A Classroom Economic Experiment: How to Estimate Unemployment Rate 33 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 EXPERIENTIAL LEARNING BASED DISCUSSION VS. LECTURE BASED DISCUSSION: HOW TO ESTIMATE THE UNEMPLOYMENT RATE Inhyuck “Steve” Ha 1 and Jessica Hollars Wisniewski 2 Abstract This pedagogical method to estimate the unemployment rate is appropriate for an undergraduate course in macroeconomics. Class instructors can use the experiment to make many macroeconomic principles readily apparent to the unskilled reader. This experiment examines the dynamics of calculating the unemployment rate by means of an assessment instrument. Students can learn that the unemployment rate is calculated using estimates of the size of the labor force, which includes individuals who are both employed and unemployed. In addition, they discover that their estimated unemployment rate agrees with the estimates made by leading economic indices. Thus, by participating in the experiment, students better understand how to calculate the unemployment rate and how to understand published unemployment estimates. JEL Codes: A22, C90, E24 Key Words: Economic Experiment, Unemployment Rate Introduction The study of economics is broad in scope. It is “the social science that deals with the production, distribution and consumption of goods and services and with the theory of management of economies or economic systems (Houghton, 2000).” Since the information is of great breadth, communicating known facts can be difficult. At the same time, while learning fact-based information is important, active learning stimulates the mind to process information more effectively. It enhances a student’s ability to solve problems by putting economic ideas and facts to use. Practical examples and experiments encourage student participation and are widely used in classroom settings to demonstrate abstract ideas. Thus, experiments are a very useful tool in economics education. Traditionally, the study of economics has not been an experimental science (Holt, 1999). Recently, however, the use of classroom experiments has grown in economics courses. As an article in the Journal of Economic Education states, “textbooks for introductory courses often come with supplements of classroom games or focus exclusively on classroom experiments” (Dickie, 2006). A test was conducted to determine whether classroom experiments increase learning in introductory microeconomics. Although the evidence was limited and mixed, results from a student survey indicated “that experiments helped half the students learn better than lectures did, compared with one-eighth who found the lectures more helpful” (Dickie, 2006). Emerson (2004) stated that, “the use of experiments in the principles classroom provides students 1 Associate Professor of Economics, College of Business, Western Carolina University, Forsyth 124, Cullowhee, NC 28723. 2 Research Associate, College of Business, Western Carolina University, Forsyth 104, Cullowhee, NC 28723. 34 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 with an experiential learning opportunity: the chance to participate in a controlled market environment and to observe market forces that are normally only talked about and described as movements on a graph.” Thus, “the use of classroom experiments provides an important connection between theories and key features of the markets and institutions being studied” (Holt, 1999). Educational tools used in economic courses go beyond the traditional textbook material. Learning economics has become easier with the help of multimedia presentations, experiments and exercises. Understanding key topics, such as inflation and the unemployment rate, can pose a challenge to students who haven’t fully mastered introductory course material. In calculating the unemployment rate, a common mistake made by students is defining the denominator. The problem arises when students use population rather than labor force as the denominator. Additionally, understanding the unemployment rate in macroeconomics differs from unemployment in microeconomics. Recognizing the dynamics of both microeconomic and macroeconomic issues when discussing the unemployment rate is more effective when interactive presentations are involved. Students can easily understand the causes of fluctuations in the unemployment rate if they understand how it is calculated. For example, based on the U.S. Bureau of Labor Statistics household survey data, U.S. nonfarm payroll employment was reduced by 17,000 persons in January 2008, while the national unemployment rate declined to 4.9% from 5.0% during the same month. Declination in both employment and the unemployment rate may not be easy to understand, but it occurred again in November 2009 when nonfarm employment declined by 11,000 persons while the national unemployment rate dropped from 10.2% to 10.0%. Without an understanding of how the unemployment rate is estimated, students were puzzled when these observations were introduced. The unemployment rate can often confuse laypeople when substantial changes in the labor force occur. In the classroom experiment discussed in this paper, students are given the opportunity to learn about the unemployment rate by engaging in a hands-on activity. The primary focus of this experiment is to estimate the unemployment rate. The students insert their own household information into a matrix. Please note that the instructor should make sure that the students are able to distinguish between a household and a family. The matrix is designed to organize a student’s records of five households. Based on the matrix totals, students are able to calculate the unemployment rate. From this experiment, students