For empirical economic analysis, data usually take the form of quantified, numerical measurements of economic variables, for example, the price of gold ($400 per ounce), the quantity of automobiles sold (8 million cars), or the unemployment rate (5 percent).

Much of this data come through official government sources. The Federal Reserve System, for example, provides abundant data on money, bank deposits, and assorted financial activities. The Department of Labor provides key employment and unemployment data. The Department of Commerce maintains a database of production and income data used to calculate gross domestic product, national income, and associated measures.

Other data come from private sources. The Conference Board, for example, compiles the leading, lagging, and coincident economic indicators used to track business cycles. The Dow Jones Company maintains and reports the popular Dow Jones Industrial Average (The Dow).

Four Types

• Total: A study of the aggregate macroeconomy is full of total data. Measuring a total is relatively straightforward--just count everything. This is the technique used for calculating gross domestic product--total production in the economy.


• Average: A second type is an average, usually the arithmetic mean of several observations. A common average is per capita income--the average income for each person in the economy.


• Percentage: A third type falls under the heading percentage, the relative proportion of a value compared to a total. The unemployment rate--the percent of the labor force unemployed--is a notable and important percentage.


• Index: The fourth and last type is index. An index is a composite number that marks changes relative to some given time period or value. A useful index is the index leading economic indicators, which combines assorted measurements like the money supply, stock prices, and unemployment compensation claims.

Objective and Subjective

• Objective: Objective data are based on cardinal measures that are largely free of subjective values or beliefs. Cardinal measures are based on a benchmark scale. Linear measures of distance provide an example. A mile is 5,280 feet, regardless of who does the measuring. Likewise, if the total labor force of the economy is 100 million people and 5 million people are unemployment, then the unemployment rate is 5 percent. Such objective data is ideally suited for testing scientific hypotheses.


• Subjective: Subjective data are based on personal beliefs, opinions, and judgments. This data likely depends on who does the collecting or possibly who is being studied. A common example of subjective data comes from opinion surveys, such as that used to compile the Consumer Confidence Index. Is the economy in good shape or bad shape? The answer, the data collected, depends on who is asked.

However, the price of hot fudge sundaes is itself subjective. Duncan Thurly might be willing to pay $3 for a hot fudge sundae because he really likes hot fudge sundaes and has plenty of income to make the purchase. Lisa Quirkenstone, however, is only willing to pay $1 for a hot fudge sundae because she is allergic to chocolate and has very little available income.

Because a great deal of economic analysis involves subjective valuations, objective and subjective data are often intertwined. What might appear to be completely objective, very likely has subjective undertones.

Data Collection

• The first is total tabulation. Much as the name indicates, total tabulation identifies every item. This is the method used to derive the Census of Population. Count everyone. Total tabulation is also the primary method for deriving gross domestic product.


• The second way is sampling. At times, counting everything is not possible, the time or resources to accomplish the task are not available. In such cases, statistical sampling is generally used. Data for a small, representative number (a sample) from the larger group is collected. When done correctly, this sample can indicate important characteristics of the larger group.

Tracking Performance

• At the macroeconomic level, data such as gross domestic product, unemployment rate, and inflation rate indicate whether the economy is expanding or contracting, doing better or doing worse.


• At the microeconomic level, data such as prices, production quantities, and profit levels indicate the performance of a particular industry or the activity in a given market.

Hypothesis Verification

Professor Grumpinkston, for example, has a hypothesis that students seated closer to the front of a classroom learn more and earn higher grades. The professor has two alternatives for this hypothesis.

• One, he could treat the hypothesis "as fact." The connection "sounds right" and seems to "make sense." He could pass along this relation "as fact" to his colleagues, who might then pass it along to others. Before long everyone accepts this hypothesis "as fact."


• Two, he could test this hypothesis with real world data. He could compare grades and seating positions for thousands of students in hundreds of classes. He could collect data, calculate percentages and indexes, and conduct study after study, each time comparing his hypothesis with real world observations. In so doing, he might be able to scientifically document this hypothesis as fact, or he might refute it. In either case, he has learned something.