightarrow}\)
\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)
\(\newcommand{\ket}[1]{\left| #1 \right>}\)
\(\newcommand{\bra}[1]{\left< #1 \right|}\)
\(\newcommand{\braket}[2]{\left< #1 \vphantom{#2} \right| \left. #2 \vphantom{#1} \right>}\)
\(\newcommand{\braopket}[3]{\left< #1 \vphantom{#2}\vphantom{#3} \right| #2 \vphantom{#1}\vphantom{#3} \left| #3 \vphantom{#1}\vphantom{#2} \right>}\)
\(\newcommand{\qmvec}[1]{\mathbf{\vec{#1}}}\)
\(\newcommand{\op}[1]{\hat{\mathbf{#1}}}\)
\(\newcommand{\expect}[1]{\langle #1 \rangle}\)
\(\newcommand{\dfn}[1]{\emph{\textbf{#1}}}\)
- Identify key components of a statistical study.
- Differentiate populations and samples.
- Differentiate parameters and statistics.
- Explain sampling variability and error.
Introduction
To understand how to calculate descriptive statistics to make inferences about the world around us, there are a few definitions that we need to understand. Many of the terms used in statistics also have non-statistical meanings. That is, in statistics, some terms have slightly different definitions than when we use those terms in everyday language or in other branches of mathematics. It is important that you notice the subtle differences and use the statistical definitions when appropriate.
Population vs. Sample
In any statistical study, the first task is to determine who is being studied and what is being measured. These two questions form the foundation of every well‑designed study. The who refers to the group of individuals (or units) under consideration—these may be people, animals, manufactured items, or any units of interest. The what refers to the variable, the specific characteristic or measurement recorded for each individual.
The population is the entire group of individuals about whom information is desired. A clear population statement must always identify both who is being measured and what is being measured. Without both components, the purpose of the study is incomplete. For example, stating that a researcher will “collect data from the senators in the U.S. Congress” does not tell the reader what attribute is being studied. Are we interested in age, years of service, educational background, voting behavior, or something else? A population description must specify the individuals and the variable to give a complete picture of the study’s intent.
In some cases, the population is small enough that measuring every individual is practical. When data are collected from the entire population, the study is called a census. For instance, determining the current ages of all 100 sitting U.S. senators is straightforward because the population is limited and accessible. However, if the goal were to determine the age at which every senator first took office throughout the entire history of Congress, the population is still finite but requires more extensive data collection.
Many populations, however, are too large or too difficult to measure in their entirety. In such situations, researchers select a smaller group from the population. This subset is called a sample. A sample serves as a manageable representation of the population—ideally similar in its characteristics but far easier to measure. By studying the sample, researchers draw conclusions about the larger population.
For a visual illustration, Figure 1 shows a fish tank containing several fish -- the population. The characteristic being studied - the variable - is the number of spots on each fish. Here, the 20 individual fish are the who and the number of spots on each of the fish is the what.
Figure 1: Population of Fish in Fish Tank
To collect a sample of the fish from the tank, perhaps we use a net to select 5 fish at random. The fish in Sample A are shaded in Figure 2. Data are being obtained from the 5 fish – not all 20 fish – in hopes that the subgroup of 5 fish resembles the larger group.
Figure 2: Sample A of 5 Fish from the Fish Tank
Realize that if we repeat this sampling process, we may get the same 5 fish again, some of the same 5 fish again and some different fish, or all different fish. Figure 3 shows Sample B, a different sample of 5 fish taken from the original population of 20 fish:
Figure 3: Sample B of 5 Fish from the Fish Tank
Two of the same fish were caught in both samples, but the samples are quite different. That is, when you take a sample you just get a “snapshot” of the population and this snapshot may be different each time. The idea is to try to collect a sample that gives an accurate overall picture, but we never know for sure if the snapshot does that. Unlike previous mathematics classes where there was always one right answer, in statistics there can be many answers, and you don’t know which are right. We use probability later in this course to estimate our likelihood of being right.
In statistics, it is convention to always use the letters \(N\) and \(n\) for the size of the population and the size of the sample, respectively. For the previous fish tank illustration, there were 20 fish in the population under study so \(N=20\). To study the fish in the fish tank, we took samples of size 5 fish so \(n=5\).
