Question: Lab Carbon Blank. The following data were provided by L. A. Currie of the National Institute of Standards and Technology (NIST). The data are preliminary and exploratory but real. The unit of measure is "instrument response" and is approximately equal to one microgram of carbon. (That is, 5.18 corresponds to 5.18 instrument units of carbon and about 5.18 micrograms of carbon.) The responses come from consecutive tests on "blank" material generated in the lab.
(a) Plot measured carbon content versus order of measurement.
(b) The data are ordered in time, but (as it turns out) time intervals between measurements were not equal (an appropriate plan for data collection was not necessarily in place). What feature of the plot in (a) might still have meaning?
(c) If one treats the measurement of lab-generated blank material as repeat measurements of a single blank, what does a trend on a plot like that in (a) suggest regarding variation of the measurement process? (Assume the plot is made from data equally spaced in time and collected by a single individual.)
(d) Make a frequency histogram of these data with categories 1.00-1.99, 2.00-2.99, etc.
(e) What could be missed if only a histogram was made (and one didn't make a plot like that in (a)) for data like these?