Deck 1: Economic Questions and Data

Question 1

Design a hypothetical ideal randomized controlled experiment to study the effect of hours spent studying on performance on microeconomics exams. Suggest some impediments to implementing this experiment in practice.

Answer 1

A randomized controlled experiment is the one that consists of a control group and a treatment group. A control group is the one on which no treatment is imposed. A treatment group is the one which is subject to a particular kind of treatment. The experiment is randomized in the sense that the treatments and groups are assigned in a random manner. Any kind of preference or pattern is not allowed to maintain unbiasedness of the results.
In order to conduct the experiment to study effect of hours spent studying on the performance in a microeconomics exam, one could create blocks 0-1 hour, 1-2 hours, 2-3 hours, 3-4 hours. …, 19+ hours if the strength of the class is big in size and randomly assign students to the newly created blocks. Then students have to study in their assigned blocks in whatever session they want to and also assigned them a stopwatch to break up their study session according to their settings. According to the distribution of performances of students in their assigned block, one might promise them that they would get a letter grade, to assure them that student with longer blocks have an advantage over their peers.
Some possible hindrances to this experiment can be,
1. The number of students would need to be taught by the same professor to remove the impact of different teaching/learning techniques.
2. It is difficult to restrict a student's studying time while studying; students learn at different paces. Some students may study consistently through the course without using the stop-watch correctly while others may cram for the exam in their assigned limits.
3. Students may not study for the required amount of time, because it is something they are asked to do in their own time.

Question 2

Design a hypothetical ideal randomized controlled experiment to study the effect on highway traffic deaths of wearing seat belts. Suggest some impediments to implementing this experiment in practice.

Answer 2

To design an ideal randomized controlled experiment, for the effects on highway traffic deaths of wearing a seat belt we would need to study the number of deaths occurring during this experiment. Of a randomly selected set of automobile occupants, half would be asked not to wear their seat belts, over a period of time while the rest would be asked to wear their seat belts all the time.
Some practical impediments to this experiment include:
- One cannot tell automobile occupants not to wear their seat belts because it is illegal.
- The people in the experiment do not take exactly the same journeys. Road conditions will differ; rural versus urban and congestion versus clear. Drivers will have different reasons for their journeys, and take them at different times of the day.
- One cannot be sure, that those asked to use their seat belt will do so and vice versa.
- The number and age of the occupants may have an effect on the driver in relation to highway deaths.
- The experiment does not stop seat belt wearers and non-seat belt wearers getting into the same automobile.

Question 3

You are asked to study the casual effect of hours spent on employee training (measured in hours per worker per week) in a manufacturing plant on the productivity of its workers (output per worker per hour). Describe:
a. an ideal randomized controlled experiment to measure this causal effect;
b. an observational cross-sectional data set with which you could study this effect;
c. an observational time series data set for studying this effect; and
d. an observational panel data set for studying this effect.

Answer 3

Study on causal effect of employee training on a manufacturing plant's productivity.
a) Describing an ideal randomized controlled experiment:
To measure the causal effect of employee training under ideal randomized controlled experiment on a manufacturing plant's productivity, we need to select randomly a set of employees, where half are given training while the rest are not. The experimental data is the output per worker per hour for each of these randomly selected employees over a period of time.
b) Describing an observational cross-sectional data set:
In an observational cross-sectional data set, to measure the causal effect of employee training on a manufacturing plant's productivity, we need to take from employee and manufacturing plant records for a specified period of time. The data set is the output per worker per hour from these records over that time where the observational number is the employee number.
c) Describing an observational time series data set:
In a time series data set, the causal effect of employee training on a manufacturing plant's productivity is the output hour where the data contains observations on two variables; rate of trained employees and their output. The observation number relates to the time period, and data is obtained for total plant's output per hour for each period.
d) Describing an observational panel data set:
In an observational panel data set, the causal effect of employee training on a manufacturing plant's productivity is the output per worker per hour for untrained and trained employees observed over two or more time periods.