will learn that the unemployment rate is determined based only on those individuals in the labor force and that the unemployment rate is directly related to those employed and unemployed, and is only indirectly related to changes in population. Furthermore, students will learn that an increase in the total unemployed population, in relation to the employed, leads to a rise in the unemployment rate. Likewise, students will understand that when the total employed population increases and the unemployed population remains steady, the unemployment rate declines. Description of the Classroom Experiment This 45-minute experiment in estimating the unemployment rate is based on data taken directly from the students’ own lives. By using personal data, students are able to extrapolate their daily experience to understand how the unemployment rate is calculated and how it fluctuates due to changes in the labor force. 35 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 The foundation of the experiment is students’ generation of the data to be used with the aid of an assessment instrument. The instructor forms groups of four or five and distributes a blank matrix into which students insert their own household information. The matrix consists of six categories. The six categories include households with individuals who are: 1) under 16 and/or institutionalized, 2) not in the labor force, 3) employed, 4) unemployed, 5) the total household population and 6) the unemployment rate (left blank until students calculate it). Students should be informed of the definition of each element before they begin. The instructor may need to introduce or review each of the concepts in the experiment by asking questions, such as “where does a full-time student belong?” and “where does a military person belong?” It is strongly recommended that the instructor explain the definition of “being employed.” Students are often confused between the concepts of “being unemployed” and “not being in the labor force.” The instructor also needs to explain how an unemployed person can be excluded from labor force under certain conditions. According to the U.S. Bureau of Labor Statistics, unemployed persons are “16 years and over who had no employment during the reference week, were available for work, except for temporary illness, and had made specific efforts to find employment sometime during the 4-week period ending with the reference week.” The unemployment rate is the number of unemployed workers divided by the total civilian labor force, which includes both the unemployed and the employed. In practice, however, measuring the number of unemployed workers actually seeking work is extremely difficult. It might be beyond the scope of an introductory macroeconomic course to explain the several different methods used to measure the number of unemployed workers in details. However, it is necessary to emphasize that the statistics are estimates and that each country has a slightly different definition of unemployment. The matrix is designed to organize the students’ records of their households. Within each group of four or five, each student records his/her data in the designated boxes under column headings HH1, HH2, HH3, HH4 and HH5. See Table 1 for the assessment instrument that students use. Table 1: Assessment Instrument Number of People (in thousands) HH 1 HH2 HH3 HH4 HH5 Group Class Note Under 16 and/or Institutionalized 71,485.8 in mental hospitals, or correctional inst. Not in Labor Force 82,316.0 homemakers, voluntary 'withdrawal' for educational, retirement, or other purposes Employed 138,864.0 A Unemployed 15,142.0 B Civilian Labor Force 154,006.0 C = A + B Population 307,807.8 Unemployment Rate 9.8% B / C Source: Bureau of Labor Statistics (September 2009) 36 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 For each column heading, a student inserts the number of members in his/her household who conform to any of the six categories. The second column shows the U.S. statistics as an example of reasonable approximations of what each is supposed to measure, while the last column contains definitions and formulas. In the case that no family member conforms to one or more descriptions, a zero should be placed in the corresponding box. Once everyone in the group has finished recording data, the group reviews all the data in the matrix. Once the students have completed their review of the designated boxes, they calculate the total number of household members in each of the six categories. The sum is written under the column heading “Group.” Based on the totals, students are able to calculate the unemployment rate for their group. Since the unemployment rate is based on those individuals in the labor force, all categories except the employed and unemployed categories are ignored. The unemployment rate is calculated based on the numbers entered in the employed and unemployed rows. After each group has calculated an unemployment rate, the instructor gathers the group totals to display the results to the class. The instructor then computes the overall unemployment rate. In this way, the class can see exactly how the unemployment rate is calculated. Results of the Classroom Experiment This in-class experiment was conducted in our introductory economics courses at Western Carolina University (WCU). The experiment was run twice, in spring 2009 and fall 2009. This paper reports the results from both semesters. Table 2: Results of the Classroom Experiment section 1 section 2 total section 1 section 2 total Under 16 and/or Institutionalized 57 30 87 34 30 151 Not in Labor Force 107 76 183 92 52 327 Employed 203 130 333 132 111 576 Unemployed 29 17 46 18 20 84 Civilian Labor Force 232 147 379 