9. A huge advantage of conducting an experiment is that the experimenter has the opportunity to control for those variables that may have an effect on the outcome. For example, in problem 8 above, blocking by gender allows to exercise control over the effect of smoking in men or women. However, in practicality, it is impossible to control for everything that may have an impact on the outcome. At that point, statisticians use randomization. Explain how and why randomization makes it possible to arrive at valid conclusions. 10. There are websites that record student's evaluations, rankings, and comments on their professors. Many students use these data to seek information on courses and professors they are considering to include in their future class schedules. a. Is the data obtained from these websites biased? Explain. b. Is the data obtained from these websites confounded? Explain. c. Can the data obtained from these websites constitute a random sample of students for a particular professor/course? Explain. 11. In section 1.4 we learned that data can generally be classified as quantitative or qualitative. An additional important aspect of dealing with data is its scale of measurement. In general, researchers consider four different scales of measurement: nominal, ordinal, interval, and ratio. A nominal scale is based on a classification system driven by categories such as blood type, gender, political affiliation, age group, etc. An ordinal scale is based on rank order with respect to some variable or criteria. The rankings are not to be confused with the magnitude of numerical differences. For instance, for the 2015 baseball 1-44 season, the San Francisco Giants finished second in the National League West. The ranking "second" says nothing about how far behind they finished (in terms of win/losses) in relation to the team that finished first (the Los Angeles Dodgers). In an interval scale, equal differences between scale values do have equal quantitative meaning. For example, numerical data, whether continuous or discrete, is measured on an interval scale. Also, when using an interval scale, the "zero" or starting point is arbitrary. For instance, the interval between the starts of years 2013 and 2014, 365 days, is the same as that between the starts of the years 2015 and 2016. For this example, note that both, 2013 and 2015, would qualify as the "zero" or starting point. A ratio scale is similar to an interval scale with one major difference: there is an absolute "zero" or starting point. The system of measurement using a common ruler is an example of a ratio scale. The zero inch mark, for example, does represent an absence of length. With this scale, we can make statements such as: an object measuring 6 inches is twice as long as an object measuring 3 inches. For each situation below, determine which measurement scale is most appropriate and explain why. a. Temperature in degrees Celsius. b. The order of finish for runners in the 2016 Boston Marathon. c. The make of cars sold in the United States. d. The annual rate of inflation in the United States starting with the year 1970. e. A person's balance in a bank checking account. f. The number of students attending your local community college whose last name starts with "S". g. Height, in inches, of all students at your local community college taking Elementary Statistics this academic term. h. Letter grade (A, B, C, D, F), obtained in their most recent math course, by all the students at your local community college taking Elementary Statistics this academic term. 1-45