In each of the following surveys, what besides the mechanism under study could have contributed to the result?

a. An estimation of per-capita wealth for a city is made from a random sample of people listed in the city's telephone directory.

b. Political preference is determined by an interviewer taking a random sample of Monday morning bank customers.

c. The average length of fish in a lake is estimated by:

i. The average length of fish caught, reported by anglers

ii. The average length of dead fish found floating in the water

d. The average number of words in the working vocabulary of first-grade children in a given county is estimated by a vocabulary test given to a random sample of first grade children in the largest school in the country.

e. The proportion of people who can distinguish between two similar tones is estimated on the basis of a test given to a random sample of university students in a music appreciation class.