150 131 660 Population 396 253 649 276 213 1,138 Unemployment Rate 12.5% 11.6% 12.1% 12.0% 15.3% 12.7% NC unemployment rate when the data were collected spring 2009 fall 2009 10.8% 11.0% Source: Class Experiments 2009 In each course section, the students were formed into groups of four. After each group finished collecting data and discussing its findings, all group results were reported. Table 2 summarizes the aggregated findings for each class section. When the experiments were conducted, in both the spring and fall semesters, the actual unemployment rate in North Carolina was 10.8% and 11.0%, respectively. Surprisingly, the spring 2009 results of section 2 and fall 2009 results of section 1 were very close to the actual 37 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 unemployment rate in North Carolina at that period in time. However, both the spring 2009 section 1 results and fall 2009 section 2 results overestimated the unemployment rate. It is surprising that the class results revealed a higher unemployment rate than North Carolina considering that, in general, college students have relatively higher family incomes and their parents tend to have higher levels of education, both of which are inversely related to unemployment. This may be caused by a sampling bias, which also can be discussed with students. Effectiveness of the Classroom Experiment The pedagogical difference between an experimental group and a control group is statistically different. For the purpose of this study, the experimental group is composed of all groups of students that engage in the unemployment rate activity. The control group is composed of all groups of students that were taught how to calculate the unemployment rate through class lecture. To alleviate room for error, the same faculty member taught both groups, including experimental and control groups, during the same semester. Table 3: Difference in Means: Experimental vs. Control Groups Experimental Control N = 59 N = 29 Midterm Exam 28.8% 24.1% 0.458 0.648 Final Exam 42.4% 7.1% 3.510 0.001 Source: Authors' estimation t value Pr > |t| Mean Value The difference in means between the two groups is shown in Table 3. The sampling distribution was computed for the fall 2009 semester only. The number of observations for the experimental group and control group are 59 and 29, respectively. Based on the sample data, the independent mean difference test was applied. Results show that students who engage in the classroom experiment retain and understand the purpose of the unemployment rate better than students who merely listen during a class lecture. To illustrate, 28.8% of the experimental group understood how to calculate the unemployment rate on the midterm exam while only 24.1% of those in the control group did. This difference in means was not statistically significant. Surprisingly, the difference based on the final exam is statistically significant one and a half months later. Forty two percent of the experimental group was able to calculate the unemployment rate correctly on the final exam, in contrast to only 7.1% of those in the control group. The results illustrate that the experiment is more beneficial to long-term memory than is the lecture style teaching method. Suggested Questions for Discussion The following questions can provide the basis for follow-up discussion. It is recommended that each student answer the questions for himself/herself, prior to open discussion. 1. How is the unemployment rate measured? How would you interpret the unemployment rate? 38 | JOURNAL FOR ECONOMIC EDUCATORS, 11(1), SUMMER 2011 2. What was the unemployment rate in your group? What was the unemployment rate for the entire class? Are they close to each other? Are they close to the state and/or national unemployment rate? If not, what does it imply? 3. What individuals make up the labor force? How does the size of the labor force affect the unemployment rate? In other words, what could happen to the unemployment rate if the labor force participation rate increases or decreases? 4. How could part-timers or discouraged workers affect the unemployment rate? Conclusion This experiment demonstrates how the unemployment rate is estimated. Students are introduced to one of the most fundamental concepts in economics – unemployment rate – by participating in an economic experiment. In doing so, they identify the labor force rather than total population as the basis of unemployment when they figure out who is employed or unemployed in the sample households. The follow-up discussion embraces policy issues such as how employment statistics can mislead people when there are substantial changes in labor force. Independent mean difference testing showed the experiential approach to teaching the unemployment rate to be significantly more effective than teaching the concept lecture style. The results suggest that the unemployment rate is more effectively taught through active learning methods. References Dickie, M. 2006. Do Classroom Experiments Increase Learning in Introductory Microeconomics? Journal of Economic Education 37(3): 267-288. Emerson, T. 2004. Comparing Student Achievement across Experimental and Lecture-Oriented Sections of a Principles of Microeconomics Course. Southern Economic Journal, 70(3): 672-693. Holt, C. 1999. Teaching Economics with Classroom Experiments: A Symposium. Southern Economic Journal. 65(3): 603-610. Houghton, Mifflin. 2000. Definition of economics. Retrieved January 28, 2008, from http://education.yahoo.com/reference/dictionary/entry/economics U.S. Department of Labor, Bureau of Labor Statistics. 2008. www.bls.gov http://education.yahoo.com/reference/dictionary/entry/economics http://www.bls.gov/