How you collect your sample can determine how accurate the results of your study are as well. We hope that the sample resembles the population as closely as possible. There are many ways to collect samples. Some of them create better samples than others. More on that will be discussed later.
Why Sample?
In the last illustration, the population of fish was quite small (only 20 fish), and it was manageable to conduct a census of the fish in the fish tank. Most of the time, however, it is extremely difficult or very costly to collect all the information about an entire population. For a study of tortoises in the Galapagos Islands, it would be very difficult and perhaps even destructive to search every square meter of the habitat to be sure that you reached every tortoise. A battery manufacturer would not want to test every battery it produces to determine the average lifetime of its batteries because it would have none left to sell. It is very expensive to get accurate and complete information about all the residents of the United States to help effectively address the needs of a changing population. For that reason, the U.S. Government conducts a census only once every ten years.
To find the most common manufacturer, a campus security officer records the manufacturer of every 5th car that enters the college parking lot until data about 40 cars has been collected. Identify the individual, the variable, the population, and the sample, \(n\), and \(N\), where possible. Also tell if the data collected in the scenario is categorical or quantitative.
Solution
- Individual: a car that enters the college parking lot
- Variable: the name of the car manufacturer
- Population: all cars entering the college parking lot
- Sample: every 5th car entering the college parking lot
- \(n=40\) because data will be collected from 40 cars
- \(N\) is unknown. We were not told how many cars will enter the college parking lot.
- Categorical data because the name of a car manufacturer in being recorded
A biologist planted 500 Variety X tomato plants and gave them a new plant food. Now, the biologist wants to estimate the average height of the plants. She randomly selects 40 of the tomato plants and measures the heights.
Identify the individual, the variable, the population, and the sample, \(n\), and \(N\), where possible. Also tell if the data collected in the scenario is categorical or quantitative.
Solution
- Individual: a Variety X tomato plant given the new plant food
- Variable: the height
- Population: all the Variety X tomato plants that were planted and given the new food
- Sample: only the plants that the biologist measured heights for
- \(n=40\) because data was collected from 40 plants in the sample.
- \(N=500\) because the entire group under study contains 500 plants
- Quantitative continuous data because the data being collected consists of numbers that come from measuring height in inches or centimeters.
A school administrator wants to estimate the average number of hours students at Lincoln High School spend on homework each week. She randomly selects 60 students and asks each one how many hours they spent on homework during the past week.
Identify the individual, the variable, the population, the sample, \(n\) and \(N\), where possible. Also tell if the data collected in the scenario is categorical or quantitative.
- Answer
-
- Individual: one student who was asked about weekly homework hours
- Variable: the number of homework hours each student reports
- Population: the number of homework hours completed by each of the students at Lincoln High School
- Sample: the number of homework hours reported by each of the 60 students selected and surveyed
- \(n = 60\)
- \(N\) is unknown. It would be the number of students at Lincoln High School.
- Quantitiative data
Parameter vs. Statistic
Once data have been collected—either through a census or from a sample—the next step is typically to summarize the information by performing calculations. Depending on the study, a researcher may wish to determine the proportion of residents who support a mayoral candidate, the average growth of a plant treated with a new fertilizer, or the variability in a set of test scores. When these calculations are based on every unit in the population, the resulting numerical value is called a parameter. Parameters are usually represented with Greek letters like ρ, µ, and σ.
As noted earlier, collecting data from an entire population is often impractical or impossible. Although the parameter is the value we ultimately wish to know, it cannot be computed directly when the full population is not observed. Instead, researchers rely on a numerical value computed from a sample, known as a statistic, to estimate the parameter. Because samples differ from one another, the statistic will vary from sample to sample. It provides an estimate of the parameter, but its accuracy cannot be guaranteed. Statistics are typically represented with Latin letters and symbols (like \(\hat{p}\), \(\bar{x}\), or \(s\)).
Use Figure 1 to find the average number of spots per fish in the fish tank. Is this average a parameter or a statistic?
Solution
Conduct a census. There are \(N=20\) data values in this population so find the total number of spots and divide by 20:
\(\begin{align*} &= \frac{\begin{aligned} (3+1+5+2+3+2+0+3+2+6 \\ {}+ 4+2+4+1+7+3+0+5+0+2) \end{aligned}}{20} \\[6pt] &= \frac{55}{20} \\[6pt] &= 2.75 \end{align*} \nonumber \)
This value is a parameter because all values from the population were used in the calculation of the average. We can write the average as \(\mu=2.75\) spots.
