In short I would go and use observation research to see what unmet needs people had with SUVs and cars because asking them would yield little if any insights. For this type of research the interaction between
people and products are critical.

Question 9: Define a hypothetical situation in which you would favor the
use of a census instead of a sample. Discuss the elements of this situation
that would influence your decision.

While censuses are quite expensive to complete and are very time-consuming,
the one circumstance that would lead me to pursue this and recommend it
would be of a new ethical drug trial and the results obtained from having
every patient who had been prescribed the medicine be tested for all side-
effects. While the Food and Drug Administration (FDA) does provide
sampling methodologies that seek to statistically validate drug testing,
for any new drug I would propose an entire census sampling approach. The
reasons would be to first limit any harmful side-effects of the new drug,
check on the accuracy of the claims made as to what the drug helps cure,
and also create a database of all early adopters to complete trending
analysis on their progress towards wellness. Further, I would also want to
track all vital signs of all patients in the...

Discuss what elements of this situation would influence your decision.
In the case of exhibiting a new product at a trade show, I would rely on
non-probability sampling as this approach would allow for the randomness of
trade show visitors completing the survey on a laptop or PC in the trade
show booth. As the research objective of this activity is to gauge the
level of interest in the product, the level of awareness of the company
itself, and the demographics of the shows' audience, in addition to what
common media materials the shows' audience relied on. I would also rely on
non-probability sampling from the standpoint of analyzing, interpreting and
using the results as merely a snapshot of the shows' activity, and the
collective level of interest in the new products shown. Given the fact
that there is not an ample opportunity to control the sampling frame, this
approach of non-random sampling along with all the caveats of being not
statistically relevant would apply to the communication of all results.