Use Figure 2 and Figure 3 to find the average number of spots per fish in Sample A and the average number of spots per fish in Sample B. Are these averages parameters or statistics?
Solution
There are 5 data values in Sample A so find the total number of spots and divide by \(n=5\):
\(\dfrac{1+3+4+4+0}{5}=\dfrac{12}{5}=2.4\)
We can write the average for Sample A as \(\bar{x}=2.4\) spots.
There are 5 data values in Sample B so find the total number of spots and divide by \(n=5\):
\(\dfrac{1+3+2+6+3}{5}=\dfrac{15}{5}=3\)
We can write the average for Sample B as \(\bar{x}=3\) spots.
Both of the calculated values in Example 1.2.4 are statistics because only some of the fish from the population were used to estimate the average number of spots for the entire population. Note that the two samples produced different statistics, but both are reasonable estimates of the population parameter – one is a bit lower and one is a bit higher and neither are exactly correct.
A health researcher wants to estimate the proportion of adults in a city who walk at least 30 minutes per day. She collects data from a random sample of 400 adults and finds that 62% meet this guideline.
Classify each value:
- The percentage of all adults in the city who walk at least 30 minutes per day.
- The percentage of the 400 sampled adults who walk at least 30 minutes per day.
- The average age of all adults in the city.
- The average age of the 400 sampled adults.
- Answer
-
- parameter
- statistic
- parameter
- statistic
Sampling Error
Researchers must recognize that statistics calculated from a sample always carry some degree of uncertainty. This uncertainty is unavoidable unless the entire population is measured through a census. Because samples vary by chance, results will differ from one sample to another.
A statistician can report the estimated average weight of a blue crab in two ways: as a point estimate (for example, 0.35 pounds) or as an interval estimate. A typical interval estimate statement might be: “I am fairly confident that the average weight of a blue crab lies between 0.3 pounds and 0.4 pounds.” This range reflects the natural variability that arises from sampling—not an error in data collection or analysis.
The difference between the true population parameter and the sample statistic is known as sampling error. These concepts of estimates and errors will be explored further in a later chapter.
Section Summary
In statistics, the entire group of data being studied is called the population. The entities (people, animals, or things) in the population are called individuals. The characteristics of interest to us are called variables.
Because of the difficulties of obtaining information about all individuals in a population, it is common to use a small, representative subset of the population called a sample. A calculated value from population data (percentage of blue crabs in the Chesapeake Bay that are female, average weight of all blue crabs, etc.) is called a parameter. We can estimate a parameter by collecting a sample and calculating a statistic.
Whenever data from samples are used instead of from a census, we must accept that our results will vary from sample to sample due to sampling error. These results are merely estimates of the parameter value.
Section 1.2 Review Questions
- In each of the following situations, identify the individual, variable, population, and sample. Also tell if the variable is categorical or quantitative.
- Sweet Tooth Candy Company makes chocolate candy bars. A quality control worker weighs every 100th candy bar the company makes to ensure that it is very close to the published weight.
- Lucio wants to know whether the hot food he serves in his restaurant is within a safe range of temperatures. He randomly selects 10 entrees and measures their temperatures just before he serves them to his customers.
- Suppose you want to know how Maryland workers age 18 or older travel to work. To estimate the percentage of people who use the different modes of travel, you take a sample containing 500 Maryland workers age 18 or older.
- You want to find the average cholesterol level of patients two days after they had a heart attack. To estimate the average, a doctor collects data from 8 heart patients in a large hospital who had a heart attack two days ago.
- Decide whether each underlined value is a parameter or statistic.
- A study of 600 adults using public rest rooms in Atlanta, Chicago, New York City, and San Francisco found that 23% did not wash their hands before exiting.
- 32% of U.S. kindergarten teachers say that knowing the alphabet is an essential skill.
- At a college, 90% of the members of the Board of Trustees approved the contract of the new president.
- The 2182 students who accepted admission offers to Northwestern University in 2009 have an average SAT score of 1442.
- A researcher wants to know the average height of women aged 20 years or older. The researcher calculates an average height of 63.9 inches from measuring 45 women.
- Ty Cobb is one of major league baseball’s greatest hitters of all time and had a career batting average of 0.366.

